Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

All subtypes of opioid receptors have been shown to negatively couple to the adenylyl cyclase/cAMP system in various tissues (Childers, 1991; Satoh and Minami, 1995; Law et al., 2000). Studies on isolated membranes from brain tissues and cultures of transformed cells have indicated that -, -, and µ-opioid receptors are coupled to specific GTP-binding proteins (G_i/o) and that receptor activation causes inhibition of adenylyl cyclase, which, in turn, suppresses cAMP formation (Loh and Smith, 1990; Reisine et al., 1996). Lai et al. (1995) reported, however, that the cloned rat ₁-opioid receptor can couple to a pertussis toxin (PTX)-insensitive G protein, G_z, to inhibit adenylyl cyclase. G proteins are characterized mainly by their -subunits, and the existence of all three G_i subtypes has been demonstrated in transformed pigmented and nonpigmented ciliary epithelial cells grown in culture (Wax, 1992). At this point, neither ciliary epithelial cell type has been shown to contain G₀ or G_z. In later studies, G_i types 1 and 3 were also detected in rabbit ciliary processes but no detectable G_i type 2 or G_o was observed (Mittag et al., 1994).

Previous evidence suggests that opioid receptors exist on autonomic nerves within the eye and its adnexa. For example, Trendelenburg (1957) demonstrated that morphine inhibited the contraction elicited by stimulating sympathetic nerves to the cat nictitating membrane in vivo. In keeping with other prejunctionally active inhibitors, morphine was more effective in reducing the size of contractions at low frequencies of electrical stimulation of sympathetic nerves than at high frequencies. Other investigators demonstrated that morphine and met-enkephalinamide produced miosis when injected intracamerally, and this miotic effect was reversed by pretreatment with naloxone (Drago et al., 1980). Thus, in these experiments, the observed miotic effect was attributable, in part, to activation of inhibitory opiate receptors on sympathetic nerve endings. These prior findings are consistent with the presence of -receptors on sympathetic innervation of both extraocular (nictitans) and intraocular (iris) structures.

Recent data from this laboratory have shown that the -opioid receptor agonist bremazocine (BRE) and the -opioid receptor agonist [D-Pen²,D-Pen⁵]-enkephalin cause dose-dependent reductions in intraocular pressure in normal, dark-adapted New Zealand White rabbits (Wang and Potter, 1996; Russell et al., 2000). Moreover, [D-Pen²,D-Pen⁵]-enkephalin was found to inhibit cAMP production and [³H]norepinephrine (NE) release from rabbit iris-ciliary bodies. The present study was conducted to extend the previous in vivo findings of the ocular actions of bremazocine on -opioid receptors (Russell et al., 2000). The specific aim of this investigation was to evaluate the cellular actions associated with activation of -opioid receptors in the iris-ciliary body. The relatively selective -agonist BRE, the relatively selective -opioid receptor antagonist norbinaltorphimine (nor-BNI), and PTX were used to evaluate the effects of BRE on -opioid receptor-mediated events in the iris-ciliary body. The end points used included electrically stimulated [³H]NE release from postganglionic sympathetic neurons and cAMP accumulation in isolated iris-ciliary body of New Zealand White rabbits.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. New Zealand adult, male, albino rabbits (2-4 kg) were maintained in individual cages under reverse cyclic lighting (12-h dark/light) conditions. These lighting conditions were used because indices of increased adrenergic activity (aqueous levels of norepinephrine and cAMP) were found to be higher in the dark than in the light (Yoshitomi et al., 1991; Kiuchi and Gregory, 1992). Animals were sacrificed and iris-ciliary bodies removed to determine the effects of BRE and a specific selective -opioid receptor antagonist, nor-BNI, alone and in combination on 1) [³H]NE overflow from sympathetic nerve endings, and 2) cAMP accumulation. Studies were conducted according to The Association for Research in Vision and Ophthalmology Resolution on the Use of Animals in Research, and all protocols were approved by the Morehouse School of Medicine's Animal Care and Use Committee.

