Oxidized Low Density Lipoprotein–Induced Inhibition of Processing of Photoreceptor Outer Segments by RPE

Abstract

purpose. To examine the effects of oxidized low-density lipoproteins (oxLDL) on phagocytosis and processing of photoreceptor outer segments (OS) by retinal pigment epithelial (RPE) cells.

methods. Confluent cultures of RPE-J cells were pretreated with oxLDL or LDL, and the effects of such treatment on the processing of added OS was determined. Processing was determined either by the degradation of ¹²⁵I-labeled OS to trichloroacetic acid-soluble label or by the cleavage of rhodopsin observed on Western blot analysis of cell lysates separated by sucrose density gradient fractionation. Binding to and uptake of OS by RPE-J cells was assessed by determining the fluorescence of FITC-labeled OS before and after quenching with trypan blue.

results. OxLDL induced a significant decrease in the degradation of ¹²⁵I-OS in RPE-J cells but no reductions in either binding or uptake, when a 24-hour recovery period was inserted between treatment with oxLDL and challenge with OS. Rhodopsin cleavage increased in a time-dependent manner after phagocytosis of OS by RPE-J cells. The small guanosine triphosphatase (GTPase), Rab5, was first found in phagosome fractions containing rhodopsin and its cleavage products, and at later times of challenge, in more dense fractions representing phagolysosomes. These were assessed by the colocalization of rhodopsin cleavage products in density fractions with cathepsin D, a marker of lysosomes. OxLDL induced a reduction in rhodopsin cleavage product formation and in phagosome-lysosome fusion (maturation). It also reduced the time-dependent shift of rhodopsin to higher density fractions containing more cathepsin D without any detectable reduction in the expression of cathepsin D or in acid protease activity.

conclusions. OxLDL induces a reduction in the processing of OS by RPE by perturbing the fusion of lysosomes with phagosomes.

The retinal pigment epithelial (RPE) cell layer acts as a support cell for photoreceptors performing such functions as nutrient and waste transport, as well as phagocytosis and processing of shed photoreceptor outer segments (OS). ^{1 2 3 4 5} However, this processing has been suggested to become perturbed by the prooxidant environment of the retina ^{6 7} and to be responsible for the intralysosomal formation and accumulation of lipofuscin, a characteristic of RPE cells in vitro ^{8 9} and in vivo. ^{4 10} Although oxidative events in the RPE have been linked to such disease states as age-related macular degeneration (AMD), ⁷ the evidence for a causative relationship remains circumstantial. We therefore sought to develop a model system to directly assess the effects of oxidized lipids on a key cell biological function of RPE cells—the processing of isolated OS by the rat RPE-J cell line—and to better understand the underlying mechanisms. We used Cu²⁺-oxidized plasma low-density lipoproteins (oxLDL) ¹¹ as the major source of oxidized lipids. Our goal was to assess whether ox-lipids affect processing of OS, and if so, whether the ox-lipid–induced effect is the result of inhibition in their binding, uptake, and/or degradation by the cells. In preliminary studies, ¹² we found that inhibition was at the level of degradation, and we therefore sought to determine whether this was due to a reduction in the total cellular acid protease activity in RPE or to an inhibition of the fusion of phagosomes containing OS and lysosomes. ^{4 5} We report that oxLDL induced a deficiency in the degradation of OS by RPE-J cells, that the pertubation was primarily at the level of OS degradation after phagocytosis, and that it was most likely due to an inhibition of the phagosome–lysosome fusion event. ^{4 5}  

Materials and Methods

RPE Cell Culture

The SV40-immortalized rat RPE-J cell line was maintained as described elsewhere. ^{13 14} In all assays RPE-J cells were plated on multiwell plates at a density of 3 × 10⁵ cells/cm² in DMEM containing 4% FBS and 2.5 nM retinoic acid (medium A) for 6 to 7 days at the permissive temperature of 32°C, followed by a switch to 40°C for 2 days in the presence of retinoic acid. ¹³  

