Effects of Lipid Peroxidation-Related Protein Modifications on RPE Lysosomal Functions and POS Phagocytosis

Abstract

purpose. Lipofuscin accumulation in the RPE is a common downstream pathogenic pathway in various monogenic and complex retinal diseases including age-related macular degeneration (AMD). Lipid peroxidation-induced modification of proteins is thought to play a role in lipofuscinogenesis and may contribute to RPE dysfunction. A prior study demonstrated that a variety of lipofuscin-associated proteins are damaged by aberrant covalent modifications of malondialdehyde (MDA) and 4-hydroxynonenal (HNE). The present study was conducted to test the hypothesis that these damaged proteins are more resistant to proteolytic attack and act as protease inhibitors.

methods. Isolated photoreceptor outer segments (POS) were radioactively labeled and in vitro modified with MDA and HNE. Pure lysosomal fractions isolated from human RPE were tested for their proteolytic activities toward modified and unmodified POS proteins. In parallel, modified and radiolabeled POS were fed to RPE cell cultures for phagocytosis and their lysosomal degradation as well as intracellular accumulation was compared with unmodified POS.

results. Both experimental approaches revealed that MDA or HNE modifications strikingly increase the resistance of POS proteins to the attack by lysosomal proteases. When cultured RPE cells were fed with modified or unmodified POS the amount of degraded POS proteins was reduced by approximately 60% to 70% for the modified POS compared with those in normal control subjects. Some of the modified proteins remained undegraded in the lysosomal compartment of cultured RPE cells and were still detectable 3 weeks after feeding, whereas unmodified POS were completely degraded within 1 week after feeding. Moreover, modified proteins had the potential to impair degradation of unmodified proteins, indicating their efficacy as proteolytic antagonists.

conclusions. The results indicate that lipid peroxidation-derived protein modifications are involved in lipofuscinogenesis and may contribute to cell damaging effects of lipofuscin in retinal diseases such as AMD.

Several lines of evidence suggest that dysfunction of the retinal pigment epithelium (RPE) is a crucial event in the molecular pathogenesis of age-related macular degeneration. ¹ The RPE fulfills metabolic functions that are essential for proper action and survival of retinal photoreceptors. These vital functions include maintenance of the visual cycle by continuous uptake, processing, transport, and release of vitamin A; generation of ion gradients within the photoreceptor matrix; mediation of active transport of nutrients between the choroid and the photoreceptors; formation of the outer blood-retinal barrier; and phagocytic uptake and degradation of the constantly shed photoreceptor outer segments (POS). ^{2 3} As a consequence of the task of permanently digesting POS, the lysosomal apparatus of postmitotic RPE cells in the macular region, compared with all other cells in the body, has to bear the largest load of material to digest. Impaired lysosomal degradative capacities of the RPE are assumed to play an essential role in the initiation of pathophysiological events finally causing RPE damage and resulting in degeneration of the corresponding neurosensory retina. ^{4 5 6} Lipofuscinogenesis is considered a consequence of the lysosome’s inability to degrade phagocytosed material completely, ^{7 8} and lipofuscin depositions in the lysosomes of the RPE are a common and characteristic feature of various blinding diseases, including age-related macular degeneration (AMD). Several lines of evidence indicate that RPE lipofuscin impairs normal RPE cell function ⁹ and clinical findings signal that excessive lipofuscin accumulations precede atrophy of outer retinal layers, including the RPE with functional loss. ^{10 11} Thus, understanding the molecular mechanism underlying ocular lipofuscinogenesis is likely to shed light on the pathogenesis of AMD. 

A proteome analysis of ocular lipofuscin granules revealed that various lipofuscin proteins are damaged by lipid peroxidation-derived modifications. ^{12 13} Lipid peroxidation is a complex process that involves the action of oxygen-derived free radicals on polyunsaturated fatty acids, which are degraded into a variety of products. Some of them, such as highly electrophilic aldehydes, are very reactive and have the potential to damage proteins. Thus, malondialdehyde (MDA) and 4-hydroxynonenal (HNE), representing the major lipid peroxidation products, are capable of easily attaching covalently to proteins by forming adducts with cysteine, lysine, or histidine residues. ¹⁴ Finally, the modified proteins tend to form aggregates by intra- or intermolecular cross-linking via MDA or HNE modification. Such protein aggregation appears to be a common feature of several neurodegenerative diseases, and decreased proteolysis of modified or cross-linked proteins has been proposed as a primary cause of lipofuscin-like deposits in disease-specific lesions. ¹⁵ Lipid peroxidation-derived aldehydes are also likely to contribute to lysosomal disorder and lipofuscinogenesis in RPE cells, thereby triggering AMD-specific RPE damage. ^{6 13 16 17} However, there is still no direct evidence that MDA or HNE modifications of phagocytosed POS proteins can induce lysosomal storage of undegraded material in RPE cells. We therefore investigated lysosomal degradation and accumulation kinetics of MDA- and HNE-modified POS in comparison to normal POS in isolated RPE lysosomes and RPE cultures. 

