Retinal Pigment Epithelium Protection from Oxidant-Mediated Loss of MMP-2 Activation Requires Both MMP-14 and TIMP-2

Abstract

purpose. Eyes with age-related macular degeneration (AMD) demonstrate accumulation of specific deposits and extracellular matrix (ECM) molecules under the retinal pigment epithelium (RPE). Metalloproteinases (MMP) are crucial regulators of basement membrane and ECM turnover. Accordingly, loss of RPE MMP activity most likely leads to excessive accumulation of collagen and other ECM, a potential mechanism for formation of deposits. A prior study showed that MMP-2 activity, but not pro-MMP-2 protein, decreases after RPE oxidative injury, indicating that oxidant injury disrupts the enzymatic cleavage of pro-MMP-2. Activation of MMP-2 requires the formation of a tri-molecular complex of pro-MMP-2, MMP-14, and tissue inhibitor of metalloproteinases (TIMP)-2. Therefore, a study was conducted to investigate the impact of oxidant injury on the interaction between these three molecules.

methods. Human GFP-RPE cells were oxidant injured by transient exposure to H₂O₂ and myeloperoxidase, and the time course of recovery determined. Supernatants and cell lysates were collected for analysis of MMP-2, MMP-14, and TIMP-2 activity, mRNA and protein expression. In some studies, overexpression with either MMP-14 or TIMP-2 was performed to revert the cells to a preinjury phenotype.

results. Transient injury resulted in a decrease of both MMP-14 and TIMP-2 activity and protein. Overexpression of each single molecule failed to prevent the injury-induced decrease of MMP-2 activity. In contrast, overexpression of MMP-14 together with the addition of exogenous TIMP-2 prevented the reduction of MMP-2 activation.

conclusions. Loss of MMP-2 activity after oxidant injury is caused by the downregulation of MMP-14 and TIMP-2. Overexpression of either MMP-14 or TIMP-2 alone before oxidant injury is not enough to prevent loss of MMP-2 activity. All three components of the tri-molecular complex must be present to preserve normal MMP-2 activity after oxidant injury.

Age-related macular degeneration (AMD) is the most important cause of lost central vision in the elderly. ¹ Histopathology of early AMD demonstrates accumulation of specific lipid-rich deposits under the retinal pigmented epithelium (RPE). Many reports have suggested a role for oxidant injury to the RPE as a putative mechanism for AMD pathogenesis, and several different potential oxidants have been suggested, such as those induced by light exposure, endogenous metabolism, and inflammatory cells. We hypothesize that two specific “nonlethal” injury responses are especially relevant to early deposit formation in AMD. In previous studies, we have demonstrated that oxidant injury to the RPE causes nonlethal blebbing, characterized by extrusion of cell membrane and cytosol. However, dysregulated production and breakdown of extracellular matrix (ECM) components in Bruch’s membrane is another specific injury response relevant to AMD. 

The normal anatomy and physiology of ECM in most tissues requires continuous turnover of collagen and other matrix components by a tightly regulated balance in production of such matrix molecules as collagen IV, matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs). ² Relatively small dysregulation of the ratio of these factors can produce profound changes in the ECM, including thickening and deposit formation. 

We and others have shown that RPE expresses high levels of MMP-2, a gelatinase important for degradation of collagen IV, fibronectin, and other matrix components relevant to Bruch’s membrane and sub-RPE deposit formation. ^{3 4 5 6} In recent work, we have shown that transient nonlethal RPE injury with the macrophage-derived pro-oxidant enzyme, myeloperoxidase (MPO), plus its substrate, hydrogen peroxide, results in cell membrane blebbing associated with increased release of pro-MMP-2 from the cell surface, but decreased activation of pro-MMP-2 into the active form. ⁷ Because most MMPs, including MMP-2, are synthesized as pro molecules that require posttranslational cleavage to produce maximum enzymatic activity, oxidant injury may alter production, expression, or activity of enzyme regulators of MMP activation, to explain the decrease in MMP-2 activity. 

