Intravitreous Injection of a Membrane Depolarization Agent Causes Retinal Degeneration Via Matrix Metalloproteinase-9

Abstract

purpose. Membrane depolarization and subsequent synaptic release of l-glutamate have been implicated in ischemic retinal damage. However, the mechanisms that lead to ischemia-induced retinal damage are poorly understood. In this study, KCl, a classic membrane depolarizing agent, was injected into the vitreous humor, and the role of matrix metalloproteinase (MMP)-9 in KCl-induced retinal damage was investigated.

methods. Normal adult CD-1 mice were treated with KCl by intravitreal injection. MMP activity in retinal protein extracts was determined by gelatin zymography. Tissue localization of MMP-9 in the retina was determined by immunohistochemistry. MMP-9, MMP-2, tissue inhibitor of MMP (TIMP)-1, TIMP-2, Bax, and BCl-2 proteins in retinal extracts were determined by Western blot analysis. Apoptotic cell death in the retina was determined by TUNEL assays. Retinal damage was assessed by immunolocalization studies with antibodies against neurofilament-light (NF-L) and calretinin.

results. Depolarizing concentrations of KCl induced a dose- and time-related upregulation in MMP-9 activity and protein in the retina. KCl-mediated MMP-9 upregulation was associated with an increase in proapoptotic protein Bax and apoptotic death of cells in the ganglion cell (GCL) and inner nuclear layer (INL), and subsequent loss of NF-L-positive ganglion cells and calretinin-positive amacrine cells. Intravitreal injection of KCl along with an N-methyl-d-aspartate (NMDA)-type glutamate receptor antagonist, MK-801, and a non-NMDA-type glutamate receptor antagonist, NBQX, resulted in a reduction in KCl-mediated MMP-9 upregulation in the retina. Furthermore, a synthetic MMP inhibitor inhibited KCl-mediated MMP-9 upregulation, which led to a significant attenuation of KCl-induced retinal damage.

conclusions. These results suggest that upregulation of MMP-9, in part, plays a causative role in KCl-induced retinal damage.

Several previous studies have implicated ischemia, in part, as playing a degenerative role in human blinding retinal diseases such as glaucoma and diabetic retinopathy. ^{1 2 3 4 5 6 7} Although the mechanisms that underlie ischemia-induced retinal damage are less clear, the wealth of information accumulated so far suggests that ischemic damage starts with a reduction in cellular adenosine triphosphate (ATP) levels that in turn result in rapid failure of the Na/K-adenosine triphosphatase (ATPase) pumps and subsequent depolarization of the cell membrane. ^{8 9 10} Membrane depolarization then leads to two additional responses in affected cells: a large increase in calcium influx through voltage-gated channels and the release of the physiological neurotransmitter l-glutamate into the extracellular space. ^{10 11 12} Accumulation of extracellular glutamate may then overstimulate (excitotoxicity) NMDA and kainate-type ionotropic glutamate receptors and lead to retinal damage, ^{13 14} due to increased calcium influx. ¹⁵ However, the precise mechanisms involved in ischemia-induced retinal degeneration are poorly understood. 

Results from this laboratory have demonstrated that an extracellular matrix (ECM)–modulating protease, matrix metalloproteinase (MMP)-9, mediates ischemia- and excitotoxicity-mediated retinal damage. ^{16 17 18} Matrix metalloproteinases (MMPs) are a family of proteases that play an important role in modulation or degradation of the ECM. So far more than 28 members of the MMP family have been identified. ^{19 20 21} MMPs play a role not only in ECM remodeling but also in cell migration, invasion, ^{22 23 24} angiogenesis, ²⁵ and wound healing. ^{26 27 28 29} MMP activity is controlled at multiple levels, such as transcriptional regulation, proenzyme activation, and inhibition by tissue inhibitors of metalloproteinases (TIMPs). ^{22 23 30 31 32} Although MMPs are essential for physiological remodeling, excessive levels of these proteases can modulate or degrade ECM ^{22 33 34 35 36} and contribute to cell death. In this regard, mounting evidence indicates that MMPs play a degenerative role in the central nervous system (CNS) ³⁷ and optic nerve. ^{38 39} In addition, recent studies have reported that MMP-9 alone causes apoptosis of neuronal cells in vitro. ⁴⁰  

