Nonlethal Oxidant Injury to Human Retinal Pigment Epithelium Cells Causes Cell Membrane Blebbing but Decreased MMP-2 Activity

Abstract

purpose. This study was undertaken to determine whether transient or sustained nonlethal oxidant injury can induce RPE cell membrane blebbing and alter RPE expression of matrix metalloproteinase (MMP)-2 and type IV collagen, two molecules that are necessary for regulation of the turnover of the RPE basal lamina.

methods. The ARPE-19 cell line stably expressing green fluorescent protein (GFP) targeted to the cell membrane was bleb injured by exposure to myeloperoxidase (MPO; 10 microunits) and H₂O₂ (100 μM). Sustained (>6 hours) or transient (up to 6 hours) exposure to MPO/H₂O₂ was evaluated. An MTS assay conversion and cell counts were used to detect cell viability. Supernatants and the cell homogenates were collected from cultured ARPE-19 to assess fluorescent GFP-derived blebs, MMP-2 protein by Western blot, MMP-2 activity by zymography, and type IV collagen accumulation by ELISA. Expression of MMP-2 was examined by real-time RT-PCR with total RNA.

results. Both sustained and transient exposure of RPE cells to nonlethal oxidant injury upregulated blebbing and increased pro-MMP2 protein, but downregulated the MMP-2 activity released into the supernatant in a time-dependent manner. Only sustained oxidant injury for 24 hours induced an increase in collagen type IV. After removal of transient oxidant exposure, blebbing resolved and RPE MMP-2 activity and protein recovered to normal levels within 48 hours.

conclusions. Sustained or transient oxidant injury causes increased cell membrane blebbing but decreased activation of MMP-2. The findings lead to the hypothesis that blebs released in the absence of active MMP-2 may become trapped between the RPE and its basal lamina as sub-RPE deposits, possibly contributing to drusen formation in age-related macular degeneration. Also, the results lead to the postulation that oxidant injury disrupts the cell-specific surface proteases necessary to cleave and activate pro-MMP-2.

Asubstantial body of literature suggests a role for oxidant injury to the retinal pigment epithelium (RPE) as a putative mechanism in the pathogenesis of age-related macular degeneration (AMD). Although intuitively obvious, oxidant injury can induce either lethal responses, leading to cell death, or nonlethal responses inducing a functional change from baseline compatible with continued life of the cell but leading to dysfunction of the tissue or organ. Most studies focus on oxidant-mediated death of RPE. ^{1 2 3 4} Yet, RPE death (so-called geographic atrophy) is a very late stage of dry AMD, resulting from a very chronic and progressive process. Subretinal deposits and thickening of Bruch’s membrane, the hallmarks of early AMD, develop decades before the RPE cells actually die. Therefore, nonlethal cellular responses to RPE oxidant injury must contribute to early AMD. 

We hypothesized that two specific “nonlethal” injury responses are especially relevant to early deposit formation in AMD: cell membrane blebbing and dysregulated turnover of the extracellular matrix (ECM). Nonlethal cell membrane blebbing is the process by which a cell can pinch off part of its plasma membrane and cytosol in an attempt to discard damaged cellular organelles, molecules, and lipid membrane, ^{5 6} and was first introduced 25 years ago as a possible pathogenic mechanism in drusen formation. ^{7 8 9 10 11} This process is different from lethal blebbing and apoptosis. ^{12 13 14} Nonlethal blebbing is an injury-induced cell response characterized by the formation of “focal adhesions” (membrane plaques of Hsp27, actin, and other proteins), aggregates of cytoplasmic actin filaments and stress fibers, followed by blebbing. ¹⁵ Nuclear fragmentation and cell death do not occur. 

Dysregulated production and breakdown of the ECM is another injury response relevant to AMD. The normal anatomy and physiology of extracellular matrix (ECM) in most tissues requires continuous turnover of collagen and other matrix components by a tightly regulated balance in production of matrix molecules like collagen IV, matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs). ^{16 17} Relatively small dysregulation in the ratio of these factors can produce profound changes in the ECM, including thickening and deposit formation. ^{18 19} Accordingly, dysregulated turnover of ECM is a major mechanism of disease pathogenesis in many tissue sites, including renal disease, atherosclerosis, lung disease, and others. ^{17 18 19 20} Unfortunately, minimal information is available concerning normal turnover in healthy Bruch’s membrane or dysregulated turnover in AMD eyes. Recently, we demonstrated that sustained, nonlethal oxidant injury can induce a wide range of changes in gene expression, especially for those genes involved in regulation of extracellular matrix. ²¹  

Another important issue pertaining to oxidant-mediated injury of the RPE that has not been thoroughly addressed is the distinction between transient and sustained exposure. In vivo, the duration of cellular exposure to oxidant injury is unknown. Intuitively, one might predict most in vivo exposures to be brief—for instance, after transient exposure to intense sunlight or to cigarette smoking. However, more sustained exposure might occur in association with diseases related to circulating plasma oxidants, such as abnormal production of circulating hormones that activate endogenous oxidant enzymes (hypertension-associated angiotensin II), or in diseases such as diabetes. Nonetheless, minimal data are available that compare differences in cellular responses in the setting of variable oxidant exposure times. 