Two important regulators of pro-MMP activation include membrane type 1 matrix metalloproteinase (MT1-MMP or MMP-14), and tissue inhibitor of metalloproteinases (TIMP-2). ^{8 9} Active cell surface bound MMP-14 binds the aminoterminal domain of TIMP-2 whereas its carboxyl terminal domain interacts with pro-MMP-2. ¹⁰ Activation of the tethered pro-MMP-2 is executed by a second active MMP-14. Conversely, binding of a second TIMP-2 results in inhibition of pro-MMP activation by MMP-14. Because we and others have shown that RPE expresses high levels of pro-MMP-2, active MMP-2, and TIMP-2, ^{3 4 5 6 11} we investigated whether MMP-14 is expressed in RPE cells and whether MMP-14 and TIMP-2 mRNA and protein expression are altered after oxidant injury, as a mechanism for changes in MMP-2 activation. 

We found that both MMP-14 and TIMP-2 mRNA and protein were expressed and active on noninjured cultured RPE cells. Six to 8 hours after removal of transient oxidative injury, both MMP-14 and TIMP-2 protein expression and activity decreased, resulting in decreased MMP-2 activation. The changes returned to normal by 96 hours after removal of oxidant injury. Of importance, transfection studies revealed that both MMP-14 and TIMP-2 were necessary to prevent oxidant-mediated loss of MMP-2 activation. Loss of regulation of MMP activation may contribute to AMD pathophysiology, and these molecules may be targets for therapeutic intervention. 

Materials and Methods

Cell Culture Conditions

ARPE-19 cells stably expressing green fluorescent protein (GFP, GFP-RPE), were grown in 10% FBS DMEM-F12 and exposed to 10% CO₂. ¹² For experiments, cells were plated in 24- or 6-well plates (Nunclon; Nalge Nunc, Rochester, NY) and allowed to grow to confluence. At the time of confluence, cells were exposed to 10% FBS DMEM-F12, phenol-red free for 48 hours, followed by 10% charcoal dextran-stripped serum (Hyclone, Logan, UT) phenol-red free for 24 hours, and 1.0% charcoal dextran stripped serum and phenol-red free medium for an additional 24 hours. All cell culture reagents were purchased from Gibco-Invitrogen (Grand Island, NY). 

Oxidant Injury

GFP-RPE cells were treated with 10 μU MPO followed by 100 μM H₂O₂, as previously described. ⁵ Six hours later, cells were washed and exposed to 0.1% charcoal stripped serum. Twenty-four hours after treatment GFP-RPE cells were collected and processed. For time course experiments, cells were collected 6, 20, 48, 72, 96, and 144 hours after injury. 

MMP-2 Activation Assay and Zymography

Ten micrograms of GFP-RPE cell lysate, 10 mU of pro-MMP-2 and reaction buffer (50 mM Tris-HCl, 5 mM CaCl₂, 1 mM ZnCl₂, and 0.05% NaN₃) were incubated for 1 hour at 37°C. Gelatin zymography was performed as previously described, with 10 μL aliquots of this reaction mixture. ¹³ Samples were applied to an SDS polyacrylamide gel copolymerized with 10% gelatin. Gels were rinsed twice after electrophoresis in 2.5% Triton X-100 and then incubated for 18 hours at 37°C in incubation buffer (50 mM Tris-HCl, 5 mM CaCl₂, 1 μM ZnCl₂, and 0.05% NaN₃). The gelatin gels were stained with Coomassie brilliant blue and destained with 10% acetic acid and 10% isopropanol and dried. 