Despite previous studies from this laboratory that showed that ischemia-mediated MMP-9 contributes to retinal damage, the mechanisms that lead to upregulation of MMP-9 under ischemic conditions are poorly understood. Because depolarization seems to be one of the earliest events during ischemic retinal damage, a hypothesis was put forward that depolarization might be one of the mechanisms that leads to an upregulation in MMP-9 in the retina and that MMP-9 in turn promotes retinal damage. Therefore, three specific questions were addressed in this study: Does a membrane depolarizing agent, KCl (potassium chloride), cause retinal damage? Does an upregulation in MMP-9, in part, play a role in KCl-induced retinal damage? Is MMP-9 upregulation mediated by glutamate receptors? These questions were investigated in an animal model in which a classic membrane depolarizing agent, KCl, was injected into the vitreous humor of normal adult CD-1 mice. 

Methods

Intravitreal Injections

All the experiments on mice were performed under general anesthesia according to institutional protocol guidelines and the ARVO Statement for the Use of Animals in Ophthalmology and Vision Research. Normal adult CD-1 mice (8–10 weeks old; Charles River Breeding Laboratories, Wilmington, MA) were anesthetized by an intraperitoneal injection of 1.25% Avertin (2,2,2-tribromoethanol; in tert-amyl alcohol; 17 μL/g body weight). One of two membrane-depolarizing agents, KCl ^{41 42 43} or tetraethyl ammonium chloride (TEA-Cl), ⁴¹ was injected into the vitreous humor in CD-1 mice, according to previously described methods. ^{44 45} These previously published procedures describe, in general, an intravitreal injection technique, and they are not specific to induction of depolarization in the retina. In control experiments, eyes (n = 6) were injected with 2.5 μL of 0.1 M phosphate-buffered saline (PBS; pH 7.4) alone; and, in treatment groups, eyes (n = 6) were injected with 2.5 μL of 8, 80, and 160 mM KCl (corresponding to 20, 200, and 400 nmoles final concentration), 2.5 μL of 160 mM TEA (corresponding to 400 nmoles), 2.5 μL of 160 mM NaCl (corresponding to 400 nmoles) prepared in PBS. In separate sets of experiments, eyes (n = 6) were injected with 2.5 μL of 160 mM KCl (corresponding to 400 nmoles) plus 100 and 200 mM NBQX (corresponding to 100 and 200 nmoles; Tocris, Ellisville, MO), or KCl plus 100 and 200 mM (+) MK801 maleate (corresponding to 200 and 400 nmoles; Tocris). 

Protein Extraction

At 6 hours, 12 hours, 1 day, and 2 days after intravitreal injection, eyes were enucleated from animals anesthetized with an overdose of Avertin. Enucleated eyes were cut in half at the equator, and the lenses were removed. Retinas were carefully peeled off with forceps and washed three times with PBS (pH 7.4) to remove vitreous that may have adhered to the retina. Three to four retinas each were placed in tubes (Eppendorf, Fremont, CA) containing 40 μL extraction buffer (1% nonidet-P40, 20 mM Tris-HCl, 150 mM NaCl, and 1 mM Na₃VO₄ [pH 7.4]) and the tissues were homogenized. Tissue homogenates were centrifuged at 10,000 rpm for 5 minutes at 4°C, and the supernatants were collected. The protein concentration in the supernatants was determined with a commercial assay (Bio-Rad Laboratories, Hercules, CA). 

