Effects of Adrenomedullin on Cyclic AMP Formation and on Relaxation in Iris Sphincter Smooth Muscle

Abstract

purpose. To determine whether iris sphincter and other tissues of the iris-ciliary body secrete adrenomedullin (ADM), a novel hypotensive peptide that is classified into the calcitonin gene–related peptide (CGRP) family and to determine the binding sites for ADM and compare the effects of ADM and CGRP in the absence and presence of their receptor antagonists on cAMP formation and relaxation in the iris sphincter.

methods. Sphincter muscle was incubated in Krebs–Ringer bicarbonate buffer in the absence and presence of ADM for 10 minutes. Accumulation of cAMP in the tissue extract was determined by radioimmunoassay (RIA). The binding of [¹²⁵I]ADM to iris sphincter membranes was carried out by rapid filtration. Distribution of ADM in the ocular tissues was determined by RIA. Changes in muscle tension were recorded isometrically.

results. Immunoreactive ADM was present in all tissues of the cat iris-ciliary body. In the isolated cat iris sphincter, ADM increased cAMP accumulation in a time- (t _1/2 = 2.2 minutes) and concentration- (EC₅₀ = 13 nM) dependent manner, and this effect was sixfold more efficacious than CGRP. ADM, CGRP, vasoactive intestinal peptide, prostaglandin E₂, isoproterenol, and forskolin increased cAMP formation in cat sphincter by 12.5-, 2-, 2.2-, 1-, 2.6-, and 2.4-fold, respectively. The rank of the effects of ADM on cAMP formation in iris sphincter isolated from different animal species was in the following order: cat > dog > bovine > human > rabbit. In the cat iris sphincter, the CGRP antagonist, CGRP(8 to 37), was more effective than the ADM antagonist, ADM (26 to 52), in inhibiting both ADM- and CGRP-induced cAMP formation. ADM and CGRP inhibited carbachol-induced contraction in a concentration-dependent manner with IC₅₀ values of 10 and 90 nM, respectively. Both ADM and CGRP displaced the binding of [¹²⁵I]ADM to sphincter membranes effectively, with IC₅₀ values of 0.81 and 1.15 nM, respectively.

conclusions. In iris sphincter isolated from cat and other mammalian species including human, ADM is a much more efficacious activator of adenylate cyclase and a much more effective relaxant than CGRP. Its biological effects may be due to direct involvement of ADM receptors, but also to activation of CGRP receptors. Activation of ADM receptors by the peptide leads to concentration-dependent increases in cAMP accumulation and subsequent inhibition (relaxation) of smooth muscle contraction. These findings suggest a role for ADM as a local modulator of smooth muscle tone. A possible function for this potent hypotensive peptide in the regulation of intraocular pressure remains to be investigated.

Adrenomedullin (ADM) is a bioactive 52-amino acid peptide that was originally discovered in acid extracts from human pheochromocytoma tissue by monitoring the elevation of cAMP in rat platelets. ¹ ADM shares structural homology with the sensory neuropeptide calcitonin gene–related peptide (CGRP), and like CGRP, it is a potent hypotensive peptide. ² ADM is found to be widely distributed in various tissues and organs, including adrenal medulla, lung, kidney, spleen, and heart. ^{1 2} Furthermore, ADM was found to circulate in blood, ³ which suggests that this peptide is different from CGRP, which functions as a neuropeptide. Several peripheral actions of ADM have been reported, including in vasculature (hypotension), heart (increased coronary blood flow), lung (vasodilation), adrenal gland (inhibition of K ⁺-stimulated aldosterone secretion), kidney (increased renal blood flow), pituitary gland (inhibition of ACTH secretion), and brain (increased cerebral blood flow). ⁴ Cloning of the cDNA encoding the ADM precursor has shown that this gene is highly expressed in the heart, adrenals, kidneys, and lungs of both rats and humans. ⁵ Binding studies have demonstrated that abundant and specific receptors for this peptide are highly present in the heart, lungs, spleen, and kidneys. ⁶ ADM was shown to induce cAMP accumulation in a variety of cultured cells, including smooth muscle cells, ⁷ endothelial cells, ⁸ and glomerular cells, ⁹ and in isolated cardiac myocytes. ¹⁰ These observations suggest that activation of the adenylate cyclase–cAMP system may mediate the physiological functions of ADM. 

There is little information on the receptors and the mechanism involved in the action of ADM in ocular tissues. Recently, Okamura et al, ¹¹ working on the mechanism of ADM-induced relaxation in isolated canine retinal arteries, concluded that endothelium-independent relaxations to ADM of canine arteries may be mediated primarily by intracellular cAMP by stimulation of CGRP receptors and partially by cGMP. More recently, we reported that cAMP mediates the relaxant action of CGRP in rabbit iris dilator. ¹² Although ADM stimulates cAMP synthesis in many tissues, ² the site of its action and signal transduction pathway in ocular tissues are unknown. The objectives of the present study were to investigate (1) whether iris-ciliary body produces ADM, (2) whether cAMP is involved in the relaxant action of ADM on the iris, and (3) whether the action of ADM is mediated by receptors for ADM or CGRP. This is the first report on the characterization of ADM effects in the iris-ciliary body. 