MMP-14 Activity

GFP-RPE cells were plated in 24-well plates and grown to confluence at which time cells were treated as just described. After treatment, the medium was removed and replaced with 250 μL of extraction buffer provided with the MMP-14 activity assay system (Biotrak; GE Healthcare, Piscataway, NJ) and incubated at 4°C for 15 minutes. The supernatant was assayed for MMP-14 according to the manufacturer’s directions for lower endogenous MMP-14 levels (assay range, 0.125–4 ng/mL). 

Reverse Zymography

TIMP expression was determined as previously described. ¹³ Briefly, conditioned medium was prepared as for zymography and loaded onto a polyacrylamide gel containing gelatin and recombinant gelatinase A. The gels were processed as for zymography and after staining with Coomassie brilliant blue, areas of inhibition of gelatinase A activity were visualized as blue-stained regions on a clear background. 

Western Blot Analysis

GFP-RPE cell lysates were extracted and the protein assessed with the bicinchoninic acid [BCA] protein assay kit (Pierce Biotechnology, Rockford, IL). Equal amounts of protein were applied to precast SDS polyacrylamide gels (Invitrogen[b]) as previously described. ⁵ Electrotransfer of proteins from the gel to the nitrocellulose was performed by electroelution and immunoblotting. Blots were exposed to antibodies against MMP-14 and TIMP-2 (Chemicon, Temecula, CA) followed by chemiluminescence solution (Santa Cruz Biotechnology, Santa Cruz, CA) and exposure to autoradiograph film (X-Omat AR; Eastman-Kodak Co., Rochester, NY). The film was scanned for densitometric analysis by using Image J software from NIH (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). For immunoprecipitation experiments, 100 μg of protein extract was incubated with MMP-14 antibody (Chemicon) or normal goat IgG for 1 hour at 4°C, followed by the addition of protein G-agarose overnight. The resultant protein G-antibody conjugate was centrifuged at 4°C and washed four times with lysis buffer (pH 7.4). The final pellet was resuspended and analyzed as described above. Human MMP-14 or TIMP-2 were used as the positive controls for the antibodies. 

Real-Time PCR

Amplification and measurement of target RNA was performed on a sequence-detection system (Prism 7700; Applied Biosystems, Inc. [ABI], Foster City, CA), as previously described. ¹⁴ Quantitative RT-PCR was performed in a single one-step buffer system according to the manufacturer’s instructions. The sequence of the primers and probes for TIMP-2 and MMP-14 have been published (see Refs. ¹⁵ , ¹⁶ , respectively). All primers and probes were synthesized by ABI. PCR assays were conducted in duplicate for each sample. 

Transfection Studies

A PCR3.1 expression vector containing a full-length, membrane-anchored MMP-14 ¹⁷ was transfected into GFP-RPE cells (Transfast; Promega Corp., Madison, WI), as previously described. ⁵ To determine whether MMP-14 was overexpressed, protein was collected, and Western blot analysis performed as described earlier. Transfected cells were exposed to increasing concentrations of TIMP-2 followed by the standard injury protocol, as described earlier. Supernatants were collected for zymography and reverse zymography and the number of cells counted for loading normalization. Control cells were transfected with an empty control plasmid. Alternatively, cells were infected with a replication-deficient adenoviral (Ad) vector encoding the cDNA for hTIMP-2 under the transcriptional control of cytomegalovirus. Ad vector infection in vitro was achieved with 5 pfu/cell (multiplicity of infection). 

Statistical Analysis

The mean ± SD of the measures were calculated and probabilities (Student’s t-test) performed (Prism; GraphPad, San Diego, CA). P < 0.05 was considered statistically significant for all forms of statistical analysis used. 