Isolation of OS

OS were freshly isolated from bovine eyes, as described previously. ³ Samples of the isolated OS were radio iodinated by the iodine monochloride technique. ¹⁵ OS were also labeled with fluorescein isothiocyanate (FITC) as reported previously. ^{16 17}  

Oxidation of LDL or OS

LDL or OS were oxidized by dialyzing for 24 hours against 5 μM copper sulfate, as reported elsewhere. ¹¹ OxLDL or oxidized OS were then extensively dialyzed against PBS to remove copper ions. Oxidation of LDL has been shown to result in an increase in lipid hydroperoxides, a measure of oxidation of unsaturated fatty acids such as linoleate present in phospholipids, cholesteryl esters, and triglycerides ^{11 18} ; in thiobarbituric acid reactive substances (TBARS), a general measure of the formation of malondialdehyde (MDA) from the decomposition of such hydroperoxides ¹¹ ; in oxysterols, ¹⁹ indicative of the oxidation of cholesterol; and in free-radical–induced fragmentation of apolipoprotein (apo)B-100, the protein moiety of LDL. ¹¹  

Assessment of the degree of oxidation was determined using the TBARS assay ¹¹ and by the formation of total lipid hydroperoxide (LPO), 13-HPODE. ^{11 18 19} When expressed as nanomoles MDA per milligram protein, TBARS results were: oxLDL, 72.4; LDL, 0.9; oxOS, 257.9; OS, 4.5. When expressed as nanomoles 13-HPODE per milligram protein, LPO concentrations were: oxLDL, 38; LDL, 0; oxOS, 169; OS, 0. 

Degradation of OS by RPE-J Cells

The degradation of ¹²⁵I-labeled proteins associated with isolated OS was measured as a general indicator of the efficiency of phagocytic processing by RPE-J cells. RPE-J cells were incubated for 21 hours with increasing amounts of LDL, oxLDL, OS, or oxOS in medium A. Cells were then washed twice with PBS containing 0.1 mM Ca²⁺ and 1 mM Mg²⁺ (PBS-CM), and 20 μg/ml of ¹²⁵I-labeled OS in medium A was added for 5 hours. Protein degradation was determined by assessing amounts of trichloroacetic acid (TCA)-soluble ¹²⁵I-label in the culture media. ^{11 20 21} Data were also obtained showing degradation in untreated cells, extracellular degradation, and degradation in cell-free control. Cytotoxicity was monitored by observing the release of [¹⁴C]adenine which had been incorporated into the cell. ¹⁹ In select experiments, we measured the in vitro degradation at an acid pH 4.5 of ¹²⁵I-OS to TCA-soluble label by lysates of RPE cells that had been pretreated with oxLDL, LDL, or untreated, and the data were expressed as micrograms degraded ¹²⁵I-ligand per milligram cell protein. 

Binding and Uptake of OS by RPE-J Cells

RPE-J cells that had been pretreated with- or without oxLDL were incubated with FITC-OS, as previously reported. ^{16 17} Binding and internalization at 40°C of FITC-OS to RPE-J cells was calculated from the difference in fluorescence measured fluorometrically, before and after quenching with trypan blue, as described elsewhere. ^{16 17}  

Sucrose Density Fractionation of RPE-J Organelles

RPE-J cells grown in 100-mm Petri dishes were incubated with 0.5 × 10⁸ OS/dish for the indicated periods, washed three times in PBS-CM, gently scraped off the dishes, and pelleted at 200g. Cells were resuspended in homogenization buffer containing 250 mM sucrose, 0.5 mM EGTA, 20 mM HEPES (pH 7.0; medium B), and a cocktail of protease inhibitors (Complete; Roche, Indianapolis, IN). Plasma membranes were disrupted by passing cells five times through a 30-gauge needle. Cell homogenates were loaded onto a 25% to 56% continuous sucrose gradient and centrifuged in a rotor (SW41; Beckman Instruments, Fullerton, CA) at 200,000g for 16 hours. One-milliliter fractions were collected from the top of the tube, diluted 1:1 with 0.5 mM EGTA and 20 mM HEPES (pH 7.0), and centrifuged at 3000g for 10 minutes. Pellets were dissolved in nonreducing Laemmli sample buffer, resolved on 4% to 20% gradient SDS-PAGE, and immunoblotted with antibodies against cathepsin D, Rab 5 (both from Santa Cruz Biotechnology, Santa Cruz, CA), and a monoclonal antibody to rhodopsin ²² (B6–30N, kindly provided by Paul Hargrave, University of Florida, Gainsville, FL). 

