Adenosine A₁ Receptor Modulation of MMP-2 Secretion by Trabecular Meshwork Cells

Abstract

purpose. Studies have shown that adenosine A₁ agonists can lower IOP in rabbits, mice, and monkeys, and this response is mediated in part by increases in outflow facility. The purpose of this project was to evaluate the response of trabecular meshwork cells to the addition of the adenosine A₁ receptor agonist N ⁶-cyclohexyladenosine (CHA).

methods. The human trabecular meshwork (HTM-3) cell line and primary cultures of bovine trabecular meshwork (BTM) cells were used in these studies. Cells were treated with CHA, and the secretion of matrix metalloproteinase (MMP)-2 or the activation of extracellular signal–regulated kinase (ERK1/2) was determined.

results. Treatment of HTM-3 and BTM cells with CHA (0.1 μM) resulted in a time-dependent secretion of MMP-2 that was measurable as early as 30 minutes after treatment and reached a maximum by 2 hours. This CHA-induced secretion of MMP-2 was inhibited by the adenosine A₁ receptor antagonist 8-cyclopentyl-1,3-dimethylxanthine (CPT) and by the ERK1/2 pathway inhibitor U0126. Treatment of HTM-3 cells with CHA produced a rapid dose-dependent activation of ERK1/2 with an EC₅₀ of 5.7 nM. The CHA-induced activation of ERK1/2 was inhibited by pretreatment with the adenosine A₁ antagonist CPT and by the ERK pathway inhibitor U0126.

conclusions. The addition of the adenosine A₁ agonist CHA stimulates the secretion of MMP-2 from trabecular meshwork cells. This secretory response involves the activation of adenosine A₁-linked stimulation of ERK1/2. These results provide evidence for the existence of functional adenosine A₁ receptors in the trabecular cells and that the activation of these receptors stimulates secretion of MMP-2.

Adenosine A₁ receptor agonists have been shown to lower intraocular pressure (IOP) in rabbits, mice, and monkeys. ^{1 2 3 4 5} This reduction in IOP can involve both a decrease in aqueous flow and an increase in outflow facility. ^{2 3 4} Previous studies have provided evidence that the reduction in aqueous flow may be mediated by postjunctional adenosine A₁ receptors in the ciliary body. ³ However, the cellular mechanisms responsible for the increase in outflow facility have not been identified. 

There is increasing evidence that the resistance to conventional aqueous outflow is in part dependent on the composition of the extracellular matrix in the trabecular meshwork (TM). ^{6 7 8 9 10} Bradley et al., ⁷ demonstrated that an increase in outflow facility could be achieved in a human anterior chamber model by perfusion with specific matrix-degrading enzymes (matrix metalloproteinases [MMPs]). Additional studies have shown that cells in the trabecular meshwork may regulate outflow resistance directly by modifying their surrounding extracellular matrix (ECM) through the secretion of ECM material, MMPs (and other matrix-degrading enzymes), and tissue inhibitors of MMPs (TIMPs). ^{6 7 11 12}  

To investigate the possibility that adenosine A₁ receptor agonists increase outflow facility and lower IOP by acting directly on TM cells, we evaluated the secretion of MMPs by trabecular cells after the addition of the adenosine A₁ agonist, N ⁶-cyclohexyladenosine (CHA). Because previous studies have shown that the secretion of specific MMPs are dependent on the activation of the extracellular signal–regulated pathway, ⁶ we also sought to determine whether the adenosine agonist activates this signaling pathway in trabecular cells. Our results provide evidence for the presence of functional adenosine A₁ receptors on trabecular cells and that the activation of these receptors stimulates the secretion of MMP-2 through the activation of the ERK1/2 pathway. 

Methods

Reagents

Solutions of 1 mM CHA, adenosine A₁ receptor antagonist 8-cyclopentyl-1,3-dimethylxanthine (CPT), and A₂ receptor antagonist 1,3 dipropyl-7-methylxanthine (DMPX), were dissolved in deionized water just before the use of each agent. Stock solutions (10 mM) of the MAP kinase kinase (MEK) inhibitor U0126 were prepared in 1% dimethyl sulfoxide (DMSO) and stored at −20°C. 