Gelatin Zymography

MMP activity was determined by gelatin-zymography assays, according to methods described previously. ^{16 17 45} Briefly, aliquots containing equal amounts of retinal protein extracts (25 μg) were mixed with SDS gel-loading buffer and loaded without heating onto 10% SDS polyacrylamide gels containing 0.2% gelatin as an MMP substrate. After electrophoresis, the gels were washed three times with 2.5% Triton X-100 (15 minutes each time), placed in activation buffer containing 10 mM CaCl₂ (pH 7.4), and incubated overnight at 37°C to allow proteolysis of the substrates in the gels. The gels were stained with 0.1% Coomassie brilliant blue-R250 and then destained with a solution containing 25% methanol and 10% acetic acid. Samples containing a mixture of murine MMP-9 and -2 were co-electrophoresed for comparison (data not shown). A reduced molecular weight size standard was also included on all gels (data not shown; Life Technologies, Gaithersburg, MD). 

Immunohistochemistry

At the indicated times after intravitreal injections (6 hours, 12 hours, 1 day, and 2 days), eyes were enucleated, fixed with 4% paraformaldehyde for 1 hour at room temperature, and embedded in optimum cutting temperature (OCT) compound (Sakura Finetek USA, Torrance, CA). Transverse, 10-μm-thick cryostat sections were cut and placed onto slides (Super-Frost Plus; Fisher Scientific, Pittsburgh, PA). Sections were subsequently processed for indirect immunofluorescence localization with antibodies against murine MMP-9 (1:1000 dilution in PBS; the kind gift of Robert Senior, Washington University School of Medicine, St. Louis, MO), neurofilament-light (NF-L, 1:100 dilution in PBS; Chemicon, Temecula, CA), and calretinin (1:100 dilution in PBS; Chemicon). The sections were incubated with appropriate Alexa Fluor-568-conjugated secondary antibodies (1:200 dilution in PBS; Molecular Probes, Eugene, OR) for 1 hour at room temperature, mounted with a coverslip, and observed under a bright-field microscope equipped with epifluorescence (Nikon, Melville, NY). Digitized images were obtained with a digital camera (Spot; Diagnostic Instruments, Sterling Heights, MI). Images were processed and compiled on computer (Photoshop, ver. 5.5 and 7.0; Adobe Systems, Mountain View, CA). 

Apoptosis Assay

Apoptotic cell death in the retina was determined as previously described. ^{16 45} Briefly, 10-μm-thick cryostat sections were prepared as described in the prior section, and apoptotic cell death was detected with a TdT-mediated dUTP nick-end labeling (TUNEL) assay kit (In Situ Cell Death Detection kit with fluorescein; Roche Biochemicals, Mannheim, Germany), according to the protocol provided by the manufacturer. Tissue sections were mounted and analyzed as described for immunohistochemistry. TUNEL-positive cells in retinal cross sections were counted in four microscope fields in three independent experiments. Statistical significance was analyzed by analysis of variance (ANOVA), followed by a post hoc Tukey/Kramer procedure (GB-Stat Software; Dynamic Microsystems, Silver Springs, MD) and expressed as the mean ± SEM. 

Western Blot Analysis

Aliquots containing an equal amount of protein (25 μg) were mixed with gel loading buffer and separated on 10% SDS-polyacrylamide gels. For Bcl-2 and Bax, aliquots containing an equal amount of protein (25 μg) were separated on 4% to 20% gradient Tris-glycine gels (Bio-Rad Laboratories). After electrophoresis, the proteins were transferred onto nylon membranes, and nonspecific binding was blocked with 10% nonfat dry milk in Tris-buffered saline containing 0.1% Tween-20 (TBS-T). Membranes were then probed with antibodies against mouse MMP-9 (1:2500 dilution in TBS; Triple Point Biologics, Forest Grove, OR), MMP-2 (1:000 dilution in TBS; NeoMarkers, Fremont, CA, data not shown), TIMP-1 (1:500 dilution in TBS; Santa Cruz Biotechnology, Santa Cruz, CA), TIMP-2 (1:1000 dilution in TBS; Sigma-Aldrich, St. Louis, MO), Bcl-2 (1:1000 dilution in TBS; BD-PharMingen, San Diego, CA), and Bax (1:1000 dilution in TBS; Cell Signaling Technology, Beverly, MA). After incubation with the primary antibodies, membranes were washed with TBS-T, and incubated with appropriate horseradish peroxidase (HRP)–conjugated secondary antibodies (1:2500 dilution in TBS) at room temperature for 1 hour. Finally, the proteins on the membranes were detected with chemiluminescence (ECL; Amersham Pharmacia Biotech, Piscataway, NJ) and the membranes exposed to x-ray film. Purified MMP-9 (Triple Point Biologics) and MMP-2 (Laboratory Vision, Fremont, CA) were co-electrophoresed as positive standards (data not shown). For Bax and BCl-2, protein bands on x-ray films were scanned by a densitometer, and the data from three independent experiments were represented as mean arbitrary units ± SEM. Statistical significance was analyzed by ANOVA, followed by a post hoc Tukey/Kramer procedure (GB-Stat Software; Dynamic Microsystems). 

