Modulation of Conventional Outflow Facility by the Adenosine A₁ Agonist N ⁶-Cyclohexyladenosine

Abstract

purpose. Studies have shown that the activation of adenosine A₁ receptors lower intraocular pressure primarily by increasing total outflow facility. The purpose of this study was to investigate the actions of the adenosine A₁ agonist N ⁶-cyclohexyladenosine (CHA) on conventional outflow facility.

methods. Conventional outflow facility was evaluated in isolated bovine anterior segments, perfused at a constant pressure of 10 mm Hg. After overnight perfusion to establish a stable baseline, the concentration- and time-dependent changes in outflow facility induced by CHA were determined. To confirm the involvement of adenosine A₁ receptors and matrix metalloproteinases (MMP) in any change in facility, the responses to CHA were evaluated in preparations treated with the adenosine A₁ receptor antagonist, 8-cyclopentyl-1,3-dimethylxanthine (CPT), or the nonselective MMP inhibitor GM-6001.

results. The administration of CHA (10 μM) to perfused anterior segments produced a 28% increase in outflow facility over basal levels. This response was relatively slow to develop with no significant change in outflow facility measured until after 60 minutes of CHA infusion. The peak response to CHA infusion occurred between 3 and 4 hours after CHA administration. Analysis of the CHA concentration–response curves demonstrated that this increase in outflow facility was concentration-dependent, with an EC₅₀ of 0.28 μM. Pretreatment with the adenosine A₁ receptor antagonist CPT (10 μM) or the nonselective MMP inhibitor GM-6001 (10 μM) blocked the response to CHA (1 μM). When compared with control eyes, no significant change in baseline facility was measured in eyes perfused with CPT or GM-6001.

conclusions. These studies demonstrate that the adenosine agonist CHA significantly increases conventional outflow facility in the perfused bovine eye. Analysis of the CHA concentration–response curve and inhibition of the CHA-induced increase in outflow facility by the adenosine A₁ antagonist confirms that this response is mediated by the activation of adenosine A₁ receptors. The inhibition of the CHA-induced increase in outflow facility by the MMP inhibitor GM-6001 provides evidence that the secretion and activation of MMPs within the conventional outflow pathway play a central role in the ocular hypotensive action of adenosine A₁ agonists.

In primary open-angle glaucoma, elevated IOP results from increased resistance within the conventional outflow pathway. ¹ Although several drugs have been developed to control this increase in IOP, most of these agents act on target tissues outside of the conventional outflow pathway. The identification of agents and their associated receptors that act on cells within the conventional pathway offer the potential to develop therapies that target the site responsible for the ocular hypertension observed in primary open-angle glaucoma. 

Adenosine is considered a reactive metabolite involved in cellular communication during periods of stress. In the eye, adenosine levels have been shown to increase during periods of retinal ischemia and elevated intraocular pressure. ^{2 3} Biochemical, pharmacological, and molecular studies have identified four adenosine receptor subtypes: A₁, A_2a, A_2b, and A₃. ⁴ Recent in vitro studies have identified functional adenosine receptors on both trabecular meshwork and Schlemm’s canal cells. ^{5 6 7} In vivo studies have shown that the administration of adenosine A₁ receptor agonists lowers IOP in rabbits, mice, and monkeys. ^{8 9 10} Although this reduction in IOP involves an early decrease in aqueous flow, most of this ocular hypotensive response results from an increase in total outflow facility. ^{10 11} In vitro studies have shown that adenosine A₁ receptors are expressed by trabecular meshwork cells and the activation of these receptors are linked to the secretion of matrix metalloproteinases (MMPs). ⁶ However, studies have not been conducted to evaluate whether the activation of adenosine A₁ receptors within the conventional outflow pathway contribute to the increase in total outflow facility. Moreover, the cellular events that mediate any functional changes in outflow have not been identified. 

To investigate whether adenosine A₁ receptor activation modulates conventional outflow facility, we evaluated the effects of the adenosine A₁ agonist N ⁶-cyclohexyladenosine (CHA) on outflow facility in isolated perfused bovine anterior segments. Our results demonstrate that the administration of CHA increased outflow facility in this model. This increase in outflow facility was dependent on adenosine A₁ receptor activation and blocked by pretreatment with the broad-spectrum MMP inhibitor GM-6001. 

