Nitric Oxide Attenuates α₂-Adrenergic Receptors by ADP-ribosylation of G_iα in Ciliary Epithelium

Abstract

purpose. To determine the mechanism by which nitric oxide (NO) regulatesα ₂-adrenergic receptor coupling to adenylyl cyclase in bovine ciliary epithelium.

methods. Ciliary epithelial explants were dissected, cultured, and labeled with[ ³H]adenine. [³H]Adenosine 3′,5′-cyclic monophosphate (cAMP) was measured under basal conditions and after exposure to forskolin, isoproterenol, clonidine, yohimbine, pertussis toxin, and the NO donor spermine-NO. Endogenous and NO-stimulated ADP-ribosylation of ciliary epithelial membrane proteins was determined by [³²P]nicotinamide adenosine diphosphate (NAD) labeling and autoradiography. The three isoforms of the G_iα protein subunit were evaluated by Western blot analysis.

results. Basal [³H]cAMP content was 13.4 ± 1.3 picomoles/mg protein (SEM). Both isoproterenol and forskolin stimulated[ ³H]cAMP accumulation to 36.0 ± 3.9 and 73.2 ± 17.5 picomoles/mg protein, respectively. Clonidine did not affect basal [³H]cAMP levels, but attenuated both isoproterenol- and forskolin-mediated [³H]cAMP accumulation to 23.2 ± 4.4 and 31.6 ± 4.6 picomoles/mg protein, respectively. Yohimbine antagonized the clonidine-mediated adenylyl cyclase inhibition. Pertussis toxin blocked the effect of clonidine. In the presence of the NO donor spermine-NO, the clonidine-mediated inhibition of forskolin- and isoproterenol-stimulated cAMP accumulation was attenuated completely. NO significantly stimulated endogenous[ ³²P]ADP-ribosylation of a 40-kDa membrane protein. Western blot analysis with specific antibodies revealed expression of all three G_i subtypes—G_i1α, G_i2α, and G_i3α—in bovine ciliary epithelium.

conclusions. NO attenuates α₂-adrenergic receptor–mediated inhibition of adenylyl cyclase in ciliary epithelium through ADP-ribosylation of the G_iα subunit. The findings demonstrate heterologous regulation between the NO and cAMP signaling pathways in ciliary epithelium.

The ciliary body is a complex tissue composed of smooth muscle, blood vessels, nerve terminals, and a uniquely organized epithelial bilayer with pigmented and nonpigmented cells. This organ is responsible for two important physiological functions: accommodation and regulation of aqueous humor secretion. Transmembrane signaling pathways involving both adenylyl cyclase and phospholipase C have been examined, and the pharmacologic effects of drugs on ciliary body functions are well known. For example, the cholinergic muscarinic agents couple to phospholipase C in ciliary smooth muscle ¹ and mediate accommodation by regulating ciliary smooth muscle tone. ² Drugs that decrease adenylyl cyclase activity, such as β-adrenergic receptor (AR) antagonists andα ₂-AR agonists, lower intraocular pressure (IOP) by decreasing aqueous humor flow, presumably at the level of the ciliary epithelium. ^{3 4} More recently, β-nicotinamide adenine dinucleotide phosphate-diaphorase (NADPH-d) immunoreactivity has been identified in the ciliary smooth muscle, ⁵ trabecular meshwork, ⁵ ciliary body capillaries, ⁶ and ciliary muscle ganglion cells. ⁶ In addition, nitric oxide synthase (NOS) immunoreactivity has been demonstrated in nonpigmented ciliary epithelium. ^{5 6} It has been proposed that the physiological role of nitric oxide (NO) in the ciliary body is primarily in relaxing smooth muscle tone ⁷ and regulating blood flow. ⁸ Although we appreciate the physiological roles within the ciliary body of these three distinct signaling pathways—that is, adenylyl cyclase, phospholipase C, and NOS, as elucidated by reductionist experimental approaches—the complexity of the heterologous regulation between these different pathways is not known. 