Results

Effect of KCl on Retinal Degeneration

Because depolarization seems to be one of the earliest events during ischemic damage, experiments were performed to determine whether intravitreal injection of a classic depolarizing agent, KCl, causes retinal damage. Three different assays were performed to assess KCl-induced retinal damage. In the first experiment PBS, 400 nmoles of KCl, TEA (a concentration sufficient to induce depolarization), or NaCl was injected (2.5 μL volume) into the vitreous humor of normal CD-1 mice. At 1 day after injection, total retinal proteins were extracted, and changes in proapoptotic protein Bax and antiapoptotic protein Bcl-2 levels were determined by Western blot analysis (Fig. 1A) . Low levels of Bax and BCl-2 proteins were observed in retinal protein extracts prepared after intravitreal injection of PBS. In contrast, Bax protein levels were increased in retinal protein extracts prepared after intravitreal injection of KCl or TEA, but not after injection of NaCl. No significant changes in Bcl-2 protein levels were observed (Fig. 1) . 

In the second experiment, retinal cross sections were prepared at various times after intravitreal injection of KCl, and apoptotic cell death was determined by TUNEL assays (Fig. 2A) . Examination of retinal cross sections indicated TUNEL-positive apoptotic cells in the ganglion cell layer as early as 6 hours after intravitreal injection of KCl. At 1 day after KCl injection, TUNEL-positive cells were observed in the ganglion cell (GCL) and inner nuclear (INL) layer. Two days after intravitreal injection of KCl, an increased number of TUNEL-positive cells were observed in the INL. TUNEL-positive cells were absent in retinal cross sections prepared from noninjected control or PBS- or NaCl-injected eyes. In contrast, TEA-injected eyes showed TUNEL-positive cells, both in the GCL and INL, similar to those observed in retinal cross sections prepared from KCl-injected eyes (Fig. 2A) . Quantitation of TUNEL-positive cells in retinal cross sections indicated a significant increase in the number of apoptotic cells, both in the GCL and in the INL, after intravitreal injection of KCl or TEA, but not after injection of NaCl (Fig. 2B) . 

In the third experiment, retinal cross sections prepared after intravitreal injection of KCl or NaCl were immunostained with antibodies against the ganglion cell marker NF-L and the amacrine cell marker calretinin (Fig. 3) . Examination of retinal cross sections indicated loss of NF-L-positive immunostaining 1 day and 2 days after KCl injection (400 nmoles) but not after intravitreal injection of a similar concentration of NaCl (Fig. 3A) . In addition, loss of calretinin-positive amacrine cells was observed in retinal cross sections prepared at 1 day and 2 days after intravitreal injection of KCl (Fig. 3B) . Taken together, the results of the three foregoing experiments indicated that intravitreal injection of KCl leads to apoptotic death of both ganglion cells and amacrine cells in the retina. 

