Alpha ^Adrenoceptors in Human Corneal Epithelium

Ronald J. Walkenbach,*t Guo-Sui Ye,* Peter 5. Reinach,^: and Frances Boney*

Specific binding of the potent, selective alphaj-adrenoceptor antagonist 3H-prazosin was demonstrated in cultured human corneal epithelial cells. Specific binding of the radioligand was concentration-dependent between 0.5 and 6 nM, with apparent saturation of receptor sites seen at higher concentrations. The cells exhibited a maximum binding capacity for 3H-prazosin of 225 fmol/mg of cellular protein and a dissociation constant of 2 nM. The binding of 3H-prazosin was competitive with known alpha,-adrenoceptor ligands and was reversible. Epithelium of intact human corneas also exhibited specific 3H-prazosin binding, as did cultures of bovine and rabbit corneal epithelium. The alpha-adrenergic agonist methoxamine significantly stimulated phosphatidylinositol 4,5-bisphosphate hydrolysis, measured as myoinositol trisphosphate accumulation in cultures of human corneal epithelium. This stimulation was inhibited by the presence of prazosin during the assays. These findings indicate the existence of specific, reversible, high-affinity receptors for alpha,-adrenoceptors that regulate inositol phosphate turnover in human, rabbit, and bovine corneal epithelial cells. Invest Ophthalmol Vis Sci 32:3067-3072,1991

The cornea is innervated by adrenergic nerve late inositol phosphate turnover in human corneal epi- fibers,1"3 but their role(s) in corneal physiology re- thelial cells, a finding analogous to that previously main poorly understood. The existence of beta-adren- reported in rabbit corneal epithelium. oceptors on corneal epithelial cells has been estab- 4 5 Methods and Materials lished ' and shown to be predominately of the beta2 subtype.6 Corneal epithelial beta-adrenoceptors have Rabbit and bovine eyes were obtained from local been associated with stimulation of adenylate cyclase slaughterhouses within 2 hr after the animals were 7 9 and cyclic AMP-dependent protein kinase, " chlo- killed. The eyes were kept on ice for up to 4 hr more, 10 14 ride secretion, " as well as inhibition of mitotic until further processing occurred. Human eyes were 15 8 rates and glycogen synthase activity. obtained from the Missouri Lions Eye Tissue Bank. The existence of alpha-adrenoceptors on corneal ep- The corneas of human eyes were dissected within 12 ithelial cells is less well understood. Some preliminary hr postmortem and stored in Dexsol (Chiron Ophthal- reports using broken cell tissue preparations have sug- mics, Irvine, CA) or a similar medium composed of gested the absence of alpha-adrenoceptors,1617 M-199 tissue culture medium supplemented with whereas other studies have used drugs with alpha- 1.35% chondroitin sulfate, 1% dextran (40,000 kDa), adrenoceptor agonist properties to demonstrate stimu- 17 raM Na bicarbonate, 20 mM HEPES buffer, 12 lation of ion transport in frog and inositol phosphate brought to a final pH of 7.4 with 1 N NaOH. Corneas 18 turnover in rabbit corneal epithelium. were stored at 4°C for up to 72 hr in one of these The direct radioligand binding studies shown here media before initiation of tissue fractionation, cell indicate that intact corneal epithelial cells from rab- culture, or binding experiments. bit, bovine, and human tissue exhibit high-affinity, Particulate fractions of native corneal epithelium specific alpha radrenoceptors. These receptors regu- from rabbit, bovine, or human corneas were prepared as previously described.19 Briefly, the corneas were rinsed with an ice-cold solution containing 10 mM From the *Missouri Lions Eye Research Foundation, Columbia, K2HPO4 in 0.9% NaCl (PBS), and the corneal epithe- Missouri; the f Departments of Ophthalmology and Pharmacology, University of Missouri, Columbia, Missouri; and the tDepartment lium was removed from the isolated cornea (human) of Physiology and Endocrinology, Medical College of Georgia, Au- or eye (bovine and rabbit) with a scalpel blade. Tissue gusta, Georgia. was homogenized in 1 ml per cornea of 25 mM gly- Supported by National Institutes of Health grants EY 02597 and cylglycine buffer (pH = 7.6) with a Teflon/glass tissue EY 04795. grinder in the cold. The homogenate was centrifuged Submitted for publication: March 20, 1991; accepted June 17, 1991. at 50,000 X g for 30 min at 4°C. The pellet was resus- Reprint requests: Ronald J. Walkenbach, PhD, The Missouri pended in fresh buffer, and the centrifugation and re- Lions Eye Research Foundation, 404 Portland Street, Columbia, suspension steps were repeated to produce each epithe- MO 65201. lial particulate fraction.