Results

Expression of Pro-MMP-2, TIMP-2, and MMP-14

MMP-2 activation requires a tri-molecular complex between pro-MMP-2, MMP-14, and TIMP-2. We have previously shown the presence of MMP-2 in GFP-RPE cells, ⁵ and, in the present study, Western analysis of GFP-RPE cell lysates revealed a band at approximately 72 kDa representing MMP-14 (Fig. 1A , lane 2) and at 22 kDa representing TIMP-2 (Fig. 1B , lane 2). To determine whether MMP-14 and TIMP-2 are bioactive, we performed an activity assay to measure the activation of exogenously added pro-MMP-2 into active MMP-2 (Fig. 2) . The data demonstrated a doubling of specific activity in the presence of GFP-RPE cell lysate compared with medium alone, which was not exposed to cells. We conclude that bioactive MMP-14 and TIMP-2 are produced by cultured GFP-RPE. 

Oxidative Injury Decreases Activity of MMP-14 and TIMP-2 in GFP-RPE Cells

To test our hypothesis that oxidative injury changed surface activation of MMP-2 and thereby led to ECM dysregulation, we injured GFP-RPE cells by exposure to MPO/H₂O₂. As previously described by our group, MMP-2 activity decreased by more than 60% after oxidant injury (33% ± 12% of control, P < 0.005 n =3). The time course of recovery (Fig. 3A , Table 1 ) indicated that the maximum loss occurred within 20 hours after removal of the oxidant, followed by a slow return to normal within 96 hours. TIMP-2 activity demonstrated a partial decrease that remained constant throughout the experiment (Fig. 3B , Table 1 ). Loss of MMP-14 activity decreased in a pattern that paralleled MMP-2 (Table 1) . Of importance, MMP-14 activity was preserved at the 6-hour time point, suggesting that direct oxidation of the molecule was not the specific mechanism for loss of activity. 

Oxidative Injury Decreases MMP-14 Protein and mRNA Expression

MMPs demonstrate complex transcriptional and translational regulation. We investigated the relative protein and mRNA expression of MMP-2, MMP-14, and TIMP-2 after oxidant injury. As shown by us previously, oxidant injury actually increased pro-MMP-2 protein expression, consistent with a loss of bioactivation. ⁵ However, TIMP-2 protein demonstrated a partial reduction in expression after injury (Table 2) . More striking, we found that oxidant injury caused a decline in MMP-14 protein expression (Table 2) . The time course of recovery indicated that decreased MMP-14 protein expression persisted for 48 hours after removal of oxidant injury and then returned to baseline (Table 2) . 

Real time RT-PCR (Table 3)demonstrated that after oxidant injury there was a decrease in MMP-14 mRNA content at 20 hours, followed by recovery. MMP-2 mRNA and TIMP-2 mRNA acted in a similar manner. Taken together, the data suggest that oxidant-induced loss of MMP-2 activity occurred concomitantly with loss of MMP-14 activity which, in turn, was related to significant downregulation of MMP-14 protein synthesis. 

Overexpression of MMP-14 or TIMP-2

MMP-14 was overexpressed in GFP-RPE cells and oxidant injury performed as described. Cell lysate and supernatant were collected approximately 24 hours after injury. MMP-14 overexpression after transfection was detected by Western blot analysis. Although overexpression of MMP-14 prevented the injury-induced decrease in MMP-14 protein and activity (Fig. 4A) , it did not prevent the injury-induced decrease of MMP-2 activity (Fig. 5 , lane 5). Therefore, increasing concentrations of human recombinant TIMP-2 were added to GFP-RPE cells before injury (Fig. 5 , lanes 6–10). TIMP-2 enhanced the activation of MMP-2 when MMP-14 was overexpressed after injury, suggesting that both TIMP-2 and MMP-14 were necessary to restore MMP-2 activation after oxidant injury. 

In complementary experiments, overexpression of TIMP-2 prevented the injury-induced decrease of TIMP-2 (Fig. 4B)as measured by reverse zymography. However recombinant TIMP-2 alone (Fig. 5 , lane 3) or overexpression of TIMP-2 in the absence (Fig. 5 , lanes 11, 12) or presence of recombinant MMP-14 did not prevent the injury-mediated decrease in MMP-2 activity (30% decrease in MMP-2 activity compared with control cells). 