The best known biochemical mechanism of NO signaling is activation of soluble guanylyl cyclase with subsequent increase in intracellular cyclic guanosine monophosphate (cGMP). ⁹ Another less appreciated consequence of NO formation is enhancement of endogenous adenosine diphosphate (ADP)-ribosyltransferases. ^{10 11 12} These enzymes catalyze the transfer of ADP-ribose from nicotinamide adenosine diphosphate (NAD) to an acceptor protein. ¹³ Endogenous ADP-ribosylation has been observed in different tissues, and several ADP-ribose acceptor proteins, such as elongation factor 2, G_iα, G_sα, and rho p21 G protein, have been identified. ¹⁴ Exogenous ADP-ribosyltransferases have been thoroughly studied because of the recognition that some bacterial exotoxins stimulate ADP-ribosylation of specific cellular target proteins. ¹⁵ For instance, the α subunit of the G proteins G_i and G_o are ADP-ribosylated by pertussis toxin, which leads to functional inactivation and uncoupling between cell surface receptors and their intracellular effectors. Cholera toxin causes ADP-ribosylation of G_sα, which leads to constitutive activation and coupling to adenylyl cyclase with subsequent increase in intracellular cyclic adenosine monophosphate (cAMP). Both endogenous and exogenous ADP-ribosyltransferases are potential modulators of receptor signal transduction. The purpose of the present study was to determine whether NO regulates theα ₂-AR–mediated inhibition of adenylyl cyclase in the bovine ciliary epithelial explant. Our findings suggest that NO disinhibits α₂-ARs coupled to adenylyl cyclase in ciliary epithelium through ADP-ribosylation of the G_iα subunit. 

Materials and Methods

Materials

Yohimbine, clonidine, 3-isobutyl-1-methylxanthine (IBMX), spermine-nitric oxide (SPER/NO), and pertussis toxin were obtained from Research Biochemicals International (Natick, MA). Isoproterenol, forskolin, SPER, Ca²⁺- and Mg²⁺-free Hanks’ balanced salt solution (HBSS), Ca²⁺- and Mg²⁺-containing HBSS, and Dulbecco’s modified Eagle’s medium (DMEM) with both normal (1.0 g/l) and high (4.5 g/l) glucose content were purchased from Sigma-Aldrich (St. Louis, MO). [³H]Adenine and[ ³²P]NAD were purchased from Amersham (Arlington Heights, IL), [³H]cAMP from NEN Life Science Products, Inc. (Boston, MA), and G50wx4 Dowex resin (100–200 mesh) from Bio-Rad Laboratories (Hercules, CA). Anti-G_i1α and anti-G_i3α antibodies were from Calbiochem (San Diego, CA), and anti-G_i2α antibody was from Santa Cruz Biotechnology (Santa Cruz, CA). 

Tissue Preparation

Bovine eyes were obtained from Wolverine Packing Company (Detroit, MI) within 3 hours after death. The tissue was prepared according to a procedure established previously. ¹⁶ After the trypsin-mediated dissection, the isolated epithelial sheets were cultured in normal glucose DMEM at 37°C in a 5% CO₂ incubator. The experiments were performed on epithelial explants cultured for 24 to 96 hours after dissection. These freely floating explants are predominantly nonpigmented epithelial cells with some pigmented cells, but without other underlying connective tissue of the ciliary body. ¹⁶  

Whole-Cell Adenylyl Cyclase Assay

Forty-eight hours after the epithelial sheets were dissected, 6-mm explants were punched from the sheets and placed into individual wells of a 96-well tissue culture plate. The explants were labeled overnight with 2 μCi of [³H]adenine in 100μ l HBSS with Ca²⁺, Mg²⁺, and 10% fetal bovine serum (FBS). The unincorporated radioactivity was aspirated and the explants were washed with HBSS with Ca²⁺ and Mg²⁺, but without FBS. The labeled explants were incubated with 100 μl DMEM (normal glucose) containing 0.1 mM IBMX at room temperature for 15 minutes. Various drugs were added to the wells at a 25-μl volume for a total incubation volume of 125 μl, and the plate was placed at 37°C in a 5% CO₂ incubator for 30 minutes. The reaction was terminated with 100 μl blocking solution (5% trichloroacetic acid, 100 μM adenosine triphosphate [ATP], and 100 μM cAMP), and the samples were transferred to polypropylene test tubes. An additional 900 μl blocking solution was added to each tube and the explants were homogenized (Tissumizer; Tekmar, Cincinnati, OH). The levels of[ ³H]cAMP accumulation were measured using standard sequential Dowex/alumina ion-exchange column chromatography. ^{17 18} Column recovery was determined by separate columns without samples by using[ ³H]cAMP tracer and was determined to be 42%. 