Cell Culture

The transformed human TM cell line (HTM-3) and primary cultures of bovine TM (BTM) cells were used in these studies. The HTM-3 cells were maintained on polypropylene cell culture plates and grown in DMEM containing 10% heat-inactivated fetal calf serum (FCS). ¹³ The cells were passaged at 3- to 4-day intervals and allowed to grow to approximately 80% confluence. Primary bovine cell cultures were established from TM explants by techniques previously established by our laboratory. ⁶ Briefly, small strips of TM tissue were dissected from one or two eyes and homogenized by means of a Teflon hand homogenizer in DMEM containing 15% FCS. The homogenized tissue was plated onto a 60-mm collagen-I–coated (Biocoat, Fort Washington, PA) cell culture plate and allowed to grow 2 weeks in DMEM containing 15% FCS. The resultant cells were harvested and plated onto polypropylene cell culture plates in DMEM containing 10% FCS. These cells were allowed to grow to approximately 80% confluence. Second- or third-passage cells were used in studies involving BTM cells. 

MMP-2 Assay

Cells were washed and then maintained in serum-free medium for 16 hours before the addition of any agents. To stimulate secretion of MMP-2, cells were treated with 100 nM CHA for 2 hours, unless otherwise noted. In experiments evaluating the actions of the MEK inhibitor U0126 or the adenosine receptor antagonists, cells were pretreated for 30 minutes with the inhibitor before the addition of CHA. At the end of the incubation period, media were collected and concentrated 10-fold (Centricon concentrators; Millipore Corp., Bedford, MA). Equivalent volumes of media were then loaded onto 12% SDS polyacrylamide gels, where secreted proteins were separated according to molecular weight by standard SDS-PAGE protocols and transferred onto nitrocellulose paper. The combined level of pro-MMP-2 and active MMP-2 was then determined by immunoblot analysis with rabbit polyclonal anti-MMP-2 antibodies (Research Diagnostics, Flanders, NJ). Bands were visualized by the addition of anti-rabbit horseradish peroxidase (HRP)–conjugated secondary antibodies (New England Biolabs, Inc., Beverly, MA) and enhanced chemiluminescence (ECL) reagents (Amersham, Buckinghamshire, UK). 

Extracellular Signal-Regulated Kinase Assay

Cells were washed and maintained in serum-free medium for 16 hours before the addition of any agents. To activate the extracellular signal-regulated kinase (ERK) pathway, cells were treated with CHA for 10 minutes, unless otherwise noted. In experiments evaluating the MEK inhibitor U0126 or the adenosine receptor antagonists, cells were pretreated for 30 minutes with the inhibitor before the addition of CHA. At the end of the incubation period, cells were rinsed with ice-cold PBS and lysed by the addition of 0.5 mL of lysis buffer (50 mM β-glycerophosphate, 20 mM EGTA, 15 mM MgCl₂, 1 mM NaVO₄, 1 mM dithiothreitol (DTT), and 1 μg/mL of a protease inhibitor cocktail (Roche Diagnostics, Indianapolis, IN). The total cell lysate was then transferred to microcentrifuge tubes and sonicated for 5 seconds and the solution clarified by centrifugation (10 minutes at 10,000g). A small aliquot of the supernatant of each sample was removed for a protein assay, and SDS-running buffer was added to the remaining fraction. Samples were heated for 5 minutes at 95°C and then stored at −80°C. Sample protein concentrations were determined with a protein assay kit (Bio-Rad, Richmond, CA). 

To determine the level of ERK1/2 activation (phosphorylation), equivalent amounts of protein were loaded onto 12% SDS polyacrylamide gels, proteins separated according to molecular weight by standard SDS-PAGE protocols and transferred to nitrocellulose paper. Total ERK levels (phosphorylated and nonphosphorylated forms) were determined by immunoblot techniques using polyclonal anti-ERK2 antibodies (New England Biolabs Inc., Beverly, MA). Bands were visualized by the addition of anti-rabbit HRP–conjugated secondary antibodies and ECL reagents (Amersham). Blots were then stripped by incubation in “stripping buffer” (62.5 mM Tris [pH 6.7], 100 mM β-mercaptoethanol, 2% SDS) for 30 minutes at 50°C. The level of phosphorylated ERK1/2 was then determined by immunoblot analysis with polyclonal antiphospho-ERK antibodies (New England Biolabs, Inc.) and visualized by the addition of anti-rabbit HRP–conjugated secondary antibodies and ECL reagents. Band densities were quantified with image-management software (Scion Imaging, Frederick, MD) and the level of phosphorylated ERK1/2 isoforms normalized for differences in loading, using the total ERK protein band intensities. 

