Human ciliary process adrenergic receptor: pharmacological characterization James A. Nathanson To better understand the nature of interaction of various amines with adrenergic receptors in the human ciliary process, beta-adrenergic -stimulated adenylate cyclase activity was characterized in broken cell preparations of this tissue from donor eyes. Among various agonists, isoproterenol was the most potent activator of enzyme activity (K a = 3.4 x 10~ 7 M), folloioed in order by epinephrine (K a = 2.7 x lo'^m), norepinephrine (K a = 2.1 x 1Q~~*M), and phenylephrine (K a > 10~ 4 M). lsoproterenol-stimulated enzyme activity ivas blocked by timolol (K, = 3.4 x 10- i} M), IPS 339 (K, = 4.4 x 10~ 9 M), H35/25 (K, = 6.9 x 10~ 7 M), and atenolol (Kj = 1.4 x 10~ 5 M). These pharmacological characteristics indicate that the human ciliary processes contain a predominance of beta 2 -adrenergic receptors. The findings are relevant to physiological studies of aqueous humor secretion and to the potential development of adrenergic agents with greater specificity for the beta-adrenergic receptor. (INVEST OPHTHALMOL VIS SCI 21:798-804, 1981.) Key words: ciliary process, adrenergic, receptor, beta-adrenergic receptor, cyclic AMP, adenylate cyclase, sympathetic nervous system Beta-adrenergic receptors are divided into two subtypes according to their relative affinities for certain catecholamines: betajadrenergic receptors (predominating in heart and adipose tissue) show approximately the same affinity for norepinephrine and epinephrine, whereas beta 2 -adrenergic receptors (predominating in lung, liver, and vascular smooth muscle) are about 10-fold more sensitive to the effects of epinephrine. 1 This distinction among beta-adrenergic receptors From the Departments of Neurology and Pharmacology, Harvard Medical School, and Massachusetts General Hospital, Boston. This research was supported by a grant from the William F. Milton Fund and a PMAF Faculty Development Award in Clinical Pharmacology. Submitted for publication Dec. 15, 1980. Reprint requests: James A. Nathanson, M.D., Ph.D., Department of Neurology, Massachusetts General Hospital, Boston, Mass. 02114. has proved useful because in recent years it has been possible to develop pharmacological agents with relative specificity toward either the beta! or beta 2 subtype. Such selective agents minimize side effects to non-target tissues that would otherwise be affected by a nonspecific beta-adrenergic agonist or antagonist. With the recent use of biochemical techniques (i.e., radioligand binding and hormone-sensitive adenylate cyclase) it has been possible to define further the distribution of beta-adrenergic receptor subtypes in different tissues and to determine the ability of various agents to distinguish between these subtypes. Such biochemical studies have revealed that (1) the relative proportion of beta r vs. beta 2 -adrenergic receptors in a particular tissue may vary among different animal species. 2 ' 3 and (2) certain beta-selective agents, which by physiological studies were previously considered to be fairly specific, show only a low degree of specificity for beta x - vs. 798 0146-0404/81/120798+07300.70/0 1981 Assoc. for Res. in Vis. and Ophthal., Inc.
Volume 21 Number 6 Human ciliary process adrenergic receptor 799 beta 2 -adrenergic receptors when measured by radioligand binding or activation of betaadrenergic sensitive adenylate cyclase. 4 Evidence from physiological and biochemical studies of the eye suggests that betaadrenergic agents may play an important role in the regulation of intraocular pressure (IOP) (for review see ref. 5). Although the effects of these agents are notably complex, a number of investigators have tried to localize their site of action by determining the localization of beta-adrenergic receptors in various ocular tissues. Waitzman and Woods 6 first identified a beta-adrenergic-stimulated adenylate cyclase in rabbit ciliary process. Later, Neufeld and coworkers 7 ' 8 characterized beta-adrenergic ligand binding to membrane preparations derived from a combined preparation of rabbit iris and ciliary body. The presence of beta-adrenergic receptor binding in rabbit ciliary process (separate from that in iris) was confirmed recently by Bromberg et al. 9 That such beta-adrenergic receptors may reside in ciliary process epithelial cells separate from (or in addition to) blood vessels has been suggested by histochemical studies in the rat 10 utilizing a fluorescent adrenergic antagonist and, more recently, by biochemical studies using isolated ciliary process epithelial cells. 