Investigative Ophthalmology & Visual Science, Vol. 33, No. 1, January 1992 Copyright Association for Research in Vision and Ophthalmology Effects of Cholinergic and Adrenergic Agonists on Adenylate Cyclase Activity of Retinal Microvascular Pericytes in Culture Gabryleda Ferrari-Dileo, E. Barry Davis, and Douglas R. Anderson Pericytes are contractile cells that might help regulate microvascular blood flow. To understand their potential role in the regulatory responses of the retina and optic nerve head vessels, the response of pericytes isolated from bovine retinal microvessels was determined to oxotremorine, isoproterenol, phenylephrine, and clonidine. Isoproterenol doubled the basal levels of cyclic adenosine monophosphate (camp) specifically through /?-adrenergic receptors, because the effect was blocked by dl-propranolol. The a, agonist phenylephrine did not induce any major change in adenylate cyclase activity. The a 2 agonist clonidine decreased basal camp synthesis and reduced the effect of isoproterenol. The cholinergic agonist oxotremorine did not modify the basal activity of adenylate cyclase but was able to decrease by almost 50% the forskolin-induced increase of camp. These results suggest that pericytes have functional adrenergic and cholinergic receptors, and they might respond to autonomic vasoactive substances present in vivo. Invest Ophthalmol Vis Sci 33:42-47, 1992 Blood flow in the retina and optic nerve head is autoregulated. 12 The optic nerve vessels are mainly capillaries; therefore, regulatory activity may reside in these microvessels. These vessels lack the smooth muscle cells found in arteries but have abundant pericytes. Pericytes are distinctive contractile cells, 3 " 5 numerous in retinal capillaries, 6 that together with endothelial cells might control microvascular blood flow. The exact role of these cells is not known, and little is known about what causes or influences their contraction. Faulty autoregulation may permit pressure-induced optic nerve damage in glaucoma. 7 ' 8 This deficient autoregulation could be caused, among other things, by circulating vasoactive substances that diffuse from the choroid into the optic nerve head. 910 Contractile tone in microvessels might limit the vessel's capacity for autoregulatory dilation, and we wished to know which types of external agents might induce such tone. The potential effects of vasoactive substances on the tone or contractility of microvessels can be judged from their biochemical responses to selected vasoactive substances, neurotransmitters, From the Department of Ophthalmology, Bascom Palmer Eye Institute, University of Miami School of Medicine, Miami, Florida. Supported in part by Public Health Service research grant R01- EY-00031 and a core center grant P30-EY-02180 from the National Institutes of Health, Bethesda, Maryland. Submitted for publication: February 27, 1991; accepted July 22, 1991. Reprint requests: Douglas R. Anderson, MD, Bascom Palmer Eye Institute, P.O. Box 016880, Miami, FL 33101. and hormones that promote physiologic responses through specific receptors. Intermediate mechanisms in the cells, such as the release of intracellular calcium ion (Ca 2+ ), changes in cyclic adenosine monophosphate (camp) or cyclic guanosine monophosphate (cgmp) levels, stimulation of phospholipase C, and opening of Ca 2+ channels, are among the biochemical tools that might be used to study responsiveness of the receptors in a living cell or tissue. Therefore, we studied the effects of cholinergic and adrenergic agents on the adenylate cyclase activity of the pericytes isolated from retinal microvessels. Changes in enzymatic activity showed that these cells respond to exposure to these vasoactive agents. Materials and Methods Agonists and Antagonists Prazosin was obtained from Pfizer (Brooklyn, NY). Isoproterenol, phenylephrine, clonidine, forskolin, dl-propranolol, carbachol, oxotremorine, and atropine were all obtained from Sigma (St. Louis, MO). Pericyte Culture Bovine retinal pericytes were isolated and cultured as described earlier.'' In brief, retinas were obtained from bovine eyes placed on ice at the time of slaughter and shipped overnight (BioResources, Irving, TX). The retinas were removed and digested with collagenase (Sigma). Dispersed cells were sieved, centrifuged, and plated in cultureflasks.they were maintained at 37 C in Dulbecco's modified Eagle's medium (Gibco, 42
