Endothelin-1-Mediated Signaling in the Expression of Matrix Metalloproteinases and Tissue Inhibitors of Metalloproteinases in Astrocytes

Abstract

purpose. Endothelin (ET)-1 levels are increased in aqueous and vitreous humor in patients with glaucoma and animal models of glaucoma. Whether the elevated ET-1 induces extracellular matrix (ECM) remodeling in the optic nerve head is still unknown. In the present study, the regulation of matrix metalloproteinases/tissue inhibitors of matrix metalloproteinases (MMPs/TIMPs) and ECM remodeling in ET-1-activated human optic nerve head astrocytes (hONAs) were determined.

methods. Primary hONAs were exposed to ET-1 for 1 day and 4 days. Incubation media were subjected to zymography and Western blot to detect activity and expression of MMPs and TIMPs. Fibronectin (FN) was monitored by Western blot and immunofluorescent staining.

results. ET-1 increased the activity of MMP-2 and the expression of TIMP-1 and -2 in hONAs. The expression of TIMP-1 and -2 induced by ET-1 was abolished by application of inhibitors of mitogen-activated protein kinase (MAPK) or PKC, leading to enhanced activity of MMP-2. Knockdown of MMP-2, by using small interfering (si)RNA, not only decreased the activity of MMP-2 but also decreased the expression of TIMP-1 and -2. ET-1 increased the soluble (s)FN expression as well as FN matrix formation. However, the accumulation of sFN did not enhance FN matrix formation. Unlike ET-1’s effects on MMP-2, blockade of MAPK and PKC did not alter the expression and deposition pattern of FN in hONAs.

conclusions. ET-1 increased the expression and activity of MMP-2 and TIMP-1 and -2. The ERK-MAPK and PKC pathways are involved in the regulation of expression of MMP-2 and TIMP-1 and -2. ET-1’s effects on MMPs/TIMPs may be important, not only in regulating the expression of MMPs and TIMPs, but also in influencing ECM remodeling.

Glaucoma is an optic neuropathy, characterized by cupping of the optic disc, progressive loss of retinal ganglion cells, and slow degeneration of the optic nerve, ultimately resulting in blindness. ^{1 2} The optic nerve head (ONH) is the major pathologic site in glaucoma. It is reported that extensive remodeling of the extracellular matrix (ECM) occurs in the ONH in patients with glaucoma and in animal models of glaucoma. ^{3 4}  

Numerous proteolytic enzymes participate in ECM degradation and remodeling; however, one class of enzymes that appear to play a pivotal role are the matrix metalloproteinases (MMPs). ⁵ MMPs are a large family of zinc-dependent enzymes that are capable of degrading the constituents of the ECM. The activity of MMPs is tightly controlled by their tissue inhibitors of matrix metalloproteinases (TIMPs). ⁶ The balance of MMPs and TIMPs is necessary for normal physiological and pathologic conditions, including embryonic development, growth, and tissue remodeling and repair. ⁷  

It is reported that differential MMP and TIMP expression profiles are part of remodeling of the ONH in response to different damage and insults to the optic nerve. For example, in laser-induced glaucoma in primates, there was a significant increase in expression of MT1-MMP (membrane type) and MMP-1 in reactive astrocytes at the ONH due to elevation of intraocular pressure (IOP), whereas in a parallel experiment in primates with optic nerve transection, there was increased expression of MT1-MMP, MMP-1, MMP-2, TIMP-1, and TIMP-2 in reactive astrocytes at the transection site. ⁸ Other investigators have shown that extensive ECM remodeling correlates with RGC apoptosis and axon loss and that the increased MMP activity may enhance ECM degradation, including collagen degradation, to facilitate the migration of astrocytes into optic nerve bundles. ^{9 10 11} Therefore, key ECM molecules not only provide mechanical support for cells, but also are involved in cell apoptosis and axon loss at the ONH. Reactive astrocytes involved in astrogliosis may be responsible for ECM remodeling in glaucoma, and recently, endothelin (ET)-1 has been implicated as a potential factor in reactivation of astrocytes, which leads to astrogliosis. ^{12 13 14}  