Materials and Methods

Chemicals

Chemicals used were obtained from the following sources: human ADM (1–52), human ADM (26–52), and [¹²⁵I] ADM (1–52) RIA kit from Phoenix Pharmaceuticals, Inc. (Mountain View, CA); CGRP (1–37), CGRP (8–37), and radioiodinated human ADM,[ ¹²⁵I]ADM (1–52) (specific activity of 1258 Ci/mmol) from Peninsula Laboratories, Inc. (Belmont, CA); Endothelin-1 from Peptides International (Louisville, KY); carbachol (CCh), 3-isobutyl-1-methylxanthine (IBMX), indomethacin, isoproterenol, forskolin, and vasoactive intestinal peptide from Sigma (St. Louis, MO); 2′-,5′-dideoxyadenosine (DDA) from Biomol Research Laboratories (Plymouth Meeting, PA); [¹²⁵I] cAMP– and[ ¹²⁵I]cGMP–RIA kits from Amersham (Arlington Heights, IL). All other chemicals used were of reagent grade. 

Preparation of Iris Sphincter

Cat and dog eyes were obtained through the courtesy of Richmond County Animal Control (Augusta, GA). Rabbit and bovine eyes were obtained from local slaughter houses. Human eyes were obtained (approximately 20 to 36 hours after death) from the National Disease Research Interchange (NDRI), and human ciliary smooth muscle (HCSM) cells were a generous gift from Parimal Bhattacherjee (Department of Ophthalmology, University of Louisville, KY) and were harvested from the tissue 15 hours after death, using the method of Weinreb et al. ¹³ In general, we obtain the eyes 1 hour after the animals are killed. Iris sphincters were dissected out from dilator and placed in a modified Krebs–Ringer bicarbonate buffer (KRB, pH 7.4) of the following composition (in mM): NaCl, 118; NaHCO₃, 25; KCl, 4.7; KH₂PO₄, 1.2; MgSO₄, 1.2; CaCl₂, 1.25; and glucose, 10. The pH of the buffer was adjusted and maintained at 7.4 with a mixture of 97% O₂–3% CO₂. The tissues were kept at 4°C and used in the following studies within 1 hour. In general, each sphincter muscle from the same eye was cut into six to eight equal strips. One strip served as a control, and the other as experimental. Indomethacin (1μ M), a cyclooxygenase inhibitor, was added to the incubation medium in all our experiments, to prevent the formation of endogenous prostaglandins. 

The methods for securing animal tissue were humane, included proper approval, and complied with the ARVO statement for the use of Animals in Ophthalmic and Vision Research. 

Assays of cAMP and cGMP

Iris sphincter muscles were incubated in 1 ml Krebs–Ringer bicarbonate buffer for 90 minutes. The medium was discarded and the muscles were then incubated in 1 ml buffer containing 0.1 mM IBMX for 10 minutes at 37°C, and incubation with ADM and other cAMP-elevating agents was continued for an additional 10 minutes (20 minutes total incubation time) or as indicated. Reaction was stopped with 1 ml ice-cold 10% trichloroacetic acid (TCA). cAMP and cGMP in the TCA-soluble extract were assayed by RIA. ¹⁴  

Measurement of Agonist-Induced Tension Response in the Sphincter Muscle

For measurement of tension response, the muscle preparations were mounted vertically in two separate 10-ml water-jacketed tissue baths that contained Krebs–Ringer bicarbonate buffer at 37°C. A mixture of O₂ (97%) and CO₂ (3%) was continuously bubbled through the solution. The tissue was allowed to equilibrate for 90 minutes under a resting tension of 37.5 mg. To inhibit the endogenous formation of prostaglandins, 1 μM indomethacin was routinely added to the tissue bath. During the equilibration period the physiologic solution was changed every 20 minutes. At the end of equilibration, the test agents were added, and the changes in tension were recorded isometrically using a force-displacement transducer (model 79D; Grass Instruments, Quincy, MA). Concentration–response curves for the mechanical responses were constructed by cumulative addition of agonist to the tissue bath. The concentration of the agonist was increased only after the effect of the previous concentration had stabilized. 