Results

To determine whether oxLDL inhibits the processing of OS by RPE-J cells, we preincubated the cells with oxLDL and assessed the subsequent degradation of the ¹²⁵I-labeled OS into TCA-soluble fragments. Treatment of RPE-J cells with oxLDL resulted in a concentration-dependent inhibition of the degradation of ¹²⁵I-OS relative to treatment with unoxidized LDL (Fig. 1a) . Differences were statistically significant, even at the lowest concentration tested (50 μg/ml; P < 0.02). Cells preincubated with no lipoproteins degraded ¹²⁵I-OS to the same extent as LDL-treated cells. To determine whether oxOS also induced a reduction in proteolytic degradation of OS, we incubated RPE-J cells with up to 500 μg/ml of oxOS or OS for 21 hours before challenge with ¹²⁵I-OS. Such pretreatment with oxOS also resulted in a dose-dependent inhibition of degradation of ¹²⁵I-OS (Fig. 1b) . To assess whether oxLDL up to 200 μg/ml or oxOS up to 500 μg/ml was cytotoxic, we incorporated[ ¹⁴C]adenine into RPE-J cells and then determined the release of label into the medium as a measure of cell membrane disruption due to the oxidized lipid. ¹⁹ Neither oxLDL nor oxOS, demonstrated significant cytoxicity at concentrations up to those used in these toxicity studies (Fig. 2) . 

To assess whether any differences could be found in the binding and/or uptake of OS by RPE-J cells pretreated with oxLDL or LDL, we used a fluorometric technique ^{14 16 17} that exploits trypan blue quenching of FITC fluorescence associated with extracellular OS, but not fluorescence of internalized OS. Binding of FITC-OS was determined fluorometrically from the difference in emission before and after quenching with trypan blue. ^{16 17} We could not detect any difference in binding of OS to cells when treated with oxLDL in comparison with pretreatment with LDL (Fig. 3a) . However, when the amount of internalized OS was determined, we found a modest (20%) reduction of internalized FITC-OS in cells pretreated with oxLDL relative to those pretreated with LDL (Fig. 3b) . We then sought to find conditions in which the subsequent perturbation of processing of internalized OS by oxLDL treatment was limited only to the modulation of intracellular degradation—for example, a condition that minimized changes in the internalization step. We reasoned that if we could still detect a reduction in degradation of OS in RPE-J cells, even when binding and uptake were unaffected, it would have to be at the level of intracellular processing. 

We had found earlier that a similar oxLDL-induced reduction in phagocytosis in mouse peritoneal macrophages could be minimized by introducing a recovery period between the pretreatment period and the phagocytosis of aggregated ligands, without affecting the reduction in intracellular degradation induced by oxLDL (Hoppe G, Hoff H, unpublished observations, 1999). When such a 24-hour recovery period, in which cells were incubated with medium alone, was introduced between the 21-hour pretreatment with oxLDL and the challenge with OS, neither binding (Fig. 3c) nor uptake (Fig. 3d) was affected by oxLDL pretreatment. This result permitted us to evaluate the effects of oxLDL on the intracellular degradation of OS by RPE-J cells. 