ET-1 is not only known as a potent vasoactive peptide, but is also involved in ECM remodeling by shifting the balance of MMPs/TIMPs. ^{15 16 17} In addition, growing evidence from other tissues suggests that ET-1 induces the expression of many types of MMPs and TIMPs as well as ECM protein expression, including collagens, laminin, and fibronectin (FN), in cell culture and animal models. ^{18 19} It has been shown that ET-1 levels are elevated, not only in the aqueous humor and plasma of patients with glaucoma, ^{20 21} but also in glaucoma animal models, including dogs, ²² rats, ²³ and rabbits. ²⁴ Our laboratory has shown that the increased ET-1 levels in aqueous humor in rats with elevated IOP correlate with the increased GFAP expression in astrocytes and increased ET_B receptor immunoreactivity. ²³ Ahmed et al. ²⁵ have also demonstrated that there is a significant increase in retinal mRNA levels of the ET receptors TIMP-1, MMP-3, ET-2, FN, and GFAP, when tested by RT-PCR in the presence of elevated IOP in Brown Norway rats. Furthermore, intravitreal injection or perfusion of ET-1 into eyes in many animal models such as primates, rabbits, and rats, produced the ONH damage similar to that in glaucoma, including optic disc cupping, axon loss, astrogliosis. ^{26 27} Therefore, ET-1 may be involved in ECM remodeling of the ONH in glaucoma as a result of the astrogliosis and reactivation of astrocytes by ET-1; however, a detailed study of ET-1’s effects on optic nerve ECM, MMPs, and TIMPs has not been delineated. 

In the present study, we investigated ET-1-induced activity of MMPs and expression of TIMPs in primary human optic nerve astrocytes (hONAs). In addition, we determined whether ET-1-induced differential activation of MMPs/TIMPs occurs through mitogen-activated protein kinase (MAPK)– and protein kinase C (PKC)–dependent pathways. This imbalance of MMPs an d TIMPs could contribute to ECM remodeling and the pathologic changes in the ONH in glaucoma. 

Materials and Methods

Dulbecco’s modified Eagle’s medium (DMEM, catalog no. 11995-040), and penicillin, streptomycin, and glutamine were obtained from Invitrogen-Gibco (Rockville, MD). Fetal bovine serum (FBS) was obtained from Hyclone Laboratories, Inc. (Logan, UT). PD98059, U0126, chelerythrine, and RO-31-8425 were purchased from Calbiochem (La Jolla, CA). Rabbit anti-ERK1/2 polyclonal antibody, rabbit anti-phospho-ERK1/2 (Thr202/Tyr 204) polyclonal antibody, rabbit anti-phospho-pan-PKC polyclonal antibody, and rabbit anti-phospho-PKC-ζ polyclonal antibody were purchased from Cell Signaling Technology (Beverly, MA); antibody against TIMP-1 (monoclonal, from mouse) from Chemicon, Inc. (Temecula, CA); antibody against TIMP-2 (polyclonal, from rabbit) from Santa Cruz Biotechnology (Santa Cruz, CA); and ET-1 from Peninsula Laboratories (Belmont, CA). All other chemicals were purchased from Sigma-Aldrich (St. Louis, MO). 

Cell Cultures

Primary hONA lines (kindly given by Abbot F. Clark, Alcon Laboratories, Fort Worth, TX) from normal donors (ages, 53, 62, and 66 years) were maintained in DMEM containing 10% fetal bovine serum supplemented with 100 U/mL penicillin, 100 μg/mL streptomycin, and 0.3 μg/mL glutamine under a humidified 5% CO₂ at 37°C. 

Western Blot Analysis

The hONAs were cultured in 100-mm dishes to confluence, and 24 hours later the media were changed to serum-free DMEM, and the cells were pretreated with different inhibitors for 30 minutes. The cells were then stimulated with 100 nM ET-1 for the various times described herein. The reaction was stopped by adding ice-cold PBS. The cells were scraped and placed in lysis buffer. Total cell lysate or concentrated supernatant of cell culture was applied to 10% SDS-PAGE with an equal amount of total proteins and transferred to nitrocellulose membranes (Protran Bioscience, Keene, NH). The transferred membranes were blocked with 5% nonfat milk in TBS/T (Tris-buffered saline) for 1 hour and incubated with primary antibody for 1 hour at room temperature or overnight at 4°C. Horseradish peroxidase–conjugated anti-mouse or anti-rabbit IgG antibodies (GE Healthcare, Piscataway, NJ) was used as the secondary antibody and chemiluminescence reagent (GE Healthcare) was used for blot analysis. x-Ray films (Eastman-Kodak, Rochester, NY) were exposed and scanned (Hewlett-Packard, Palo Alto, CA). 