Extraction and Radioimmunoassay of Adrenomedullin in Tissues of the Cat Iris-Ciliary Body

Extraction of ADM from the cat ocular tissues (sphincter, dilator, ciliary muscle, and ciliary processes) and aqueous humor was performed as described by Kitamura et al. ¹ Briefly, each of the ocular tissues was placed in 5 vol of water and then boiled for 10 minutes to inactivate the intrinsic proteases. After cooling, glacial acetic acid was added to make 1 M, and the mixture was homogenized with a siliconized glass homogenizer at 4°C. The homogenate was then centrifuged at 24,000g for 30 minutes, and the supernatant obtained was loaded onto a Sep-Pak C-18 cartridge (Phoenix Pharmaceuticals, Inc., Mountain View, CA) that was preequilibrated with 1 M acetic acid. The adsorbed materials were eluted with 3 ml of 60% acetonitrile in 0.1% trifluoroacetic acid, and the tubes containing the eluent were placed on ice and evaporated under N₂. The dried material obtained was dissolved in RIA buffer, and the ADM concentration quantified by means of RIA according to a procedure described by the supplier (Phoenix Pharmaceutical, Inc.). The sensitivity of the RIA used to quantify ADM was 0.017 fmol. The amount of ADM was calculated as fmol ADM/mg of wet weight tissue. 

Measurement of [¹²⁵I]ADM Binding

Microsomal fractions were prepared from cat iris sphincter muscle by differential centrifugation as described previously. ¹⁵ The binding of [¹²⁵I]ADM to the microsomal membranes was assayed according to the method of Abel et al. ¹⁶ Briefly, the binding was studied at 4°C in 250μ L assay volumes containing 20 mM HEPES (pH 7.4), 5 mM MgCl₂, 10 mM NaCl, 4 mM KCl, 1 mM EDTA, 1 μM phosphoramidon, 0.3% bovine serum albumin, protease inhibitors (0.1 mM PMSF, 0.1 mM bacitracin, 10 μg/mL each of leupeptin, antipain, and aprotinin), 100 pM [¹²⁵I]ADM, and 100 to 200μ g protein of microsomal suspension. The total concentrations of ADM were adjusted by the addition of nonradioactive ADM. The reaction was initiated by the addition of the membranes. After 30 minutes, the binding assay was terminated by the addition of 4 ml ice-cold 20 mM HEPES, pH 7.4. The relative amounts of membrane-bound and free ligand were determined by rapid filtration (GF/B filters; Whatman, Clifton, NJ) with the use of a cell harvester (model M-24R; Brandel Laboratories, Gaithersburg, MD). The filters were washed four times with the wash buffer (20 mM HEPES, pH 7.4), and the radioactivity was determined with a gamma counter (model COBRA II, auto gamma; Packard Instruments Co., Meriden, CT). Nonspecific binding of[ ¹²⁵I]ADM to the membranes was defined as the amount bound in the presence of 100 nM ADM, and the specific binding was calculated by subtracting the nonspecific binding from the total binding. The binding parameters were analyzed by the GraphPad Prism program (GraphPad Software Inc., San Diego, CA). 

Determination of Proteins

Protein content was determined by the method of Lowry et al., ¹⁷ with bovine serum albumin as standard. 

Calculation of Data and Statistics

Results are expressed as means ± SEM. Values for cAMP are reported as picomoles per milligram of protein and for ADM as fentomoles per milligram of wet weight tissue. The EC₅₀ value is defined as that concentration of the agonist that produces 50% of maximum response. Statistical differences between the two means were determined by a paired Student’s t-test. When P was < 0.05, the values were considered to be significantly different. 

Results

Effects of ADM and Other cAMP-Elevating Agents on cAMP Formation in Cat Iris Sphincter

To determine the efficacy of ADM stimulation on cAMP accumulation in the cat iris sphincter, we compared its effects on the cyclic nucleotide with those of CGRP and other cAMP-elevating agents. As can be seen from Table 1 ADM (0.1 μM) increased cAMP accumulation by 12.5-fold compared with a 2-fold increase by CGRP. Vasoactive intestinal peptide, endothelin-1, substance P, isoproterenol, forskolin, and prostaglandin E₂ increased cAMP formation by 2.2-, 3.1-, 0.91-, 2.57-, 2.45-, and 1-fold, respectively. N ^G -nitro-l-arginine (10 μM), a nitric oxide synthase inhibitor, and indomethacin (1 μM), a cyclooxygenase inhibitor, had no effect on ADM-induced cAMP accumulation (data not shown). Furthermore, the basal formation of cGMP in this tissue was 0.59 ± 0.02 pmol/mg protein, and the increase in the intracellular levels of this nucleotide due to ADM was < 35% (data not shown). These data demonstrate that ADM is the most efficacious agonist for adenylate cyclase stimulation in this tissue. 

Time-Course Effect of ADM on cAMP Accumulation in Cat Iris Sphincter

Figure 1 shows the time-course of cAMP accumulation produced by ADM in the cat iris sphincter. The peptide increased cAMP formation in a time-dependent manner with a t _1/2 value of 2.2 minutes. The intracellular cAMP level increased significantly by ADM at 1 minute and peaked at 10 minutes. Thereafter, cAMP accumulation leveled off. In the following experiments, we used 10 minutes of incubation with ADM. 