ADP-ribosylation Assay

The assays were performed on membrane fractions of both ciliary body and ciliary epithelial explants. To prepare the membranes, the tissue was homogenized in 100 mM Tris (pH 7.4); 0.1 mM phenylmethylsulfonyl fluoride (PMSF); 10 μg/ml each of aprotinin, leupeptin, and trypsin inhibitor; 50 μg/ml benzamidine; and 10 mM dithiothreitol (DTT). This homogenate was centrifuged at 1000g for 10 minutes to remove cellular debris. The resultant supernatant was centrifuged for 1 hour at 100,000g, and the membrane pellet was washed once and suspended in the homogenization buffer. Endogenous ADP-ribosylation was measured in membranes essentially as described earlier. ¹¹ Forty micrograms of membrane protein was incubated in a 50-μl volume of reaction mixture containing 100 mM Tris (pH 7.4), 2 mM MgCl₂, and 10 mM DTT and, when desired, 2 mM SPER/NO. The reactions were initiated with the addition of [³²P]NAD to a final concentration of 0.6 μM and blocked after 3 hours at 30°C with the addition of electrophoresis sample buffer. The samples were electrophoresed in 15% SDS-polyacrylamide gels, ¹⁹ proteins were visualized by staining with Coomassie blue, and the gels were dried and exposed to x-ray film (XAR-5; Eastman Kodak, Rochester, NY) to detect labeled proteins. 

Western Blot Analysis with Anti-G_iα Antibodies

Membrane proteins (40 μg) from ciliary bodies and ciliary epithelial explants were electrophoresed in 15% SDS-polyacrylamide gel in several pairs of lanes. The proteins were then transferred to nitrocellulose membrane. One pair of lanes on the nitrocellulose strip was stained with amido black, and others were used for reaction with polyclonal antibodies specific for G_i1α (1:500 dilution), G_i2α (1:200 dilution), and G_i3α (1:1000 dilution). Nitrocellulose strips were incubated with the primary antibodies for 2 hours at room temperature, followed by washing, and incubation with peroxidase-coupled anti-rabbit goat IgG for 1 hour at room temperature. After another wash, the nitrocellulose strips were developed for peroxidase reaction using 4-chloro 1-naphthol as the substrate. 

Protein Assay

Protein content was measured with a protein assay (Bio-Rad Laboratories), according to the manufacturer’s instructions. 

Data Analysis

Data generated in [³H]cAMP accumulation were analyzed by computer (Prism; GraphPad Software, Inc., San Diego, CA). Comparisons between two experimental conditions were made using the unpaired Student’s t-test. Comparisons between various drug treatments were made by ANOVA with Tukey’s multiple-comparison test at P < 0.05. All data are presented as the mean ± SEM. 

Results

The viability and integrity of the dissected bovine ciliary epithelial explants have been previously demonstrated by trypan blue staining and histology. ¹⁶ To evaluate receptor-mediated pathways coupled to adenylyl cyclase,[ ³H]adenine was added to bovine ciliary epithelial explants at various time points after dissection, and its conversion to [³H]cAMP was determined. Labeling with [³H]adenine was initiated at 0, 24, 48, and 72 hours after dissection. After labeling for 24 hours, some of the explants were stimulated for 30 minutes with isoproterenol (3 μM), which corresponded to 24, 48, 72, and 96 hours after dissection. As shown in Figure 1 , basal [³H]cAMP accumulation varied, with the highest level found in explants labeled immediately after dissection ([³H]adenine added at 0 hours and[ ³H]cAMP measured at 24 hours) and the lowest level in those labeled beginning at 72 hours after dissection. The largest isoproterenol-stimulated increase in[ ³H]cAMP accumulation was observed in explants labeled 48 hours after dissection and stimulated at 72 hours. All subsequent whole-cell adenylyl cyclase experiments were conducted with 24-hour [³H]adenine labeling of the explants beginning at 48 hours after dissection, and then the effects of drugs were tested for 30 minutes. 