Drugs. Bremazocine hydrochloride (BRE), a relatively selective -opioid receptor agonist, was obtained from Sigma/RBI (Natick, MA). Norbinaltorphimine dihydrochloride (nor-BNI), a relatively selective -opioid receptor antagonist, was obtained from Tocris (Ballwin, MO). PTX was obtained from Sigma (St. Louis, MO). These pharmacological agents were dissolved in nanopure water and were prepared on the day of the experiment.

Prejunctional Activity: [³H]NE Overflow from Sympathetic Nerves in Rabbit Iris-Ciliary Body (ICB). Perfused rabbit ICBs were used to investigate [³H]NE overflow in response to electrical field stimulation, in the presence or absence of BRE, nor-BNI, or PTX. Previous studies have shown that this preparation is useful in determining prejunctional activity of a variety of agonists (Jumblatt et al., 1987). Nor-BNI was used to antagonize the activity of -opioid receptors and PTX was used to inactivate G_i/o proteins.

Postjunctional Activity: cAMP Production in Rabbit ICB. The postjunctional action of the -opioid receptor agonist BRE was evaluated in freshly isolated rabbit ICBs by determining its effect on cAMP accumulation. ICB tissue segments were incubated in indomethacin (2.2 mg/100 ml in modified Earles-Ringer solution; 115 mM NaCl, 5 mM KCl, 1.8 mM CaCl₂, 0.8 mM MgSO₄, 0.9 mM NaH₂PO₄, 25 mM NaHCO₃, 10 mM glucose, 10 mM HEPES, 1 mM sodium ascorbate). Incubation of ICBs was carried out for 30 min at 37°C in a humidified incubator in an oxygenated (95% O₂, 5% CO₂) environment. The tissues were then treated with 3-isobutyl-1-methylxanthine (1 mM, a nonselective phosphodiesterase inhibitor) for an additional 10 min under conditions described above. BRE concentrations (0.01-10 µM), with or without the -adrenoceptor agonist isoproterenol (ISO; Sigma), were added to the tissues at 10-min intervals in the presence or absence of the -opioid receptor antagonist nor-BNI, which was used to confirm the -receptor activity of BRE. ISO was used to activate adenylyl cyclase in order to evaluate the inhibitory actions of BRE. Separate experiments were conducted in which the tissues were incubated with PTX (150 ng/ml, 4 h) to ribosylate G_i prior to a challenge with BRE.

Statistical Analysis. Statistical analyses of the experimental data used analysis of variance or the Student's t test for paired data. Values reported are mean ± S.E. and a probability factor (p) of <0.05 was established as the minimum level for statistical significance.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Prejunctional Activity: NE Overflow from Rabbit ICB. The isolated, perfused rabbit ICB, subjected to electrical field stimulation, was used as the in vitro model to localize a possible peripheral, prejunctional site of action for BRE. The end point of agonist activity in these experiments was the magnitude of concentration-related suppression by BRE of [³H]NE release induced by electrical stimulation. As shown in Figs. 1 and 2, after consecutive electrical stimulations (S2-S4), BRE caused a concentration-related (0.1, 1, and 10 µM) inhibition of [³H]norepinephrine release from ICBs to 77, 57, and 36% of the control response, respectively. Figure 1 shows the magnitude of suppression by BRE in the presence of the electrical stimulations, while Fig. 2 demonstrates the mean percentage of ratios of areas under the curves in four experiments. Only the inhibition induced by BRE at 1 µM (57%) and 10 µM (36%) was statistically different from the control.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Effect of BRE on [3H]NE release from sympathetic nerves of the ICB. This figure shows the magnitude of the stimulation.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Effect of nor-BNI and PTX on BRE-induced suppression of [3H]NE release from sympathetic nerves of the ICB. Each column represents the percentage of means of ratios of areas under the curves for Sx/S1. BRE (0.1, 1.0, and 10.0 µM) caused inhibition of electrically stimulated [3H]NE release of 77, 57, and 36%, respectively (n = 4). Nor-BNI (10.0 µM) antagonized the BRE-induced inhibition of electrically stimulated [3H]NE release (n = 4). When pretreated with PTX, the inhibitory effect of BRE was reduced (n = 5). *p < 0.05 compared with the control.