Concentration-Response Effects of ADM and CGRP on cAMP Accumulation in Cat Iris Sphincter

The effects of ADM and CGRP on cAMP formation in cat iris sphincter are shown in Figure 2 . ADM-stimulated cAMP accumulation was concentration dependent (10⁻¹⁰–10⁻⁶ M) with an EC₅₀ of 13 nM. In contrast, the EC₅₀ for CGRP stimulation was 59 nM, which is about five times as high as that for ADM. These data demonstrate that both ADM and CGRP increase cAMP formation in this tissue in a concentration-dependent manner and that ADM was about fivefold more efficacious than CGRP. 

Effects of ADM and CGRP on cAMP Formation in Iris Sphincter Isolated from Cat and Other Mammalian Species

Studies on the effects of ADM and CGRP on cAMP formation in iris sphincter isolated from cat and other animal species are given in Table 2 . ADM increased cAMP formation in cat, dog, bovine, and human by 883%, 578%, 203%, and 178%, respectively, with little effect on that of rabbit. ADM (0.1 μM) increased cAMP formation in human ciliary muscle cells by 496% (basal level of cAMP was 297 ± 16 pmoles/mg protein/10 min) (Yousufzai SYK, Abdel-Latif AA, unpublished observations). In contrast, CGRP increased cAMP formation in bovine, rabbit, cat, human, and dog by 515%, 170%, 119%, and 41%, respectively. However, the basal levels of cAMP were in the following order: rabbit > human > dog > cat > bovine. These data demonstrate major species differences in the effects of ADM and CGRP on cAMP formation in the iris sphincter. 

Effects of ADM and CGRP on cAMP Formation in Cat Iris Sphincter, Dilator, Ciliary Muscle, and Ciliary Processes

In addition to the iris sphincter, ADM and CGRP increased cAMP formation in all tissues of the cat iris-ciliary body, including dilator, ciliary muscle, and ciliary processes. As can be seen from Table 3 , ADM increased cAMP formation in the sphincter, dilator, ciliary muscle, and ciliary processes by 911%, 773%, 50%, and 307%, respectively, and CGRP increased it by 89%, 160%, 110%, and 549% respectively. 

Distribution of Immunoreactive ADM in Tissues of Cat Iris-Ciliary Body and Aqueous Humor

Tissue ADM contents are given in Table 4 . Highest concentrations of immunoreactive ADM were present in ciliary processes (1.34 ± 0.12 fmol/mg wet tissue). In addition, ADM also was detected in sphincter, dilator, ciliary muscle, and aqueous humor. ADM contents in sphincter, dilator, ciliary muscle, and ciliary processes were 14.2%, 24.2%, 20.1%, and 41.5%, respectively, of the total ADM content of the whole iris-ciliary body. It is interesting to note that low amounts of immunoreactive ADM (0.16 fmoles/ml) were also detectable in aqueous humor of the cat eye (Table 4) . 

Effects of Receptor Antagonists on ADM- and CGRP-Induced cAMP Formation in Cat Iris Sphincter

There is abundant evidence that suggests that ADM and CGRP interact, at least in part, with the same receptor to evoke a physiological response. ² Thus, it has been suggested that the vasodilator effect of ADM is due to its interaction with the CGRP receptor, since vasodilator responses in the rat isolated perfused mesentery evoked by ADM in vitro were inhibited by the CGRP receptor antagonist CGRP (8–37). ¹⁸ In the present work, we used the receptor antagonists ADM (26–52) and CGRP (8–37) to determine the role of their respective agonists in the activation of adenylate cyclase. The data obtained are presented in Figure 3 . The ADM- and CGRP-induced increases in intracellular cAMP levels were inhibited by the antagonists in a concentration-dependent manner. The IC₅₀s for ADM (26–52) inhibition of ADM- and CGRP-stimulation of cAMP formation were 1.1 × 10⁻⁷ and 3.8 × 10⁻⁸ M, respectively, and for CGRP (8–37) inhibition were 3.2 × 10⁻⁸ and 2.2 × 10⁻⁸ M, respectively. CGRP (8–37) was more effective than ADM (26–52) in inhibiting both ADM- (Fig. 3A) and CGRP- (Fig. 3B) induced cAMP formation. These results suggest that ADM and CGRP may interact with the same receptor in this tissue. 

Lack of Additive Effects of ADM and CGRP on Intracellular cAMP Levels in Cat Iris Sphincter

To determine whether ADM and CGRP interact with the same receptor, additive experiments for the activation of adenylate cyclase were performed with ADM (100 nM) and CGRP (100 nM) either alone or in combination. As shown in Figure 4 , addition of these peptides together did not result in an additive effect in cAMP accumulation, suggesting that these peptides act through a common receptor. 