We next asked whether any reduction in the degradation of OS was due to a defect in the transition from phagosome to a phagolysosome (maturation). ^{4 5 23 24} To this end, we developed a sucrose density gradient technique that permitted us to separate OS-containing phagosomes from RPE-J cells based on phagosome density. Furthermore, to evaluate the degree of intracellular breakdown of OS, cleavage of the major OS protein, rhodopsin, ²² was assessed by Western blot analysis of these fractions using an anti-rhodopsin monoclonal antibody. ²² The degree of degradation of OS that had been subjected to the fractionation procedure immediately after addition to RPE-J cells was negligible (0 hours in Fig. 4 ). Although after a 1-hour incubation of OS with RPE-J cells we observed the formation of only minor cleavage products, after a 4-hour challenge with OS, we observed two major degradation products, one with the apparent mass of slightly less than that of the rhodopsin monomer (41 kDa), ²² and one at 23 kDa (Fig. 4) . The lower molecular weight cleavage product predominated in the higher density fractions (Fig. 4 ; 8 through 12), especially at the 8-hour time point. We conclude that the time-dependent formation and increase in buoyant density of these cleavage products of rhodopsin is consistent with maturation of phagosomes. 

If fusion of lysosomes with phagosomes occurs, one would anticipate detecting cathepsin D, the major lysosomal protease in RPE, ²⁵ normally found in higher density fractions, in densities intermediate between the less dense phagosomes and the more dense lysosomes, after phagocytosis of OS by RPE-J cells. When we performed immunoblots of cathepsin D, we found that before phagocytosis of OS, cathepsin D was detected primarily in fractions 12 and 13 (Figs. 5a 5b) . However, after a 2-hour challenge with OS, the peak of cathepsin D shifted slightly to fractions 11 and 12, and after 6 hours was also discerned in density fraction 10. These fractions contained cleavage products of rhodopsin (Fig. 4) . 

Rab5, a low-molecular-weight GTPase has been shown to be a marker of early endosomes and immature phagosomes in macrophages. ^{23 24} We asked whether we could demonstrate the association of Rab5 with phagosomes in RPE-J cells. When the different density fractions were immunoblotted for both Rab5 and cathepsin D, before challenge with OS, we found Rab5 distributed in low amounts in density fractions 8 through 13. However, after a 2-hour challenge with OS, there was an increase in amount and a shift in the presence of Rab5 to include density fractions that contained phagosomes (fractions 6 through 8; Fig. 5 ). At later times of challenge (6 hours), the most intense Rab5 bands disappeared from the less dense fractions, but increased in intensity in the more dense bands 10 through 12. This result differs from results obtained on RPE phagosomes containing magnetic latex beads in which we found a time-dependent decrease in Rab5 expression as maturation of phagosomes occurred (data not shown). This latter result is more consistent with analogous studies reported on macrophages in culture. ^{26 27}  

Because pretreatment of RPE-J cells with oxLDL or oxOS led to a decrease in the subsequent degradation of internalized ¹²⁵I-OS to TCA-soluble fragments, we asked whether such pretreatment of RPE-J cells with oxLDL would lead to a discernible retardation in the maturation of phagosomes, as evidenced by a reduced formation of rhodopsin cleavage products and a decreased shift of these products to higher density fractions. When RPE-J cells were pretreated with oxLDL for 21 hours, and with medium alone for an additional 24-hour recovery period, and challenged with OS for 4 hours, we detected a reduction in the formation of rhodopsin cleavage products and a decreased appearance of such products in higher density fractions, relative to corresponding results with cells pretreated with LDL (Fig. 6) . This perturbation of processing of OS occurred without any apparent reduction in cathepsin D protein, based on the density of immunoblots, or on total acid protease activity in cell lysates, used as an approximation of lysosomal cathepsin D activity. ²⁵ The in vitro degradation at pH 4.5 of ¹²⁵I-labeled OS by lysates of RPE cells treated with oxLDL, LDL, or untreated, expressed as milligrams degraded ¹²⁵I-ligand per milligram cell protein per 5-hour incubation was: ox-LDL, 162, LDL, 146, untreated, 164. 