Gelatin and Casein Zymography

In general, MMP-2 and -9 were analyzed on 7.5% polyacrylamide gels containing 2 mg/mL gelatin(Sigma-Aldrich), and MMP-3 on 7.5% polyacrylamide gels containing 2 mg/mL β-casein (Sigma-Aldrich). Incubation medium samples for analysis by zymography were concentrated by filtering membranes (Microcon YM-10; Millipore, Bedford, MA) and adjusted to equal volume. Thirty-five microliters of concentrated medium was applied to 7.5% SDS-polyacrylamide gels containing 2 mg/mL gelatin or casein. After electrophoresis, SDS was removed from the gels by washing for 3 hours in 2.5% Triton X-100 solution (50 mM Tris [pH 7.6], 1 μM ZnCl₂, and 5 μM CaCl₂), followed by washing for 16 hours in 1% Triton X-100 solution at 37°C. After incubation, the gel was stained with a solution of 0.25% Coomassie blue R250, 40% methanol, and 10% acetic acid for 1 hour at room temperature and destained with 40% methanol and 10% acetic acid until the white bands became clear. Positive media (C1) containing MMP-2, -9, and -3 and TIMP-1 and -2 were purchased from Triple Point Biologics, Inc. (Forest Grove, OR) and used in zymography. 

Immunocytofluorescent Staining

Cells were fixed in 4% paraformaldehyde in PBS for 30 minutes at room temperature, followed by permeabilization with 0.1% Triton X-100 for 15 minutes. The cells were rinsed in PBS and incubated twice in 200 mM glycine, 15 minutes per incubation. Each coverslip was carefully inverted (cell-side facing solution) onto 200 μL of blocking solution containing 5% BSA+5% normal goat serum in PBS for 30 minutes. The coverslips were then incubated with mouse anti-FN at room temperature for 2 hours. Coverslips were rinsed and allowed to incubate with Alexa 488–conjugated donkey anti-rabbit (2 μg/mL) for 1 hour in the dark at room temperature. Nuclei were stained with 4′,6′-diamino-2-phenylindole (DAPI; 300 nM; Invitrogen-Molecular Probes, Eugene, OR) for 10 minutes. Coverslips were mounted on glass slides in antifade medium (FluorSave; Calbiochem, La Jolla, CA) and allowed to dry for 20 minutes in the dark. The cells were then viewed and images taken with a confocal microscope (LSM 410; Carl Zeiss Meditec, Dublin, CA). 

siRNA and Transfection of siRNA

We used vector-based siRNA techniques to suppress MMP-2 and TIMP-2 expression (pGeneClip U1 Hairpin hMGFP; Vector, Promega, Inc., Madison, WI) The design of siRNAs was based on the assistance provided by Promega, Inc. The sequences were: MMP-2: 5′-TCTCGAAGATGCAGAAGTTCTTTAAGTTCTCTAAAGAACTTCTGCATCTTCCT3′; 5′-CTGCAGGAAGATGCAGAAGTTCTTTAGAGAACTTAAAGAACTTCTGCATCTTC-3′; and TIMP-2: 5′-TCTCGACTCTGGAAACGACATTTAAGTTCTCTAAATGTCGTTTCCAGAGTCCT-3′, 5′-CTGCAGGACTCTGGAAACGACATTTAGAGAACTTAAATGTCGTTTCCAGAGTC-3′. 

The fragment was inserted into the vector and identified by enzyme digestion and sequencing. The scrambled siRNA was also constructed as a control for our experiments. 

For plasmid cDNA transfection, cells were freshly seeded to 80% confluence 1 day before transfection, and serum-free medium was used without penicillin and streptomycin. For each well in a six-well plate, 4 μg plasmid cDNA was used for transfection with 10 μL transfection reagent (Lipofectamine 2000; Invitrogen Inc., Carlsbad, CA), according to the manufacturer’s instructions. Eight hours after transfection, the cells were washed with serum-free DMEM and subjected to a second transfection. For the second transfection of hONAs, 4 μg plasmid cDNA was used for transfection with 4 μL of the reagents. After 8 hours, the cells were washed with serum-free DMEM and incubated for another 24 hours before treatment. 

Statistical Analysis

The density of the bands in Western blot and zymography was analyzed by NIH Image J software (available by ftp at zippy.nimh.nih.gov/ or at http://rsb.info.nih.gov/nih-image; developed by Wayne Rasband, National Institutes of Health, Bethesda, MD). For immunocytostaining, five regions (top left, top right, center, bottom left, and bottom right) in each image was quantitated using the Image J software. Images were obtained at the same focal plane, to minimize the variation in fluorescence. Data obtained by Image J are represented as the mean ± SD. Data were analyzed with one-way analysis of variance (ANOVA) for multiple comparisons. Statistical significance was set at P < 0.05. 