[¹²⁵I]ADM Binding to Membranes Isolated from Cat Iris Sphincter

[¹²⁵I]ADM binding was present in the cat iris sphincter as expected (Fig. 5) . The specific binding of [¹²⁵I]ADM was shown to be saturable for the receptor (Fig. 5A) . Scatchard analysis revealed the presence of a single population of binding sites (Fig. 5A , inset), and the calculated K _d value (mean ± SEM) of the receptor was 0.506 ± 0.171 (n = 3) nM with a B _max of 93 ± 20 fmol/mg protein (n = 3). Figure 5B shows the displacement of[ ¹²⁵I]ADM by unlabeled ADM, CGRP, and their antagonists. Displacement of [¹²⁵I]ADM specific binding by the unlabeled peptides showed that ADM tracer was replaced in a dose-dependent manner. Both ADM and CGRP displaced the binding of[ ¹²⁵I]ADM to iris sphincter membranes effectively, with IC₅₀ values of 0.81 and 1.15 nM, respectively (Fig. 5B) . The antagonists ADM (26–52) and CGRP (8–37) displaced the tracer with IC₅₀ values > 10 nM. The small difference between the IC₅₀ values of these peptides suggests that ADM and CGRP may compete for the same receptor. 

Muscle Relaxation Induced by ADM and CGRP

Figure 6 shows that ADM and CGRP inhibited carbachol-induced muscle contraction in a concentration-dependent manner with IC₅₀ values of 1.0 × 10⁻⁸ and 9.0 × 10⁻⁸ M, respectively. Increasing concentrations of the peptides resulted in increased relaxation (inhibition of contraction) of sphincter muscle precontracted with carbachol, and the maximal relaxant effects of ADM and CGRP were observed at 2.5 × 10⁻⁷ and 5 × 10⁻⁷ M, respectively. These results demonstrate that ADM is considerably more efficacious than CGRP in relaxing the cat iris sphincter muscle. 

Effects of DDA and ADM on cAMP Accumulation and on Carbachol-Induced Contraction

Previously, we have reported that DDA, a specific inhibitor of adenylate cyclase, significantly inhibited CGRP-induced cAMP accumulation and relaxation in rabbit iris dilator muscle. ¹² In the cat iris sphincter, DDA inhibited ADM-induced cAMP formation in a concentration-dependent manner with an IC₅₀ value of 13 nM (data not shown). To add further support to the notion that the relaxant action of ADM in the cat iris sphincter is mediated via cAMP formation, we investigated the effects of DDA and ADM on carbachol-induced contraction. As shown in Figure 7A , DDA inhibited ADM-induced cAMP accumulation by 82%, and when the muscle was preincubated in the presence of the adenylate cyclase inhibitor, the relaxant action of ADM was reduced by 74%. Addition of DDA (10 μM) alone or after carbachol (0.1 μM) had no effect on the contractile response in the cat iris sphincter (data not shown). These data suggest that the inhibitory effects of ADM on carbachol-induced contraction are mediated by cAMP in this tissue. 

Discussion

In the present study, we have demonstrated for the first time that ADM is a potent activator of adenylate cyclase and an effective relaxant in iris sphincter isolated from cat and other mammalian species, including humans. The magnitude of adenylate cyclase stimulation was up to sixfold greater than those seen with CGRP, vasoactive intestinal peptide, isoproterenol, forskolin, or prostaglandin E₂ (Table 1 , Fig. 2 ). Similarly, the magnitude of relaxation with ADM was much greater, and the IC₅₀ was at least nine times less than that seen with CGRP (Fig. 6) . These effects of ADM were not influenced by N ^G -nitro-l-arginine or by indomethacin, suggesting that neither nitric oxide nor prostanoids are involved in these responses. ADM increased cAMP accumulation in a time[ t _1/2 = 2.2 minutes (Fig. 1) ] and concentration- [EC₅₀ = 13 nM (Fig. 2) ] dependent manner. ADM and CGRP produced relaxation in a concentration-dependent manner with IC_50s = 10 and 90 nM, respectively (Fig. 6) . These values correlate well with their potency to stimulate cAMP accumulation; EC₅₀ values for ADM and CGRP were 13 and 59 nM, respectively (Fig. 2) . The involvement of cAMP in the relaxant action of ADM is supported by the finding that DDA, an adenylate cyclase inhibitor, inhibited both ADM-induced cAMP accumulation and ADM-induced muscle relaxation (Fig. 7) . In canine retinal arteries, the vasodilator potencies of these peptides were comparable (EC_50s for ADM and CGRP were 0.71 and 2.62 nM, respectively). ¹¹ The relaxant potency of ADM is less than that of CGRP in canine basilar, mesenteric, coronary, or femoral arteries. ¹⁹ However, ADM is more potent in vasodilating the cat pulmonary vascular bed than CGRP, ²⁰ and the opposite is true in the rat mesenteric bed. ¹⁸ Furthermore, ADM increased cAMP accumulation more potently than CGRP in rat renal tubular basolateral membranes, ²¹ rat vascular smooth muscle cells, ⁷ cultured human breast cancer cells, T47D, ²² human vascular endothelial cells, ²³ and cultured rat astrocytes. ²⁴ However, in both rat cardiac myocytes and nonmyocytes ²⁵ and in cultured human neuroblastoma SK-N-MC cells, ²⁶ CGRP is more potent in stimulating cAMP production than ADM. These differences in the physiological and biochemical responses to ADM may be due to differences in species or tissue or to affinity and population of receptors. In the present work, ADM-induced cAMP accumulation in the iris sphincter varied with the species (Table 2) as well as tissue of the iris-ciliary body (Table 3) . Furthermore, there were significant differences in the distribution of immunoreactive ADM in tissues of the cat iris-ciliary body: ciliary processes > dilator > sphincter > ciliary muscle (Table 4) . The immunoreactive ADM concentrations found in these tissues are comparable to those reported for other tissues, including (fmol/mg wet tissue) the following: aorta = 0.42 ± 0.09; small intestine = 0.97 ± 0.45; heart ventricle = 0.15 ± 0.02; heart atrium = 1.68 ± 1.58; kidney = 0.35 ± 0.12; brain cortex = 0.31 ± 0.15 and lung = 0.80 ± 0.37. ²  

