Inhibition by Brimonidine of Forskolin-Induced Nitrite Production in Isolated Pig Ciliary Processes

Abstract

purpose. To investigate by which mechanism the ocular hypotensive drug brimonidine (selective α₂-adrenoreceptor agonist) inhibits the production of nitrite induced by forskolin in isolated porcine ciliary processes.

methods. Nitrite (a nitric oxide metabolite) was measured by Griess reaction in the medium surrounding the ciliary processes, before and after exposure to different drugs. Tissues were exposed for 120 minutes to forskolin (0.1 μM; an adenylylcyclase activator) or 8-bromo-cAMP (10 μM; a cAMP analogue). Some experiments were conducted in the presence of brimonidine (0.01–10 μM), yohimbine (0.1–10 μM; α₂-adrenoreceptor antagonist), prazosin (10 μM; α₁-adrenoreceptor antagonist), nicergoline (10 μM; an α₁-adrenoreceptor antagonist), propranolol (10 μM; a β-adrenoreceptor antagonist or β-blocker), and/or pertussis toxin (2 μg/mL; PTX, a G_i-protein inhibitor).

results. Nitrite production induced by forskolin (133% ± 6%), but not that induced by 8-bromo-cAMP (133% ± 6%), was inhibited in a concentration-dependent manner by brimonidine (10 μM: 103% ± 4%, P < 0.001; EC₅₀: 0.05 μM). The inhibitory effect of brimonidine was prevented by PTX (119% ± 7%, P < 0.01) and, in a concentration-dependent manner, by yohimbine (10 μM: 134% ± 9%; P < 0.01), but not by prazosin, nicergoline, or propranolol.

conclusions. Reduction of the formation of aqueous humor in the ciliary body’s epithelium (and thus an intraocular pressure decrease) after α₂-adrenergic receptor stimulation by brimonidine is known to be associated with a G_i-protein-mediated inhibition of adenylylcyclase activity. The present study indicates that, through a similar α₂-adrenoreceptor/G_i-protein pathway, brimonidine can also inhibit nitrite production after adenylylcyclase activation (forskolin-induced) in isolated porcine ciliary processes.

Brimonidine is a selective α₂-adrenoreceptor agonist that decreases intraocular pressure in part by reducing aqueous humor production by ciliary processes. ¹ In ciliary processes, the reduction of aqueous humor production induced by brimonidine is associated with inhibition of G_i-protein–mediated adenylylcyclase activity, which leads to a decrease in cyclic adenosine-3′,5′ monophosphate (cAMP) production. ²  

Nitric oxide (NO) is a cellular messenger. In some organs, NO is involved in transepithelial fluid transport associated with transmembrane ionic currents (e.g., lung, kidney, and gut). ^{3 4 5} In the epithelium of porcine ciliary processes, where aqueous humor is produced, NO induces depolarization of the membrane potential, reflecting indirectly the ionic transmembrane currents. ⁶ Furthermore, in porcine ciliary processes, the β-adrenoreceptor antagonist (β-blocker) propranolol, which also decreases formation of aqueous humor, inhibits NO production induced by isoproterenol (a β-adrenoreceptor agonist), but not that induced by forskolin (an adenylylcyclase activator), or 8-bromo-cAMP (a membrane-permeable cAMP analogue). ⁷ These observations suggest that, as in some organs involved in transepithelial fluid transport, NO may increase aqueous humor production in ciliary processes. 

In the present study, we investigated whether, in isolated porcine ciliary processes, brimonidine inhibits the forskolin-induced production of nitrite (a stable metabolite of NO), and whether this inhibition can, in particular, be reversed by yohimbine (an α₂-adrenoreceptor antagonist) or by pertussis toxin (PTX; a G_i-protein inhibitor). 

Materials and Methods

Tissue Preparation

In adherence to the provisions of the ARVO Statement for the Use of Animals in Ophthalmology and Vision Research porcine eyes were obtained from a slaughterhouse immediately after death and transported to the laboratory in ice-cold Krebs-Ringer’s physiologic solution. Within 3 hours after death of the animal, the ciliary processes were dissected in cold Krebs-Ringer’s solution. Tissues were then stored at −70°C until nitrite measurements were conducted. 