Materials and Methods

Cell Culture and Isolation of Lysosomes

Primary cultures of human RPE cells were isolated and maintained according to published standard procedures. ¹⁸ Lysosomes were isolated from cultured RPE cells as described previously. ¹⁹  

POS Isolation and Radiolabeling

POS were isolated from porcine eyes (obtained from a local abattoir) according to the method of Schraermeyer et al. ²⁰ Radioactive labeling of POS proteins was conducted as described. ²¹ Specific radioactivity was 4.5 MBq/mg protein. Protein was determined by the Lowry procedure. After radiolabeling, the POS were divided into three portions. One was left unmodified (normal control), whereas the others were modified with either MDA or HNE. 

MDA and HNE Modification of POS

For synthesis of MDA-adducted proteins of POS these were reacted with 20 mM MDA, prepared from 1,1,3,3-tetramethoxypropane (Sigma-Aldrich, Munich, Germany) by acid hydrolysis, exactly as detailed by Lung et al. ²² Modification was 39 nanomoles MDA/mg protein, as measured by thiobarbituric acid activity. ²² HNE-modified POS proteins were prepared by reacting with 5 mM trans-4-hydroxy-2-nonenal (HNE), produced from (E)-4-hydroxynonenal-dimethylacetal (Sigma-Aldrich) by acid hydrolysis, according to a published procedure. ²³ Modification was a 48-nanomole/mg protein as determined by HNE-specific ELISA. ²³  

Proteolytic Activity of Isolated Lysosomes toward POS Proteins

Isolated lysosomes (1 μg protein) were mixed with radiolabeled POS (10 μg protein; normal controls as well as either MDA- or HNE-modified) in the presence of 100 mM sodium acetate (pH 5), in a total volume of 100 μL and incubated at 37°C for 0, 30, 60, 90, and 120 minutes. The incubation was stopped by the addition of an equal volume of 10% trichloroacetic acid. Bovine serum albumin (50 μg) was added as a precipitation aid. After centrifugation (10,000g, 4°C, 15 minutes) the supernatant was counted for radioactivity (dpm1). The precipitate was solubilized in 0.5 N NaOH and also counted for radioactivity (dpm2). Protein degradation rates were calculated in the following manner: % degradation = [dpm1/(dpm1 + dpm2)] × 100. 

Degradation of POS Proteins in RPE Cultures

Cells were seeded in 24-well plates at an initial density of 10⁵ cells per well and cultured for 72 hours. Then, radioactively labeled POS (1 μg/well; unmodified and either MDA- or HNE-modified) were added to the culture medium for 24 hours. The medium was then withdrawn and protein precipitated by the addition of an equal volume of 10% trichloroacetic acid. After centrifugation, the supernatant was counted for radioactivity (dpm1). Cells were solubilized by adding 500 μL 0.5 N NaOH per well. This solution was treated with 10% trichloroacetic acid and centrifuged as just described. The supernatant (dpm2) and the precipitate (dpm3), after solubilization with NaOH, were also counted for radioactivity. Intracellular protein degradation was calculated as: % degradation = [(dpm1 + dpm2)/(dpm1 + dpm2 + dpm3)] × 100. 

Determination of Phagocytosis Rates

Cells were seeded in 24-well plates at an initial density of 10⁵ cells per well and cultured for 72 hours. Then, 20 mM ammonium chloride for inhibition of lysosomal function and, 15 minutes later, radiolabeled POS (100,000 dpm/well; unmodified as well as MDA- or HNE-modified) were added. After incubation for 0, 30, 60, 120, 180, and 240 minutes, medium was removed, and the cells were solubilized with 0.5 N NaOH. The solution was treated with trichloroacetic acid and centrifuged as described earlier. The supernatant was discarded, and the pellet was redissolved in 0.5 N NaOH and counted for radioactivity. 

Determination of Intracellular Protein Storage

Cells were seeded and cultured as just described. The cells were then fed with radiolabeled POS (100,000 dpm/well; unmodified as well as either MDA- or HNE-modified) in the presence of 20 mM ammonium chloride for 24 hours (pulse phase). Thereafter, the cells were cultured in normal medium without ammonium chloride and POS for 1, 2, or 3 weeks (chase phase). Intracellular TCA-unsoluble radioactivity was determined after solubilization of the cells and subsequent trichloroacetic acid precipitation as just described. 