To determine whether the deficient cleavage of rhodopsin by RPE-J cells ensued over a longer period of incubation with oxLDL, in a separate experiment we studied the time-course of OS degradation by RPE-J cells. The data in Figure 7 demonstrate that incubation of OS with RPE-J cells induced the formation of rhodopsin cleavage products as early as 2 hours after addition of OS. No cleavage was observed at this time point in cells treated with oxLDL. Furthermore, RPE-J cells treated with oxLDL were unable to produce any appreciable cleavage of rhodopsin even after prolonged (8 hour) coincubation with OS. The amount of immunoreactive cathepsin D in the cells was not affected by oxLDL, as was found in the experiment described in Figure 6 . 

Discussion

In this study we showed that initial binding of OS to RPE, previously shown to be mediated by the integrinα _vβ₅ on RPE cells, ^{16 17} was not affected by preincubation of RPE with oxLDL. Similarly, the phagocytosis step was unaffected, provided a 24-hour recovery period was introduced between the treatment with oxLDL and the challenge with OS. We had found previously that such a recovery period abrogated an oxLDL-induced inhibition of phagocytosis in macrophages (Hoppe et al. unpublished studies). Based on these observations, we concluded that the oxLDL-induced inhibition in the degradation of ¹²⁵I-OS by RPE-J cells was indeed occurring at the level of intracellular degradation of OS and not at the binding and/or uptake steps. 

We developed a sucrose density gradient centrifugation technique that allowed us to isolate fractions containing phagosomes at different levels of maturation, to better define the step in processing of OS being affected by oxLDL. This could be assessed by monitoring the degree of OS degradation by determining the formation of cleavage products of rhodopsin, the major protein in OS by Western blot analysis. ²² This approach allowed us to map the transition of early phagosomes to mature phagosomes after challenge of RPE-J cells with OS, based on the observation that cleavage products of rhodopsin were present in specific density fractions when using a monoclonal antibody to rhodopsin. The shift in density fractions containing significant amounts of the lysosomal protease, cathepsin D, ²³ to a lower density, together with the appearance of rhodopsin degradation products in these fractions, suggests the formation of phagolysosomes. ^{4 5 23 24}  

To our knowledge, this is the first time such a cell fractionation technique has been used on RPE cells that have phagocytosed OS to address questions regarding intracellular trafficking of such phagosomes. We also used this fractionation technique to describe for the first time the association of the small GTPase, Rab5, considered a marker of phagosomes and early endosomes in macrophages, ^{26 27} with RPE phagosomes containing OS. We also found a shift of Rab5 to higher density fractions containing the bulk of rhodopsin cleavage products, with increasing times of phagocytosis. These results differ from the reduction in amounts of Rab5 associated with phagosomes with increasing times after phagocytosis (indicative of phagosome maturation) that we found in RPE cells after phagocytosis of latex particles. This result further highlights the differences in the events occurring during the phagocytosis by RPE cells of OS, as contrasted to that of other large particles mimicked by latex beads. 

Using this new approach, we also found that degradation of the major protein in OS, rhodopsin, ²² was retarded in cells pretreated with oxLDL, when compared with cells pretreated with LDL. This result was consistent with our data showing a reduction in the degradation of ¹²⁵I-labeled OS by oxLDL relative to LDL, and by oxOS relative to OS to TCA-soluble label. However, the latter approach measures only the formation of small degradation products, namely protein hydrolysis to TCA-soluble fragments (iodotyrosine) released by the cell into the medium. By contrast, rhodopsin cleavage, used as a marker of OS processing in RPE cells, represents a more subtle change in processing by demonstrating the initial degradative steps. The reduced degradation of OS protein could be due to a reduction in lysosomal protease mass and/or activity. However, because we found that total acid protease activity in RPE, primarily cathepsin D, in RPE cells, ²⁵ was not reduced by oxLDL, and that cathepsin D mass actually increased, the reduced degradation was most likely due to a perturbation in phagosome maturation to form phagolysosomes. This is further indicated by the reduction in rhodopsin cleavage products found in higher density fractions containing lysosomes, as indicated by the presence of cathepsin D. 