3067

Downloaded from iovs.arvojournals.org on 09/28/2021 3068 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / November 1991 Vol. 32

Rabbit, bovine, or human coraeal epithelial cells as the difference between the measured total 3H-pra- were cultured in six-well multiplates as described pre- zosin bound and nonspecific bound in each set of par- viously.19 Briefly, epithelial tissue pieces (with some allel assays. residual stroma) were placed in 5 ml per well as Ea- Binding studies with intact human corneas were gle's minimum essential medium (MEM) with D-va- performed analogously except that incubations were line to inhibit keratocyte contamination of the epithe- performed in culture media. After incubation, the lial cultures20 and 10% newborn calf serum. Media corneas were rinsed briefly with ice-cold PBS, and the were changed after 1 week of culture and twice weekly epithelium was removed quickly with a surgical scal- thereafter. Each well typically contained 150-200 ng pel blade and placed in 2 ml of 2 N NaOH. After of cellular protein when used for experiments after digestion, the samples were neutralized before mea- 3-4 weeks of culture. suring their protein and radioactivity. Paniculate fractions of cultured epithelium were Binding protocols using intact cultured cells were prepared as described except that a rubber spatula was identical except that 1 ml of 1 N NaOH was added to used to remove cells from the wells. each well after the final PBS rinse. The potent alphaj-adrenoceptor antagonist 3H- Inositol phosphate turnover experiments in cul- prazosin (87 Ci/mmol; NEN Research Products, Bos- tured human corneal epithelial cells were performed ton, MA) was used to assess receptor binding activity using the basic technique described by Martin.21 The in these tissue preparations. The protocol used for growth medium was removed and the cells washed binding to epithelial particulate fractions was analo- several times with inositol-free minimum essential gous to that described for 3H-quinuclidinyl benzilate medium (IFMEM). This medium was prepared by binding to muscarinic cholinoceptors in this tissue.19 mixing Hank's salt solution (Sigma Chemical Co., St. Briefly, total binding of the radioligand was assessed Louis, MO) with MEM amino acid mixture (Sigma) by incubating the particulate fractions with the indi- and adding the individual vitamins as listed by the cated concentration of 3H-prazosin in 25 mM glycylg- MEM formulation. The cells were cultured for 48 hr lycine buffer, pH = 7.6 at 37°C for 30 min (unless with 3 ml of fresh IFMEM, supplemented with 2% otherwise indicated). The assays were filtered and dialyzed sterile calf serum (Sigma) and 33 nM 3H- washed free of unbound radioligand with three 5-ml myoinositol (3H-inositol; 15.6 Ci/mmol, NEN Re- aliquots of ice-cold buffer. Each filterwa s placed in 10 search Products). The labeling medium was removed, ml of Scintiverse BOA cocktail (Fisher Scientific, and the cells were washed with PBS and preincubated Springfield, NJ), shaken for at least 1 hr in the dark, for 5 min at 37°C with serum-free IFMEM with 10 and counted by standard liquid scintillation tech- mM LiCl to block dephosphorylation of inositol-1- 3 22 niques. Nonspecific binding of H-prazosin was mea- PO4 to inositol. After the preincubation period, the sured by running parallel assays with 100 nM norepi- medium was replaced with fresh serum-free IFMEM nephrine added to the reaction mixtures during the containing the drugs indicated in Table 1 and incu- incubation. Specific binding of 3H-prazosih is defined bated for an additional 5 min. Reactions were termi-

Table 1. Comparison of 3H-prazosin binding in different tissue preparations of corneal epithelium from the rabbit, bovine, and human 3H-prazosin bound (fmol/mg)

Tissue preparation Total Nonspecific Specific Significance

Rabbit Fresh, particulate fraction 435 ± 34 456 ± 33 <0 NS Cultured, particulate fraction 189 ±23 193 ± 15 <0 NS Cultured, intact cells 295 ± 15 259 ± 19 36 P < 0.05 Bovine Fresh, particulate fraction 243 ±31 238 ± 25 5 NS Cultured, particulate fraction 210 ± 17 201 ± 27 9 NS Cultured, intact cells 299 ± 25 154 ±20 145 P < 0.005 Human Fresh, particulate fraction 117± 13 126 ± 19 <0 NS Cultured, particulate fraction 165 ±21 155 ± 18 10 NS Cultured, intact cells 275 ± 36 182 ±33 93 P<0.01 Fresh, intact cornea 105 ± 7 64 ± 13 41 P < 0.01

Intact cells or particulate fractions were assayed using 2 nM 3H-prazosin 20 assays, using tissue from at least three different harvest dates. Statistical without or with 100 jtM norepinephrine. Each tissue preparation's total and significance was determined using the Student's t-test for unpaired samples. nonspecific 3H-prazosin binding value represents the mean ± SEM of 12 to Protein levels ranged from 75 to 100 ng per assay.