11 Most of the above biochemical studies have indicated a greater potency of epinephrine than norepinephrine, and more extensive characterization of the rabbit ciliary process, using selective beta x and beta 2 agents, has shown that the tissue contains a predominance of beta 2 -adrenergic receptors. 12 As noted above, there is evidence that the subtype of beta-adrenergic receptor present in a particular tissue may vary among different species. Thus, although the above animal studies are of considerable importance, it would be of some use for the development of clinically effective adrenergic agents in humans to have a more detailed knowledge of beta-adrenergic receptor characteristics in the human ciliary process. In a prior report, Neufeld and Sears 13 showed that intact pieces of primate and human ciliary process increase their cyclic AM P content when exposed to 10" 5 M epinephrine or norepinephrine, thereby providing indirect evidence for the presence of an adrenergic receptor. The present study was designed to study more directly the interaction of several amines with adrenergic receptors linked to adenylate cyclase in broken cell preparations of human ciliary process. Through the evaluation of complete doseresponse curves and the use of specific betajand beta 2 -adrenergic blocking agents, it has been possible to determine that the pharmacological profile of beta-adrenergic receptors in the human ciliary process is similar to that of beta-adrenergic receptors in the lung. Materials and methods All common reagents were from Sigma Chemical Co., St. Louis, Mo. Drugs not available commercially were kindly supplied by the following sources: atenolol from Stuart Pharmaceuticals, Wilmington, Del.; H 35/25 (l-(4-methylphenyl)-2- isopropylaminopropranolol) from Hassle, Sweden; IPS 339 ((f-butyl-amino-3-ol-2-propyl)oximino-9- fluorene) from Hassle and Prof. G. LeClerc. Human ciliary processes were obtained from donor eyes supplied by the New England Eye Bank. The donors, aged 38 to 85, had died of cardiac arrest. Eyes were enucleated within 4 hr of death and stored at 4 C between 12 and 18 hr. Some additional experiments were performed with ciliary processes obtained from monkeys of two species: Macaca fasicularis (courtesy New England Primate Center), in which tissue was obtained immediately after sacrifice and used fresh, and Cercopithecus aethiops sabaeus (vervet) (kindly supplied by D. E. Redmond of the Yale Primate Research Facility), in which tissue was obtained immediately after sacrifice, frozen in liquid N 2, and later thawed for use. Each eye was opened through a circumferential incision 3 mm posterior to the Iimbus, and the vitreous was bluntly dissected from the posterior surface of the lens. The opened, anterior third of the eye was placed, cornea-down, in a solution of artificial aqueous humor (in mmol/7: NaCl 130, KC1 2.7, NaH 2 CO ;i 18.3, MgCl 2 1.33, CaCl 2 1.5, glucose 10, HEPES 10; ph 7.4), and the lens was removed by dividing the zonules at the level of the lens capsule. The villuslike ciliary processes were transected at their attachments to the ciliary body, removed, washed in 150 mm NaCl, and homogenized (10 mg/ml) by hand in an all-glass homogenizer in 6 mm Tris maleate buffer, ph 7.4. Adenylate cyclase activity was measured in test
800 Nathanson Invest. Ophthalmol. Vis. Sci. December 1981 100 80 w 60 S 40 u 20-8 "7 "6 "5 LOG AGONIST (M) -3 Fig. 1. Effects of various catecholamine agonists on human ciliary process adenylate cyclase activity. Activity for each agonist is expressed as a percentage of the maximal stimulation observed with isoproterenol. Values shown here and in Fig. 2 represent the mean ± mean deviation of replicate samples, each assayed for cyclic AMP content in duplicate. At those points which lack error bars, the range was within the size of the symbol. In the experiments shown, basal activity was 2.23 ± 0.12 pmol/mg of protein per minute, and maximal isoproterenol stimulation was 41.3 ±2.1 pmol/mg of protein per minute. ISO, Isoproterenol; EPI, epinephrine; NE, norepinephrine. tubes containing (in 0.3 ml) 80 mm Tris maleate (ph 7.4), 10 mm theophylline, 0.5 mm ethyleneglycol-fois-(/3-aminoethyl ether)-a',iv'-tetraacetic acid (EGTA), 8 mm MgCl 2, 0.03 mm guanosine triphosphate (GTP), 2 mm ATP, and ciliary process homogenate, plus or minus test substances as indicated. The amount of GTP present was that which (as determined in preliminary experiments) gave optimal enzyme activity. The enzyme reaction (4 min at 30 C) was initiated by addition of ATP, and stopped by heating to 90 for 2 min; the mixture was then centrifuged at 1000 x g for 15 min to remove insoluble material. Cyclic AMP in the supernatant was measured by protein-binding assay. 