Experimental Procedures

As previously described, ^{7 8} frozen tissues were thawed at room temperature and washed with Hanks’ balanced salt solution (HBSS). Ciliary processes were placed in a 24-well plate (ciliary processes from one eye/well), covered with 700 μL HBSS-l-arginine medium (HBSS supplemented with 10 μM l-arginine), and placed in a humidified incubator (37°C). Thirty minutes later, 100 μL of medium was taken from each well for a first nitrite measurement. After the first sampling, tissues were either left quiescent or incubated with different adrenoreceptors antagonists and/or agonists for 30 minutes. Then, tissues were either exposed to forskolin or 8-bromo-cAMP for 120 minutes before a second sample of 100 μL of medium was collected from each well for a second nitrite measurement. For each set of experiments, control experiments, in which tissues were not exposed to any tested drugs, were run in parallel. 

Griess Reaction

Nitrite’s concentrations were measured by Griess reaction. ⁹ In brief, ^{7 8} samples were centrifuged at 1000g for 15 minutes. From each sample, 45 μL of the supernatant was mixed with 45 μL of Griess reagent and incubated at room temperature for 10 minutes. Optical density was then measured on a microplate reader at 540 nm. Concentrations of nitrite were determined by comparisons with a standard sodium nitrite curve. 

Drugs

l-Arginine, 8-bromo-cAMP, brimonidine, dimethyl sulfoxide (DMSO), forskolin, dl-propranolol, nicergoline, pertussis toxin, prazosin, sodium nitrite, and yohimbine were purchased from Sigma (Buchs, Switzerland); Griess reagent from Merck (Darmstadt, Germany); and HBSS from Gibco BRL Life Technologies (Basel, Switzerland). The day of the experiments, drugs were prepared in HBSS, except brimonidine and forskolin, which were dissolved in DMSO (maximum DMSO concentration in the medium, 1%). 

Statistical Analysis

The level of nitrite production (the difference between the first and the second concentration of nitrite measured) was expressed as a percentage of the mean nitrite production in the quiescent control experiments run in parallel with each set of study experiments. The concentration that induced 50% of the maximal inhibition (EC₅₀) induced by 10 μM brimonidine was calculated with a four-parameter, logistic function, curve-fitting model. Results are expressed as the mean ± SEM with n corresponding to the number of eyes assessed. For each set of experiments (each of them shown in a different figure) a separate statistical analysis of the data was conducted by one-way ANOVA multiple-comparison (with Bonferroni correction) with P < 0.05 considered to be significant. 

Results

Nitrite Production

In isolated porcine ciliary processes, the basal nitrite production (7.5 ± 0.6 μg/mg of tissue per hour; n = 18) was significantly increased after exposure to 0.1 μM forskolin (139% ± 10%, P < 0.01; Fig. 1 ). The production induced by forskolin was not significantly affected by propranolol (10 μM; β-blocker) and/or prazosin (10 μM; α₁-adrenoreceptor antagonist). These results confirm ⁷ that at the concentration of forskolin used, it is unlikely that the increase in nitrite production is due to a prejunctional release of the α- and β-adrenoreceptor agonist norepinephrine, because it has been observed at higher concentrations of forskolin in the rabbit iris-ciliary body. ¹⁰ If this were the case, the release of norepinephrine would ultimately lead to an increase in formation of nitrite that would be inhibited by the β-adrenoreceptor propranolol. ⁸  

Modulation of Forskolin-Induced Nitrite Production

To investigate whether the α₂-adrenoreceptor agonist brimonidine can modulate nitrite production, ciliary processes were exposed to forskolin (0.1 μM) in the absence or presence of increasing concentrations (0.01–10 μM) of brimonidine (Fig. 2) . In the presence of brimonidine, this increase (133% ± 6%) was significantly blunted (10 μM, 103% ± 4%, P < 0.001) in a concentration-dependent manner (EC₅₀, 0.05 μM). Basal nitrite production was not significantly affected by 10 μM brimonidine (108% ± 7%; P = 0.37). These results indicate that the α₂-adrenoreceptor agonist brimonidine can inhibit the forskolin-induced production of nitrite in isolated porcine ciliary processes. 