Downloaded from iovs.arvojournals.org on 09/28/2021 No. 12 ALPHA,-ADRENOCEPTOR5 IN HUMAN CORNEAL EPITHELIUM / Wolkenboch er ol 3069

nated by aspirating the incubation medium and add- 200 ing 1 ml of ice-cold 10% HC1O4 to each well. The cells in the wells were frozen, thawed, and kept on ice for 30 min to complete the extraction of 3H-inositol phosphates from the cells, then decanted. The residue in each well was saved for subsequent protein assays; the extract was neutralized using approximately 1 ml of 75 mM HEPES in 2 N KOH. The neutralized extract was centrifuged at 2,000 rpm for 2 min at 4°C. One milliliter of each supernatant fraction was applied to columns with a 3 cm X 0.7 cm diameter bed of Dowex 1x8-200 resin. 3H-inositol, 3H-inositol-l-phosphate, 3H-inositol-l,4-bisphosphate, and 3H-inositol-1,4,5- trisphosphate were isolated from each column by re- 1 10 100 taining the effluents after sequential applications of 6 log 3H-prazosin cone. (nM) ml of 0.1 M formic acid; 10 ml of 0.2 M ammonium Fig. 2. Concentration-dependence and saturation of 3H-prazo- formate in 0.1 M formic acid; 10 ml of 0.6 M ammo- sin-specific binding to human cornea epithelial cells in culture. nium formate in 0.1 M formic acid; and 7.5 ml of 1.0 Cells were incubated with the concentration of 3H-prazosin indi- M ammonium formate in 0.1 M formic acid, respec- cated under otherwise standard binding conditions. Specific bind- tively. A sample (2.5 ml) of each effluent was mixed ing was calculated for each 3H-prazosin concentration as described with 10 ml of Scintiverse BOA cocktail, stored in the in Figure 1. The data were treated according to the method of Ro- senthal (inset) to determine a maximal binding capacity (Bmax) for dark for ^ 1 hr and counted using liquid scintillation 3 3 H-prazosin of 225 fmol/mg and the concentration at which one- techniques. In preliminary experiments, H-inositol half of the receptors are occupied (KD), which was 2 nM. and standard samples of each of the 3H-inositol phosphate derivatives (NEN Research Products) were used to determine the optimal separation conditions and Protein was measured using the method of Lowry 23 the efficiency of isolating each product from the col- et al for all tissue preparations. umns. Unless otherwise indicated, the data are presented as the means ± SEM of three experiments, each using triplicate assays per experimental condition. All reagents were purchased from Sigma Chemical Co., and all supplies were obtained from Fisher Scien- 1 50 - tific unless specifically indicated. CD is O Results y—__ The kinetics of total, nonspecific, and specific bind- 3 T3 1 00 - ing of H-prazosin to human corneal epithelial cells in

O culture are shown in Figure 1. A steady state of binding was reached within 20 min of incubation at 37°C, and binding levels remained stable for at least 30 min thereafter. Although nonspecific binding of 3H-prazosin under these conditions was substantial, a signifi- X cant difference between total and nonspecific binding (specific binding) was consistently observed. The level of specific 3H-prazosin binding to cultured human corneal epithelial cells was proportional 20 30 to the radioligand concentration between 0.5 and 6 assay time (min) nM, as shown in Figure 2. Rosenthal analysis24 of Fig. 1. Time course of 3H-prazosin binding to human corneal these data (inset, Fig. 2) indicated a maximal binding epithelial cells in culture. Cells were incubated using 1 nM 3H-pra- capacity (B ) for 3H-prazosin of 225 fmol/mg pro- 3 max zosin alone (—•—) to assess total binding or with 1 nM H-prazo- tein and a dissociation constant (KJ of 2 nM. sin + 100 fiM norepinephrine in parallel assays to measure nonspe- Agents with known alpha-adrenoceptor potency cific binding (—•—) of the radioligand. Specific binding (—A—) 3 was calculated as the difference between the levels of total and non- were able to compete with H-prazosin binding dur- specific binding. Each symbol represents the mean of nine determi- ing incubation with cultured human corneal epithe- nations. Brackets indicate the SEM. lium (Fig. 3). Unlabeled prazosin was the most effec-