Accordingly, in this study our goals were to examine the ability of nonlethal oxidant injury to regulate cell membrane blebbing and molecules relevant to extracellular matrix accumulation (MMP-2 and collagen type IV) and to compare sustained with transient exposure to the oxidant stimulus. We found that both transient and sustained oxidant injury caused nonlethal blebbing and increased release of pro-MMP-2, but decreased release of active MMP-2. All these effects recovered to baseline after cessation of transient oxidant exposure. These responses favor accumulation of the sub-RPE deposits after oxidant injury. 

Materials and Methods

Cell Culture Conditions

ARPE-19 cells stably expressing green fluorescent protein (GFP) targeted to the inner leaflet of the plasma membrane (GFP-c’-rRas-ARPE-19) were generated as described. ²² Cells were plated at subconfluent density on plastic in T-75 (75 cm²) flasks and grown to confluence in maintenance medium (Dulbecco’s modified Eagle’s medium-Ham’s F12 [DMEM]/F12 [1:1 vol/vol] supplemented with 10% fetal bovine serum [FBS], 1 mM l-glutamine, 100 μg/mL penicillin/streptomycin, and 0.348% Na₂HCO₃) in a 10% CO₂ humidified air incubator at 37°C. All cell culture reagents were purchase from Invitrogen-Gibco (Grand Island, NY). The cells were then subcultured, propagated, and maintained in the same medium. For the experiments, confluent cells were split and plated at subconfluent density in plates or flasks coated with 0.5 mg/mL collagen IV (Sigma-Aldrich, St. Louis, MO) and 0.5 mg/mL laminin (Invitrogen, Carlsbad, CA) mixed 1:1 (vol/vol) and grown to confluence. 

Cell-Viability Assay

Confluent GFP-ARPE-19 cells were split and 1 × 10⁴ cells were plated on collagen IV/laminin-coated 96-well culture plates, as described in Cell Culture Conditions. The cells were then grown for 4 days to confluent density. At the time of confluence (day 0), the cells were prepared for the experiment by changing the maintenance medium to the assay medium (maintenance medium without phenol red and penicillin/streptomycin) for 3 days. This medium was then replaced with assay medium that was supplemented with 1% FBS instead of 10% for 2 days. Subsequently, the medium was changed to the assay medium described in Cell Culture Conditions but with a supplementation of 0.1% in FBS for 1 day. On day 7, cells were treated with 10 microunits myeloperoxidase (MPO: Sigma-Aldrich) for 90 minutes in Earle’s balanced salt solution (EBSS). After that, H₂O₂ was added at different final concentrations for 2 hours. For some experiments, 100 μM H₂O₂ was added for 2, 6, 12, and 24 hours, respectively. The number of surviving cells was measure by cell count (Coulter ZI cell counter; Beckman Coulter, Hialeah, FL), and by MTS (a tetrazolium salt) assay (Cell Titer 96 AQueous One Solution kit; Promega, Madison, WI) 24 hours after removal of the oxidant. MPO (and H₂O₂) were chosen for oxidant injury, because the combination represents a biologically relevant macrophage-derived pro-oxidant enzyme and substrate. Macrophages and monocytes have been linked to the progression of AMD. MPO has been implicated as a major cause of lipid peroxidation and protein oxidation of vascular cells and deposits in atherosclerotic plaques. 

RPE Membrane Blebbing

Confluent GFP-ARPE-19 cells were split and 2 × 10⁵ cells were plated at subconfluent density in six-well plates coated with collagen IV/laminin as described previously. They were grown for 4 days to confluent density in maintenance medium. At this time, cells were prepared for the experiment, as described earlier. Subsequently, on day 7, the cells were incubated with 10 microunits MPO for 90 minutes in EBSS, followed by exposure to different concentrations of H₂O₂ (0, 100, 250, and 350 μM respectively) for 2 hours. After exposure to MPO followed by H₂O₂, cells were examined under a dual-channel laser scanning confocal microscope for blebbing (LSM-510; Carl Zeiss Meditec, Thornwood, NY). 

Sustained and Transient Oxidant Injury

Confluent GFP-ARPE19 cells were plated onto collagen IV/ laminin–coated six-well plates at subconfluent density (2 × 10⁵ cells) and grown for 4 days to confluence. At the time of confluence, the cells were prepared as described in previous sections. On day 7, 10 microunits MPO in EBSS were added for 90 minutes, followed by exposition to 100 μM H₂O₂ in assay medium supplemented with 0.1% FBS. The oxidant exposure was either sustained or transient. For sustained injury, MPO and H₂O₂ were allowed to remain in the medium for the duration of the experiment (2, 6, 12, and 24 hours). For transient exposure, MPO and H₂O₂ were removed after 6 hours (acute transient injury phase), followed by reassessment during the subsequent 6 to 72 hours (recovery phase). The culture medium was removed, and then the cells were washed two times with 1× PBS. After that, fresh assay medium supplemented with 0.1% FBS was added for 24 hours. The cells were harvested for protein collection and/or RNA and for quantification of collagen type IV accumulation, whereas the supernatants obtained to collect proteins and measure MMP-2 activity were collected after centrifugation. Three independent experiments were performed in triplicate, with reproducible results. 