Results

Radioactive labeling of isolated POS yielded high specific radioactivities of 4 to 5 MBq/mg protein allowing sensitive measurement of POS protein degradation. Quantitation of MDA- or HNE-modifications after in vitro modification of POS with the aldehydes yielded 39 nanomoles MDA/mg protein and 48 nanomoles HNE/mg protein. Western blot analysis with MDA- or HNE-specific antibodies after SDS-gel electrophoresis of the modified proteins indicated broad spreading of the modification over the bulk of POS proteins (data not shown). To measure directly the proteolytic activity of lysosomal proteases toward radioactively labeled normal and MDA- or HNE-modified POS, these were mixed with a purified lysosomal fraction and protein degradation rates were measured. Degradation rates of MDA- or HNE-modified POS were reduced by 52% and 61%, respectively, compared with normal POS (Fig. 1) . When cultured RPE cells were fed with modified or unmodified POS the amount of degraded POS proteins was reduced by approximately 60% to 70% for the modified POS compared with the normal control (Fig. 2) . Blocking of lysosomal (ammonia-sensitive) protein degradation totally abolished POS protein degradation (Fig. 2) . To detect the potential effects of MDA or HNE modifications on phagocytosis rates of POS in cultured RPE cells, lysosomal protein degradation was blocked by ammonium chloride, and intracellular accumulation of radioactively labeled POS was quantified. Approximately the same phagocytosis rates were measured in the control and modified POS (Fig. 3) . Finally, long-term lysosomal storage of the POS was tested. Some of the modified proteins remained undegraded in the lysosomal compartment of cultured RPE cells and were still detectable 3 weeks after feeding, whereas unmodified POS were completely degraded within 1 week (Fig. 4) . 

The observed increased stability of MDA- or HNE-modified proteins toward proteolytic attack by lysosomal proteinases may elicit a competitive inhibitory effect on degradation of normal POS. This possibility was tested in an experimental setting in which the lysosomal degradation of radioactively labeled normal POS was measured in the presence of increasing amounts of “cold” POS that were either unmodified or MDA or HNE modified. As shown in Figure 5MDA- or HNE-modified POS exerted a striking inhibitory effect on degradation of normal POS substrates by isolated lysosomes. Thus, normal POS degradation was reduced by 70% to 80% in the presence of an excess of aldehyde-modified proteins. In an analogous experiment, the inhibitory action of aldehyde-modified proteins on degradation of normal POS proteins was also tested in RPE cultures. The modified proteins had a potent inhibitory effect on degradation of normal phagocytosed POS proteins in cultured RPE cells (Fig. 6) . 

Discussion

Lipid peroxidation-derived cellular damage has been related to normal cellular aging and various age-related degenerative diseases. Tissues containing large numbers of postmitotic cells, as in the brain, skeletal muscles, or heart, are particularly at risk for age-related deterioration in function by lipid peroxidation products. The initiation of lipid peroxidation also requires the generation of oxygen-derived free radicals and the presence of polyunsaturated fatty acids. The outer retina is exposed to light in an oxygen-rich environment, and unsaturated fatty acids are present in high concentrations in the photoreceptor membranes of the retina. ²⁴ Accordingly, evidence for light-induced lipid peroxidation reactions in the retina has been reported in several studies. ^{25 26 27} Exposure to intense light is thought to acutely induce retinal damage by generating the production of high doses of lipid peroxidation-derived DNA-reactive aldehydes that trigger photoreceptor cell apoptosis. ¹⁶ Besides forming DNA lesions, the reactive aldehydes resulting from lipid peroxidation are also capable of easily forming protein adducts. ^{14 27} However, if such protein damage occurs in the photoreceptor outer region, the permanent renewal of the outer segments will clear away and replace the damaged proteins. Thus, persistent damage of the retinal outer region by such protein modifications appears unlikely. The retinal pigment epithelium, which has to phagocytose and degrade all material shed from the photoreceptor outer region, may be affected by damaged POS proteins. In vitro studies suggest that proteins adducted with lipid peroxidation-derived aldehydes have reduced susceptibility to proteolysis by α-chymotrypsin. ²⁸ The degree of protein modification used in this study corresponds to the range of in vivo carbonyl modification detected in aged human erythrocytes. ²⁹ Quantification of carbonylation of human POS has not been accomplished yet, but is also considered to be relatively high. Thus, the degree of modification used in our study was chosen after the preceding investigations. ^{28 29} Our results in testing the proteolytic activity of isolated lysosomal fractions from RPE cells indicated that MDA- or HNE-modified POS proteins are also stabilized toward hydrolytic attack by lysosomal proteinases. Because endoproteinases, in particular the cathepsins D, B, L, and H, are initially necessary to cleave proteins into TCA-soluble fragments, ³⁰ these results suggest that lipid peroxidation-modified POS proteins are at least partially stabilized toward cathepsin-mediated proteolysis. A similar mechanism has been suggested as a cause of oxidized low-density lipoprotein accumulation within macrophages. ³¹ In our experiments, cultured RPE cells were fed with POS for 24 hours, and degradation rates for MDA- or HNE-modified POS proteins were only 30% compared with unmodified POS. The presence of ammonium chloride, a blocker of intralysosomal proteolysis, ³² completely abolishes degradation of normal as well as modified POS, which indicates that the lysosomal compartment is the exclusive site of POS degradation. Thus, in accordance with the results obtained with isolated lysosomes, also in intact cells, lipid peroxidation-modified POS proteins appear to be more resistant to intracellular proteolysis than do their normal counterparts. 