Results

Effect of ET-1 on the Active Form of MMP-2 and Expression of TIMP-1 and -2

Our experience has shown that primary hONAs grow slowly in cell culture. ¹³ Therefore, we selected 24 to 96 hours of ET-1 treatment for time course experiments. The concentrated incubation media were collected from the hONA culture and subjected to gelatin and casein zymography. There was an increased level of total MMP-2 activity when hONAs were treated with ET-1 during the longer treatment of 48 to 96 hours; however, maximum levels of the active form of MMP-2 was increased to a greater extent than total MMP-2 (Fig. 1A) . Similarly, an upregulation of TIMP-1 and -2 was found with ET-1 treatment from 72 to 96 hours. The MMP-3 level was very low in our experiments, but was also upregulated after the longer treatment time (data not shown). The expression of TIMP-2 was not detectable by Western blot at the first day of ET-1 treatment. Therefore, for hONAs, a 4-day (96-hour) treatment was used for further experiments with ET-1, unless otherwise noted. The results suggest that ET-1 increases not only the total activity of MMP-2, but also the conversion of active MMP-2 from the latent form. 

ET-1’s effects on the expression of MMPs and TIMPs were tested at different concentrations from 1 to 1000 nM (Fig. 1B) . The results showed that at ET-1 concentrations from 100 to 1000 nM the highest level of expression of MMP-2 and TIMP-1 and -2 and the highest level of active MMP-2 were achieved, which is consistent with our previous reports. ¹³  

Effect of ET-1 on Phosphorylation of ERK1/2 and PKC-βI/βII and δ in hONAs

Because ET-1 was shown to increase activity of MMP-2 and expression of TIMP-1 and -2 in hONAs, the signaling pathways activated by ET-1 including phosphorylation of ERK1/2 and activation of PKC isoforms were identified. A significant increase in phosphorylated ERK1/2 was detected in hONAs treated with 100 nM ET-1 for 5 minutes (Fig. 2A) . ET-1-induced phosphorylation of ERK1/2 and basal phosphorylation levels of ERK1/2 were completely abolished by the application of U0126 (an inhibitor of MEK1/2 which are the upstream kinases of ERK1/2), whereas phosphorylation was not affected by application of chelerythrine (a general inhibitor of PKC isoforms; Fig. 2B ). For further assessment of PKC activation by ET-1, isolated membrane fractions of hONAs were probed with antibodies against several phosphorylated isoforms of PKCs. There was an increase in phosphorylation of PKC βI/βII and δ in hONAs treated with ET-1 for 5 and 30 minutes, whereas no such phosphorylation was seen in PKC-α, -ε, and -ζ (Fig. 2B) . Pretreatment with chelerythrine decreased the basal phosphorylated level of all PKC isoforms in hONAs. 

Blockade of MAPK-ERK or PKC on the Activity of MMP-2 and the Expression of TIMP-1 and -2

When hONAs were pretreated with U0126 (10 μM) and PD98059 (25 μM) (inhibitors of MEK1/2), the activity of MMP-2, especially the active form of MMP-2, was even further enhanced than that seen with ET-1 alone (Fig. 3A) . However, the expression of TIMP-1 and -2 was decreased when inhibitors were applied (Fig. 3A) . The application of U0126 to hONAs inhibited the basal expression of TIMP-1 and -2. A very low level of MMP-3 was seen in casein zymography, and there was also increased activity of MMP-3 in inhibitor-treated groups (data not shown). 

As described earlier, increased activity of total MMP-2 was detected with blockade of PKC signaling by chelerythrine (2 μM) or RO-31-8425 (1 μM; PKC inhibitors), although the active MMP-2 was increased much more than total MMP-2 (Fig. 3B) . In addition, chelerythrine decreased the expression of TIMP-1 and -2 (Fig. 3B) . 

The increased activity of MMP-2 and -3 may be caused by the decreased expression of TIMP-1 and -2, which are endogenous inhibitors of MMP-2 and -3. Therefore, it is likely that the balance of MMPs and TIMPs was altered. 

Knockdown of MMP-2 on Expression of TIMP-1 and -2

To evaluate the regulation of expression of MMPs and TIMPs and the relationship of the sensitive balance in MMPs and TIMPs to substrate degradation, siRNA was used to knock down MMP-2 and TIMP-2. There was a decrease in MMP-2 activity after treatment with either MMP-2 siRNA or TIMP-2 siRNA. The decreased expression of TIMP-1 and -2 was also observed in hONAs transfected with MMP-2 siRNA (Fig. 4) . These results are in agreement with the preliminary experiments in U373MG astrocytoma cells, which were transfected with siRNA constructs and used to test siRNA design and knockdown efficiency. The higher knockdown level of MMP-2 and TIMP-2 that was observed in U373MG cells was probably due to the higher transfection efficiency in this cell line. Similarly, the decreased TIMP-1 and -2 was found in treatment with MMP-2 siRNA in U373MG cells (data not shown). The introduction of TIMP-2 siRNA lowered the activity of MMP-2 in hONAs. Alteration in the expression of either MMPs or TIMPs appears to affect each other, leading to MMPs and TIMPs’ reaching a new ratio. 