Quantification of GFP-Modified Blebs Released into the Culture Medium

Confluent cells were split and plated at subconfluent density (2 × 10⁵ cells) in six-well plates coated with collagen IV/laminin at concentrations described in the first protocol. The cells were then grown for 4 days to confluence and prepared for the experiment during 6 days, as described previously. On day 7, they were treated with 10 microunits MPO followed by 100 μM H₂O₂. The oxidant exposure was transient or sustained as described earlier in the article. The GFP-modified blebs released into the cultured medium were collected and concentrated by centrifugation. All samples were stored at −80°C until protein quantification by the bicinchoninic (BCA) protein assay. Three micrograms of proteins extract was used to measure GFP-modified blebs by Western blot. 

MMP-2 Activity

Culture medium was collected 24 hours after treatment and centrifuged 30 minutes at 15,000g at 4°C. At the time the supernatant was collected, protein quantification was determined, and MMP-2 activity was assessed using 10% zymography gels, as described previously. ²³ Ten micrograms of protein extracts from each experimental condition were used. Standards were electrophoresed in parallel. Gels were incubated 18 hours in 50 mM of Tris buffer, allowing determination of total proteolytic MMP-2 activity with no interference from their associated tissue inhibitors. ^{24 25} Densitometry was performed using the ImageJ 1.17 densitometry program (available by ftp at zippy.nimh.nih.gov/ or at http://rsb.info.nih.gov/nih-imageJ; developed by Wayne Rasband, National Institutes of Health, Bethesda, MD), to determine relative MMP-2 activity. Each zymography assay was repeated at least three times. Inhibition of gelatinase activity was assayed by incubating gels with 1 mM EDTA, a specific metalloproteinase inhibitor (data not shown). 

Western Blots

MMP-2 and GFP expression was assessed by Western blot, as described previously. ²³ Supernatant and confluent cell layers were collected. Protein concentration was determined by BCA protein assay. Three to 10 μg protein extracts from each experimental condition were denatured with SDS sample buffer followed by 5 minutes of boiling and then were separated on a 10% to 12% polyacrylamide gel (Novex, San Diego, CA). After electrophoresis the proteins were transferred in 1× transfer buffer (25 mM Tris, 192 mM glycine, 0.1% SDS, and 20% methanol [pH ∼8.4]) to a 0.45-μm polyvinylidene difluoride membrane (Immobilon-P; Millipore Corp., Bedford, MA) in a transfer cell (mini-PROTEAN II; Bio-Rad Laboratories, Inc., Hercules, CA) set at a constant voltage of 120 mV for 2 hours. Membranes were then blocked in a 5% nonfat dry milk PBS solution for at least 1 hour at room temperature. Incubation with the primary antibody (monoclonal antibody against GFP, 1 μg/mL [BD Biosciences-Clontech, Palo Alto, CA], or anti-MMP-2 antibody, 1 μg/mL [Chemicon International]) proceeded overnight at 4°C. Membranes were washed four times with PBS, incubated with horseradish-peroxidase–linked donkey anti-mouse antibodies (1:1000 dilution; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) for 2 hours at room temperature and then washed four times with PBS. Immunoreactive bands were determined by exposing the nitrocellulose blots to a chemiluminescent solution and exposing to autoradiograph film (X-Omat AR; Eastman Kodak Co., Rochester, NY). Relative MMP-2 and GFP concentration was determined by the ImageJ 1.17 densitometry program, as described earlier in the article. Three independent experiments were performed in triplicate. Results are expressed as a percentage of control (untreated cells). 

Degradation of GFP by Oxidant in the Medium

We determined whether degradation of GFP induced by MPO and H₂O₂ in the medium might have caused underestimation of bleb release. For this experiment, confluent GFP-ARPE-19 cells were prepared for the experiment described previously. The cells were exposed to a lethal concentration of MPO and H₂O₂ for 12 hours. Then, GFP-modified blebs released into the medium were collected, and the GFP was concentrated by centrifugation. Protein quantification was determined by BCA protein assay. The GFP extract (normalized for total protein content) was added to medium containing 10 microunits MPO and 100 μM H₂O_2, and the medium was sampled at 0, 6, 12, 24, and 48 hours. Total protein quantification was determined, and immunoreactive (intact) GFP was assessed by Western blot analysis with 3 μM of protein extracts from each experimental condition. We found that Western blot detection of GFP decreased in a time-dependent manner with a half-life of approximately 12 hours (data not shown). Thus, bleb release as measured by GFP Western blot underestimates the magnitude of the response. 

Degradation of Active MMP-2 Induced by an Oxidant

We evaluated whether degradation of active MMP-2 induced by an oxidant might have caused decreased release of active MMP-2 into the supernatant. For this study, equal amount of active MMP-2 was placed on top of cell culture inserts (BD Biosciences, Bedford, MA) and 10 microunits of MPO and 100 μM H₂O₂ were added in the bottom well for 0, 2, 6, 12, and 24 hours. Then, 25 μg/mL catalase (Sigma-Aldrich, St. Louis, MO) were added for 15 minutes at 25°C to stop the residual H₂O₂ activity. Protein quantification was determined by BCA assay once the medium was collected and centrifuged at 15,000g for 30 minutes at 4°C, and the supernatant was collected. MMP-2 activity was assessed by gelatin zymography using 10 μg protein extracts from each time point. 