No. 1 AUTONOMIC AGONISTS IN PERICYTE5 / Ferrari-Dileo er ol 43 Grand Island, NY), supplemented with 10% fetal calf serum (Gibco), 50 Mg/ml amphotericin B (Sigma), and 1.25 Mg/ml gentamicin (Sigma). At confluency (after 2-3 weeks), pericytes were identified by their slow growth, their morphology, the presence of both muscle and nonmuscle actin 12 (using rabbit antiactin immunoglobulin G and rabbit muscle-specific antiactin immunoglobulin G [Biomedical Technologies, Stoughton, MA] reacted with rhodamine-labeled goat anti-rabbit immunoglobulin G [Organon Teknika, West Chester, PA]), and the absence of acetylated low-density lipoprotein (Ac-LDL) uptake 13 (labeled with 1,l'-dioctadecyl 1-2, 3,3',3'-tetramethyl-indocarbocyanine perchlorate; Biomedical Technologies). camp Determination Intracellular levels of camp were determined by radioimmune assay using the Dual Range camp kit commercially available (Amersham, Arlington Heights, IL). Primary cultures or first-passage pericytes were grown in multiwell tissue culture plates (Corning Glass Works, Corning, NY) and allowed to reach confluency. Before each experiment, the cells were rinsed with balanced saline solution (BSS, 135 mm NaCl, 2.7raMKC1, 8.4 mm Na 2 HPO 4, 1.4 mm KH 2 PO 4, 1.1 mm d-glucose; Fisher, Fair Lawn, NJ) at room temperature. For a typical experiment, at least four multiwell plates were used, and each pair of wells represented one condition (eg, control, forskolin with or without agonists, agonists in different concentration, or agonists with or without antagonists). On each day for each set of conditions, controls and forskolin were included. Isobutylmethyl xanthine (IBMX, Sigma) was added to all experiments, including the controls, to inhibit endogenous and carbachol-stimulated phosphodiesterase activity. All agents were added in equal volumes; the wells had afinalvolume of 1.8 ml. The IBMX, forskolin (when needed), and antagonists were added 10 min before the agonists. After 15 min, the medium was removed, and the cells were washed with drug-free BSS. In the set of experiments that included prazosin, 0.01% ascorbic acid was present in the BSS as an antioxidant. Intracellular camp was extracted with 0.5 ml 0.1 N HC1 and determined in duplicate together with the standards also in 0.1 N HC1. Proteins were extracted with 0.5 ml of 0.1 N NaOH and determined by the Lowry method, 14 with bovine serum albumin as the standard. Statistical Analysis Statistical significance was analyzed by unpaired analysis of variance. Differences were considered significant when P <, 0.05. Results The retinal microvascular pericytes had a slow doubling time, as expected, and reached confluency in 35-mm wells in 2-3 weeks. The staining with musclespecific antiactin (not shown) was diffuse, and with the nonselective antiactin, the bundles of nonmuscle actin became evident (Fig. 1). These cells did not take up the fluorescent Ac-LDL (not shown). The growth features, morphology, and staining characteristics of these cells were identical with those obtained by other authors. 12-15 Cholinergic Agonists Basal production of camp varied between 12-25 pmol/mg protein/15 min. Neither carbachol nor oxotremorine induced a major change in basal camp levels in 15-min incubations (Fig. 2, open circles) in any of the concentrations tested (0.1-100 fim). Because these two agonists mediate their responses mainly through the M2 cardiac type of muscarinic receptor, which is linked to an inhibitory effect on adenylate cyclase activity, 16 we artificially elevated the enzyme activity with the diterpene, forskolin, so that inhibitory effects could be recognized. Forskolin (1 /im) increased basal levels pf camp to 66 pmol/mg protein/15 min (3-4.5-fold over control). Carbachol appeared to decrease somewhat the activity of forskolin-stimulated adenylate cyclase, but the effects were not statistically significant (Fig. 2, closed circles). Oxotremorine decreased forskolin-stimulated levels of camp by almost one half at all doses tested. Preincubation with the muscarinic antagonist, atropine (1 /xm) selectively blocked the inhibitory effect of oxotremorine at 10 jtm and 100 /*M (not shown). Adrenergic Agonists Isoproterenol increased the basal adenylate cyclase activity by 2.3-fold at all concentrations tested (Fig. 3, open circles). When coincubated with 1 pm forskolin, isoproterenol in concentrations of 0.01-1 /um boosted the already stimulated activity of the enzyme by almost 30% (Fig. 3, closed circles). The effect of isoproterenol was mediated specifically through adrenergic receptors, since dl-propranolol (a 0, and 3 2 antagonist) blocked the effects of the highest doses of isoproterenol tested (Fig. 3, closed diamonds). Phenylephrine in concentrations ranging from 0.1-100 /zm had a tendency to stimulate adenylate cyclase activity, although the effect was not statistically different from control values (Fig. 4, open circles). When the pericytes were preincubated with the 8 blocker dl-propranolol, the observed tendency of phenylephrine to increase camp levels was abolished (Fig. 4, diamonds). However, after preincubation with 1 mm