To determine the type of receptors involved in the actions of ADM in the cat iris sphincter, we examined the effects of receptor antagonists and performed binding studies. At present, there is controversy about a receptor(s) responsible for ADM-induced vasorelaxation. ^{18 23} An issue we tried to address in the present work is whether the observed effects of ADM and CGRP are mediated by the same or by different receptors. Two types of ADM receptors have been characterized. ²⁷ One type of ADM receptor, which binds ADM but not CGRP or CGRP (8–37) with high affinity, is present in rat tissue endothelial cells and rat vascular smooth muscle cells. ^{7 28} This is consistent with the cloned ADM receptor, which bound [¹²⁵I]ADM with high affinity and elevated cAMP threefold. In contrast, 1 μM CGRP had no effect on cAMP, and the increase in cAMP caused by 10 nM ADM was not antagonized by CGRP (8–37). ²⁹ A second type of ADM receptor was identified in bovine aortic endothelial cells, neuroblastoma cells, and L6 cells, which binds ADM, CGRP, and CGRP (8–37) with high affinity. ^{8 26} In this case, ADM elevates cAMP, but the increase in cAMP is antagonized by CGRP (8–37). This may represent the CGRP type 1 receptor, which was cloned and has 56% homology with the calcitonin receptor. ³⁰ The data presented here suggest that cat iris sphincter contains the second type of ADM receptor, namely the CGRP₁ receptor. This conclusion is supported by the following findings in the present work: (1) The cAMP responses to ADM and CGRP are inhibited markedly by treatment with the competitive CGRP antagonist, CGRP (8–37) (Fig. 3) . CGRP (8–37) inhibited ADM- and CGRP-induced cAMP formation in a concentration-dependent manner with IC₅₀ values of 32 nM (Fig. 3A) and 22 nM (Fig. 3B) , respectively. CGRP (8–37) was considerably more potent in its inhibitory effects than ADM (26–52) (Fig. 3) . CGRP (8–37) inhibited the vasodilator response to ADM in the isolated perfused mesenteric bed ¹⁸ and in the isolated rat heart, ³¹ indicating that the effect of ADM in these tissues is via CGRP₁ receptors. (2) Addition of ADM and CGRP together did not result in an additive effect on cAMP accumulation, suggesting that these peptides exert their effects through a common receptor (Fig. 4) . (3) The present study has for the first time shown the presence of ADM binding sites in ocular tissues (Fig. 5) . The Scatchard plot was linear indicating that[ ¹²⁵I]ADM bound with high affinity (K _d = 0.51 ± 0.17 nM) to a single class of sites (B _max = 93 ± 20 fmol/mg protein) (Fig. 5A) . Both ADM and CGRP displaced the binding of[ ¹²⁵I]ADM to iris sphincter membranes effectively, with IC₅₀ values of 0.81 and 1.15 nM, respectively (Fig. 5B) . In the present work we have observed an apparent difference between the IC₅₀ values of ADM (26–52) and CGRP (8–37) on ADM-binding and on ADM-induced cAMP formation (Figs. 3 4) . This could be explained by the fact that the binding experiments were carried out on isolated membranes (Fig. 5) , whereas the cAMP studies were performed on whole muscle (Fig. 3) . The finding that the specific [¹²⁵I]ADM binding was strongly inhibited by CGRP suggest that most of the[ ¹²⁵I]ADM binding detected here is primarily due to interaction with CGRP receptors. In rat spinal cord microsomes[ ¹²⁵I]ADM binding showed high affinity (K _d = 0.45 ± 0.06 nM) and sites were abundant (B_max = 723 ± 71 fmol/mg protein). ³² Scatchard plots of[ ¹²⁵I]ADM binding in human brain (cerebral cortex) gave a K _d of 0.17 ± 0.03 nM and maximal binding of 99.3 ± 1.9 fmol/mg protein. ³³ In smooth muscle, pharmacologically distinct binding sites for ADM were reported in rat uterus (K _d = 0.08 ± 0.006 nM; B _max = 21 ± 2 fmol/mg protein), ³⁴ and in cultured vascular smooth muscle cells. ³⁵ ADM interacts with the CGRP receptor in several tissues, including cultured human neuroblastoma SK-N-MC cells, ²⁶ cultured human breast cancer cells, ²² rat cardiac myocytes and nonmyocytes, ²⁵ rat vascular smooth muscle cells, ³⁵ and rat mesenteric beds. ⁷ In contrast, cultured endothelial cells of human umbilical vein possess specific ADM receptors coupled with the adenylate cyclase that may have little affinity with CGRP, ²³ the renal action of ADM is not mediated via the activation of CGRP₁ receptors, specific ADM binding sites that differ from CGRP receptors ³⁶ exist in brain, ³³ and cultured mouse astrocytes possess specific ADM receptors that are coupled to adenylate cyclase but do not interact with CGRP. ³⁷ These observations indicate that the distribution, characteristics, and biological effects of ADM and CGRP overlap in many tissues. (4) ADM and CGRP inhibited, presumably via stimulation of adenylate cyclase, carbachol-induced contraction in the cat sphincter in a concentration-dependent manner with IC₅₀ values of 10 and 90 nM, respectively (Fig. 6) . These results suggest that in the iris sphincter signal transduction mechanisms responsible for relaxation caused by ADM are similar to those caused by CGRP. 