Discrimination between cell surface-bound and internalized POS is a well-known problem in the field of phagocytosis of POS by RPE cells. However, our results are based on the measurement of the generation of TCA-soluble degradation products representing proteolytically degraded POS proteins. The observed increase of low-molecular-weight radioactivity with time derives from lysosomal POS degradation, as confirmed by the ammonium chloride control. Therefore, the presented data on POS degradation in cultured RPE cells exclusively reflect catabolism of internalized POS. 

Because of the very high concentrations of cathepsins in lysosomes the rate-limiting step for intralysosomal proteolysis of phagocytosed material in intact cells is normally its uptake into the cell. ³³ We therefore had to exclude the possibility that the phagocytic step preceding lysosomal proteolysis—namely, binding of POS to cell surface-associated uptake receptors or the intracellular transport to the lysosomal compartment, is influenced by the lipid peroxidation aldehyde modification of the substrates. Ammonia is considered to be a specific inhibitor of intralysosomal hydrolytic activity without short-term effects on phagocytosis or autophagic sequestration. ³² Thus, in the presence of ammonium chloride, the intracellular accumulation of TCA-insoluble material resulting from its uptake from the culture medium represents phagocytosis rates. In this experimental setting for measuring phagocytic uptake rates for POS in cultured RPE cells, similar rates for unmodified and HNE- or MDA-modified POS were observed, indicating that POS uptake by the RPE cell is not affected. A recent study suggested that intense light exposure promotes “oxidative tagging” of POS with structurally defined choline glycerophospholipids that may serve as a physiological signal for CD36 scavenger receptor-mediated phagocytosis under oxidative stress conditions. ³⁴ Our results provide no indication that MDA or HNE modifications may also serve as specific ligands for uptake via CD36. Altogether, our experiments exploring the effects of lipid peroxidation-derived aldehydes on the degradability of phagocytosed POS show that MDA or HNE modifications increase the resistance of POS proteins to lysosomal proteases, whereas no direct effects on phagocytic uptake were detectable. 

As shown in the pulse-chase experiments in which RPE cells were loaded with radioactively labeled POS and the label chased for 3 weeks, normal POS proteins were found to be totally degraded within 1 week. In contrast undegraded MDA- and HNE-modified proteins were still detectable after 3 weeks of chase, indicating long-term storage of protease-resistant POS proteins. Such intracellular deposition of undegraded material is considered seminal for lipofuscinogenesis. ^{7 35}  

In addition to their altered susceptibility to proteolysis carbonylated proteins may be also effective as proteolytic antagonists, thereby acting as general proteolytic inhibitors. Thus MDA-modified albumin has been shown to act as a potent inhibitor of the proteolytic activity of the protease α-chymostatin. ²⁸ We therefore tested the potential effects of MDA- or HNE-modified POS proteins on the lysosomal degradation of normal POS. Indeed, lipid peroxidation-modified POS proteins hampered degradation of normal POS, as observed in the experiments with highly purified RPE lysosomal fractions as well as with cultured RPE cells. Accordingly, lipid peroxidation damage of POS proteins may be considered the initiator of a vicious cycle in which undegraded aldehyde-modified proteins accumulate intralysosomally thereby causing reduction of lysosomal degradative capacities by antagonizing lysosomal endoproteinases which in turn accelerates lysosomal storage of undegraded material. 