Statistical Analysis

Statistical comparisons were made using the Student’s t-test for nonpaired data or one-sample t-test. P ≤ 0.05 was considered significant. The dose–response curve was analyzed by nonlinear regression analysis (GraphPad Software, Inc., San Diego, CA). 

Results

Treatment of both the HTM-3 and the BTM cells with CHA (100 nM) resulted in a time-dependent increase in levels of MMP-2 above those observed after the addition of vehicle alone. An increase in MMP-2 levels was evident within 30 minutes after addition of CHA, with a maximum increase in MMP-2 levels measured by 2 hours after addition (Fig. 1) . In selected experiments, the addition of CHA for up to 6 hours did not lead to any further increase in MMP-2 levels above that measured after the 2-hour incubation period (data not shown). The addition of CHA for 2 hours did not stimulate the secretion of MMP-3 or -9 (data not shown). 

Pretreatment of HTM-3 or BTM cells with the relatively selective adenosine A₁ receptor antagonist CPT for 30 minutes (10 μM) significantly inhibited (P < 0.05) CHA-induced secretion of MMP-2 (Fig. 2) . The addition of CPT alone did not produce any significant change in secretion of MMP-2 from that measured in control cells. 

To determine whether the activation of ERK1/2 is involved in the CHA-induced secretion of MMP-2, HTM-3 and BTM cells were pretreated with the ERK pathway inhibitor U0126 (0.1 μM) for 30 minutes before the addition of CHA for 2 hours. Pretreatment with U0126 significantly inhibited (P < 0.05) the secretion of MMP-2 in response to CHA (Fig. 3) . The addition of U0126 alone did not significantly alter the secretion of MMP-2 in these experiments from that measured in control cells. 

The addition of CHA (0.1 μM) produced a time-dependent phosphorylation (activation) of ERK1/2 with a maximal activation occurring by 10 minutes that gradually returned to control levels by the 2-hour time point (Fig. 4) . This increase in activation of ERK1/2 was dose-dependent (Fig. 5) , and regression analysis of these data calculated an EC₅₀ of 5.7 nM. 

Pretreatment of HTM-3 cells with the adenosine A₁ receptor antagonist CPT (10 μM) for 30 minutes significantly inhibited (80%; P < 0.05, n = 4; Fig. 6A ) the CHA-induced activation of ERK1/2. However, pretreatment with the A₂ receptor antagonist DMPX (n = 3) did not significantly alter CHA-induced activation of ERK1/2 (Fig. 6B) . Pretreatment with the MEK inhibitor U0126 (1 μM) for 30 minutes also significantly inhibited (88%; P < 0.05, n = 5; Fig. 6C ) the CHA-induced activation of ERK1/2. 

Discussion

Adenosine is a common signaling molecule often associated with cellular responses to stressful situations. Ischemia, in many tissues including the eye, can lead to rapid increases in adenosine concentrations. ¹⁴ Recent studies have also shown that the ciliary epithelium contains releasable stores of ATP that may be degraded to adenosine by the action of ecto-adenosine triphosphatase (ATPase). ¹⁵ Experiments in our laboratory have shown that adenosine levels in the aqueous humor are positively correlated to IOP. ¹⁶  

In rabbits, mice, and primates, the activation of adenosine A₁ receptors lowers IOP, ^{5 17} and the decrease can result in part from an increase in outflow facility. ^{3 4} Previous studies have shown that reductions in aqueous flow are regulated by postjunctional A₁ receptors in the ciliary body. ² The purpose of this study was to begin to characterize a cellular mechanism associated with the adenosine A₁ receptor–mediated decrease in outflow resistance. Because the TM and associated extracellular matrix are thought to play a central role in the regulation of outflow resistance, ¹⁸ this investigation was focused on evaluating the action of adenosine A₁ agonist on cells from this region. 