To examine the α₂-AR transmembrane signaling pathway, the effect of clonidine, an α₂-AR agonist, was examined on the[ ³H]adenine-labeled epithelial explants. As shown in Figure 2 , clonidine (10 μM) had no effect on basal[ ³H]cAMP accumulation (15.8 ± 1.6 picomoles/mg protein), but decreased the isoproterenol-mediated[ ³H]cAMP accumulation by 36% (36.0 ± 3.9 picomoles/mg protein with isoproterenol and 23.2 ± 4.4 picomoles/mg protein with isoproterenol and clonidine). Similarly, clonidine decreased the forskolin-stimulated (10 μM) accumulation by 57% (73.2 ± 17.5 picomoles/mg protein with forskolin and 31.6 ± 4.6 picomoles/mg protein with forskolin and clonidine). The α₂-AR antagonist, yohimbine (1 μM) antagonized the effect of clonidine on both the forskolin- and isoproterenol-stimulated [³H]cAMP accumulation. These results demonstrate that the α₂-AR pathway regulates adenylyl cyclase activated by either forskolin or isoproterenol in the isolated bovine epithelial explant. 

Pertussis toxin was used to determine the involvement of G_iα subunit in theα ₂-AR–mediated inhibition of adenylyl cyclase in our organ culture system. Basal levels of[ ³H]cAMP were unchanged (14.6 ± 0.9 picomoles/mg protein) after an overnight incubation of the ciliary explants with pertussis toxin (100 ng/ml; Fig. 3 ). Pertussis toxin had no effect on forskolin-stimulated[ ³H]cAMP accumulation (45.1 ± 6.4 picomoles/mg protein with forskolin and 41.5 ± 2.2 picomoles/mg protein with forskolin and pertussis toxin). However, pertussis toxin completely blocked the inhibitory effect of clonidine on forskolin-stimulated [³H]cAMP accumulation (26.9 ± 2.7 picomoles/mg protein with forskolin and clonidine and 41.8 ± 5.6 picomoles/mg protein with forskolin, clonidine, and pertussis toxin). This finding demonstrates the presence of a pertussis toxin–sensitive G_iα subunit that couples theα ₂-AR negatively to adenylyl cyclase in the ciliary epithelial explant. 

Using the NO donor SPER/NO (2 mM), we examined the potential role of NO on regulating the inhibitory adenylyl cyclase pathway. SPER/NO had no effect on basal [³H]cAMP accumulation (10.2 ± 1.8 picomoles/mg protein basal and 11.4 ± 0.9 picomoles/mg protein SPER/NO; Fig. 4 ). However, SPER/NO abolished the clonidine-mediated inhibition of both isoproterenol-stimulated [³H]cAMP accumulation (21.6 ± 4.4 picomoles/mg protein with isoproterenol and clonidine versus 40.5 ± 10.9 picomoles/mg protein with isoproterenol, clonidine, and SPER/NO), and clonidine and forskolin-stimulated[ ³H]cAMP accumulation (28.0 ± 4.1 picomoles/mg protein with forskolin and clonidine versus 58.5 ± 15.5 picomoles/mg protein with forskolin, clonidine, and SPER/NO). Spermine alone, without the NO moiety, had no effect on clonidine-mediated inhibition of [³H]cAMP accumulation from either isoproterenol or forskolin (18.8 ± 2.4 picomoles/mg protein for isoproterenol, clonidine, and spermine and 20.6 ± 3.2 picomoles/mg protein for forskolin, clonidine, and spermine). These results suggest that NO from an NO donor uncouples theα ₂-AR-mediated inhibition of adenylyl cyclase, similar to the effect observed with pertussis toxin (Fig. 3) . Because it is well established that pertussis toxin’s effect is mediated by ADP-ribosylation of G_iα, it is possible that NO acts similarly. 