Postjunctional Activity: cAMP Production from Rabbit ICB. In studies designed to assess -opioid receptor activation on cAMP levels in ICBs, it was observed that the magnitude of the effects of BRE alone were concentration-related, but it was observed that the response was elevation of cAMP levels. Interestingly, the stimulatory effect of BRE alone was inversely related to the concentration of the -agonist. As the concentrations were raised incrementally from 0.01 to 10 µM, BRE alone caused obvious stimulation of cAMP above basal levels. However, the response declined as the concentration increased to 1 µM. BRE (10 µM) caused only a 50% stimulation in cAMP accumulation, whereas at 0.01 µM there was a stimulation of 130% (basal, 149 pmol/mg of protein; 0.01 µM BRE, 342 pmol/mg of protein; 10 µM BRE, 224 pmol/mg of protein; Fig. 3).

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Effect of BRE on cAMP accumulation in the iris-ciliary body. BRE (0.01-10.0 µM) caused significant increases (130, 124, 67, and 50%) in cAMP levels compared with basal. Each column represents the mean of seven experiments. *p < 0.05 compared with basal.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Effect of nor-BNI on BRE-induced increases in cAMP accumulation. When tissues were pretreated with nor-BNI (0.01 µM), the cAMP levels were significantly reduced to lower than basal levels. Nor-BNI (0.01 µM) therefore antagonized the stimulatory effects of BRE on cAMP levels. Each column represents the mean of five experiments. +p < 0.05 compared with basal. *p < 0.05 compared with BRE (0.01 µM).

View larger version (30K): [in this window] [in a new window]   Fig. 5.   Effect of BRE on ISO-induced stimulation of cAMP levels in the iris-ciliary body. ISO (1 µM) caused a 3-fold stimulation in cAMP levels. BRE (0.01-10.0 µM) caused significant inhibitory effects on ISO-stimulated cAMP levels by 32, 51, 59, and 64%, respectively, while BRE (0.0001 µM) caused a potentiation of the ISO-stimulated levels. Each column represents the mean of five experiments. +p < 0.05 compared with basal. *p < 0.05 compared with ISO (1 µM).

View larger version (26K): [in this window] [in a new window]   Fig. 6.   Effect of nor-BNI on BRE-induced suppression of ISO-stimulated cAMP accumulation in the iris-ciliary body. Pretreatment with nor-BNI antagonized the inhibition induced by BRE (10.0 µM). Each column represents the mean of five experiments. *p < 0.05 compared with BRE + ISO. +p < 0.05 compared with ISO (1 µM).

View larger version (27K): [in this window] [in a new window]   Fig. 7.   Effect of PTX on BRE-induced suppression of ISO-stimulated cAMP accumulation in the iris-ciliary body. Pretreatment with PTX (150 ng/ml) significantly reduced the cAMP levels (compared with basal) but followed by treatment with BRE and ISO, the inhibition induced by BRE (1 µM) was abolished. Each column represents the mean of seven experiments. +p < 0.05 compared with basal. *p < 0.05 compared with BRE + ISO.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Inhibition of adenylyl cyclase and suppression of hormone secretion or neurotransmitter release are relatively common biological consequences of agonist occupancy of receptors linked to G_i/o proteins (Limbird, 1988; Majewski et al., 1990). In the experiments described in this study, the isolated iris-ciliary body was used to test for the effects of a -receptor activation by BRE on an index of adenylyl cyclase (cAMP accumulation) activity and release of [³H]norepinephrine from sympathetic nerves. The effects of BRE on cAMP accumulation and norepinephrine release were examined using the isolated iris-ciliary body because the ciliary epithelium is the site of aqueous humor formation. Moreover, the sympathetic innervation of the ICB represents a recognized site of modulation of aqueous flow by the ciliary epithelium. Thus, both the ciliary epithelium and sympathetic nerves in the ICB represent potential sites of action of -agonists. For these reasons, this tissue is a particularly useful model in localizing prejunctional and postjunctional drug effects more precisely at the tissue level and in providing clues to cellular site(s) and mechanism(s) of action. Previously published data from this laboratory had shown that BRE reduces intraocular pressure in rabbits, at least in part, by reducing aqueous humor flow (Russell et al., 2000). Therefore, it was assumed that changes in the dynamics of cAMP formation and NE release in the iris-ciliary body caused by stimulating -opioid receptors with BRE could mediate the suppression of the secretion of aqueous humor and the lowering of intraocular pressure.