Methods

Reagents

Stock solutions (1 mM) of CHA, and adenosine A₁ receptor antagonist (8-cyclopentyl-1,3-dimethylxanthine [CPT]), were dissolved in deionized water just before the use of each agent. A stock solution (10 mM) of the nonselective MMP inhibitor GM-6001 was prepared in dimethyl sulfoxide just before use. CHA, CPT, and GM-6001 were purchased from Sigma-Aldrich (St. Louis, MO). Dulbecco’s modified Eagle’s medium (DMEM) was purchased from Invitrogen (Carlsbad, CA). 

Anterior Segment Perfusion

Bovine eyes were obtained from a local abattoir. After removal, eyes were transported in ice-cold phosphate-buffered saline (pH 7.4). Eyes were bisected at the equator, and the lens was removed. The remaining choroid, iris, and ciliary body were gently teased away. Isolated corneoscleral shells were then attached to a perfusion chamber and perfused with DMEM supplemented with 50 U/mL penicillin and 50 μg/mL streptomycin. The entire perfusion apparatus was placed in an incubator at 37°C and 5% CO₂. Perfusion pressure was maintained at a constant level of 10 mm Hg, and the rate of fluid outflow was monitored continuously by measuring the rate of fluid flow from a reservoir by means of an analytical balance (Model ACCU 124; Fisher, Pittsburgh, PA). The rate of fluid flow was recorded by computer (Dell, Round Rock, TX; with Collect XL software LabTronics, Inc., Guelph, Ontario, Canada). Outflow facility was calculated every 2 minutes as the ratio of flow rate to perfusion pressure (in microliters per minute × mm Hg). Preparations were allowed to stabilize overnight (12–14 hours), and baseline facilities were then recorded for 40 to 60 minutes. Only preparations with stable baselines and basal facilities ranging from 0.4 to 2.5 (in microliters per minute × mm Hg) were used in these studies. The adenosine A₁ agonist CHA was then introduced into the perfusion system by media exchange. In experiments evaluating the effects of the adenosine receptor antagonist CPT or the MMP inhibitor GM-6001, these agents were included in the medium during the stabilization periods and after media exchange with CHA. The overnight pretreatment with CPT or GM-6001 allowed us to determine whether adenosine A₁ receptor and MMP activation are involved in CHA-induced changes in outflow facility, and whether endogenous adenosinergic or MMP activity modulates basal facility measured in these preparations. 

Data Analysis

To calculate mean results ± SE, individual facility measurements were normalized to baseline values and expressed as the percentage change in facility at the time of CHA addition (t = 0). Data were analyzed by the unpaired t-test or analysis of variance followed by the Duncan multiple-range test. Comparisons yielding P < 0.05 were considered significant. Concentration–response curves were analyzed by nonlinear regression analysis (Prism; GraphPad Software, Inc., San Diego, CA). The ED₅₀ and Hill coefficient were entered as variables in the dose–response equation and reported as best-fit values. Starting values for the regression analyses were determined by visual inspection of the data. 

Results

Bovine anterior segments perfused overnight with control medium (14–16 hours) exhibited outflow facilities ranging from 0.41 to 2.5 μL/min · mm Hg. The mean baseline facility for control preparations was 0.97 ± 0.22 (n = 10). No evidence of washout was observed in these preparations when perfused at 10 mm Hg. Figure 1shows the response of perfused bovine anterior segments to medium exchange with control medium or medium containing CHA (10⁻⁵ M). Control exchange of perfusion media did not produce any consistent change in outflow facility. However, the introduction of medium containing CHA produced a slow increase in outflow facility that became evident 40 to 60 minutes after media exchange and reached a maximum between 3 and 4 hours after administration. The mean time-dependent changes in outflow facility from baseline induced by CHA (10⁻⁶ M) are shown in Figure 2 . When compared with control exchanges, the addition of CHA induced a significant increase in outflow facility at 60 minutes after administration. Mean outflow facility continued to increase through 160 minutes, achieving a maximum increase of 27% ± 3.9% above baseline. No significant change in outflow facility from basal levels was measured in control preparations up to 4 hours after media exchange. 

The CHA-induced increases in outflow facility were concentration-dependent (Fig. 3) . Regression analysis of the data yielded an EC₅₀ of 0.28 μM and a maximum increase in outflow facility of 28%. The Hill coefficient for regression analysis was not significantly different from 1.0. Overnight perfusion with the adenosine A₁ receptor antagonist CPT (10⁻⁵ M), blocked the CHA-induced increase in outflow facility (Fig. 4) . In these preparations, no significant change in baseline outflow facility was measured when compared with control preparations. 