Assessment of Collagen Type IV

Cell layers were collected after 24 hours of incubation, and an ELISA was performed as described. ¹⁸ Briefly, the medium was incubated for 2 hours at 37°C and then in blocking solution for an additional 30 minutes. Incubation with antibody against collagen type IV diluted 1:3000 (Biodesign International, Saco, ME) in blocking solution was performed overnight at 4°C. After washes, a biotinylated goat anti-rabbit IgG (Sigma-Aldrich) was applied for 2 hours. The concentrations of the type IV standards were 0 to 3 ng/well. Three independent experiments were performed in duplicate. Final values were expressed as nanograms per 10⁵ cells, and results are expressed as the percentage of control (untreated cells). 

Isolation of mRNA and Real-Time RT-PCR

Total RNA was extracted from confluent cell cultures as previously described. ²³ Commercial software (Primer Express; Applied Biosystems, Inc. [ABI], Foster City, CA) was used to design the primer pair and probe sequences for human MMP-2. The primer pair was selected so that they amplified different exons, to prevent the amplification of contaminating genomic DNA. The sequence of probe for MMP-2 was 5′-CGCCAAATGAACCGGTCCTTGAAG-3′ and labeled FAM (6-carboxyfluorescein) fluorescence spectrum as a reporter. The amplification primer pair was: 5′-TTGATGGCATCGCTCAGATC-3′ and 5′-TGTCACGTGGCGTCACAGT-3 for MMP-2. RT-PCR reaction was performed using a kit of RT-PCR master mix reagents (TaqMan One-Step; ABI); and a sequence detection system (Prism 7700; Applied Biosystems) in a total volume of 50 μL of reaction mixture. The ribosomal RNA control reagent kit was used to detect the 18S ribosomal RNA gene, which represented an endogenous control. Each sample was normalized to the 18S transcript content. The primer probe mixture was purchased from ABI and used as specified by the manufacturer. The standard curve for MMP-2 and 18S were generated with serially diluted solutions (0.001–100 ng) of mRNA from cultured RPE cells. PCR assays were conducted in duplicate for each sample. Data are expressed as a percentage of untreated cells and represent the mean ± SEM of four independent experiments run in triplicate. 

Statistical Analyses

All experiments were performed three or four times on cultured cells, with reproducible results. Data are expressed as a percentage of the control. Results are the mean ± SEM of three or four independent experiments, performed in duplicate or triplicate (as indicated). One-way ANOVA and the Dunnett multiple comparison post hoc tests were performed. 

Results

Lethal and Nonlethal Injury after Transient or Sustained Oxidant Exposure

The GFP-c’-rRas-ARPE-19 cells were used to evaluate cell viability after MPO-mediated injury. The viability of ARPE-19 was determined by MTS assay 24 hours after transient oxidant injury (3.5 hours) with 10 μM MPO and various concentrations of H₂O₂. As shown in Figure 1 , the 50% lethal dose (LD)₅₀ and LD₉₀ were approximately 275 and 500 μM H₂O₂, respectively. However, concentrations of 100 μM H₂O₂ or less were nonlethal within 24 hours. 

In vivo, the duration of exposure to oxidant injury is unknown. Thus, we exposed the cells to different durations of 10 microunits MPO and 100 μM H₂O₂. Cell counts were performed 24 hours after oxidant exposure was initiated that lasted 2, 6, 12, and 24 hours. We found that increasing the duration of exposure to this concentration of oxidant injury did not decrease the cell count (Fig. 2) . Comparable results were observed for cell viability measured by MTS assay (data not shown). Thus, transient or sustained exposure to 10 microunits MPO and 100 μM H₂O₂ was nonlethal. 

Cell Membrane Blebbing during Oxidant Injury

The GFP-c’-rRas-ARPE-19 cells were used to evaluate cell membrane blebbing after transient exposure to nonlethal and lethal oxidant injury. Figures 3A 3B 3C 3Ddemonstrate the extensive blebbing that occurred in RPE cells 3.5 hours after exposure to nonlethal and lethal concentrations of MPO and H₂O₂. At an H₂O₂ concentration less than 200 μM, small surface blebs were apparent. At a concentration greater than 200 μM many RPE cells showed large protruding cell membrane blebs, as well as morphologic characteristics of cell death, such as cell shrinkage or detachment (Figs. 3C 3D) . 

We collected the GFP-modified blebs released into the medium after oxidant-mediated bleb injury and determined the amount of GFP by Western blot analysis (Fig. 3E) . We found significant increases of GFP release in the supernatant at all concentrations tested, with an apparent linear increase of GFP release with increased H₂O₂ concentration. During the remaining experiments, we focused our attention on the nonlethal concentration of 10 microunits MPO and 100 μM H₂O₂. 