Discussion

AMD, the most important cause of central vision loss in the elderly, is characterized by progressive thickening and accumulation of ECM deposits under the RPE. ^{1 18} We have previously shown that loss of RPE MMP-2 activity correlates with deposit severity in vivo when mice are exposed to blue light, a surrogate for sunlight and acute oxidant injury. ¹⁹ We therefore hypothesized that dysregulated production of MMP-2 after oxidant injury could contribute to matrix accumulation. The regulation of MMP-2 activation is well studied in other tissues; however, little information is available regarding MMP-2 activation in human GFP-RPE cells. ^{5 7}  

Activation of MMP-2 is a membrane associated event requiring the presence of TIMP-2 and surface MMP-14 binding. This complex acts as a receptor for pro-MMP-2 and subsequent conversion to its active form. In this study, we demonstrated that GFP-RPE cells have active MMP-14 on the cell surface and that it is required for MMP-2 activation. Although the presence of MMP-14 has been reported in human RPE cells, ^{4 20} to our knowledge, there have been no reports of its ability to activate MMP-2 in vitro in retinal epithelial cells. Because TIMP-2 performs dual roles of either promoting or inhibiting activation of MMP-2, we also investigated its presence in human RPE cells. We were able to detect TIMP-2 protein by Western analysis and reverse zymography, as previously reported in rat and human RPE cells. ^{6 21}  

To mimic an oxidant-mediated injury relevant to AMD, we exposed the cells to MPO/H₂O₂. MMP-14 protein expression and activity were significantly decreased to 50% of control 6 hours after injury and correlated with a decrease of MMP-2 activity. Not unexpectedly, TIMP-2 activity and protein expression were also significantly decreased by 6 hours. 

Regulation of MMPs is controlled by gene transcription, activation of pro-MMPs, and interaction of secreted MMPs with inhibitors. ²² Therefore, time course experiments were performed to assess transcriptional and translational responses to injury. We found the maximum decrease in MMP-14 activity and protein expression at 48 hours, at which time recovery began and returned to near baseline levels by 96 hours. Of note, mRNA expression began to increase at 20 hours before recovery of activity. These data partially confirm that the injury, although limited to the cell surface, was not merely damaging cell surface–associated proteins. 

MMP-14 activity is crucial for both physiological and disease processes. It was initially identified as the activator of MMP-2, ^{8 23} but recently has been shown to be essential for wound healing, angiogenesis, and inflammation ²⁴ . It is overexpressed in cancers leading to migration, invasion, and metastasis. ²⁵ MMP-14 activation occurs via a proprotein convertase, ^{26 27} and recent data suggest that its function is modified by glycosylation, internalization, and recycling. ^{28 29} Finally, MMP-14 directly degrades ECM molecules including collagen type I and III, laminin, and fibronectin. ^{30 31} MMP-14 in our system clearly plays a crucial role in regulating MMP-2 and subsequent ECM changes, although its direct influence on accumulation of deposits in AMD remain to be explored in future experiments. 

TIMP-2 protein and activity recovered in a coordinate manner, although in our experiments, activity levels never fully reached control levels. Despite this, TIMP-2 activity was sufficient to form a ternary complex with pro-MMP-2 and MMP-14 and to promote recovery of MMP-2 activity. 

TIMP-2 has a dual role as mediator of activation and inhibitor of MMP-2. ³² Translational regulation of TIMP-2 is not well described, as its expression is constitutive. Sp1 and NF-Y have been shown to be involved in a cAMP-dependent TIMP-2 gene transcription activation. ³³ Yamamoto et al. ³⁴ showed that oxidant injury to an RPE cell line increased cAMP production 36 hours later, which could lead to increased TIMP-2 gene expression. This appears consistent with our mRNA expression data. Alternatively, TNFα activation of the ERK-MAPK pathway after oxidant injury may play a role in TIMP-2 protein regulation. Alexander and Acott ³⁵ showed that treatment of human and porcine trabecular cells with TNFα increases MMP expression but decreases TIMP-2 activity. This finding may be relevant, as oxidative stress has been shown to cause ERK activation in RPE cells. ³⁶  

It is likely that the amount of TIMP-2 protein produced after injury is precisely regulated for the RPE to maintain ECM balance. In fact it has been proposed for TIMP-1, that there is an exquisite equilibrium between the rate of mRNA degradation and the rate of translational initiation and elongation. 