To investigate whether the secretion and activation of matrix metalloproteinases are involved in the response to CHA, we perfused anterior segments overnight with the broad-spectrum MMP inhibitor GM-6001 (10⁻⁵ M). In anterior segments perfused with GM-6001, the mean basal outflow facility was 0.71 ± 0.17 μL/min · mm Hg (n = 7). Although statistical analyses indicated a trend toward lower basal outflow facilities in GM-6001-treated preparations, the difference was not significant (P = 0.10). However, overnight pretreatment with GM-6001 significantly inhibited the CHA-induced increases in outflow facility at all time points (Fig. 5) . 

Discussion

The topical administration of adenosine A₁ agonists has been shown to lower IOP in rabbits, mice, and primates. ^{8 9 10} Although the initial reduction in IOP induced by these agents involves a reduction in aqueous flow, the reduction in pressure results primarily from an increase in total outflow facility. ^{10 11} Cell culture studies have identified adenosine A₁ receptors on both trabecular meshwork and Schlemm’s canal cells. ^{5 6} The purpose of this study was to determine whether the activation of adenosine A₁ receptors can modulate outflow facility in the conventional outflow pathway. 

The results, presented in Figures 1 2 and 3 , demonstrate that administration of the relatively selective adenosine A₁ agonist CHA significantly increased the conventional outflow facility in the perfused bovine anterior segment. This response was concentration dependent and blocked by the A₁ receptor antagonist CPT. Taken together, these data provide the first functional evidence that adenosine A₁ receptor activation can increase in the conventional outflow facility. However, overnight pretreatment with CPT did not alter baseline facility, indicating that, in the perfused anterior segment, no significant endogenous adenosinergic tone exists within the conventional outflow pathway of these preparations. 

Previous studies have shown that CHA produces a dose-related reduction in IOP in vivo and induces expression of mitogen-activated protein (MAP) kinase in trabecular cells in vitro. ^{6 8} It is difficult to compare dose–response characteristics from the present study to the dose-related changes in IOP induced by CHA, as this agonist has been shown to act at multiple sites to lower pressure, there are species differences, and there is a lack of data on the effective concentration of CHA within the outflow pathways. However, in vitro studies on bovine trabecular meshwork cells have shown that for CHA-induced activation of the ERK1/2 signaling pathway, the EC₅₀ is 5.7 nM. ⁶ Compared with the present results, CHA was approximately two times more potent in stimulating ERK1/2 than in increasing conventional outflow facility. This difference in potency may reflect that a functional response to CHA within the conventional outflow pathway involves cross-talk between multiple signaling systems or that CHA induces multiple pharmacological effects within the outflow pathway, some of which may inhibit CHA-induced increase in outflow facility. 

In vitro studies have shown that the activation of adenosine A₁ receptors in trabecular or Schlemm’s canal cells are linked to changes in ion conductance, increases in intracellular Ca²⁺, and MMP secretion. ^{5 6 7} As a result, investigators have speculated that adenosine A₁ agonist–induced changes in IOP and outflow facility may be related to enhanced ion and water transport, changes of cell volume, cell contraction, or modification of extracellular proteins within the conventional outflow. ^{5 6 7} Studies have shown that adenosine receptor-mediated changes in membrane ion currents, intracellular Ca²⁺, and cell volume are relatively fast, requiring only 2 to 5 minutes for the full response to be expressed. ⁵ In addition, procedures inducing cell volume changes in perfused eyes result in changes in outflow facility within 5 to 10 minutes. ¹² Hence, the delayed onset and subsequent slow increase in outflow facility after administration of CHA observed in this study indicates that changes in cell contraction or volume are unlikely as potential mechanisms responsible for the increase in conventional outflow facility. 