Inhibition of Effect of Brimonidine on Forskolin-Induced Nitrite Production

To evaluate whether the inhibitory effect of brimonidine on the forskolin-induced production of nitrite involves the activation of an α₂-adrenoreceptor, the following experiments were conducted (Fig. 3) . Ciliary processes were exposed to forskolin (0.1 μM) alone, or to both forskolin and brimonidine (10 μM) in the absence or in the presence of prazosin (10 μM; an α₁-adrenoreceptor antagonist), nicergoline (10 μM; a selective α₁-adrenoreceptor antagonist), or increasing concentrations of yohimbine (0.1–10 μM; an α₂-adrenoreceptor antagonist). 

Forskolin-induced nitrite production (136% ± 10%) was again significantly inhibited by brimonidine (117% ± 7%, P < 0.05). The inhibition by brimonidine was reversed in a concentration-dependent manner by yohimbine (10 μM, 134% ± 9%, P < 0.01; Fig. 3 , top). 

Such an effect was not observed with prazosin (113% ± 7%) or nicergoline (112% ± 4%, Fig. 3 , bottom). It should be mentioned that yohimbine alone did not significantly affect the basal production level of nitrite (data not shown). The results shown in Figure 3 suggest that the inhibitory effect of brimonidine on forskolin-induced nitrite production is probably mediated by the activation of an α₂-adrenoreceptor in porcine ciliary processes. 

Effect of PTX on Forskolin-Induced Nitrite Production

To further assess whether a G_i-protein mediates the inhibitory effect of brimonidine on the increase in forskolin-induced nitrite production, experiments were also performed with the G_i-protein inhibitor PTX (Fig. 4) . In this set of experiments, after the first nitrite sampling, preparations were either incubated with PTX (2 μg/mL) or left quiescent for 3 hours. Then, according to the same experimental protocol used in all other experiments, ciliary processes were exposed to forskolin for 120 minutes (0.1 μM) in the presence or absence of brimonidine (10 μM, incubated for 30 minutes). In the absence of PTX, the increase in forskolin-induced nitrite production (128% ± 6%) was again significantly (P < 0.001) inhibited by brimonidine (94% ± 4%). The inhibitory effect of brimonidine on forskolin-induced production was prevented by PTX (119% ± 7%, P < 0.01). In the presence of brimonidine and PTX, or in the absence of these two drugs, forskolin-induced nitrite production did not significantly (P = 0.21) differ. PTX had no significant effect on basal nitrite production (101% ± 5%; P = 0.74). These data indicate that a PTX-sensitive G_i-protein probably mediates the inhibitory effect of brimonidine on the production of nitrite induced by forskolin in porcine ciliary processes. 

8-Bromo-cAMP–Induced Nitrite Production

Forskolin stimulates adenylylcyclase and consequently increases cAMP production. To show that brimonidine does not interfere with the production of nitrite induced by cAMP (Fig. 5) , ciliary processes were exposed to the membrane-permeable cAMP analogue, 8-bromo-cAMP (10 μM), in the presence or absence of brimonidine (10 μM). In agreement with previously published data, ⁷ 8-bromo-cAMP (10 μM) significantly (P < 0.001) increased nitrite production (133% ± 6%). The increase in nitrite production induced by 8-bromo-cAMP was not significantly (P = 0.40) affected by the presence of brimonidine (124% ± 8%). These results indicate that brimonidine probably does not inhibit forskolin-induced nitrite production by interfering with cAMP or one of its by-products. 

Discussion

In the present study, in isolated porcine ciliary processes, the ocular hypotensive α₂-adrenoreceptor agonist brimonidine inhibited the forskolin-induced production of nitrite through a pathway that apparently involves an α₂-adrenoreceptor and a G_i-protein. 