Effect of KCl on MMP Activity in the Retina

To determine whether KCl modulates MMP levels in the retina, various concentrations of KCl (20, 200, and 400 nmoles) were injected into the vitreous humor of CD-1 mice. Retinal proteins were extracted at 2 days after intravitreal injection, and aliquots containing an equal amount of protein (25 μg) from each treatment were then loaded onto SDS-polyacrylamide gels containing gelatin as a substrate for MMPs. Gelatin zymography assays indicated low and constitutive levels of MMP-9 in retinal proteins extracted from noninjected control or PBS-injected eyes (Fig. 4A) . Intravitreal injection of KCl resulted in an upregulation in MMP-9 activity but only at concentrations higher than 20 nmoles. Intravitreal injection of 400 nmoles KCl resulted in an increase in MMP-9 activity over noninjected control or PBS-injected eyes. Upregulation of MMP-9 was also observed in retinal extracts after intravitreal injection of TEA (400 nmoles). Western blot analysis confirmed the increase in MMP-9 protein levels in response to both KCl and TEA injections (Fig. 4A) . Similar concentrations of NaCl failed to upregulate MMP-9 activity in the retina (Fig. 4A) . 

To determine whether KCl mediates time-related upregulation in MMP-9 activity in the retina, 400 nmoles of KCl was injected into the vitreous humor of CD-1 mice, retinal proteins were extracted at various times after injection, gelatin-zymography assays were performed to determine MMP activity, and Western blot analysis was performed to determine MMP-9 protein levels. Gelatin zymography and Western blot analysis indicated a time-related upregulation in MMP-9 activity and protein levels in the retina after intravitreal injection of KCl (Fig. 4B) . MMP-9 activity and protein levels were increased as early as 6 hours, reached peak levels at approximately day 1, and then returned to lower levels by day 2 (Fig. 4B) . No other gelatinolytic MMPs were observed after KCl injection. Immunolocalization studies performed on retinal cross sections prepared 1 day and 2 days after intravitreal injection of KCl indicated that astrocytes in the ganglion cell/nerve fiber layer express MMP-9 protein (Fig. 4C) . Western blot analysis indicated no major changes in TIMP-1 and -2 protein levels in the retina after intravitreal injection of KCl (Fig. 4B) . In addition, no significant changes in MMP-2 protein levels were observed (data not shown). These results indicate that KCl induces MMP-9 upregulation in the retina. 

Effect of Glutamate Receptor Antagonists on MMP-9 Activity

Previous reports have suggested that membrane depolarization causes the release of endogenous glutamate into the extracellular space, which then overstimulates postsynaptic retinal neurons (excitotoxicity). Although extracellular glutamate levels in the retina were not determined in this study, it was reasoned that use of glutamate receptor antagonists might help to determine whether glutamate (which may be released after depolarization) plays a role in KCl-induced MMP-9 upregulation in the retina. Therefore, using an indirect approach, the role of glutamate receptors in MMP-9 upregulation was determined by injecting 400 nmoles KCl along with an NMDA-type glutamate receptor antagonist, MK-801 (200 and 400 nmoles), or a non-NMDA-type glutamate receptor antagonist, NBQX (100 and 200 nmoles), into the vitreous humor in CD-1 mice. Total retinal proteins were extracted at 1 day after injection, and MMP activity was determined by gelatin zymography assays. An increase in MMP-9 activity was observed in retinal proteins extracted after KCl injection, as expected (Fig. 5) . In contrast, a decrease in MMP-9 activity in the retina was observed after intravitreal injection of KCl along with MK-801 or NBQX. Intravitreal injection of receptor antagonists alone (without KCl) failed to upregulate MMP-9 activity in the retina. These results indicate that KCl-induced MMP-9 activity is mediated in part by both NMDA and non-NMDA-type glutamate receptors. 