Lysosomal dysfunction and extensive intralysosomal storage of undigested material is known to impair overall cellular function severely, as most strikingly exemplified by the inherited lysosomal storage disorders, including mucopolysaccharidosis, oligosaccharidosis, or lipidosis. Lipofuscin is considered a product of intralysosomal storage material in postmitotic cells and represents a hallmark of various blinding retinal diseases, including AMD. Lipid peroxidation-derived aldehydes, such as MDA and HNE, were suggested to contribute considerably to lipofuscinogenesis. ^{13 36} Our present results corroborate the concept that these aldehydes may initiate lipofuscinogenesis by forming protease-resistant protein adducts. 

Besides being a product of lipid peroxidation processes, lipofuscin itself may also represent a source of lipid peroxidation damage. It has been shown that exposure of lipofuscin-fed RPE cells to short-wavelength-visible light causes photo-oxidative processes generating several oxygen-reactive species, which also produce increased levels of MDA and HNE. ³⁷ Thus, the RPE not only has to deal with lipid peroxidation-modified substrates resulting from phagocytosis but also with substrate modifications generated directly in the lysosomal compartment. Photoreactivity of lipofuscin is also likely to inactivate lysosomal hydrolases and impair lysosomal stability. ³⁸ Currently, studies are being performed in our laboratory with the purpose of elucidating the effects of lipid peroxidation-derived aldehydes on catalytic function of lysosomal hydrolases in RPE cells. 

 Supported by German Research Council (DFG) Priority Program “Age-Related Macular Degeneration” (SPP 1088), DFG Grants Ho 1926/2-1 and Ko 1663/2-3, and Bonn University Bonfor Grant Kr O-137.0004.

 Submitted for publication May 18, 2006; revised September 29, 2006; accepted January 5, 2007.

 Disclosure: E. Kaemmerer, None; F. Schutt, None; T.U. Krohne, None; F.G. Holz, None; J. Kopitz, None

 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.

 

Degradation of POS by isolated lysosomes. A lysosomal fraction was isolated from cultured human RPE cells and mixed with radioactively labeled POS. Degradation rates of normal POS (○) were compared to POS that were either (▾) MDA- or (▪) HNE-modified. Results are the means ± SD of three independent measurements.

Figure 2.

 

Degradation of POS in cultured RPE cells. Radioactively labeled POS were included in the culture medium of human RPE cells for 24 hours, and POS degradation was determined. Radioactive POS were either nonmodified (open box), MDA-modified (light grey box) or HNE-modified (dark grey box). In parallel, the effect of 20 mM NH₄Cl was tested: (striped box) normal POS + 20 mM NH₄Cl; (light grey striped box) MDA-modified POS+20 mM NH₄Cl; (dark grey striped box ) HNE-modified POS + NH₄Cl. Results are the mean of eight measurements ± SD.

Figure 3.

 

Phagocytosis rates for POS in cultured RPE cells. Radioactively labeled POS were included in the culture medium, whereas lysosomal degradative function of the cells was blocked by 20 mM ammonium chloride. Intracellular accumulation of protein-bound (TCA-insoluble) radioactivity was determined at the indicated time points. (○) Normal POS; (▾) MDA-modified POS; (▪) HNE-modified POS. Results are the means of four measurements ± SD.

Figure 4.

 

Intracellular accumulation of MDA- or HNE-modified POS in cultured RPE-cells. Radioactively labeled POS were included in the culture medium for 24 hours. The cells were cultured in nonradioactive medium for 3 weeks, and intracellular TCA-insoluble radioactivity was determined at the indicated time points. Radioactive POS were nonmodified (□), MDA-modified ( ), or HNE-modified (▪). Results are the means of four measurements ± SD.

Figure 5.

 

Inhibition of proteolytic activities by MDA- or HNE-modified POS proteins in isolated lysosomes. Isolated lysosomes were mixed with 10 μg radioactively labeled POS and unlabeled 5, 10, or 50 μg POS, resulting in labeled/nonlabeled ratios as indicated on the abscissa. Unlabeled POS were nonmodified (□), MDA-modified ( ), or HNE-modified (▪). POS protein degradation in the unmodified control samples was set 100%. Results are the means of three measurements ± SD.

Figure 6.

 

Inhibition of lysosomal proteolytic activities by MDA- or HNE-modified proteins in cultured RPE cells. Radioactively labeled POS were mixed with a fivefold excess of unlabeled POS, included in the culture medium of human RPE cells for 24 hours, and POS degradation was determined. Unlabeled POS were nonmodified (□), MDA-modified ( ), or HNE-modified (▪). POS protein degradation in the unmodified control samples was set 100%. Results are the means of four measurements ± SD.