In conclusion, the data reported here show that in cat iris sphincter ADM is a much more potent activator of adenylate cyclase and a much more effective relaxant than CGRP. Immunoreactive ADM is present in all tissues of the iris-ciliary body, and its biological effects may be due to direct involvement of ADM receptors, but also to activation of CGRP receptors. It is therefore plausible that ADM, like CGRP, serves a possible novel neurotransmitter/neuromodulator role in the ocular tissues. In the eye, CGRP causes vasodilation and reduces intraocular pressure (IOP) by facilitating the aqueous outflow. ³⁸ Furthermore, intravitreal administration of CGRP into rabbit eyes leaves the blood aqueous barrier intact and causes an increase in the outflow facility of aqueous humor with a concomitant long-lasting decrease in IOP. ³⁹ In the present work, functional receptors for ADM have been demonstrated in iris-ciliary body isolated from several mammalian species including human (Table 2) , activation of these receptors by the peptide leads to concentration-dependent increases in cAMP accumulation and subsequent inhibition (relaxation) of smooth muscle contraction. These findings suggest a role for ADM as a local modulator of smooth muscle tone. A possible function for this potent hypotensive peptide in the regulation of IOP remains to be investigated. 

 Supported by National Institutes of Health Grants R37-EY-04171 and EY-04387.

 Submitted for publication February 24, 1999; revised June 8 and July 12, 1999; accepted August 2, 1999.

 Commercial relationships policy: N.

 

Table 1.

 

Effects of Adrenomedullin and Other cAMP-Elevating Agents on cAMP Formation in Cat Iris Sphincter

Additions	Concentration (μM)	cAMP Formation (pmol/mg protein/10 min)	% of Control
None	—	180 ± 12	100
ADM	0.1	2440 ± 180*	1356
CGRP	1	536 ± 29*	298
Vasoactive intestinal peptide	1	580 ± 52*	323
Endothelin-1	0.1	740 ± 56*	411
Substance P	1	344 ± 30, †	191
Isoproterenol	1	644 ± 52*	357
Forskolin	1	620 ± 40*	345
Prostaglandin E2	1	372 ± 28, †	206

 Iris sphincter muscles were incubated in 1 ml Krebs–Ringer bicarbonate buffer, pH 7.4, for 90 min. The medium was discarded, and the muscles were then incubated in 1 ml buffer containing 0.1 mM IBMX for 10 min at 37°C. Incubation with ADM and other cAMP-elevating agents was continued for an additional 10 min (20 min total incubation time). cAMP in the TCA-soluble extract was determined by RIA as described in Materials and Methods. The results are means ± SEM of three experiments, each run in triplicate.

 Significant differences from control,

*   P < 0.001;

†   P < 0.002.

Figure 1.

 

Time-course effect of ADM on cAMP accumulation in cat iris sphincter. Conditions of incubation were the same as described in Table 1 , except that incubations were carried out for various time intervals as indicated. The basal value for cAMP formation was 145 ± 13.5 pmol/mg protein. The results are means ± SEM obtained from three different experiments, each run in triplicate.

Figure 2.

 

Concentration-response effects of ADM and CGRP on cAMP accumulation in cat iris sphincter. Conditions of incubation were the same as described in Figure 1 , except that different concentrations of the peptides were added as indicated and the time of incubation with the peptides was 10 minutes. The results are means ± SEM obtained from four different experiments, each run in triplicate.

Table 2.