Recent investigations have established that MMPs can decrease outflow resistance in the conventional outflow pathway. ^{7 8} The TM has been shown to increase the expression and secretion of other MMPs in response to a number of different stimuli, including phorbol esters, growth factors, cytokines, and mechanical stress. ^{8 9 10 11} Treatment periods during which these changes have been observed are generally 24 hours or more, indicating that MMPs may not be involved in the acute response to the agent that enhances conventional outflow facility. However, recent results have shown that secretion of MMP-2 from TM cells can occur within 2 hours. ⁶ Unlike other MMPs, studies have shown that MMP-2 is constitutively expressed in and is regulated primarily at the level of secretion. ¹⁷ Hence, we hypothesized that activation of adenosine A₁ receptors would lead to a similar rapid secretion of MMP-2 from TM cells. 

Our results demonstrate that CHA-induced secretion of MMP-2 from TM cells occurred as early as 30 minutes after treatment with CHA and reached a maximum level by 2 hours. This response was blocked by pretreatment with the adenosine A₁ receptor antagonist CPT or the ERK1/2 pathway inhibitor U0126. These data, along with the dose-dependent activation of the ERK1/2 pathway by CHA and inhibition of this response by the A₁ adenosine receptor antagonist, provide the first evidence for the presence of functional adenosine A₁ receptors in TM cells. The increase in secretion of MMP-2 observed in these studies is consistent with the time frame of the decrease in IOP observed in vivo. ^{3 4} Although it is difficult to make comparisons between cells in culture and in vivo physiological responses, these data provide a potential cellular mechanism to explain the decrease in outflow resistance observed after treatment with an adenosine A₁ agonist. 

No increases in MMP-3 and -9 were observed in these cells after 2 hours of treatment with CHA. Previous studies have shown that a number of cytokines and growth factors can stimulate the secretion of these MMPs. Our results may indicate that the activation of adenosine A₁ receptors in trabecular cells produces a selective secretion of MMP-2 in relation to MMPs; however, it should be noted that most MMPs are regulated at the level of transcription. ¹⁷ Hence, the absence of any increase in secretion of MMP-3 or -9 may reflect the short treatment periods that did not allow sufficient time for expression and secretion of these proteins. Although previous studies have provided evidence that activation of adenosine A_2b receptors can decrease expression of collagenase, ¹⁹ our report is the first to demonstrate an adenosine A₁–mediated increase in secretion of MMP. 

The results presented in this report focus only on the increase in secretion of MMP-2; MMP-2 activity was not investigated in the study. MMP-2 is secreted as a proenzyme and is activated at the extracellular cell surface through its association with TIMP-2, MT1-MMP, and α3βv 2integrin. ^{17 20 21} Although these experiments establish that adenosine A₁ receptor activation increases secretion of MMP-2, additional experiments are necessary to determine the level of MMP-2 activity, as well as changes in TIMP-2, MT1-MMP, and α3βv integrin expression, and to understand how these factors may work in a coordinated fashion to regulate trabecular function. 

The activation of the ERK1/2 pathway is an important cell-signaling mechanism regulating multiple cell functions. ²² Adenosine A₁ receptors specifically have been shown to activate ERK1/2, ^{23 24} and previous results from this laboratory have shown that secretion of MMP-2 is dependent on activation of ERK. ⁶ As shown in Figure 3 , administration of CHA induced a rapid activation of ERK1/2 in the TM cells, with the maximum activation occurring at the 10-minute time point, then returning to control levels by the 2-hour time point. This CHA-induced activation of ERK1/2 was inhibited by the adenosine A₁ receptor antagonist CPT and by the MEK inhibitor U0126, confirming that activation of ERK1/2 is stimulated by adenosine A₁ receptors in trabecular cells. The relatively rapid activation of ERK1/2, when compared with the MMP-2 secretory response, indicates that activation of ERK1/2 is an upstream regulator of the secretion of MMP-2. These results, along with previously published reports, demonstrate the significance of the ERK1/2 pathway in the regulation of trabecular cell function. 

In summary, our data provide functional evidence for the presence of adenosine A₁ receptors on trabecular cells. From the results presented herein, we conclude that activation of these receptors leads to a rapid secretion of MMP-2 that is dependent on the activation of the ERK1/2 pathway. These results support the idea that the increase in outflow facility involves the activation of adenosine A₁ receptors on trabecular cells. It is tempting to speculate that the secretion of MMP-2 contributes to an adenosine A₁–mediated increase in outflow facility. However, additional studies investigating purinergic modulation of MMP activity and outflow facility are needed to make this determination. 