ET-1 and sFN in Cultured hONAs

Typically, two forms of FN are present when the newly synthesized dimer of FN is soluble and the polymerized FN forming an fibrillar matrix network is insoluble. ^{28 29} In the present study, both forms of FN were investigated after ET-1 treatment. The soluble (s)FN was monitored by Western blot analysis of culture media from hONAs and insoluble network FN by immunocytostaining. The culture incubation media were collected on days 1 to 4 and concentrated by filters (YN-10; Millipore) after ET-1 administration, with or without pretreatment of U0126 (10 μM) or chelerythrine (2 μM) for 30 minutes. In dose-dependency experiments, the highest expression of sFN was observed with 100 and 1000 nM of ET-1 after a 4-day treatment of hONAs (Fig. 5A) . The result was consistent with ET-1’s effects on MMP-2 and TIMP-1/2 expression, which reached the highest levels with 100 nM ET-1 after a 4-day treatment (Fig. 1) . In a time course experiment, sFN accumulated with increasing time with ET-1 treatment and reached the highest level on day 4 (Fig. 5B) . 

To investigate the effects of MAPK-ERK and PKC on sFN expression, hONAs were pretreated with either U0126 (10 μM) or chelerythrine (2 μM). The application of inhibitors did not attenuate ET-1-induced sFN expression nor the basal level of sFN (Fig. 5C) , suggesting that the expression of FN is not under the control of MAPK-ERK and PKC pathways. ET-1 induced sFN secretion, but it was reduced after treatments with siRNA for MMP-2 but not with siRNA for TIMP-2 (Fig. 5D) . 

ET-1 and Insoluble FN in hONAs

The experiments described thus far show that there was a significant increase of sFN in hONAs treated with ET-1; therefore, we wanted to investigate ET-1’s effects on insoluble FN that typically forms a matrix network with other matrix proteins. ET-1-treated hONAs, with or without U0126 (10 μM) and chelerythrine (2 μM), were used to probe FN by immunofluorescent staining. The immunoreactivity of FN was assessed in hONAs treated with ET-1 for 1 and 4 days (Fig. 6A) . The immunoreactivity in 1-day ET-1 treatment was similar to that with the 4-day ET-1 treatment. There was a significant increase in sFN in the 4-day ET-1 treatment, but it seemed that not all the increased sFN seen after 4 days of ET-1 treatment participated in the formation of the insoluble FN network. 

To investigate the regulatory effects of MAPK and PKC pathways on insoluble FN deposition, we performed the same immunostaining of FN in hONAs in the presence of U0126 and chelerythrine (Fig. 6B) . The application of either U0126 or chelerythrine did not attenuate ET-1-induced FN deposition in matrix network and this was in agreement with the results shown in Figure 6C , in that the application of both inhibitors did not attenuate ET-1-induced sFN expression. 

Discussion

In the regulation of ECM components in the brain and eyes, astrocytes are a major source of MMPs and provide ECM components. ³⁰ Reactive astrocytes involved in astrogliosis may be responsible for ECM remodeling in glaucoma. ³ ET-1 has been shown to be present in astrocytes in brain ³¹ and eye. ³² In the present study, ET-1 not only upregulated the activity of active MMP-2 and the expression of TIMP-1 and -2 in hONAs, but also increased the deposition of FN and formation of an ECM network in an hONA culture model. ET-1-mediated regulation of ECM in hONAs appears to be involved in two signaling pathways, ERK-1/2 and PKC, since the blockade of these pathways increased the activity of MMP-2, but inhibited the expression of TIMP-1 and -2. Therefore, the balance of MMP/TIMP was shifted toward increased MMP production and activity (Fig. 7) . 