The results of this study demonstrated that BRE can act at prejunctional (neuronal) sites in the iris-ciliary body to inhibit the release of [³H]NE. Activation of -opioid receptors has been shown to cause a decrease in voltage-dependent calcium conductance, resulting in suppression of neuronal activity (North, 1993). The ion channel effects produced by opioid agonists are believed to be similar to those produced by ₂-adrenoceptor agonists, thereby leading to diminished transmitter release (Thayer et al., 1987; Kongsamut et al., 1989). Activation of these effector pathways in nerves and other tissues may involve the inhibition of cAMP generation, the suppression of N-type calcium channels through the G_i/G_o proteins, and/or the activation of potassium channels (Sharma et al., 1975).

The inhibition of norepinephrine release in the nerve endings of ICB by BRE is in agreement with published data from other tissues. For example, opioid agonists have generally been shown to inhibit the release of a variety of neurotransmitters, including norepinephrine, dopamine, and acetylcholine (Lambert et al., 1993). Bremazocine and other opioids were found to inhibit sympathetic neurotransmission in bovine iris by acting via prejunctional -opioid receptors (Anderson et al., 1994). The activation of presynaptic - and -opioid receptors on postganglionic sympathetic nerve fibers also resulted in a decrease in sympathetic neurotransmitter release and consequently, the response to sympathetic nerve impulses in isolated cardiovascular tissues of the rabbit (Knoll, 1976; Illes et al., 1985). Experiments using electrically stimulated sympathetic outflow in pithed rabbits have indicated that opioid agonists, like BRE, decreased the action potential-evoked release of norepinephrine and secondarily, lowered blood pressure (Ensinger et al., 1986; Szabo et al., 1986, 1988). Collectively, these data provide additional evidence that -receptor agonists, such as BRE, can act at prejunctional opioid receptors on postganglionic sympathetic endings in the rabbit iris-ciliary body.

Data suggesting the involvement of G proteins in the presynaptic control of acetylcholine release from rat myenteric plexus (Dolezal et al., 1989) and norepinephrine from brain neurons (Allgaier et al., 1989) have been described. To investigate the nature of G proteins linked to -receptors in the ICB, PTX was used to catalyze the transfer of the ADP-ribose moiety of NAD to the active subunit of G_i/G_o protein; this treatment results in disruption of G_i protein function (Ui et al., 1984). The interaction of PTX with G_i proteins is highly specific and leads to inactivation by the ribosylation of G protein subunits. The alteration of BRE-induced suppression of norepinephrine release from the iris-ciliary body by pretreatment with PTX indicates that one or more PTX-sensitive G proteins (G_i/G_o) are involved in mediating the action of BRE.