Although in this study we have used oxLDL in a model system to ask fundamental questions of the role of ox-lipids on specific aspects of the cell biology of RPE cells, oxidative damage by products of lipid peroxidation has been considered to be a major player in the etiology of AMD. ⁷ In one hypothesis, the pro-oxidative environment of the retina leads to oxidation of lipids such as the highly unsaturated docosahexaenoic acid in membranes of OS. ^{7 18 28} This results in greater formation of reactive aldehydes ^{18 29 30} that can cross-link proteins, ³⁰ leading to the formation of lipofuscin, a fluorescent product ⁸ that also contains A2E, the major fluorophore of RPE lipofuscin. ^{8 31 32} An ox-lipid–induced reduction in the degradation of components of OS that had been phagocytosed by RPE, similar to our earlier findings in macrophages, ³³ would be expected to lead to greater accumulations of lipofuscin in RPE lysosomes. Because lipofuscin at higher concentrations was shown in model systems to be toxic to RPE cells, ^{9 34 35 36} greater accumulations of lipofuscin might lead to RPE cell death, a characteristic of advanced AMD. 

In another hypothesis, oxLDL plays a direct role in the etiology of AMD. One mechanism linking LDL to this disease process is one in which LDL filters from the choriocapillaris to Bruch’s membrane where the LDL particles undergo oxidation, because of the pro-oxidative milieu. ⁷ If oxLDL could penetrate Bruch’s membrane and come into contact with the basolateral surface of the RPE, as might occur with increasing age, ³⁷ some ox-lipids could enter the cell and perturb OS degradation. This could occur through receptor-mediated endocytosis, if scavenger receptors are present on the basolateral side of the RPE, ³⁸ or by direct transfer of the ox-lipids to the cell membrane. Several lines of published evidence link oxLDL to AMD. Clinical correlative studies have shown an association of AMD and coronary heart disease due to atherosclerosis, ³⁹ a disease in which oxidation of LDL plays a major role. ⁴⁰ In addition, drusen were shown to contain plasma components and lipids, ⁴¹ including cholesteryl esters, ⁴² whereas Bruch’s membrane showed an increase in cholesterol content with age ⁴³ and contained lipid-rich particles that resembled those present in atherosclerotic lesions. ⁴³ Collectively, this latter group of results suggests that LDL particles become trapped in or around Bruch’s membrane where they may eventually interact with the basolateral side of the RPE to modulate ROS processing. 

In conclusion, in this study processing of phagocytosed OS was reduced in RPE-J cells that had been pretreated with oxLDL, when compared with corresponding processing of cells pretreated with LDL. Our data suggest that this perturbation of processing occurs at the level of phagosomal maturation, potentially by inhibiting fusion with lysosomes. The underlying mechanism responsible for a reduced fusion of phagosomes with lysosomes remains speculative. Lipid peroxidation is known to lead to the production of reactive aldehydes that can induce the formation of lipid–protein adducts ^{18 28 29 33} which cause oxidative modification. ^{29 44} If such adduct formation were to occur with key fusion proteins such as Rab5 and Rab7, ^{26 27} this could perturb their functions. Similarly, products formed during lipid peroxidation ^{18 29 30} could indirectly affect the level of key proteins in the fusion event by controlling the level of expression, perhaps by modulating signal transduction events. It will be important to evaluate in future studies any perturbation in their expression by oxLDL treatment. 

 Presented at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May 2000.

 Supported by Scientist Development Grant 9930181N from the American Heart Association (GH) and funds from the Foundation Fighting Blindness (HFH), National Eye Institute Grant EY13160 (ADM), and a Kirchgassner Foundation Research Grant (ADM).

 Submitted for publication March 26, 2001; revised June 12, 2001; accepted June 19, 2001.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked“ advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Pretreatment with oxLDL or oxOS reduced the degradation of OS in RPE-J cells. RPE-J cells were treated for 21 hours with the indicated concentrations of unoxidized or oxidized LDL (a), or unoxidized or oxidized OS (b). Cells were then washed and incubated for 5 hours with fresh medium containing 20 μg/ml ¹²⁵I-labeled OS. The amount of degraded ¹²⁵I-ligand was determined as iodine-free TCA-soluble radioactivity in medium.