ET-1 has been considered an important factor that may contribute to the formation and progression of glaucoma. ^{33 34} However, it remains unclear whether ET-1 exerts any effect on the ECM remodeling seen in the ONH during progressive glaucoma. Our results showed that ET-1 upregulated the activity of the active form of MMP-2 and -3 as well as the expression of TIMP-1 and -2 in either a dose- or time-dependent manner, suggesting that these cells reach a new balance of MMPs/TIMPs in response to ET-1. In addition, the rapid increase of MMP-2 and -3 was observed as early as 24 hours after ET-1 treatment in hONAs, whereas upregulated TIMP-1 and -2 was detected only between 72 to 96 hours, indicating that the upregulation of MMPs was the early and immediate response to ET-1 followed by the late increased TIMP-1 and -2 to counteract MMPs and thereby reaching a new balance. Another target in this study was MMP-9, which was considered a regulatory enzyme compared with MMP-2. Monitoring the changes in MMP-9 could give us more information on ET-1’s effects in ECM remodeling, but the basal level of MMP-9 in hONAs was too low to be investigated in the present study. In an elevated IOP model of glaucoma in rats, there were increased levels of ET-1 ²³ and ET_B receptor ²⁵ and upregulation of MMP-9, TIMP-1, and collagen I ¹⁰ in retina and collagen IV and VI at the ONH, ¹¹ whereas there was a downregulation of laminin at the RGC layer. ¹⁰ In addition, apoptosis of RGCs correlated significantly with the upregulation of MMP-9 and reduced laminin staining in rats after the intravitreal injection of the cytotoxic agent, kainic acid. The death of RGCs was rescued by application of the MMP inhibitor GM6001. ³⁵ Such observations suggest that ECM remodeling contributes to the deleterious pathologic changes in the retina and ONH. It is important to distinguish that although MMP-9 may be relevant to RGC apoptosis, activation of astrocytes may be dependent on MMP-2 and TIMP 1/2 activation. It is presently unknown whether ET-1-induced RGC apoptosis also involves upregulation of MMP-9. 

Increased activity of MMP-2 and -3 and the attenuated expression of TIMP-1 and -2 were observed in media collected from hONAs, in which MAPK-ERK and PKC pathways were blocked by application of several different inhibitors. This is the first report stating that the blockade of MAPK or PKC pathways can result in increased activity of MMP-2 and 3 in hONAs. Several reports have shown that these two pathways control the expression and activity of MMPs. Blocking one or both of them decreases the level of MMPs. ^{36 37 38} For instance, phorbol ester (12-myristate 13-acetate [PMA])–induced MMP-9 was inhibited by curcumin through MAPK and PKC pathways in human astroglioma cells, ³⁶ and the expression and activity of MMP-2 and -9 induced by IL-1 were regulated by differential PKC isoforms via JNK or ERK1/2 MAPK pathways in cardiac fibroblasts. ³⁹ In our study, pretreatment of hONAs with MAPK-ERK and PKC inhibitors decreased TIMP-1 and -2, endogenous inhibitors of MMPs, that resulted in an imbalance between MMPs and TIMPs, since TIMPs bind tightly with MMPs to inhibit the enzymatic activity at 1:1 molar ratio. Consequently, the activity of MMPs was increased in hONAs, as shown in our results. 

In this study, ET-1 increased the sFN in the incubation media of cell cultures in a time-dependent pattern, whereas matrix-formed FN increased to a maximum at day 1, and the level was maintained on day 4 in hONAs (Figs. 5B 6A) . These findings suggest that cells may use a certain amount of FN to form a network to maintain their physiological state and avoid forming the excess network that may prevent cells from migrating when necessary (i.e., during development, injury, or neurotrauma). Cells need balance in the ECM, in that they not only must have ECM formation for mechanical support and signal transduction, but also they must control the amount of FN matrix and limit the extent of the matrix for physiological functions, including cell migration. Evidence exists for ET-1-induced cell migration in many types of cells, including human ovarian carcinoma cell lines ⁴⁰ and neural stem cells. ⁴¹ Therefore, matrix formation could be regulated to a certain extent to facilitate cell migration. In the present study, hONA migration in response to ET-1 was not assessed but is the subject of ongoing studies. Although excess sFN was secreted by hONAs, it did not form an insoluble matrix, perhaps due to increased MMP-2 production. Therefore, substrate synthesis and degradation are also associated in this system. Furthermore, FN is necessary to bind with its receptor integrin, to induce or mediate signal transduction. Preventing FN binding with integrin by application of an antibody against FN inhibits outgrowth of neurites in dorsal root ganglion neurons. ⁴² The defective scaffold formation and a failure of normal vascular development in the retina were also observed in mice null for orphan nuclear receptor-tailless, due to impaired formation of FN matrix. ⁴³ This finding suggests that FN mediates a wide variety of cellular interactions with the ECM and plays important roles in cell adhesion, migration, growth, and differentiation. ⁴⁴ In addition to being an important constituent of ECM and structural support, FN exerts its diverse biological functions by binding and interacting with integrins, heparin, collagen-gelatin, elastin, and fibrin. These interactions are also important for communication between cells. ²⁸ Thus, altering the balance of sFN to FN matrix could affect these cells significantly, leading to communication dysregulation. 