Downloaded from iovs.arvojournals.org on 09/28/2021 3070 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / November 1991 Vol. 32

tured bovine corneal epithelium showed the highest specific 3H-prazosin binding activity tested—145 fmol/mg protein. Cultured human corneal epithelium displayed less 3H-prazosin binding activity (93 fmol/mg) than bovine tissue, but the level of specific binding was still highly significant. Moreover, specific binding of 3H-prazosin also could be observed in fresh, intact human corneal epithelium, if whole corneas were incubated with the radioligand before isolation of the epithelium.

In most tissues studied, alpharadrenoceptors are associated with regulating the formation of two intra- -11 -9 -7 -5 log unlabelled ligand (M) cellular second messengers: inositol-1,4,5-trisphosphate and 1,2-diacylglycerol.25 Accordingly, the effect Fig. 3. Inhibition of 3H-prazosin binding to cultured human cor- of an alpha adrenoceptor agonist and antagonist were nea epithelial cells by alpha adrenergic agents. Control cells were 3 tested on the inositol phosphate turnover in cultured incubated with 1 nM H-prazosin alone. Other cells were incubated human corneal epithelial cells. Table 1 shows that in- with 1 nM 3H-prazosin and the indicated concentration of either unlabeled prazosin (—•—), yohimbine (—*—), norepinephrine cubation of the cells with 100 nM methoxamine signif- (—A—), methoxamine (—0—), or phenylephrine (—•—). Each icantly elevated levels of inositol-1,4,5-trisphosphate, symbol represents the mean binding level of 3H-prazosin at its re- as well as its bisphosphate and monophosphate metab- spective experimental condition, expressed as a percent of the con- olites in these cells. When cells were incubated with trol level of binding. methoxamine in the presence of 0.1 nM prazosin, the level of inositol-1,4,5-trisphosphate was significantly reduced compared with that seen with methoxamine tive agent tested, requiring only 1 nM to reduce the alone. 3 level of 1 nM H-prazosin binding by 50% (IC50). Yo- himbine, a relatively selective alpha2-adrenoceptor antagonist, was much less effective than prazosin at in- 3 100 hibiting H-prazosin binding, exhibiting an IC50 of 100 nM. At somewhat higher concentrations, the alpha-adrenoceptor agonists norepinephrine, methoxamine, and phenylephrine also competed with 3H- prazosin binding, exhibiting IC50 values of 2 fiM, 10 nM, and 50 /iM, respectively. The beta-adrenoceptor antagonist timolol did not compete with 3H-prazosin binding, using concentrations up to 1 mM (not shown). Human corneal epithelial specific binding of 3H- prazosin was readily reversible, as shown in Figure 4. When excess unlabeled norepinephrine was added to assays after 3H-prazosin had reached a steady state of binding, the quantity of bound radioligand decreased in a time-dependent manner for approximately 15 min, until it reached a new steady state at the nonspecific 3H-prazosin binding level. 10 20 30 The levels of 3H-prazosin binding were compared in several types of epithelial preparations from rabbit, time (min) bovine, and human corneas. Table 2 shows that signif- 3 3 Fig. 4. Reversibility of specific H-prazosin binding to human icant levels of specific H-prazosin binding could not cornea epithelial cells in culture. All cells were incubated with 2 nM be observed in paniculate fractions of fresh or cul- 3H-prazosin for 30 min under otherwise standard conditions. Con- tured corneal epithelium from any species tested. trol assays were terminated at this time. All other cells received 100 However, specific binding was seen in all of the intact ^M norepinephrine and incubation continued. At the times indicated, appropriate assays were terminated and the cells processed cell preparations tested. Cultured cells from rabbit epi- for measurement of radioactivity, as usual. Symbols portray the thelium exhibited relatively small, but statistically sig- mean level of 3H-prazosin bound at the indicated time after norepi- nificant, levels of specific 3H-prazosin binding. Cul- nephrine addition, expressed as a percent of the control value.