Evidence from studies in a variety of tissues suggests that the biochemical changes affected by receptors linked to inhibition of adenylyl cyclase may result, in part, from the ability of these receptors to lower cAMP generation in the target cell (Limbird, 1988). The current results also indicate that, at concentrations of 0.01 to 10 µM BRE, activation of -opioid receptors by BRE inhibits ISO-stimulated cAMP levels. However, in the absence of stimulation by ISO, these same concentrations of BRE elevated cAMP levels above basal levels. The stimulation of cAMP levels in the iris-ciliary body by BRE suggests that this -agonist has a dual effect: inhibiting ISO-stimulated cAMP production at high concentrations and, in the absence of ISO, enhancing the cAMP levels, particularly at low concentrations of BRE. A similar response was demonstrated in slices of rat hippocampus where the effect of opioid receptor activation on cAMP level was studied (Dziedzicka-Wasylewska and Przewlocki, 1995) using the -opioid receptor agonist U50,488H as well as morphine (µ-opioid receptor agonist). Thus, it was suggested that the increase in cAMP level might result from direct coupling of -receptors to adenylyl cyclase via a stimulatory G protein (G_s). A role for calcium ions in -receptor activation of adenylyl cyclase was also suggested. Cruciani et al. (1993) also presented evidence to suggest that a subset of opioid receptors may be linked directly to G_s and thereby mediate stimulation by adenylyl cyclase. Another receptor (₂-adrenoceptor) that is negatively coupled to adenyl cyclase also showed a biphasic response when stimulated by an ₂-agonist (Eason et al., 1992). Thus, previous work in other tissues from other groups suggests that ₂-adrenoceptors and opioid receptors couple to both G_i and G_s.

There have been other reports that the -subunits of two molecular species of G_i, G_i-1, and G_i-2A, cause stimulation rather than inhibition of adenylyl cyclase in a reconstituted system (Newton and Klee, 1990). Thus, it is possible that stimulation of adenylyl cyclase by BRE could occur as a result of -receptors linked to multiple G protein-subunit mediated biochemical pathways. It is of interest that all three G_i subtypes (G_i1, G_i2, and G_i3) have been shown to reduce levels of adenylyl cyclase. In contrast, stimulation of adenylyl cyclase has been shown to occur through -subunits of G_i/o (Federman et al., 1992) that synergize with G_s to elevate cAMP levels (Mhaouty-Kodja et al., 1997). Other groups (Tang and Gilman, 1991) have also shown that the -subunit complex of G proteins can differentially affect the activity of different isoforms of adenylyl cyclase. For example, it has been proposed that receptors coupled to pertussis toxin-sensitive G proteins may initiate adenylyl cyclase activity by promoting interaction with the -subunit complex (Tang and Gilman, 1991).

Taken together, the experimental data from the cAMP experiments in this study suggest that -opioid receptors can evoke both stimulatory and inhibitory processes at postjunctional sites such as the ciliary epithelium. The effect of BRE depended on the concentration used and is probably dependent also on the G protein subunit(s) linked to the -opioid receptor. The inhibitory effect observed was demonstrated by way of a PTX-sensitive (G_i/o) protein in the ICB of the rabbit. The augmentation of cAMP levels in the ICB was observed under two experimental situations: at low concentrations of BRE in an ISO-stimulated system and also in the absence of ISO at a variety of concentrations of BRE.

nor-BNI has been shown to be a highly selective -antagonist both in vivo and in receptor binding studies (Portoghese et al., 1987; Takemori et al., 1988). It was shown to be approximately 170 and 150 times more selective for - than for µ- and -receptors, respectively. As a result of the finding that nor-BNI (10 µM) was able to antagonize the suppressive effect of BRE (10 µM) on ISO-stimulated cAMP and also to inhibit the stimulatory effect of BRE in the absence of ISO, these data indicate the possibility that -receptors, present in the ciliary epithelium of the rabbit, are linked to more than one type of G protein and more than one isoform of adenylyl cyclase. Further investigation is required to determine the types of adenylyl cyclases linked to G proteins and -opioid receptors in the iris-ciliary body.

In summary, the data presented here provide additional evidence that the -opioid receptor agonist bremazocine acts on prejunctional (neuronal) and postjunctional (ciliary epithelium) receptors in the ciliary body to modify aqueous humor dynamics. The regulation of adenylyl cyclase(s) at post- and prejunctional sites in the rabbit iris-ciliary body could involve multiple subtypes of -opioid receptors linked to more than one type of G protein and more than one isoform of adenylyl cyclase. Additional studies will be needed to investigate this hypothesis.