In summary, ET-1 increased the activity of MMP-2 and the expression of TIMP-1 and -2 as well as the expression of sFN and FN matrix formation in hONAs. In addition, the ERK-MAPK and PKC pathways appear to be involved in the regulation of MMP-2 activity and TIMP-1 and -2 expression, but not in ET-1-mediated FN expression. A balance of MMPs/TIMPs may be important, not only in regulating the expression of MMPs and TIMPs but also in influencing ECM remodeling. Our current data show that ET-1 shifted the balance of MMP activity and substrates in a temporal fashion, and that this may lead to ECM remodeling in activated hONAs. Astrogliosis occurs after neurotrauma, hypoxia-ischemia, and other diseases and is manifested by a dramatic change in the expression of ECM profile, which in most cases results in a glial scar. ³⁰ Migration of the activated astroglia into optic nerve bundles may induce axon loss. ^{3 45} Because the axon loss is one of the characteristics of glaucoma, the ONH is considered to be an important site for glaucomatous pathologic changes. The putative effect of elevated IOP is thought to induce pathologic changes at the ONH that causes physical compression on optic nerve axons. Such actions would lead to ischemia of the optic nerve and also block axoplasmic flow, specifically anterograde and retrograde transport between axons and retinal ganglion cell bodies. ^{3 30} ET-1 has been shown to affect axonal transport in the optic nerve. ⁴⁶ Therefore, the present study provides additional insight into ECM remodeling, another critical component of the ET–glial–axon interactions that may be involved in the genesis of glaucoma as well as that of other neuropathies. 

²  Current affiliation: Ocular Biology, Pfizer Global R & D, La Jolla, California.

 Supported by National Eye Institute Grant EY11979 (TY) and a Texas Higher Education Coordination Board Advanced Technology Grant (TY).

 Submitted for publication September 21, 2006; revised March 10, 2007; accepted May 22, 2007.

 Disclosure: S. He, None; G. Prasanna, None; T. Yorio, 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.

 

ET-1 increased the active form of MMP-2, TIMP-1, and TIMP-2. (A) In time-course studies, gelatin zymography was used to detect the activity and expression of MMP-2 in incubation media collected from different time points after 100 nM ET-1 treatment. Western blot analysis was use to detect the expression of TIMP-1 and -2 in incubation media collected from different time points after 100 nM ET-1 treatment. PMA (1 μM) treatment was used as the positive control. There was greater activity of MMP-2 and expression of TIMP-1 and -2 with a longer treatment with ET-1. (B) Dose–response studies: gelatin zymography was used to detect the activity of MMP-2 in incubation media collected from hONAs treated with different concentration of ET-1 for 4 days. Western blot was used to detect the activity of TIMP-1 and -2 in incubation media collected from hONAs treated with different concentration of ET-1 for 4 days. The highest activity of MMP-2 and expression of TIMP-1 and -2 were observed in treatment with 100 nM ET-1. MMP-3 was not detectable in Western blot. *P < 0.05, ET-1 treatment versus the corresponding control; one-way ANOVA. The data are from a representative of three individual experiments that had consistent results.

Figure 2.

 

ET-1 induced the phosphorylation of ERK1/2, PKCβI/βII/δ, but not PKCε/α/ζ in hONAs. (A) Western blot (WB) showed phosphorylation of ERK1/2 induced by 5-minute ET-1 (100 nM) treatment in hONAs after cells were pretreated with U0126 (10 μM), chelerythrine (2 μM, CHELE), or dimethyl sulfoxide (vehicle) for 30 minutes. ET-1 induced the rapid phosphorylation of ERK1/2, which was abolished by U0126, but not chelerythrine. (B) The phosphorylated isoforms of PKC were detected by Western blot in hONAs, which were treated with ET-1 (100 nM) for 5 and 30 minutes. One set was pretreated with chelerythrine (2 μM) for 30 minutes before ET-1 treatment. Phosphorylation of PKCβI/βII/δ was induced by ET-1 in hONAs. *P < 0.05, ET-1 treatment versus corresponding control, one-way ANOVA. #P < 0.05, inhibitor treatments versus vehicle control, one-way ANOVA. The data are representative of two individual experiments with consistent results.

Figure 3.