The mode of action by which some ocular hypotensive drugs, such as α₂-adrenoreceptor agonists or β-blockers, inhibit production of aqueous humor in the ciliary processes is still unclear. It has recently been reported that the β-blocker propranolol inhibits NO production in isolated porcine ciliary processes, thus suggesting that NO plays a role in the modulation of aqueous humor production. ^{7 8} This hypothesis is supported to a certain extent by a series of additional observations. Immunohistochemistry studies have revealed the presence of marked NO synthase staining, both in the pig and in the rabbit ciliary epithelium (where aqueous humor is produced). ^{11 12} By electrophysiology, NO donors were able to induce, in porcine ciliary epithelium, a membrane potential depolarization (indicating ciliary epithelium transmembrane ionic currents) that could be blocked by anion channel inhibitors (suggesting the possible presence of an NO-mediated transepithelial chloride transport). ⁶ Finally, a decrease in intraocular pressure has been reported in rabbits after systemic administration of an inhibitor of the formation of NO. ¹³ The present study indicating that the ocular hypotensive drug brimonidine can inhibit the production of nitrite (a metabolite of NO) is in agreement with the conjecture made that, in ciliary processes, a decrease in NO concentration could lead to a decrease in formation of aqueous humor. 

Not only in ciliary processes, but also in other tissues, NO has been reported to modulate transepithelial fluid transport (e.g., lung epithelium, ³ kidney, ⁴ colon, ⁵ salivary acinar cells, ¹⁴ and brain choroid plexus ¹⁵ ). In addition, reports have indicated that NO can mediate the modulation of epithelial chloride transport induced by adrenergic agents. ^{16 17} For example, it has been shown that, in the trachea, NO plays a role in isoproterenol-mediated chloride secretion, ¹⁶ or that clonidine (α₂-adrenoreceptor agonist) regulates NO synthesis in the kidney’s loop of Henle, and, consequently, chloride transepithelial transport. ¹⁷  

It is well known that the reduction in formation of aqueous humor induced by brimonidine after activation of the α₂-adrenoreceptor is associated with a G_i-protein–mediated inhibition of adenylylcyclase activity in rabbits, ² and that α₂-adrenoreceptor stimulation inhibits forskolin-induced cAMP production in human ciliary processes. ¹⁸ The present study, conducted in isolated porcine ciliary processes, shows that the inhibition of nitrite production induced by brimonidine involved a similar pathway. Indeed, the effect of brimonidine was prevented by the α₂-adrenoreceptor antagonist yohimbine but was unaffected by the α₁-adrenoreceptor antagonists prazosin and nicergoline. In addition, the effect of brimonidine was abolished in the presence of the inhibitor of G_i-protein, PTX. 

The results of the present study complement and support other observations indicating the potential role played by cAMP in the modulation of ciliary NO production. Indeed, isoproterenol and forskolin, which are known to activate adenylylcyclase, were, in isolated porcine ciliary processes, not only able to increase the production of cAMP, but also that of nitrite. ⁷ Furthermore, another study has shown that the increase in nitrite production is prevented either by the nitric oxide synthase (NOS) inhibitor l-NAME or by the cAMP-dependent protein kinase (PKA) inhibitor KT 5720, suggesting that cAMP increases NO production through the activation of a PKA and a NOS in porcine ciliary processes. ⁷ Therefore, the observation made that brimonidine is able to inhibit forskolin-induced nitrite production and that this effect can be prevented by PTX further supports the role played by cAMP in the modulation of NO production in porcine ciliary processes. 