To further test our hypothesis that MMP-2 changes occur due to changes in its regulator molecules, we transiently transfected human RPE cells to overexpress MMP-14 before oxidant injury. Overexpression of MMP-14 prevented the injury-induced decrease in MMP-14 activity and protein, but did not prevent the injury-induced decrease in MMP-2 activity. Therefore, we pretreated transfected cells with increasing concentrations of recombinant TIMP-2 and found that both MMP-14 and TIMP-2 were necessary to restore MMP-2 activation to preinjury levels. This effect of TIMP-2 was concentration dependent. When we induced overexpression of TIMP-2, we were unable to restore MMP-2 activity after injury, even with the addition of recombinant MMP-14. We presume that in our system overexpression of TIMP-2 produced a ratio of MMP-14 to TIMP-2 that favored inhibition rather than activation of MMP-2. These paradoxical effects of TIMP-2 have been shown in other cell types, and it appears that, in RPE cells as well, prevention of dysregulation of ECM requires a precise ratio of MMP-14 to TIMP-2. 

A recent report eloquently showed that there is an alternative pathway for MMP-2 activation through another surface MMP, MMP-15, which is TIMP-2 independent. ³⁷ The expression of MMP-15 has been reported in human RPE cells, but in our studies this molecule clearly did not compensate for the lack of MMP-14 and TIMP-2 after injury. ⁴  

Although there have been reports of other MMP and TIMP molecules produced by RPE cells, we focused on the components of the tri-molecular complex. In fact, MMP-14 was found to be the most abundant molecule in microarray analysis of human RPE samples. ²¹ This does not exclude the relevance of RPE synthesized MMP-9 and TIMP-1, which are regulated when exposed to cytokines. ⁴ In fact, we detected a decrease in TIMP-1 activity by reverse zymography (data not shown) after injury. In addition, we were able to visualize TIMP-3, although there were no changes in activity after injury. Mutations in the gene for TIMP-3 cause Sorsby’s fundus dystrophy, a degenerative disease of the macula, ³⁸ and possible implications for its role in wet AMD have been suggested. ³⁹  

We propose that in dry AMD RPE injury or stimulation with macrophage-derived factors, especially MPO and TNF-α lead to dysregulated RPE production of matrix molecules that may influence turnover of Bruch’s membrane, accumulation of deposits, and development of corneal neovascularization. Prevention of dysregulation of the components of the tri-molecular complex may lead to development of therapeutic interventions. 

 Supported in part by National Eye Institute Grant R01 EY14477-02 (SE, SWC).

 Submitted for publication September 23, 2005; revised December 15, 2005, and January 3, 2006; accepted February 21, 2006.

 Disclosure: S. Elliot, None; P. Catanuto, None; W. Stetler-Stevenson, 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.

 

MMP-14 and TIMP-2 is expressed by GFP-RPE cells. Human GFP-RPE cell lysates were collected and analyzed by Western blot for the presence of MMP-14 (A) and TIMP-2 (B). (A) MMP-14 protein expression. Lane M: marker; lane 1: positive control expressing only the catalytic domain; lane 2: endogenous expression of MMP-14 in GFP-RPE cell lines. (B) TIMP-2 protein expression. Lane M: marker; lane 1: positive control human TIMP-2 protein; lane 2: endogenous expression of TIMP-2 in GFP-RPE cells. Western blots are representative of results in three individual experiments.

Figure 2.