 

Effects of ADM and CGRP on cAMP Formation in Iris Sphincter Isolated from Different Mammalian Species

Additions	cAMP Formation (pmol/mg protein/10 min)
		Cat	Dog	Bovine	Rabbit	Human*
None	162 ± 15	194 ± 20	105 ± 9	318 ± 24	287 ± 25
ADM (0.1 μM)	1593 ± 152, † (983)	1315 ± 126, † (678)	318 ± 28, † (303)	366 ± 38 (115)	799 ± 47, † (278)
CGRP (0.1 μM)	356 ± 28, ‡ (220)	274 ± 21, § (141)	640 ± 58, † (615)	859 ± 76, ‡ (270)	629 ± 38, † (219)

 Conditions of incubation were the same as described in Table 1 and Materials and Methods. Data are means ± SEM of three experiments, each run in triplicate, with % of control in parentheses.

*  Data from the human are averages of four determinations.

 Significant differences from control,

†   P < 0.001;

‡   P < 0.002;

§   P < 0.005.

Table 3.

 

Effects of ADM and CGRP on cAMP Formation in Cat Iris Sphincter, Dilator, Ciliary Muscle, and Ciliary Processes

Additions	cAMP Formation (pmol/mg protein/10 min)
		Sphincter	Dilator	Ciliary Muscle	Ciliary Processes
None	203 ± 18	386 ± 31	292 ± 30	327 ± 31
ADM (0.1 μM)	2052 ± 196* (1011)	3213 ± 241* (873)	438 ± 38, § (150)	1330 ± 121* (407)
CGRP (0.1 μM)	384 ± 41, ‡ (189)	957 ± 81, † (260)	614 ± 48, ‡ (210)	2112 ± 183* (649)

 Conditions of incubation were the same as described under Table 1 and Materials and Methods. Data are means ± SEM of five experiments, each run in triplicate, with % of control in parentheses.

 Significant differences from control,

*   P < 0.001;

†   P < 0.002;

‡   P < 0.01;

§   P < 0.02.

Table 4.

 

Distribution of Immunoreactive ADM in Tissues of Cat Iris-Ciliary Body and Aqueous Humor

Tissue	Immunoreactive ADM (fmol/mg wet tissue)	Immunoreactive ADM (fmol/ml)
Iris Sphincter	0.46 ± 0.035 (11)	—
Iris Dilator	0.78 ± 0.083 (7)	—
Ciliary Muscle	0.65 ± 0.083 (11)	—
Ciliary Processes	1.34 ± 0.12 (4)	—
Aqueous Humor	—	0.16 ± 0.022 (3)

 Data are means ± SEM of 3–11 experiments. Numbers in parentheses are the number of experiments.

Figure 3.

 

Concentration dependency for ADM antagonist [ADM (26–52)] and CGRP antagonist [CGRP(8–37)] inhibition of ADM- (A) and CGRP- (B) induced cAMP formation in cat iris sphincter. Conditions of incubation were as described in Figure 1 , except that the muscles were preincubated with different concentrations of either ADM (26–52) or CGRP (8–37) as indicated for 10 minutes, and then incubation continued in the absence (control) or presence of 0.1 μM ADM (A) or 0.2 μM CGRP (B) for 10 minutes The results are means ± SEM obtained from three different experiments, each run in triplicate.

Figure 4.

 

Lack of additive effects of ADM (10⁻⁷ M) and CGRP (10⁻⁷ M) on cAMP levels in cat iris sphincter. Bar graph shows the effects of ADM, CGRP, and the combined effect of ADM and CGRP. The results are means ± SEM for three experiments, each run in triplicate. Significant differences from control,* P < 0.001; **P < 0.01.

Figure 5.

 

Representative saturation curves, Scatchard analysis, and displacement of [¹²⁵I] ADM binding to cat iris sphincter membranes. (A) Concentration dependence of [¹²⁵I]ADM binding to membranes. Inset: Scatchard analysis of specific [¹²⁵I]ADM binding. (B) Displacement of [¹²⁵I]ADM to the membranes by increasing concentrations of various peptides. Each point represents the mean of triplicate determinations.

Figure 6.

 

Concentration–contraction response curves of the effects of ADM and CGRP on carbachol-induced contraction in cat iris sphincter. Muscles were preequilibrated in buffer for 90 minutes and then contracted by CCh (0.1 μM) for 2.5 minutes, followed by addition of ADM cumulatively (10⁻⁹–2.5 × 10⁻⁷ M) or CGRP cumulatively (10⁻⁹–10⁻⁶ M). Each point represents the mean ± SEM of 3 to 4 observations.

Figure 7.

 

Effects of 2′,5′-dideoxyadenosine (DDA) and ADM on cAMP accumulation and on CCh-induced contraction. Conditions of incubation for cAMP accumulation and for contraction were the same as described in Figures 1 and 6 , respectively, except that DDA (10 μM) was added 20 minutes before the addition of CCh. Each data point represents the mean ± SEM of three different experiments, each run in triplicate. The stimulatory effect of ADM on cAMP formation and inhibitory action of ADM on contraction are significantly reduced by DDA compared with their respective controls, P < 0.001, respectively.

The authors thank Phattra Volarath and Jacqueline Negron for technical assistance and Jennifer Hatfield for typing the manuscript.