14 Under the assay conditions used, adenylate cyclase activity was linear with respect to time and protein concentration, and cyclic nucleotide phosphodiesterase activity was nearly completely inhibited. Other experiments indicated that heat inactivation (vs. other methods of reaction termination) did not affect the results obtained. Protein concentration was determined by the method of Lowry et al., lo with bovine serum albumin used as a standard. Activation constants (KJ and IC 50 values were determined with a computerized sigmoidal curvefitting program 16 utilizing eight to 12 data points per dose-response curve. Inhibitory constants (K ( ) for the various adrenergic blockers were calculated from the equation, I7 K ( = (IC 50 )/(l + S/KJ, where IC 50 is the concentration of antagonist necessary to give 50% inhibition of activity in the presence of isoproterenol, S is the concentration of isoproterenol present, and K a is the concentration of isoproterenol (3.4 X 10" 7 M) necessary for half-maximal activation of human ciliary process adenylate cyclase activity. Results The beta-adrenergic agonist isoproterenol was a potent stimulator of adenylate cyclase activity in the human ciliary process. At maximally effective concentrations, enzyme ac-
Volume 21 Number 6 Human ciliary process adrenergic receptor 801 V 00 80 < 60 Atenolol 40 2 o 20 "8-7 -6 "5 "4 LOG ANTAGONIST (M) -3 Fig. 2. Effects of various adrenergic antagonists on human ciliary process adenylate cyclase activity. Shown is the degree of enzyme activity observed in the presence of 3 x 10~ 6 M isoproterenol with increasing concentrations of antagonist. Stimulation (above that seen in the presence of antagonist alone) is expressed as a percentage of that seen with isoproterenol alone (38.4 ± 1.0 pmol/mg of protein per minute). tivity was increased by as much as 1800% over basal levels (Fig. 1). Concentrations as low as 10~ 8 M caused some stimulation, and halfmaximal activation (K a ) occurred at 3.4 X 10~ 7 M. Epinephrine (K a = 2.7 X 10-6 M) was about eightfold less potent than isoproterenol, and norepinephrine (K a = 2.1 x 10" 5 M) was about eightfold less potent than epinephrine. (Statistical analysis of these agonist curves 16 showed that there was no significant (p < 0.05) overlap.) At high concentrations, maximal stimulation (V max ) by both norepinephrine and epinephrine was similar to that caused by isoproterenol. Phenylephrine, an alpha-adrenergic agonist, caused little stimulation of enzyme activity at concentrations below 10~ 4 M (data not shown). The relative order and ratios of potency for the above amines in activating human ciliary process adenylate cyclase activity (i.e., isoproterenol > epinephrine > norepinephrine > phenylephrine) are similar to those reported for mammalian lung, which is known to contain a predominance of beta 2 - adrenergic receptors. In addition to the human ciliary process, that from monkey eye also demonstrated beta-adrenergic-stimulated adenylate cyclase activity. At maximally effective concentrations of isoproterenol, enzyme activity was stimulated to 264% of control in the Macaca fasicularis and to 223% in the vervet (K a = 2 X 10~ 7 M). The stimulation of human ciliary process adenylate cyclase by isoproterenol was inhibited by low concentrations of the nonspecific beta-adrenergic antagonist, timolol (Fig. 2). The calculated inhibitory constant (Kj) for timolol was 3.4 X 10" 9 M. IPS 339, a compound which has been shown by both physiological 18 and biochemical 3 studies to have considerable specificity in blocking beta 2 - adrenergic receptors, was also a very potent inhibitor of isoproterenol stimulation, with a calculated Kj of 4.4 X 10" 9 M. This calculated
802 Nathanson Invest. Ophthahnol. Vis. Sci. December 1981 Kj was not significantly different from that of timolol. H35/25, a beta-adrenergic blocker with some beta 2 selectivity, 3 was also effective in inhibiting ciliary beta-adrenergic-sensitive adenylate cyclase activity, with a Kj of 6.9 X 10" 7 M. On the other hand, atenolol, which is known to be a potent beta r adrenergic blocker in other tissues, 3 was much less potent in the human ciliary process, with a calculated Kj of 1.4 X 10" 5 M. To quantitate the degree of similarity of the beta-adrenergic receptor in human ciliary process with either the beta 2 -adrenergic receptor predominating in lung or the betajadrenergic receptor predominating in heart, the K a 's and Kj's of the above agonists and antagonists were compared with the published constants for the same agonists and antagonists in activating or inhibiting adenylate cyclase in membrane fractions from lung and heart. 