To determine the mechanism for NO-mediated disinhibition ofα ₂-ARs, the effect of NO on endogenous ADP-ribosylation of G_iα was examined in the ciliary epithelial explants. In prior studies, NO was shown to highly activate endogenous ADP-ribosylation of G_iα in bovine ciliary body. ¹¹ Furthermore, endogenous ADP-ribosylation of G_iα in human platelets uncouples epinephrine-mediated inhibition of adenylyl cyclase. ²⁰ Consequently, we investigated the possibility that G_iα is ADP-ribosylated in the epithelial explants, and whether this posttranslational protein modification is influenced by NO. As shown in Figure 5 , a 40-kDa protein was ADP-ribosylated in the membranes of the explants as well as in the membranes prepared from ciliary body. The ADP-ribosylation was greatly enhanced by NO in both cases. Because the ADP-ribosylation of the 40-kDa protein was thoroughly investigated in ciliary body and the protein was identified as G_iα, ¹¹ it is highly likely that the ADP-ribosylated 40-kDa protein in the ciliary explant is also G_iα and that it is equally sensitive to stimulation by NO. 

The G_iα subunit has at least three different subtypes: G_i1α, G_i2α, and G_i3α. ^{21 22} In simian virus (SV)40–transformed human pigmented and nonpigmented ciliary epithelial cell lines, all three members of G_iα were identified by Western blot analysis. ²³ In rabbit ciliary processes, only G_iα types I and III were found, not G_iα type II, by Western blot analysis. ²⁴ In a bovine SV40-transformed pigmented ciliary epithelial cell line, all three G_iα subunits were identified by immunoprecipitation of adenosine A₁-G protein complex using G_iα-subtype–specific antibodies. ²⁵ Because we observed that G_iα coupled α₂-AR to adenylyl cyclase (Fig. 3) , we investigated the expression of the different G_iα subunits in the bovine ciliary explants by probing membranes with antibodies specific for the various G_iα subunits. As shown in Figure 6 , the G_i1α and G_i3α subunits appeared to be more abundant than the G_i2α subunit in the ciliary explants and ciliary body. In addition, all the G_iα subunits were enriched in isolated epithelial membranes compared with membranes of the ciliary body, which contain a heterogeneous source of cell types. 

Discussion

Elevated IOP is a risk factor for glaucomatous optic neuropathy, and at present it is the main treatable risk factor for glaucoma. Consequently, both medical therapy and surgical management of glaucoma have focused on lowering IOP to minimize this risk factor. Although the pharmacology of drugs used to lower IOP is rather well understood, we still cannot explain why responses to these drugs vary in patients. The variable responses may be due to a combination of compliance, environmental factors, genetics, and biological mechanisms. One well-characterized biological mechanism identified for variable ocular drug response for mydriasis is related to pigmentation of the iris. ²⁶ In the present study, our results suggest another biological mechanism that may account for variable IOP response to theα ₂-AR agonists brimonidine and apraclonidine. Our findings suggest that NO enhances endogenous ADP-ribosylation of a 40-kDa membrane-associated protein (Fig. 5) that leads to disinhibition of the α₂-AR to adenylyl cyclase (Fig. 4) similar to pertussis toxin (Fig. 3) . 

Given our results, we propose the following model summarized in Figure 7 . Adenylyl cyclase activity in the ciliary epithelium is regulated by both the β₁- and β₂-AR and α₂-AR G protein–coupled pathways (Fig. 7A) . ^{27 28} Our new contribution to this model is that NO disinhibits the α₂-AR pathway by posttranslational modification of G_iα—specifically by enhancing endogenous ADP-ribosylation (Fig. 7B) . This G_iα-inhibitory pathway is also sensitive to exogenous ADP-ribosylation by pertussis toxin. In rabbits, pertussis toxin treatment attenuated brimonidine-induced lowering of IOP and aqueous humor flow. ²⁹ In addition, this pertussis toxin sensitivity has also been shown to attenuate clonidine-induced lowering of IOP and to diminish the clonidine- and neuropeptide Y–mediated inhibition of adenylyl cyclase activity in rabbits. ³⁰ Hence, this biological regulatory mechanism is physiologically relevant in regulating aqueous humor flow. This posttranslational modification results in elevated intracellular cAMP and potentially in sustained aqueous humor secretion. In our system, we determined that the G_iα subunits were enriched in bovine ciliary epithelium in comparison to ciliary body (Fig. 6) . Based on our present study, we are not able to deduce whether a particular G_iα isoform preferentially couples the activated receptor to adenylyl cyclase or whether a particular isoform is selectively ADP-ribosylated. 