We determined the time course of bleb release after initiation of oxidant injury (Fig. 4) . We found that GFP release into the supernatant was significantly increased in comparison with the control within 2 hours (P < 0.05), and was at its maximum by 12 hours (approximately a sixfold increase over control; P < 0.001). After 24 hours of treatment, the GFP level was less than at earlier time points. This decline was still observed, even when the data were corrected for potential GFP oxidative degradation (not shown). 

We also evaluated the disappearance of blebbing after discontinuation of transient (6 hours) oxidant injury. Within 18 hours after the termination of transient oxidant injury, the release of GFP-modified blebs had declined dramatically, and it returned to baseline within 48 hours (Fig. 5) . 

Effect of Sustained Oxidant-Injury on MMP-2 Activity and Protein Expression in RPE Cells

Oxidant injury has been shown to regulate MMP-2 in nonocular tissues. ^{26 27 28} In the present study, we examined whether oxidant injury modulates MMP-2 activity and protein. We exposed RPE cells to MPO followed by H₂O₂ injury for 2, 6, 12, and 24 hours. By zymography, we found that sustained oxidant-mediated injury downregulated MMP-2 activity released into the supernatant in a time-dependent manner, with no detectable activity after 24 hours (Fig. 6A) . In the presence of 10 microunits MPO and 100 μM H₂O₂ for 6 hours, MMP-2 activity diminished approximately 1.8-fold (P < 0.05), and a further decrease in MMP-2 activity was observed after 12 hours of sustained injury (P < 0.01). No active MMP-2 was detected after 24 hours of sustained oxidant exposure (Fig. 6A) . 

To explain the decrease of active MMP-2 in the supernatant during times in which we expected high amounts of surface MMP-2 to be blebbed into the culture medium, we performed two analyses. First, we confirmed that persistence of the oxidant within the culture medium did not degrade the enzymatic activity of exogenous MMP-2. As shown in Figure 6B , the presence of MPO and H₂O₂ did not cause a decline in activity of exogenously added active MMP-2. Second, we used Western blot analysis to compare the impact of oxidant injury on the ratio of secreted pro-MMP-2 protein (72 kDa) and cleaved, active MMP-2 (68 kDa). At both 6 and 12 hours after oxidant injury, a large increase of pro-MMP-2 (72 kDa) was present in the supernatant, which continued to increase by 24 hours after injury (Fig. 7A) . In contrast, no proportionate increase in cleaved MMP-2 (68 kDa) was observed. Also, we observed a 61% decline of cell-associated pro-MMP-2 and cleaved MMP-2 in the cellular lysate (Fig. 7B)at 6 hours after oxidant injury. Paradoxically, oxidant injury for 12 and 24 hours seemed to induce a modest upregulation of cell-associated pro-MMP-2 latent and active protein (Fig. 7B) . 

We also performed real-time RT-PCR on total RNA extracts to determine the impact of oxidant injury on MMP-2 mRNA expression. Minimal modification in levels of mRNA was observed after acute oxidant injury for 2, 6, and 12 hours, and only a small (but statistically significant) increase of 41.8% was observed after exposure of RPE cells to oxidant injury for sustained 24 hours (Table 1) . In summary, sustained oxidant injury greatly diminished MMP-2 activity but increased release of pro-MMP-2 protein in the supernatant. The large increase in the ratio of pro-MMP-2 to active MMP-2 suggests the loss of cell surface proteases responsible for cleavage and activation. 

Regulation of MMP-2 by Transient Oxidant Injury in RPE Cells

We exposed RPE cells to MPO followed by H₂O₂ injury for 6 hours of oxidant exposure (acute transient oxidant injury phase), followed by replacement with maintenance medium (reassessment after removal of injury stimulus during the subsequent 6 to 72 hours). By zymography, we found that transient exposure to oxidant injury was associated with diminished MMP-2 activity (∼1.86-fold; 50.4% ± 7.38% of control; P < 0.05) 6 hours after removal of the oxidant, but quickly recovered to normal levels 24 hours after injury (Fig. 8A) . Thereafter, MMP-2 activity released into the culture medium remained at control levels up to 72 hours after injury. Thus, oxidative damage to the enzyme activity of released MMP-2 was unlikely to explain the observed decrease in MMP-2 activity. 

Similar to the results observed with sustained oxidant injury for pro-MMP-2 protein, we also observed a significant increase of the ratio of pro-MMP-2 to active MMP-2 in the supernatant 6 hours after acute injury (1.5-fold increase; P < 0.05), which quickly recovered to normal levels 24 hours after injury (Fig. 8B) . Also, cell-associated pro-MMP-2 was downregulated by 88% during this time point (Fig. 8C) . Minimal changes in MMP-2 mRNA expression were observed 24 and 48 hours after transient injury, although a small increase in mRNA expression was observed at 72 hours after injury. (Table 2) . In summary, transient oxidant injury with removal of the oxidant stimulus resulted in similar changes of decreased MMP-2 activity and increased pro-MMP protein release, as observed with sustained exposure, although they were of smaller magnitude and duration. 