44 INVESTIGATIVE OPHTHALMOLOGY G VISUAL SCIENCE / January 1992 Vol. 33 Fig. 1. Phase contrast (A) and fluorescent photomicrographs (B) of retinal microvascular pericytes that had been incubated with the rabbit anti-actin IgG (nonselective) and rhodamine-labeied goat anti-rabbit IgG. B prazosin (an a, antagonist), phenylephrine produced a marked increase in adenylate cyclase activity; the levels almost reached those observed in the presence of the /3 agonist isoproterenol (Fig. 4, closed circles). Clonidine decreased basal production of camp when tested in concentrations ranging from 1-100 i*m (Fig. 4, open circles) and reduced the isoproterenol (10~5 M) response by 30% (Fig. 4, closed circles). Discussion This study showed the presence of muscarinic and adrenergic receptors linked to adenylate cyclase in cultured retinal microvessel pericytes. Muscarinic acetylcholine receptors are a family of distinct receptor subtypes associated with guanine nucleotide proteins (G proteins). These are classified inm, and M2 types according to their pharmacologic and biochemical characteristics. The M{ type has a high affinity for the antagonist pirenzepine and is associated with phosphatidyl inositol (PI) turnover, whereas the M2 type has a low affinity for pirenzepine and is associated with inhibition of adenylate cyclase. At least five genes encode for five distinct subtypes of receptors. The m b m3, and m5 subtypes belong to the M,
No. 1 AUTONOMIC AGONISTS IN PERICYTES / Ferrori-Dileo er ol 45 701 60-i 50- c 40- E to E in -^j=30h o a 30-20 o-* 10 60-50- 20-10- o -8-7 -6-5 Isoproterenol 4 I Log Ml ^ 5 Carbachol llogml _?. 5. 4 Oxotremorine Fig. 2. Effect of carbachol and oxotremorine on adenylate cyclase activity. Control conditions (O) and in the presence of 1 ^M forskolin ( ). Points are the average of three to four independent experiments run in duplicate. Bars represent SEM. *P < 0.001 (ANOVA). Note that only oxotremorine significantly reduces forskolin-increased adenylate cyclase activity. Fig. 3. Effect of isoproterenol on adenylate cyclase activity of retinal pericytes. Control conditions (O), in the presence of 1 /xm forskolin ( ) and in the presence of 0.1 mm dl-propranolol ( ). Each point is the mean of three to four experiments run in duplicate. Bars are SEM. *P < 0.001 (ANOVA) (different from control); **P < 0.001 (ANOVA) (different from isoproterenol alone). type, whereas the m 2 and m 4 belong to the class associated with agonist-induced inhibition of adenylate cyclase. 17 The distribution of these receptors among different tissues is not clear but they may have more than one subtype. The final effect in any tissue may result from the fine tuning of an agonist acting on more than one receptor. Carbachol, for example, is able both to stimulate PI hydrolysis and to inhibit camp production in a narrow concentration range. The stimulated hydrolysis of inositol lipids results in an increased intracellular Ca 2+ with a possible indirect stimulation of adenylate cyclase and an increase of camp formation. 16 At the same time, carbachol is able to decrease camp level in cells by stimulating phosphodiesterase activity. 18 Although we cannot discriminate the subtypes of acetylcholine muscarinic receptors present in retinal microvascular pericytes, active muscarinic receptors, perhaps more than one type, are present in these pericytes. In addition, at least one subtype is associated with inhibition of adenylate cyclase. This can be concluded from the results obtained with oxotremorine, which has a less marked effect on PI turnover than carbachol. 50-i 40-30- 20 10-7 -6-5 -4 Phenylephrine I Log Ml -6-5 -4 Clonidine Fig. 4. Effects of phenylephrine and clonidine. Left, Phenylephrine alone (O), phenylephrine in the presence of 1 mm prazosine ( ), phenylephrine in the presence of 0.1 mm dl-propranolol ( ). *P < 0.001 ANOVA (different from phenylephrine alone). Right, Clonidine (O), and clonidine in the presence of 10 nm isoproterenol ( ). *P < 0.001 ANOVA (different from control); **P < 0.002 ANOVA (different from isoproterenol alone). Each point is the mean of three to four experiments run simultaneously in duplicate. Bars are SEM.