The administration of MMPs can increase outflow facility in the conventional outflow pathway. ¹³ In addition, trabecular meshwork cells have been shown to increase the secretion of MMP-2 in response to adenosine A₁ receptor activation. ⁶ The time-course for the increase in MMP-2 secretion was similar to the increases in outflow facility measured in this study, with detectable changes in MMP-2 secretion being observable by 30 minutes and reaching a maximum 2 hours after the addition of CHA. ⁶ To investigate whether secretion and activation of the MMPs were responsible for changes in conventional outflow facility after administration of the adenosine A₁ antagonist, preparations were treated overnight with the broad-spectrum MMP inhibitor GM-6001 (Galardin; Glycomed, Inc., Alameda, CA; or ilomastat). GM-6001 is a hydroxamic acid derivative originally synthesized as an inhibitor of human skin collagenase, ¹⁴ and has been shown also to block MMP-1, -2, -3, and -9. ^{15 16} Pretreatment with GM-6001 blocked the increase in outflow facility induced by administration of CHA. These data, together with in vitro studies demonstrating that adenosine A₁ agonists induce MMP secretion from trabecular meshwork cells, support the idea that adenosine-mediated changes in outflow facility involve the secretion and activation of MMPs within the conventional outflow pathway. Overnight pretreatment with GM-6001 also produced a trend toward reduced basal outflow facility (P = 0.01). Although these data indicate that MMPs may also regulate basal outflow facility within the conventional outflow pathway, the physiological and pathophysiological role of these enzymes in the regulation of conventional outflow facility in vivo needs additional investigation. 

In vivo studies evaluating adenosine A₁ agonist–meditated changes in total outflow facility have shown that these agents increase total outflow by 70% to 85%. ^{10 11} In the present study, the maximum increase in conventional outflow facility was only 28%. This difference in magnitude between the CHA-induced increase in total outflow facility measured in the in vivo studies and the conventional outflow facility as measured in the perfused bovine anterior segment may reflect species differences. However, MMPs are thought to be major mediators in modulating outflow resistance within the uveoscleral pathway. ^{17 18} Hence, it is tempting to speculate that an adenosine-receptor–induced increase in MMPs within the ciliary body also regulates uveoscleral outflow facility. However, additional studies investigating purinergic modulation of MMP secretion and activation within the ciliary body and uveoscleral outflow are needed to make this determination. 

In summary, these data provide functional evidence that CHA, an adenosine A₁ agonist, can increase conventional outflow facility. The inhibitory actions of CPT and GM-6001 provide evidence that this increase in facility is due to the stimulation of adenosine A₁ receptors and involves the secretion and activation of MMPs from cells within the conventional outflow pathway. 

 Supported in part by National Eye Institute Grants EY09741 (CEC) and EY14793 and an unrestricted grant to from Research to Prevent Blindness (SEI).

 Submitted for publication April 1, 2005; revised May 31, 2005; accepted August 2, 2005.

 Disclosure: C.E. Crosson, None; C.F. Sloan, None; P.W. Yates, 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.

 

Effect of administration of CHA on outflow facility and perfused bovine anterior segments. Isolated anterior segments were perfused overnight (14–16 hours) before the beginning of facility measurements. (A) The response to controlled media exchange. (B) The response to media exchange containing CHA (10 μM). Note the delayed response to administration of CHA and the subsequent slow increase in outflow facility.

Figure 2.

 

Mean changes in outflow facility induced by administration of CHA. Individual outflow facilities were normalized to baseline levels (T = 0) and calculated as the percentage change from baseline. Bovine anterior segments were perfused overnight and the perfusion medium exchanged with control medium or medium containing CHA (10 μM). Data are the mean ± SE. *Significant difference (P < 0.05) from control results (n = 5–10).

Figure 3.

 

Concentration-dependent increases in outflow facility induced by CHA. Individual outflow facilities were normalized to baseline levels (t = 0) and calculated as the percentage change from baseline. Data are the mean ± SE of normalized facility measurements at 3 hours after administration of CHA (n = 3–5).

Figure 4.

 

Inhibition of outflow facility response by the adenosine A₁ antagonist CPT. Bovine anterior segments were perfused with control medium or medium containing CPT (10 μM) overnight (14–16 hours). Media were then exchanged with medium containing CHA (1 μM) in the presence or absence of CPT. Individual outflow facilities were normalized to baseline levels (t = 0) and calculated as the percentage change from baseline. Data are the mean ± SE of normalized facility measurements at 3 hours after CHA administration. *Significant difference (P < 0.05) from corresponding control results (n = 4–10).

Figure 5.

 

Mean changes in outflow facility induced by CHA administration in the presence or absence of GM-6001 (10 μM). Bovine anterior segments were perfused overnight with control medium or medium containing GM-6001. Media were then exchanged with medium containing CHA (1 μM) in the presence or absence of GM-6001. The effect of media exchange on the GM-6001-treated preparation was also determined. Individual outflow facilities were normalized to baseline levels (t = 0) and calculated as the percentage change from baseline. Data are the mean ± SE. *Significant difference (P < 0.05) from data for preparations perfused only with GM-6001 (n = 5–7).