3 ' 4 The calculated r coefficient (0.88) indicated a good correlation (p < 0.025) between the pharmacological characteristics of the ciliary and lung enzymes as compared with a poor correlation (r = 0.25; not significant) between the ciliary process and heart enzymes. (By utilizing published K d values for ligand binding studies, a similarly good correlation was found between the pharmacological characteristics of the human ciliary process adrenergic receptor and that in lung (r = 0.99; p < 0.01) as compared with a poor correlation between ciliary process and heart (r = 0.26; not significant).) Of interest, too, were calculations indicating that the beta-adrenergic-sensitive adenylate cyclase present in mammalian choroid plexus (a tissue that, like the ciliary process, is involved in fluid secretion, which may be under adrenergic control 19 ' 20 ) also shows receptor characteristics very much like those found in the ciliary process (r = 0.92; p < 0.01). Discussion Although the above data indicate that human ciliary process beta-adrenergic receptors have pharmacological characteristics similar to those of beta 2 -adrenergic receptors present in mammalian lung, it should be pointed out that many tissues contain a mixture of receptor subtypes 4 ' 21 and it is not possible from the present experiments to rule out entirely a small percentage of beta r adrenergic receptors. Furthermore, because the human material used in this study was obtained from donor eyes, the possibility of postmortem tissue change affecting the results should be considered. In this regard, it is reassuring that the K a and Kj values obtained with the human material are quite similar to those values obtained for beta 2 -adrenergic receptors in a number of other tissues obtained without postmortem delay, 3> 4> 21 including the rabbit ciliary process. 12 In addition, the K a value for isoproterenol obtained for the human was quite similar to that obtained in the monkey ciliary processes. Thus, if postmortem changes did occur, it is likely that they were comparatively small. This is consistent with previous experiments of hormonesensitive adenylate cyclase in animal and human brain, which have shown that the receptor characteristics of tissue kept in the intact state prior to homogenization is surprisingly resistant (4 hr or more at room temperature) to loss of activity. 22 In the present study, the percent stimulation of basal enzyme activity by isoproterenol was greater in the human than monkey ciliary process. Although this could represent a species difference, further experiments will have to be done to determine whether postmortem change (in the human) also contributed. Further studies are also needed to determine whether the subtypes of beta-adrenergic receptor in the monkey are similar to or different from that in the human and rabbit. 12 The ciliary processes contain a number of different cell elements, including pigment and secretory epithelium, vascular endothelium, connective tissue cells, nerve fibers, and some intravascular blood elements. Because, under similar assay conditions, mammalian blood elements contain little betaadrenergic-sensitive adenylate cyclase, 20 it appears unlikely that the activity described in the present studies was due to retained blood. It has been reported that, in the rab-
Volume 21 Number 6 Human ciliary process adrenergic receptor 803 bit, there is a substantial enrichment of beta 2 -adrenergic receptors in epithelial cells obtained through selective enzymatic digestion of intact ciliary processes. 11 If the human ciliary process is similar, then a localization in epithelium is possible, although further studies will be needed to confirm this. These results suggest that it may be of interest to investigate the possible effects on IOP of specific and highly potent beta 2 - adrenergic antagonists. Unfortunately, most existing beta 2 -adrenergic antagonists demonstrate much less potency than nonspecific beta-adrenergic blockers such as propranolol or timolol. 3 ' 4 Furthermore, recent biochemical studies indicate that some beta 2 -adrenergic antagonists (such as butoxamine and H35/25), which were previously thought to be fairly specific, have, in fact, only a low degree of selectivity between betaj- and beta 2 -adrenergic receptors. 3 ' 4 This finding could explain, in part, the somewhat inconsistent results on IOP that have been obtained in animals with prior trials of butoxamine and H35/25. 23-24 IPS 339, on the other hand, represents one of the first examples of a beta 2 -adrenergic antagonist that is both potent and highly selective. 