 

Pro-MMP-2 is activated by RPE cells. There was a twofold increase in pro MMP-2 activation in the presence of GFP-RPE cell lysates compared with medium alone. Ten micrograms of GFP-RPE cell lysate and 10 mU of pro-MMP-2 were incubated and analyzed by zymography. Data are expressed as the percent of activation of pro-MMP-2. Results are representative of two individual experiments performed in duplicate on cultured cells.

Figure 3.

 

(A) Time course of MMP-2 and TIMP-2 activity after injury. (A) Representative zymography of MMP-2 activity time course before (C) and after injury (I), from 6 to 72 hours. Data are representative of results in three individual experiments performed in duplicate. (B) Representative reverse zymography of TIMP-2 activity time course before (C) and after injury (I) from 20 to 114 hours. R, recombinant TIMP-2 protein. Data are representative of results in three individual experiments performed in duplicate.

Table 1.

 

Time Course of Recovery of MMP-2, MMP-14, and TIMP-2 Activity

Time after Removal of Oxidant Injury (h)	% of Control Activity MMP-2	% of Control Activity MMP-14	% of Control Activity TIMP-2
6	38 ± 12	77 ± 18	ND
20	73 ± 16	49 ± 26	70 ± 4.5
48	81 ± 16	68 ± 0.2	72 ± 10
72	78 ± 13	81 ± 13	78 ± 2.3
96	NC	84 ± 17	66 ± 7.9

 Data are mean ± SD of results in three experiments performed in duplicate. NC, no change from control; ND, not determined.

Table 2.

 

Time Course of Recovery of MMP-2, MMP-14, and TIMP-2 Protein Expression

Time after Removal of Oxidant Injury (h)	% of Control Protein Content MMP-2	% of Control in Protein Content MMP-14	% of Control Protein Content TIMP-2
6	64 ± 0.8	51 ± 3.7	75 ± 15
20	87 ± 1.2	ND	63 ± 2
48	NC	62 ± 24	70 ± 15
72	ND	85 ± 1.4	66 ± 14
96	NC	92 ± 7.7	91 ± 8

 Data are mean ± SD of results in three experiments performed in duplicate. NC, no change from control; ND, not determined.

Table 3.

 

Time Course of Recovery of MMP-2, MMP-14, and TIMP-2 mRNA Expression

Time after Removal of Oxidant Injury (h)	% of Control mRNA Content MMP-2	% of Control mRNA Content MMP-14	% of Control in mRNA Content TIMP-2
6	ND	ND	ND
20	55 ± 13	76 ± 14	61 ± 18
48	77 ± 13	70 ± 24	72 ± 15
72	70 ± 19	NC	75 ± 27

 Data are mean ± SD of results in three experiments performed in triplicate. NC, no change from control; ND, not determined.

Figure 4.

 

Overexpression of MMP-14 and TIMP-2. Activity of MMP-14 (A) as measured with an enzyme activity kit (Biotrak; GE Healthcare) and TIMP-2 (B) as measured by reverse zymography was not significantly decreased after overexpression and injury (+ I). Data represent results in two individual experiments performed in duplicate.

Figure 5.

 

Restoration of MMP-2 activity requires both MMP-14 and TIMP-2. Representative zymogram (A) and graph depicting the mean ± SEM of duplicate experiments (B) of RPE cells transiently transfected with either a full-length transcript of MMP-14 (lanes 4–10) or TIMP-2 (lanes 11, 12). Cells were treated with decreasing concentrations of recombinant TIMP-2 (rTIMP-2, nM) and injured according to a standard protocol (lanes 6–10). Only those cells with MMP-14 and the appropriate concentration of TIMP-2 showed restored MMP-2 activity (lanes 9, 10). Lane 3: recombinant TIMP-2 treatment of cells without injury. C, control; I, injury. Data are expressed as arbitrary densitometry units and represent results in three individual experiments performed in duplicate.