Our findings add evidence to the role of NO in modulating neurotransmission and the to growing list of other physiological roles for NO. Unlike the other cellular signaling molecules, cAMP, cGMP, and inositol 1,4,5-trisphosphate, NO diffuses across biological membranes, which allows NO to participate in both intra- and intercellular signaling pathways. ³¹ Given its labile nature ³² and that it is not compartmentalized by cellular boundaries, ³¹ the regulation of NO relies on synthesis on demand by NOS using the substrate l-arginine. Three distinct forms of NOS have been identified, and of these, two isoforms are considered constitutively expressed, and the third is considered inducible. ³³ The nomenclature for the constitutive forms are NOSI, also known as neuronal NOS or nNOS, and NOSIII, also called endothelial NOS or eNOS. The inducible form, iNOS, has been designated NOSII. The cofactors, the differential regulation of the constitutive versus inducible NOS isoforms by calcium-calmodulin, and transcriptional regulation of NOSII have been well established. ³⁴ NO has been demonstrated to be involved in vasodilation in arterial smooth muscle, ³⁵ cytotoxicity mediated by activated macrophages, ³⁶ attenuation of vascular endothelial proliferation, ³⁷ neurotransmission, ³⁴ and vision. ^{10 38 39} More recently, upregulation of NOSII was demonstrated in the optic nerve heads of patients with glaucoma ⁴⁰ and in rats with experimental glaucoma. ⁴¹ Both of these latter observations suggest a potential role for NO in the pathogenesis of glaucomatous optic neuropathy. 

Our study also supports the potential regulatory role for NO on aqueous humor dynamics. Given the distance between the ciliary body and optic disc and the normal aqueous humor dynamics, it is highly unlikely that NO derived from one of these two regions of the eye has any influence on the other. NO has to be generated locally to influence the physiology of the ciliary processes. Although we used an exogenous NO source in our studies, there is evidence for NO production in ciliary body. In isolated ciliary processes, formation of nitrate, an NO metabolite, is regulated by a cAMP-dependent protein kinase pathway. ^{42 43} High levels of NADPH-d activity, which is an index of NO production capacity, ⁴⁴ and NOS immunoreactivity have been demonstrated in ciliary epithelium. ⁵ NO may also reach the ciliary epithelium by local diffusion from blood vessels ⁶ or conceivably from smooth muscle. ^{6 7 42} In the setting of uveitis, additional sources of NO include activated macrophages and neutrophils that infiltrate the ciliary body. ^{45 46} Hence, there is evidence that the ciliary body has NO produced by both constitutive NOS from endogenous tissues—that is, ciliary smooth muscle and ciliary epithelium—and inducible NOS from invading macrophages and neutrophils in the setting of uveitis. 

In summary, these results demonstrate that our organ culture of isolated bovine epithelial explants may be used to investigate heterologous regulation between transmembrane signaling pathways. In a previous study, we have shown the presence of receptor-mediated phospholipase C, ¹⁶ and in the present study, we demonstrate regulation of adenylyl cyclase activity involving G_sα and G_iα proteins. The new finding that NO disinhibits the α₂-ARs in ciliary epithelium suggests further complexity in the regulation of aqueous humor secretion by transmembrane signaling pathways. This posttranslational modification involving ADP-ribosylation of the G_iα subunit may also account for the variable IOP response in patients to α₂-AR agonists. 

 Supported by National Eye Institute Grant EY00353 (SEM), Core Grants EY07003, EY05230, and EY07158 (AS), and a grant from the National Glaucoma Research programs of the American Health Assistance Foundation (AS).