3 ' 18 In addition, IPS 339 has been reported to lower IOP in the normal rabbit. 25 It will be of interest to investigate further the possible physiological effects of IPS 339 and of other newly developed, potent, and specific beta 2 -adrenergic antagonists. I thank E. J. Hunnicutt for technical assistance, Ms. Gail Green of the New England Eye Bank, and D. E. Redmond of the Yale Primate Research Facility. REFERENCES 1. Lands AM, Arnold A, McAuliff JP, Luduena FP, and Brown TG: Differentiation of receptor systems activated by sympathomimetic amines. Nature 214: 597, 1967. 2. Hancock AA, DeLean AL, and Lefkowitz RJ: Quantitative resolution of beta-adrenergic subtypes by selective ligand binding: application of a computerized model fitting technique. Mol Pharmacol 16:1, 1979. 3. Minneman KP, Hedberg A, and MolinoffPB: Comparison of beta-adrenergic receptor subtypes in mammalian tissues. J Pharmacol Exp Ther 211:502, 1979. 4. Minneman KP, Hegstrand LR, and Molinoff PB: The pharmacological specificity of beta! and beta 2 adrenergic receptors in rat heart and lung in vitro. Mol Pharmacol 16:21, 1979. 5. Sears ML: Catecholamines in relation to eye. In Handbook of Physiology. Vol. VI. Endocrinology, Astwood E and Creep R, editors. Washington, D. C., 1979, American Physiological Society, pp. 553-590. 6. Waitzman MB and Woods WD: Some characteristics of an adenyl cyclase preparation from rabbit ciliary process tissue. Exp Eye Res 12:99, 1971. 7. Neufeld AH and Page ED: In vitro determination of the ability of drugs to bind to adrenergic receptors. INVEST OPHTHALMOL VIS SCI 16:1118, 1977. 8. Neufeld AH, Zawistowski KA, Page ED, and Bromberg BB: Influences on the density of beta adrenergic receptors in the cornea and iris-ciliary body of the rabbit. INVEST OPHTHALMOL VIS SCI 17:1069, 1978. 9. Bromberg BB, Gregory DS, and Sears ML: Betaadrenergic receptors in ciliary processes of the rabbit. INVEST OPHTHALMOL VIS SCI 19:203, 1980. 10. Lahav M, Melamed E, Dafna Z, and Atlas D: Localization of beta receptors in the anterior segment of the rat eye by a fluorescent analogue of propranolol. INVEST OPHTHALMOL VIS SCI 17:645, 1978. 11. Nathanson JA: Beta-adrenergic receptors in ciliary process epithelium. Soc Neurosci Abst 6:356, 1980. 12. Nathanson JA: Adrenergic regulation of intraocular pressure: identification of beta 2 -adrenergic-stimulated adenylate cyclase in ciliary process epithelium. Proc Natl Acad Sci USA 77:7420, 1980. 13. Neufeld AH and Sears ML: Cyclic AMP in ocular tissues of the rabbit, monkey, and human. INVEST OPHTHALMOL 13:475, 1974. 14. Brown BL, Elkins RP, and Albano JDM: Saturation assay for cyclic AMP using endogenous binding protein. Adv Cyclic Nucleotide Res 2:25, 1972. 15. Lowry OH, Rosebrough NJ, Fair AL, and Randall RJ: Protein measurement with the Folin phenol reagent. J Biol Chem 193:265, 1951. 16. Rodbard D and Hutt DM: Iterative least squares method for logistic curve fitting. In Proceedings of the Symposium on Radioimmunoassay and Related Procedures in Medicine. Vienna, 1974, International Atomic Energy Agency, pp. 165-192 (available from Unipub, New York). 17. Chen Y-C and Prusoff WH: Relationship between the inhibition constant (KJ and the concentration of an inhibitor which causes 50 percent inhibition (IC 50 ) of an enzymatic reaction. Biochem Pharmacol 22:3099, 1973. 18. Imbs JL, Miesch F, Schwartz J, Velly J, LeClerc G, Mann A, and Wermuth G: A potent new beta 2 - adrenoceptor blocking agent. Br J Pharmacol 60: 357, 1977. 19. Lindvall M, Edvinsson L, and Owman C: Sympathetic nervous control of cerebrospinal fluid production from the choroid plexus. Science 201:176, 1978. 20. Nathanson JA: Beta-adrenergic-sensitive adenylate
804 Nathanson Invest. Ophthalmol. Vis. Sci. December 1981 cyclase in choroid plexus: properties and cellular localization. Mol Pharmacol 18:199, 1980. 21. Minneman KP, Hegstrand LR, and Molinoff PB: Simultaneous determination of beta-1 and beta-2 adrenergic receptors in tissues containing both receptor subtypes. Mol Pharmacol 16:34, 1979. 22. Clement-Cormier YC, Kebabian JW, Petzold GL, and Greengard P: Dopamine-sensitive adenylate cyclase in mammalian brain: a possible site of action of antipsychotic drugs. Proc Natl Acad Sci USA 71:1113, 1974. 23. Rowland JM and Potter DE: Adrenergic drugs and intraocular pressure: suppression of ocular hypertension induced by water loading. Exp Eye Res 30:93, 1980. 24. Colasanti BK and Trotter RR: Effects of beta, and beta 2 agonists and antagonists on intraocular pressure in the cat. INVEST OPHTHALMOL VIS SCI 18(ARVO Suppl.):24, 1979. 25. Nathanson JA: Effects of a potent beta 2 -adrenoceptor antagonist on intraocular pressure. Br J Pharmacol 73:97, 1981.