Effect of MMP Inhibitor on KCl-Induced Retinal Degeneration

Although the results described herein implied an association between KCl-mediated upregulation of MMP-9 and subsequent retinal degeneration, it was not completely clear whether MMP-9 plays a causative role in KCl-induced retinal degeneration. Therefore, a synthetic MMP inhibitor (MMP-I; N-hydroxy-1,3-di-[4-metoxybenzenesulfonyl)-5,5-dimethy;-[1,3]-piperazine-2-carboxamide; MMP inhibitor II; Calbiochem, San Diego, CA; IC₅₀ 2.7 nmoles for MMP-9) was used to investigate whether inhibition of MMP-9 activity attenuates KCl-induced retinal damage. The MMP-I (0.6 and 6 nmoles) and KCl were injected into the vitreous humor, and retinal proteins were extracted 1 day after injection. Aliquots containing equal amount of proteins (25 μg) were then subjected to gelatin-zymography assays to determine MMP activity in the retina (Fig. 6A) . Zymography assays indicated an upregulation in MMP-9 in the retina after intravitreal injection of KCl but not after injection of dimethyl sulfoxide (DMSO; a vehicle used to dissolve the MMP-I). In contrast, intravitreal injection of 6.0 nmoles of MMP-I along with KCl resulted in a reduction in MMP-9 activity levels, whereas intravitreal injection of DMSO along with KCl failed to inhibit KCl-induced MMP-9 activity in the retina (data not shown). To determine whether inhibition of MMP-9 observed in retinal proteins extracted after intravitreal injection of MMP inhibitor also attenuates retinal degeneration, retinal cross sections were prepared 1 day after intravitreal injection of KCl, with or without an MMP-I, and apoptotic cell death was determined by TUNEL assay (Fig. 6B) . Examination of retinal cross sections indicated the appearance of few TUNEL-positive cells in the GCL and many in the INL after KCl injection, as expected. In contrast, a significant decrease in the total number of TUNEL-positive cells (including cells in the GCL and INL) was observed after intravitreal injection of KCl with MMP-I (Figs. 6B 6C) . TUNEL-positive cells were absent in retinal cross sections prepared from both noninjected control and DMSO-injected eyes. In addition, intravitreal injection of MMP-I along with KCl attenuated loss of NF-L-positive ganglion cells and calretinin-positive cells in the retina (Fig. 6D) . These results indicate that MMP-9, in part, plays a causative role in KCl-induced retinal degeneration. 

Discussion

Although membrane depolarization and subsequent synaptic release of l-glutamate have been implicated in ischemic retinal damage, the mechanisms involved in retinal damage are poorly understood. In this study, we report a novel finding that injection of a classic depolarizing agent, KCl, into the vitreous humor in CD-1 mice upregulates MMP-9 activity and protein in the retina and that MMP-9 in turn promotes KCl-induced retinal damage. In support of this hypothesis, depolarizing concentrations of KCl (>20 nmoles) injected into the vitreous humor led to a transient upregulation of MMP-9 activity in the retina. TEA-Cl, another depolarizing agent also upregulated MMP-9 activity, but similar concentrations of NaCl failed to induce this upregulation. Although the MMP-9 pro form observed in the zymograms can be converted into an active form during retinal damage, our zymography results showed an increase in the pro form of MMP-9, consistent with findings in several previous investigations, and due in part to the instability and rapid degeneration of activated MMP-9. Nevertheless, KCl not only induced upregulation of MMP-9 activity, but it was also associated with an increase in the proapoptotic protein Bax and the subsequent apoptotic death of both ganglion cells and amacrine cells in the retina. 

Because MMP-9 was upregulated in response to KCl, we reasoned that KCl may act by releasing endogenous l-glutamate into the extracellular space after membrane depolarization of retinal neurons (mediated by KCl). Although glutamate levels may provide some information about whether glutamate directly plays a role in MMP-9 upregulation, determination of total glutamate levels do not provide enough information about whether glutamate exists extracellularly to act on its receptors. In addition, local concentrations of glutamate seem to be more important in promoting retinal damage than the average glutamate levels present in the retina or vitreous humor. Therefore, we used an indirect approach in this study based on the notion that endogenous glutamate released into the extracellular space (by KCl-mediated depolarization) might act on both NMDA and non-NMDA glutamate receptors to upregulate MMP-9 activity in the retina. Therefore, we used two specific glutamate receptor antagonists to test this possibility: MK-801 (a specific NMDA glutamate receptor antagonist) and NBQX (a specific non-NMDA glutamate receptor antagonist). Intravitreal injection of MK-801 or NBQX along with KCl, indeed, resulted in reduced MMP-9 activity, compared with KCl injection alone, indicating that both types of glutamate receptors play a role in KCl-induced MMP-9 upregulation. 