 Submitted for publication December 13, 2000; revised March 21, 2001; accepted April 3, 2001.

 Commercial relationships policy: N.

 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 [³H]cAMP accumulation under basal conditions and under stimulation with 3 μM isoproterenol. The ciliary epithelial explants were labeled for 24 hours with 2 μCi[ ³H]adenosine at the indicated time points after dissection. Data are mean ± SEM of three experiments performed in quadruplicate. *Significant increase in [³H]cAMP accumulation over basal by unpaired Student’s t-test (P = 0.02).

Figure 2.

 

Effect of 10 clonidine (Cln) on [³H]cAMP accumulation in ciliary epithelial explants. Data are mean ± SEM of four experiments performed in quadruplicate. Clonidine (10 μM) decreased the isoproterenol-mediated (Isop, 3 μM) [³H]cAMP accumulation. Significant forskolin-mediated (Frsk) increase in[ ³H]cAMP accumulation compared with basal (*P < 0.001), and clonidine-mediated decrease in 10 μM forskolin stimulation (**P < 0.05) by ANOVA and Tukey’s multiple-comparison test. Yohimbine (Yoh, 1 μM) antagonized the effect of clonidine.

Figure 3.

 

Effect of pertussis toxin (PTX, 100 ng/ml) on [³H]cAMP accumulation in ciliary epithelial explants. Data are mean ± SEM of two experiments performed in quadruplicate. Significant increase in 10 μM forskolin-stimulated (Frsk) [³H]cAMP accumulation compared with basal (*P < 0.001) and 10 μM clonidine (Cln) inhibition (**P < 0.05) by ANOVA and Tukey’s multiple-comparison test.

Figure 4.

 

Effect of NO on clonidine-mediated inhibition of stimulated[ ³H]cAMP accumulation in ciliary epithelial explants. Data are mean ± SEM of four experiments performed in quadruplicate. Significant increase in [³H]cAMP accumulation compared with basal (*P < 0.01 for Iso and P < 0.001 for Iso + Cln + SP/NO [2 mM], Frsk, and Frsk + Cln + SP/NO) and 10 μM clonidine-mediated inhibition of 10 μM forskolin (**P < 0.05 for Frsk + Cln and P < 0.01 for Frsk + Cln + SP) by ANOVA and Tukey’s multiple-comparison test. Iso, isoproterenol; Cln, clonidine; SP/NO, SPER/NO; forsk, forskolin; and SP, SPER.

Figure 5.

 

Autoradiogram showing effect of NO on endogenous[ ³²P]ADP-ribosylation of 40 μg of membrane proteins from bovine ciliary body (A) and ciliary epithelial sheets (B). In both (A) and (B) lane 1 is control and lane 2 shows membranes treated with 2 mM SPER/NO. The 40-kDa marker is based on the location of molecular weight markers electrophoresed simultaneously.

Figure 6.

 

Western blot analysis of G_iα subunits in membrane proteins (40 μg) from ciliary body (lane 1) and ciliary epithelial sheets (lane 2) using polyclonal antibodies specific for (A) G_i1α, (B) G_i2α, and (C) G_i3α. Nitrocellulose strips were probed with the primary antibodies, incubated with peroxidase-coupled anti-rabbit goat IgG, and developed for peroxidase reaction using 4-chloro 1-naphthol as the substrate.

Figure 7.

 

Model of heterologous regulation between NO and α₂-AR G-protein–coupled pathways. Adenylyl cyclase activity in the ciliary epithelium is regulated in part by the β₁-,β ₂-, and α₂-AR G-protein–coupled pathways (A). The novel contribution of the present study to this model is that NO disinhibits the α₂-AR pathway by posttranslational modification of G_iα, specifically by ADP-ribosylation (B). The potential sources of NO are endogenous from the ciliary epithelium (i.e., constitutive NOSI and -III) and possibly exogenous from ciliary smooth muscle or vascular endothelium.

 

The authors thank Nikolay Pozdnyakov for performing the ADP-ribosylation and Western blot analysis experiments.