Regulation of Collagen Type IV by Oxidant-Mediated Injury in RPE Cells

Regulation of collagen synthesis and secretion by oxidants has been shown in a variety of nonocular ^{29 30 31} and ocular tissue. ³² Various studies have shown that RPE cells can synthesize collagen type IV, the most important collagen in the basal lamina. ^{33 34 35} Sustained oxidant injury for 24 hours induced an approximately 1.7-fold increase (175.12% ± 6.25%; P < 0.05) in collagen type IV (Fig. 9A) . In contrast, transient oxidant-injury did not induce changes in collagen accumulation (Fig. 9B) . 

Discussion

In this study, exposure of RPE cells to nonlethal oxidant injury induced cell membrane blebbing, increased the release of pro-MMP-2 protein, but greatly diminished MMP-2 activity. These responses were most pronounced on sustained oxidant exposure for 12 to 24 hours, but were also apparent after shorter, transient oxidant exposure. However, RPE recovered to normal within 24 to 48 hours after oxidant injury was terminated, although some long-term dysregulation in MMP-2 mRNA was suggested. 

Cell membrane blebbing is a well-defined injury response after cellular exposure to a wide range of injury stimuli, including toxic drugs, oxidants, and physical agents. ^{6 12 36} Nonlethal blebbing is recognized as one of the normal cellular responses to sublethal injury, ^{37 38} and may provide a mechanism by which an injured cell can discard damaged plasma membrane, organelles, and cytosolic proteins. Moderate amounts of blebbing can be well tolerated by some cells, including the RPE as shown in this and other studies by our group. ²² Nonlethal blebbing occurs in vivo and may be a common cellular injury response in certain diseases characterized by extracellular deposit accumulation, such as glomerulonephritis. ^{39 40} Several observations of human AMD and in animal models suggest that blebbing may contribute to sub-RPE deposits. ⁴¹  

In this work, we demonstrated that both transient and sustained oxidative injury to the RPE induced membrane blebbing, similar to our past findings and those of other groups. ^{22 37 38} The use of GFP to label the inner leaflet of the cell membrane is a novel and convenient method for the evaluation of blebbing, since we used Western blot analysis to measure GFP release as a surrogate of blebbing. 

Strong evidence supports the hypothesis that MMPs and their tissue inhibitors play an central role in the pathogenesis of deposit accumulation in diverse disorders such as renal disease and atherosclerosis. ^{17 18 19 20} Not surprisingly, dysregulation of these molecules in AMD pathogenesis is the topic of recent research. The RPE synthesizes collagens, fibronectin, and many other molecules crucial for the formation of its basement membrane and repair of Bruch’s membrane. ³³ Furthermore, we and others have shown that RPE synthesizes MMPs, especially MMP-2, crucial for the degradation and turnover of extracellular matrix, and that MMP-2 synthesis, release, and activity can be regulated by physiological stimuli. Active MMP-2 is the major RPE enzyme for the degradation of collagen I, collagen IV, and laminin, all essential components of Bruch’s membrane. ^{42 43} This study expands our previous work and demonstrates the capacity of oxidant injury to dysregulate MMP-2 activity. Dysregulation of MMP-2 activity is likely to play a role in AMD. For example, a recent study by Leu et al., ⁴⁴ revealed that, in AMD eyes, areas of normal Bruch’s membrane contain demonstrably active MMP-2, but drusen and sub-RPE deposits are “cold” spots for MMP-2 activity, correlating the decline in MMP-2 activity with deposit accumulation. 

These data also indicate a complex interrelationship between oxidant-induced injury and MMP-2 activation. Oxidant injury caused significant release of pro-MMP-2, but greatly diminished active MMP-2. Although direct oxidation of MMP-2 may contribute to some of the observed diminished enzymatic activity after bleb injury, we believe that the accumulation of large amounts of extracellular pro-MMP-2 indicates a more complex form of dysregulation. MMP regulation occurs by gene transcription, translational regulation and posttranslational activation of proenzymes. MMP-2 is secreted in a latent pro form in which the prodomain folds over the catalytic site. ⁴⁵ Activation of pro-MMP-2 occurs when the prodomain is cleaved by cell surface proteases, especially membrane type 1-matrix metalloproteinase (MT1-MMP). Preliminary data suggest that bleb injury interferes with the activation of latent pro-MMP-2 into active MMP-2, perhaps by downregulation of the expression of cell surface MT1-MMP and/or its accessory protein TIMP-2 (Elliot S, et al. IOVS 2004;45:ARVO E-Abstract 1816). 

Our results demonstrate surprisingly good agreement between the responses to transient and sustained oxidant exposure, although the magnitude and duration of blebbing, released pro-MMP-2, and diminished MMP-2 activity were obviously greater after sustained injury. Although we believe that transient exposure to oxidant injury followed by a period of recovery probably represents conditions more physiologically relevant to bleb-inducing injury in vivo, experimental protocols that use sustained but nonlethal oxidant exposure are likely to produce comparable findings. 