46 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / January 1992 Vol. 33 With regard to adrenergic receptors, the fi type seems to be ubiquitous. Our results show that isoproterenol, which activates both /?, and /3 2 adrenoreceptors, increases both basal and forskolin-stimulated adenylate cyclase activity in pericytes in a wide range of concentrations. The increase of camp produced by 0.1 mm isoproterenol has been described in retinal pericytes, 5 and the effect was blocked by dl-propranolol. The observed increase in camp by isoproterenol therefore was an effect mediated specifically through 0-adrenergic receptors. Phenylephrine is considered a "selective" a,-adrenergic agonist, but it also exhibits some prejunctional a 2 and postjunctional 0-agonist activity in several tissues. 19 Our results suggest that phenylephrine has an effect on adenylate cyclase activity, probably through a nonspecific effect on /3-adrenergic receptors. Even if a,-adrenergic receptor stimulation leads to an increase in adenylate cyclase activity in cerebrovascular smooth muscle as reported, 20 we did not observe any marked a, response (after dl-propranolol /3 blockade). Clonidine, however, had opposite effects on the same biochemical process as isoproterenol, when acting alone or in its presence. The predominance of a particular response through any of these postjunctional adrenergic receptors is difficult to predict, but the demonstration of various receptors shows the complexity of the control of the microvascular pericytes by circulating or locally produced catecholamines. The idea that pericytes may contract under different stimuli comes from observations made almost 80 years ago. These reports describe changes in size or shape of capillaries and/or microvascular pericytes. 21>22 Microvascular pericytes have many features of contractile cells. They contain all proteins implicated in contraction events (ie, actin, myosin, and tropomyosin). 1223 " 25 Like other contractile cells, pericytes show changes in intracellular camp through receptor activation. The presence of a cgmp-dependent protein kinase in pericytes implicates cgmp as another second messenger. 26 Probably the most direct evidence for a potential role in microvascular tone control comes from the observations that pericytes in culture enter a contracted state as they grow. 527 This state can be modified by agents known to affect vascular contractibility, such as camp derivatives, isoproterenol, and histamine. 5 Our results support two main conclusions. One is that pericytes have functional receptors for autonomic neurotransmitters. It is possible, therefore, that there is an effect of circulating or locally produced autonomic substances on the tone of this microvascular bed under normal or pathologic conditions. The second is that several receptors are linked to multiple effector systems, even in one type of cell. This makes it difficult to predict the net final physiologic effect on intact tissues or organisms when in contact with agonist drugs. Key words: adrenergic, retinal pericytes, cyclic AMP, adenylate cyclase, cell culture References 1. Geijer C and Bill A: Effects of raised intraocular pressure on retinal, pre-laminar, laminar, and retrolaminar optic nerve blood flow in monkeys. Invest Ophthalmol Vis Sci 18:1030, 1979. 2. Weinstein JM, Funsch D, Page RB, and Brennan W: Optic nerve bloodflowand its regulation. Invest Ophthalmol Vis Sci 23:640, 1982. 3. Kelley C, D'Amore P, Hechtman HB, and Shepro D: Microvascular pericyte contractility in vitro: Comparison with other cells of the vascular wall. J Cell Biol 104:483, 1987. 4. Das A, Frank RN, Weber ML, Kennedy A, Reidy CA, and Mancini MA: ATP causes retinal pericytes to contract in vitro. Exp Eye Res 46:349, 1988. 