 

Blockade of MAPK-ERK and PKC increased the activity of MMP-2 and -3 and decreased the expression of TIMP-1 and -2. (A) Gelatin zymography was used to detect the activity of MMP-2 and -9 in incubation media collected from hONAs, which were pretreated with 25 μM PD98059 and 10 μM U0126 (both are MAPK-ERK inhibitors) for 30 minutes followed with or without a 100-nM ET-1 treatment for 4 days. Western blot was used to detect the expression of TIMP-1 and -2 in incubation media collected from hONAs. Application of inhibitors of MAPK reduced the expression of TIMP-1 and -2, but increased the activity of MMP-2. (B) Gelatin zymography was used to detect the activity of MMP-2 and -9 in incubation medium collected from hONAs, which were pretreated with 2 μM chelerythrine (CHE) and 1 μM RO-31-8425 (RO) for 30 minutes, with or without a 100-nM ET-1 treatment for 4 days. Western blot analysis was used to detect the expression of TIMP-1 and -2 in incubation media collected from hONAs. Application of inhibitors of PKC reduced the expression of TIMP-1 and -2, but increased the activity of MMP-2. *P < 0.05, ET-1 treatment versus corresponding control, one-way ANOVA. #P < 0.05, inhibitor treatments versus vehicle control, one-way ANOVA. The data are representative of three to four individual experiments with consistent results.

Figure 4.

 

The activity MMPs and the expression TIMPs decreased in hONAs transfected with siRNA of MMP-2. Gelatin zymography was used to detect the activity of MMP-2 and Western blot to detect expression of TIMPs in incubation media collected from hONAs, which were transfected with siRNA of MMP-2 48 hours before treatment with 100 nM ET-1. The incubation media were collected 24 hours after cells were treated with ET-1. Knockdown of MMP-2 by using siRNA not only decreased the activity of MMP-2 but also decreased the expression of TIMP-1 and -2 (SCR: scrambled siRNA). *P < 0.05, ET-1 treatment versus corresponding control, one-way ANOVA. The data are representative of two individual experiments with consistent results.

Figure 5.

 

ET-1 increased sFN in cell culture of hONAs. Western blot was used to detect the sFN in concentrated media collected from hONA culture. (A) hONAs were treated with ET-1 from 1 to 1000 nM for 4 days. Expression of sFN increased in treatment with ET-1 from 10 to 1000 nM. (B) hONAs were treated with 100 nM ET-1 for different periods from 1 day to 4 days. Expression of sFN increased with the longer treatment time. (C) hONAs were pretreated with 10 μM U0126 or 2 μM chelerythrine (CHE) for 30 minutes followed by ET-1 treatment for 24 hours. Inhibition of MAPK or PKC pathways did not reduce the expression of sFN. (D) hONAs were transfected with siRNA two times for 48 hours followed by ET-1 treatment for 24 hours. Knockdown of MMP-2 decreased the expression of sFN (SCR: scrambled siRNA). *P < 0.05, ET-1 treatment versus corresponding control, one-way ANOVA. The data are representative of two to three individual experiments with consistent results.

Figure 6.

 

ET-1 increased the deposition of FN in hONAs. (A) Immunofluorescent (IF) staining was used to detect the deposition of FN in hONAs, which were treated with 100 nM ET-1 for 1 to 4 days (green: FN; blue: DAPI). Treatment of ET-1 increased the FN matrix, which did not accumulate like sFN. (B) IF showed the deposition of FN in hONAs, which were pretreated with 10 μM U0126 or 2 μM chelerythrine (CHELE) for 30 minutes followed by 100 nM ET-1 treatment for 24 hours (green: FN; blue: DAPI). Application of inhibitors of MAPK or PKC did not affect the deposition of FN matrix. *P < 0.05, ET-1 treatment versus corresponding control, one-way ANOVA. The data are representative of three individual experiments with consistent results.

Figure 7.

 

ET-1 increased the deposition of FN in hONAs. ET-1 activated the rapid phosphorylation of ERK1/2 and PKC βI/βII/δ, which play important roles in cell proliferation and reactivation of hONAs. ET-1 not only upregulated the activity of MMP-2 and the expression of TIMP-1 and -2 in hONAs, but also increased the deposition of FN and formation of an ECM network in an hONA culture model in a time-dependent manner. PKC and MAPK pathways are not involved in the expression and deposition of FN. However, they are involved in ET-1-mediated regulation of ECM remodeling in hONAs, since the blockade of ERK-1/2 or PKC inhibited the expression of TIMP-1 and -2, resulting in an increased level of activity of MMP-2. Therefore, the balance of MMP/TIMP shifted toward increased MMP activity.