MPO was used as a pro-oxidant enzyme in these studies based on the strong emerging role for macrophages and inflammation in the pathogenesis of AMD. ^{46 47} The most widely implicated RPE oxidants are those induced by RPE exposure to visible light or those derived from endogenous metabolism. ^{48 49} Environmental oxidants derived from cigarette smoke, pollution or industrial byproducts have also been proposed to contribute to AMD (Cousins SW, et al. IOVS 2003;44:ARVO E-Abstract 1619). Macrophage infiltration under drusen has been observed frequently in eyes with dry AMD, suggesting the possibility that oxidants produced by macrophages or monocytes are released locally. ^{50 51} In this regard, macrophage-derived MPO is a well-characterized macrophage-derived pro-oxidant enzyme. ^{50 52}  

MPO converts its substrate, H₂O₂, into an active oxidant. H₂O₂ by itself is a weak oxidant and is neutralized by catalase and other antioxidant enzyme systems. ^{53 54} However, in the setting of MPO release, H₂O₂ is converted into powerful oxidants such as hydroxyl radicals, hydroperoxides, hypochlorous acid, and tyrosyl radicals. ⁵⁵ Among their actions, MPO-derived oxidants induce injury to the cell membrane and modify cell surface proteins and receptors. ^{50 55 56 57 58 59} Macrophage-derived MPO has been implicated as the major pro-oxidant enzyme to cause lipid peroxidation in atherosclerotic plaques. Of note, the RPE synthesizes large amounts of H₂O₂. To our knowledge, no one has sought evidence to confirm or refute the presence of MPO in drusen. 

Taken together, the data suggest that even transient oxidant exposure can induce cellular responses that might promote sub-RPE deposit accumulation. Active MMP-2 regulates the breakdown and turnover of type IV collagen in the RPE basement membrane and regulates the turnover of other collagens in the inner Bruch’s membrane. In the absence of active MMP-2, RPE blebs containing cell membranes, cytosolic proteins, and organelles may be expected to accumulate as deposits between the RPE cell membrane and its basal lamina. Also, excessive amounts of new basement membrane may accumulate over these trapped blebs, causing drusen. Experiments in which repetitive oxidant injury is used to create basal deposit accumulation in vitro will be performed to test this hypothesis. Furthermore, it is possible that therapies that preserve the function of RPE-derived MMPs after oxidant injury may promote deposit clearance and diminish the progression of AMD. 

 Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May 2001.

 Supported by National Institutes of Health Grants 1RO1EY13318, 1RO1EY014477, 1RO1EY015249, and by an unrestricted grant to the University of Miami from Research to Prevent Blindness, Inc.

 Submitted for publication October 15, 2004; revised March 23, 2005; accepted April 21, 2005.

 Disclosure: M.E. Marin-Castaño, None; K.G. Csaky, None; S.W. Cousins, 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.

 

Viability of GFP-ARPE-19 cells to MPO followed by H₂O₂. Confluent cells were exposed to 10 microunits MPO for 90 minutes, followed by various doses of H₂O₂ (100, 200, 300, 400, 500, and 600 μM) for 2 hours. Cell viability was determined by MTS assay. Data are expressed as the mean ± SEM (n = 3). Significant differences at *P < 0.05, **P < 0.01 and ***P < 0.001, compared with the control.

Figure 2.

 

Time–response curve for the number of cells in human GFP-ARPE-19 cultured cells treated with 10 microunits MPO and 100 μM H₂O₂. Cells were treated with MPO followed by H₂O₂ for 2, 6, 12, and 24 hours, then incubated in assay medium (0.1% FBS) for 24 hours. The number of cells was calculated as the percentage of control. Data are expressed as the percentage of control and are shown as the mean ± SEM of results in three independent experiments run in triplicate.

Figure 3.

 

Morphology of cells and quantification of GFP in cell-derived blebs evaluated by immunoblot analysis. (A) Fluorescent GFP-ARPE-19-derived blebs before and after exposure to oxidant-mediated injury with MPO and H₂O₂. GFP-ARPE-19 were exposed to 10 microunits MPO for 90 minutes, followed by 100, 250, or 350 μM H₂O₂ for 2 hours and observed immediately by confocal fluorescence microscope. (A) Control GFP-ARPE-19 cells in which GPF was localized to the membrane (arrowheads). (B) Membrane blebs (arrows) and GFP in the membrane (arrowheads) after exposure to 10 microunits MPO and 100 μM H₂O₂. (C, D) Prominent cell membrane blebbing (arrows), protrusions, and detachment, after exposure to 10 microunits MPO and 250 or 350 μM H₂O₂. Images represent three independent experiments. Magnification, ×40. (E) Western blot analysis with a monoclonal antibody against human GFP (27 kDa) showed GFP expression in RPE-derived blebs before and after injury. Three micrograms of total protein extract were loaded in each lane.

Figure 4.

 

Quantification of 27-kDa GFP-modified blebs released into the culture medium by Western blot analysis after sustained oxidant exposure to 10 microunits MPO followed by 100 μM H₂O₂. Top: Western blot from a representative experiment. Lane M: molecular weight standard. Lane C: control. Bottom: averages of results of three independent experiments run in triplicate on cultured cells. Statistically significant differences (*P < 0.05, **P < 0.01 and ***P < 0.001), compared with the control.

Figure 5.