Although the precise mechanisms that lead to MMP upregulation after KCl-injection are not clear at this time, previous studies have suggested that KCl induces changes in the synthesis of mRNAs encoding c-Fos protein. ^{46 47} c-Fos controls the transcription of several immediate early genes, including MMP-9. ^{48 49} Once MMP-9 is upregulated, it can proteolytically modulate the ECM protein laminin in the retina, and can lead to apoptotic cell death. ^{16 45} Therefore, we reasoned that inhibition of MMP-9 may have a protective effect on KCl-induced retinal damage. To test this possibility, a synthetic MMP-I was injected into the vitreous humor to inhibit MMP activity in the retina. Although the inhibitor used in this study is not specific in inhibiting a particular MMP, we used it because KCl upregulated only MMP-9 in the retina. Intravitreal injection of MMP-I along with KCl not only reduced MMP-9 activity in the retina but also attenuated KCl-induced retinal damage. This was evident by a reduction in the number of TUNEL-positive apoptotic cells and by the number of remaining NF-L- and calretinin-positive cells in the retina after intravitreal injection of MMP-I along with KCl. These observations support the hypothesis that MMP-9 in part plays a causative role in promoting KCl-induced retinal damage. It is possible that cell loss could be due only to the toxicity of KCl. However, the reduction in KCl-induced apoptosis and protection against KCl-induced retinal damage by an MMP-I rules out this possibility. Although, additional mechanisms such as inflammation may play a role in KCl-induced retinal damage, our immunolocalization studies using antibodies against macrophages (data not shown) failed to detect inflammatory cells in the retina, ruling out the possible role of inflammation in KCl-induced retinal degeneration. 

There is one caveat in this study. Although, we used KCl, a classic depolarizing agent, to induce retinal damage, we did not determine whether KCl directly induces membrane depolarization in the retina (by electrophysiological studies). The results presented in this study, however, show that MMP-9, in part, plays a role in the KCl-induced retinal damage that in part is mediated by ionotropic glutamate receptors, indicating the classic depolarizing actions of KCl in the retina. The results presented in this study are also consistent with those in recent studies of the CNS, in which depolarization and subsequent upregulation in MMP-9 have been shown to play a causative role in cortical damage. ⁵⁰ In addition, a recent study indicated that the cortical spreading depression (associated with depolarization) also activates and upregulates MMP-9 in the CNS. ⁵¹  

In conclusion, the data presented herein provide, to our knowledge, the first evidence that intravitreous injection of a membrane depolarization agent, KCl, causes retinal degeneration by upregulating MMP-9. 

 Supported by National Eye Institute project Grant EY13643 (SKC) and Vision Research Infrastructure Development Grant EY014803.

 Submitted for publication November 24, 2004; revised February 3, 2005; accepted February 28, 2005.

 Disclosure: R.S. Mali, None; M. Cheng, None; S.K. Chintala, 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.

 

KCl induces changes in proapoptotic regulators in the retina. (A) At 1 day after intravitreal injection of PBS or KCl, TEA, or NaCl (400 nmoles each) total retinal proteins were extracted, and aliquots containing an equal amount of retinal proteins (25 μg) from each treatment were subjected to Western blot analysis with antibodies against Bax and Bcl-2. (B) Bax and Bcl-2 protein bands were scanned by a densitometer, and results from three independent experiments were represented as arbitrary densitometric units per 25 μg protein. Results indicate that intravitreal injection of KCl and TEA but not NaCl increased Bax protein levels (but not Bcl-2) in the retina. Bax, *P < 0.01 PBS versus KCl; Bax, **P < 0.01 PBS versus TEA; ANOVA, followed by a post hoc Tukey/Kramer procedure.