It also should be mentioned that brimonidine inhibited only forskolin- and not 8-bromo-cAMP–induced nitrite production, further indicating that the effect of brimonidine is not caused by an interaction of brimonidine with cAMP (or one of its by-products) but rather by a direct effect on adenylylcyclase. In addition it is also unlikely that the effect of brimonidine on forskolin-induced nitrite production would be the result of the activation of a phosphodiesterase (PDE) leading through this pathway to a decrease in cAMP concentration. Indeed, if brimonidine inhibits forskolin-induced nitrite production by activating a PDE, then the inhibitory effect of brimonidine on forskolin-induced nitrite production would probably not have been prevented by the G_i protein inhibitor PTX. This assumption is supported by another observation made in the isolated rabbit iris-ciliary body where it has been shown that brimonidine decreases cAMP production through the activation of G_i protein, which inhibits adenylylcyclase activity. ²  

In the rabbit iris-ciliary body it has also been reported that forskolin can, at a high concentration, evoke a presynaptic release of norepinephrine (α-, β-adrenoreceptor agonist). ⁹ Such a release could lead to an increase in nitrite production through the stimulation of a β-adrenoreceptor. ⁸ It appears that this is not the case at the concentration of forskolin used and in the experiments in the present study. Indeed, forskolin-induced nitrite production was not significantly affected by the presence of the β-blocker propranolol. This finding is in agreement with another observation showing that propranolol has no effect on forskolin- and 8-bromo-cAMP–induced nitrite production in isolated porcine ciliary processes. ⁷  

In conclusion, the mode of action by which the α₂-adrenoreceptor agonist brimonidine inhibits aqueous humor production is still unclear, except that it is associated with a G_i-protein–mediated inhibition of adenylylcyclase activity and an eventual decrease in cAMP production. The present study further indicates that, at least in isolated porcine ciliary processes, brimonidine is also able to inhibit nitrite production through a pathway that involves an α₂-adrenoreceptor and activation of G_i-protein. These findings are in agreement with other observations supporting the hypothesis that NO could be involved in the modulation of transepithelial fluid transport ^{3 4 5 14 15 16 17} and in particular the formation of aqueous humor. ^{6 7 8 11}  

 Supported by Swiss National Science Foundation Grants 32-52783.97 and 32-61495.00, Bern, Switzerland; Velux Foundation, Zurich, Switzerland; Schwickert Foundation, Basel, Switzerland; and Allergan Inc., Irvine, California.

 Submitted for publication January 22, 2002; revised April 2, 2002; accepted April 12, 2002.

 Commercial relationships policy: F, R (JF, IOH); N (all others).

 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 forskolin on the production of nitrite in isolated porcine ciliary processes. The increase in nitrite production was affected by neither propranolol nor prazosin. Analysis of variance with Bonferroni comparison versus forskolin alone without any antagonists (**P < 0.01).

Figure 2.

 

Inhibitory effect of brimonidine (filled circles) on the nitrite production induced by the forskolin (open circle) in isolated porcine ciliary processes. In a concentration-dependent manner, brimonidine significantly inhibited nitrite production. A four-parameter, logistic function curve-fitting model is shown. Analysis of variance with Bonferroni comparison brimonidine versus no brimonidine (*P < 0.05; ***P < 0.001).

Figure 3.

 

Effect of different α-adrenoreceptor antagonists on the inhibitory effect of brimonidine on forskolin-induced nitrite production in isolated porcine ciliary processes. Top: the inhibitory effect of brimonidine was reversed in a concentration-dependent manner by yohimbine. Bottom: the inhibitory effect of brimonidine was reversed by yohimbine but not by prazosin or nicergoline. Analysis of variance with Bonferroni comparison versus no brimonidine (*P < 0.05, **P < 0.01, ***P < 0.001) and versus forskolin and brimonidine alone (†P < 0.05, ††P < 0.01).

Figure 4.

 

Effect of PTX on the inhibitory effect of brimonidine on forskolin-induced nitrite production in isolated porcine ciliary processes. The presence of PTX prevented the inhibitory effect of brimonidine. Analysis of variance with Bonferroni comparison versus forskolin alone (***P < 0.001) and versus forskolin and brimonidine without PTX (††P < 0.01).

Figure 5.

 

Effect of brimonidine on the production of nitrite induced by 8-bromo-cAMP. Brimonidine had no significant (P = 0.40) effect on 8-bromo-cAMP–induced nitrite production.

The authors thank Andreas Schötzau, BS, for supervision of the statistical analyses.