 Supported in part by National Eye Institute Grant EY09741 (CEC) and Research to Prevent Blindness.

 Submitted for publication September 10, 2001; revised April 23, 2002; accepted May 7, 2002.

 Commercial relationships policy: F.

 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.

 

Time course of secretion of MMP-2 after treatment with CHA. Serum-deprived HTM-3 or BTM cells were treated with vehicle or 0.1 μM CHA for 30, 60, or 120 minutes. The media were then collected, concentrated, and analyzed for MMP-2. (A) Data are the mean ± SE of densitometry measurements from Western blots of concentrated media (n = 3) and are normalized to measurements from control (vehicle treated) cells. *Significant difference (P < 0.05) from control (100%). (B) Representative MMP-2 (72 kDa) immunoblot of concentrated media from HTM-3 cells. (C) Representative MMP-2 (72 kDa) immunoblot of concentrated media from primary BTM cells.

Figure 2.

 

Inhibition of secretion of MMP-2 by the adenosine A₁ antagonist CPT. Serum-deprived HTM-3 or BTM cells were treated with vehicle or 10 μM CPT for 30 minutes. Cells were then treated with 0.1 μM CHA for an additional 2 hours. The media were collected, concentrated, and analyzed for MMP-2. (A) Data are the mean ± SE of densitometry measurements from Western blots of concentrated media (n = 4) and are normalized to measurements from control (vehicle treated) cells. *Significant difference (P < 0.05) from corresponding cells treated with CHA alone. (B) Representative MMP-2 (72 kDa) immunoblot of concentrated media from HTM-3 cells. (C) Representative MMP-2 (72 kDa) immunoblot of concentrated media from primary BTM cells.

Figure 3.

 

Inhibition of secretion of MMP-2 by the MEK inhibitor U0126. Serum-deprived HTM-3 or BTM cells were treated with vehicle or 0.1 μM U0126 for 30 minutes. Cells were then treated with vehicle or 0.1 μM CHA for an additional 2 hours. The media were then collected, concentrated, and analyzed for MMP-2. (A) Data are the mean ± SE of densitometry measurements from Western blots of concentrated media (n = 4) and are normalized to measurements from control (vehicle treated) cells. *Significant difference (P < 0.05) from corresponding cells treated with CHA alone. (B) Representative MMP-2 (72 kDa) immunoblot of concentrated media from HTM-3 cells. (C) Representative MMP-2 (72 kDa) immunoblot of concentrated media from primary BTM cells.

Figure 4.

 

Time course of CHA-induced activation of ERK1/2. Serum-deprived HTM-3 or BTM cells were treated for 2, 10, 30, 60, or 120 minutes with 0.1 μM CHA. (A) Data are the mean ± SE of densitometry measurements from Western blots of HTM-3 cell lysates (n = 5) and are normalized to measurements from control cells (t = 0 hours). *Significant difference (P < 0.05) from control (100%). (B) Representative immunoblot of phospho-ERK and total ERK from HTM-3 cell lysates. (C) Representative immunoblot of phospho-ERK and total ERK from primary BTM cell lysates.

Figure 5.

 

Dose-dependent activation of ERK1/2 induced by CHA. Serum-deprived HTM-3 cells were treated for 10 minutes with vehicle (control) or various concentrations of CHA. Data ± SE are means of densitometry measurements from Western blots of HTM-3 cell lysates (n = 5) and are normalized to measurements from vehicle-treated cells.

Figure 6.

 

Inhibition of ERK1/2 activation by the MEK inhibitor U0126 and the adenosine receptor antagonists. Representative immunoblots of phospho-ERK1/2 and total ERK1/2 from HTM-3 cell lysates. (A) Serum-deprived cells were treated with vehicle or CPT for 30 minutes. Cells were then treated for an additional 10 minutes with either vehicle or 0.1 μM CHA. (B) Serum-deprived cells were treated with vehicle or 10 μM DMPX for 30 minutes and then with either vehicle or 0.1 μM CHA an additional 10 minutes. (C) Serum-deprived cells were treated with vehicle or U0126 for 30 minutes. Cells were then treated for an additional 10 minutes with either vehicle or 0.1 μM CHA.

The authors thank I.-H. Pang, Alcon Laboratories, Fort Worth, Texas, for providing the HTM-3 cells; and Luanna Bartholomew for critical reading of and comments on the manuscript.