 

Quantification of 27-kDaGFP-modified blebs released into the culture medium by Western blot analysis after transient oxidant exposure to 10 microunits MPO followed by 100 μM H₂O₂. Top: Western blot from a representative experiment. Lane C: control. Bottom: averages of results of three independent experiments run in triplicate on cultured cells. Statistically significant differences (*P < 0.05 and **P < 0.01), compared with the control.

Figure 6.

 

Sustained MPO oxidant-mediated injury downregulated MMP-2 activity. GFP-ARPE-19-derived MMP-2 protein activity evaluated by zymography in presence of 10 microunits MPO and 100 μM H₂O₂ for different hours (0, 2, 6, 12, or 24 hours). (A) Top: gelatin zymogram from a representative experiment. Lane M: molecular weight standard. Lane C: control. Bottom: averages of results of three independent experiments run in triplicate on cultured cells. Statistically significant differences (*P < 0.05 and **P < 0.01), compared with the control. (B) Sustained MPO oxidant-mediated injury did not affect degradation of active MMP-2. An equal amount of active MMP-2 was treated with 10 microunits MPO and 100 μM H₂O₂ for 0, 2, 6, 12, or 24 hours. At the end of these times, 25 μg/mL catalase was added to the reaction mixture. The reaction was further incubated at 25 °C for 15 minutes Top: Zymography analysis of MMP-2 activity from a representative experiment. Lane C: control. Bottom: average of results of three independent experiments run in triplicate.

Figure 7.

 

Sustained exposure to MPO oxidant mediated injury-regulated MMP-2 protein. GFP-ARPE-19-derived MMP-2 protein expression evaluated by Western blot analysis in presence of 10 microunits MPO and 100 μM H₂O₂ for 0, 2, 6, 12, or 24 hours. Top: Western blot analysis of MMP-2 expression from a representative experiment in the supernatant (A) and in the cellular lysate (B). Lane C: control. Bottom: average of results of three independent experiments run in triplicate. Data represent the relative amount of the 72-kDa form. Statistically significant differences (*P < 0.05 and **P < 0.01), compared with the control.

Table 1.

 

Regulation of MMP-2 mRNA by Sustained Oxidant-Mediated Injury in Cultured GFP-ARPE-19 Cells

Times	MMP-2 mRNA
		Amount (ng)	% Change
Control 2 h	97.1
2 h	97.3	+0.2
Control 6 h	98.4
6 h	99.43	+1.05
Control 12 h	99.35
12 h	108.98	+9.69
Control 24 h	101.12
24 h	143.4	+41.81*

 MMP-2 mRNA expression was normalized to 18S transcript content. Data are expressed as a percentage of control. Results are mean nanograms ± SEM of results in three independent experiments run in triplicate on cultured cells.

*  Statistically significant at P < 0.05.

Figure 8.

 

Transient exposure to oxidant injury was associated with diminished MMP-2 activity (in the supernatant), protein (in the cellular lysate), and increased MMP-2 protein (in the supernatant) during acute injury. GFP-ARPE-19 cells were treated with 10 microunits MPO and 100 μM H₂O₂ for 6 hours of (acute transient oxidant injury phase), and then reassessed after removal of the injury stimulus during the subsequent 6 to 72 hours (recovery phase). (A) Top: gelatin zymogram from a representative experiment. Lane M: molecular weight standard. Lane C: control. Bottom: average of results of three independent experiments run in triplicate on cultured cells. Statistically significant difference (*P < 0.05) compared with the control. (B, C) Top: Western blot analysis of MMP-2 expression from a representative experiment in the supernatant (B) and in the cellular lysate (C). Bottom: the average of results of three independent experiments run in triplicate. Data represent the relative amount of the 72-kDa form. Statistically significant difference (*P < 0.05 and **P < 0.01) compared with the control.

Table 2.

 

Regulation of MMP-2 mRNA by Transient Exposure to Oxidant Injury in Cultured GFP-ARPE-19 Cells

Times	MMP-2 mRNA
		Amount (ng)	% Change
Control 6 h	143.6
6 h	145.38	+1.2
Control 24 h	148.2
24 h	155.02	+4.6
Control 48 h	139.98
48 h	151.8	+8.4
Control 72 h	153.1
72 h	235.7	+53.95*

 MMP-2 mRNA expression was normalized to 18S transcript content. Data are expressed as a percentage of control. Results are mean nanograms ± SEM of results three independent experiments run in triplicate on cultured cells.

*  Statistically significant at P < 0.05.

Figure 9.

 

MPO-sustained oxidant-mediated injury upregulated collagen type IV accumulation in GFP-ARPE-19 cells after 24 hours of treatment but did not induce changes in collagen type IV accumulation in these cells during the transient exposure to MPO followed by H₂O₂. Quantification of collagen type IV accumulation in GFP-ARPE-19 cells by ELISA after sustained (A) or transient (B) exposure to oxidant injury. Data are expressed as a percentage of the control and shown as the mean ± SEM of three independent experiments run in duplicate. Statistical significance (*P < 0.05) compared with the control.

The authors thank Paola Catanuto, Simone Pereira-Simon, and Philippa Smit for assistance in the study.