Figure 2.

 

KCl induces retinal degeneration. (A) Retinal cross sections were prepared at 6 hours, 1 day, and 2 days after intravitreal injection of KCl (400 nmoles), at two days after injection of NaCl (400 nmoles), TEA (400 nmoles), PBS or from noninjected control eyes and assayed for apoptotic cell death by TUNEL assays. (B) At various times after intravitreal injection of KCl, TEA, or NaCl, retinal cross sections were prepared, and TUNEL assays were performed to determine apoptotic cell death in the retina. The number of TUNEL-positive cells in the GCL and INL were quantified. Results indicate that both of the depolarizing agents KCl and TEA induced apoptotic cell death in the retina. *P < 0.01, PBS versus KCl in the GCL; **P < 0.01, PBS versus KCl/TEA in the INL. ANOVA, followed by a post hoc Tukey/Kramer procedure.

Figure 3.

 

KCl induces loss of ganglion and amacrine cells. Retinal cross sections prepared after intravitreal injection of KCl or NaCl (400 nmoles each) were immunostained with antibodies against NF-L (A) and calretinin (B). Note a decrease in NF-L and calretinin immunoreactivity after intravitreal injection of KCl but not after intravitreal injection of NaCl. Magnification, ×40.

Figure 4.

 

KCl upregulates MMP-9 activity in the retina. (A) Aliquots containing an equal amount of retinal proteins (25 μg) extracted at 2 days after intravitreal injection of KCl, TEA, NaCl, or PBS or from noninjected control eyes were subjected to gelatin zymography (top) and Western blot analysis (bottom). (B) At various times after intravitreal injection of KCl (400 nmoles), retinal protein extracts were extracted, and aliquots containing an equal amount of retinal protein (25 μg) were subjected to gelatin zymography (top) and Western blot analysis (bottom). (C) Retinal sections prepared from PBS- or KCl-injected eyes were immunostained with antibodies against MMP-9. Note a dose- and time-related increase in MMP-9 activity and protein levels in the retina after intravitreal injection of KCl and MMP-9 tissue localization in the ganglion cell layer. Also, note that TIMP-1 and -2 levels remained unchanged.

Figure 5.

 

Glutamate receptor antagonism inhibits KCl-induced MMP-9 activity. Retinal proteins extracted at 1 day after intravitreal injection of KCl (400 nmoles), with or without indicated concentrations of the NMDA receptor antagonist, MK-801, or the non-NMDA receptor antagonist, NBQX, were subjected to gelatin zymography. Both receptor antagonists inhibited KCl-induced MMP-9 activity in the retina.

Figure 6.

 

Inhibition of MMP activity attenuates KCl-induced retinal degeneration. (A) Retinal proteins were extracted at 1 day after intravitreal injection of KCl (400 nmoles) along with indicated concentrations of an MMP-I. Aliquots containing an equal amount of proteins (25 μg) were subjected to gelatin zymography. Note a decrease in KCl-mediated MMP-9 activity after intravitreal injection of the MMP-I. (B) Retinal cross sections prepared 1 day after intravitreal injection of KCl (400 nmoles) with or without the MMP-I (6 nmoles; MMP-I) were subjected to TUNEL assays and compared with noninjected control or vehicle (DMSO)-injected eyes. (C) Quantification of TUNEL-positive cells indicates a significant reduction in the number of TUNEL-positive cells in retinal cross sections after intravitreal injection of KCl along with the MMP-I. *P < 0.01, DMSO versus KCl; **P < 0.01 KCl versus KCl+MMP-I (6 nmoles); ANOVA, followed by the post hoc Tukey/Kramer test. (D) Retinal cross sections prepared after intravitreal injection of KCl, with or without the MMP-I, were immunostained with antibodies against NF-L and calretinin. Compared with KCl-injected eyes, retinal cross sections prepared after intravitreal injection of KCl along with the MMP-I showed attenuation in ganglion and amacrine cell loss.