5. Kelley C, D'Amore P, Hechtman HB, and Shepro D: Vasoactive hormones and camp affect pericyte contraction and stress fibers in vitro. J Muscle Res Cell Motil 9:184, 1988. 6. Frank RN, Dutta S, and Mancini MA: Pericyte coverage is greater in the retinal than in the cerebral capillaries of the rat. Invest Ophthalmol Vis Sci 28:1086, 1987. 7. Ernest JT: Autoregulation of blood flow in the distal segment of the optic nerve. In Glaucoma Update, International Glaucoma Symposium, Nara/Japan, May 7-11, 1978, Krieglstein GK and Leydhecker W, editors. Berlin, Springer-Verlag, 1979, pp. 33-979. 8. Ernest JT: Pathogenesis of glaucomatous optic nerve disease. Trans Am Ophthalmol Soc 73:366, 1975. 9. Anderson DR: The posterior segment of glaucomatous eyes. In Basic Aspects of Glaucoma Research, Lutjen-Drecoll E, editor. New York, FK Schattauer Verlag, 1982, pp. 167-190. 10. Anderson DR: The mechanisms of damage of the optic nerve. In Glaucoma Update II, Krieglstein GK and Leydhecker W, editors. New York, Springer-Verlag 1983; pp. 89-93. 11. Gitlin JD and D'Amore PA: Culture of retinal capillary cells using selective growth media. Microvasc Res 26:74, 1983. 12. Herman IM and D'Amore PA: Microvascular pericytes contain muscle and non-muscle actin. J Cell Biol 101:43, 1985. 13. Voyta JC, Via DP, Butterfield CE, and Zetter BR: Identification and isolation of endothelial cells based on their increased uptake of acetylated low-density lipoproteins. J Cell Biol 99:2034, 1984. 14. Lowry OH, Rosebrough NJ, Farr AL, and Randall RJ: Protein measurement with the Folin phenol reagent. J Biol Chem 193:265, 1951. 15. Capetandes A and Gerritsen ME: Simplified methods for consistent and selective culture of bovine retinal endothelial cells and pericytes. Invest Ophthalmol Vis Sci 31:1738, 1990. 16. Mei L, Roeske WR, and Yamamura HI: Molecular pharmacology of muscarinic receptor heterogeneity. Life Sci 45:1831, 1989. 17. Bonner TI: The molecular basis of muscarinic receptor diversity. Trends in Neuroscience 12:148, 1989.
No. 1 AUTONOMIC AGONISTS IN PERICYTES / Ferrori-Dileo er ol 47 18. Meeker RB and Harden TK: Muscarinic cholinergic receptormediated activation of phosphodiesterase. Mol Pharmacol 22:310, 1982. 19. McGrath JC: Evidence for more than one type of post-junctional a-adrenoceptor. Biochem Pharmacol 31:467, 1982. 20. Wroblewska B, Spatz M, Merkel N, and Bembry J: Cerebrovascular smooth muscle culture: II. Characterization of adrenergic receptors linked to adenylate cyclase. Life Sci 34:783, 1984. 21. Clark ER and Clark EL: The relation of "Rouget" cells to capillary contractility. Am J Anat 35:265, 1925. 22. D'Amore PA: Culture and study of pericytes. In Cell Culture Techniques in Cardiovascular Research, Piper HM, editor. New York, Springer-Verlag, 1990, pp. 299-314. 23. Joyce NC, Haire MF, and Palade GE: Contractile proteins in pericytes: I. Immunoperoxidase localization of tropomyosin. J CellBiol 100:1379, 1985. 24. Joyce NC, Haire MF, and Palade GE: Contractile proteins in pericytes: II. Immunocytochemical evidence for the presence of two isomyosins in graded concentrations. J Cell Biol 100:1387, 1985. 25. Chan LS, Li W, Khatami M, and Rockey JH: Actin in cultured bovine retinal capillary pericytes: Morphological and functional correlation. Exp Eye Res 43:41, 1986. 26. Joyce NC, DeCamilli P, and Boyles J: Pericytes, like vascular smooth muscle cells, are immunocytochemically positive for cyclic GMP dependent protein kinase. Microvasc Res 28:206, 1984. 27. Schor AM and Schor SC: The isolation and culture of endothelial cells and pericytes from the bovine retinal microvasculature: A comparative study with large vessel vascular cells. Microvasc Res 32:21, 1986.