Figure 2.

 

LDL and OS were not cytotoxic before and after oxidation at concentrations used in these studies. RPE-J cells were labeled with[ ¹⁴C]adenine for 21 hours, followed by washing and the addition of one of the following: 500 μg/ml OS, 500 μg/ml oxOS, or lipid extracts (chloroform-methanol) obtained from the equivalent protein amounts of OS or oxOS; 200 μg/ml LDL or oxLDL; 0.5% Triton X-100 (TrX); or medium alone (ctr). After a 24-hour incubation at 37°C, aliquots of culture media were removed to measure the release of ¹⁴C radioactivity from the cells.

Figure 3.

 

Binding and internalization of OS by RPE-J cells is not affected by treatment with oxLDL when a 24-hour recovery period is included between the oxLDL treatment phase and the challenge with OS. RPE-J cells were first treated for 21 hours with 100 μg/ml of either LDL or oxLDL added in medium A. Cells were then washed and incubated with FITC-OS. OS binding (a, c) and internalization (b, d) were determined immediately (a, b) or after an additional 24-hour incubation (recovery period) with medium A alone (c, d). Internalization was determined by measuring the fluorescence of samples by fluorometry after trypan blue quenching of extracellular fluorescence. Binding was calculated from the difference in fluorescence before and after trypan blue quenching.

Figure 4.

 

Rhodopsin cleavage products distribute to higher density fractions with time after OS phagocytosis. RPE-J cells were grown in petri dishes and incubated with 10⁸ OS/dish for the indicated times. The cells were then washed and homogenized by inducing shear stress. The 0-hour time point represents OS that was added to the cells, which were then immediately subjected to the fractionation procedure without a washing step. Subcellular fractionation was performed by density gradient centrifugation (25%–56% sucrose). One-milliliter fractions were collected from the top of each tube, diluted, and spun at 3000g. Proteins recovered in the pellet were separated on 4% to 20% SDS-PAGE and probed with an anti-rhodopsin monoclonal antibody (B6-30N). Fractions 1 to 12 represent those isolated with increasing density.

Figure 5.

 

Cathepsin D and Rab5 shifted to lower density fractions with time after OS phagocytosis. RPE-J cells were incubated with OS and fractionated into density fractions, as described in Figure 4 . A 3000g pellet was resolved on SDS-PAGE and probed with an anti-cathepsin D antibody and an anti-Rab5 antibody. Integrated optical density of the Western blot bands (a) was analyzed by computer (Image software; Scion, Frederick, MD), and sucrose gradient density distribution of cathepsin D (b) or Rab5 (c) was plotted as a percentage of the maximal value in each individual time point.

Figure 6.

 

OxLDL treatment of RPE-J cells reduced the formation of rhodopsin cleavage products in phagosomes and decreases their presence in higher density fractions. RPE-J cells were first treated for 21 hours with 100μ g/ml LDL or oxLDL, followed by an additional 24-hour incubation with medium A alone. RPE-J cells were then incubated with OS for 4 hours and fractionated as described in Figure 4 . A 3000g pellet was resolved on SDS-PAGE and probed with an anti-rhodopsin monoclonal antibody and an anti-cathepsin D antibody.

Figure 7.

 

OxLDL induces a prolonged block of OS degradation by RPE-J cells. RPE-J cells were first treated for 21 hours with 100 μg/ml of either LDL or oxLDL, followed by an additional 24-hour incubation with medium A alone. RPE-J cells were then incubated with OS for 2, 4, or 8 hours and fractionated as described in Figure 6 . A 3000g pellet was resolved on SDS-PAGE and probed with an anti-rhodopsin monoclonal antibody.

The authors thank Silvia Finnemann, Ph.D., for helpful advice during the preparation of the manuscript.