Downloaded from iovs.arvojournals.org on 09/28/2021 No. 12 ALPHA,-ADRENOCEPTORS IN HUMAN CORNEAL EPITHELIUM / Walkenbach er al 3071

Discussion However, the corneal epithelium may be a unique tissue in this sense because our laboratory has ob- The experiments described here indicate that cor- tained a similar finding with muscarinic cholinocep- neal epithelial cells exhibit physiologically relevant al- tor binding activity in intact versus broken cells.19 pha ,-adrenoceptors that are associated with the regula- The possibility that the corneal epithelium releases tion of inositol-l,4,5-trisphosphate formation. The protein inactivating agents upon cellular disruption receptors displayed a high affinity for a well-recog- 3 seems plausible and may be a physiologically impor- nized alpha radrenoceptor antagonist, H-prazosin tant defense mechanism to protect the rest of the cor- (Fig. 2), and exhibited saturation of the receptors at nea from foreign infiltration. radioligand concentrations greater than 6 nM. Corneal epithelial alpha ,-adrenoceptors appear to Known alpha-adrenoceptor antagonists and agonists be associated with regulation of inositol-1,4,5-tris- competed for binding sites with 3H-prazosin during phosphate formation because the alpha adrenoceptor incubations with epithelial cells (Fig. 3). The relative agonist methoxamine increased turnover of inositol- affinities and the order of potencies seen among the 1,4,5-trisphosphate and its metabolites in human cor- competing alpha-adrenoceptor ligands in Figure 3 are neal epithelial cells (Table 2) in the same concentra- consistent with those seen in many other tissues with 25 tion range that it was able to effectively compete with demonstrated alpha-adrenoceptor systems. 3 H-prazosin (Fig. 3). Moreover, prazosin inhibited In Figure 3, the selective alpharadrenoceptor antag- the methoxamine-induced effects on inositol phos- onist yohimbine was a much less potent competitor 3 phate turnover at a concentration expected from its for H-prazosin binding sites than was the selective receptor-binding potency shown in Figures 2 and 3. alpha ,-adrenoceptor antagonist prazosin. Moreover, The magnitudes of inositol phosphate turnover by me- our laboratory has been unable to demonstrate spe- 3 thoxamine (Table 2) were similar to those previously cific H-yohimbine binding (not shown) using the reported in rabbit corneal epithelial tissue in response same tissue preparations and experimental conditions to norepinephrine.18 employed for these experiments. Thus, it appears that The findingo f alpha ,-adrenoceptors in bovine, rab- corneal epithelial alpha-adrenoceptors are primarily, bit, and human corneal epithelium and their regula- if not exclusively, of the alpha, subtype. tion of inositol phosphate turnover in human tissue is The data in Table 1 clearly show that particulate in accord with their previously defined role in this fractions of fresh or cultured epithelium, unlike intact 17 3 tissue as modulators of active ion transport and ino- cells, failed to exhibit specific H-prazosin binding. 18 3 sitol phosphate turnover using drugs with alpha- Similarly, there was no specific H-prazosin binding adrenergic potency. Alpha,-adrenoceptors appear to in other subcellular fractions of this tissue or after play a role in corneal epithelial cell homeostasis by treatment of intact cells with trypsin (data not regulating levels of intracellular second messengers shown). It appears that cellular disruption, at least as such as Ca2+, inositol 1,4,5-trisphosphate, and 1,2- employed in these studies, destroys or masks the cor- 3 diacylglycerol, as previously demonstrated in many neal epithelial cell's ability to specifically bind H- other tissues.26 prazosin. This is an unusual finding because many The physiologic responses to alpha,-adrenoceptor tissues show greater alpha-adrenoceptor binding ac- 26 activation and the second messengers described here tivity in broken cell preparations versus intact cells. have not been determined in the corneal epithelium. They may contribute to the regulation of corneal de- Table 2. Inositol phosphates in human corneal turgescence, although this effect would not appear to epithelial cultures be quantitatively important because our laboratory has not detected an effect of alpha,-adrenergic agents Inositol phosphate content (fmol/mg protein) on rabbit corneal thickness when the corneas were cultured in vitro (data not shown). Nevertheless, these Condition IP IP, IPs agents may have other physiologically important Control 470.1 ±70.9 221.2 ±38.7 27.4 ± 3.2 transport effects within the epithelium. It also is possi- Methoxamine, 100 MM 558.1 ± 52.8 265.0 ± 50.6 39.8 ± 7.4 ble that alpha ,-adrenoceptors contribute to the regula- Methoxamine, 100 nM tion of epithelial mitosis or the migration of epithe- + prazosin, 0.1 MM 506.0 ± 67.4 170.3 ± 56.8 32.8 ± 4.1 lium toward the anterior surface. The latter possibility Human corneal epithelial cultures were incubated without drugs (control) is being investigated in our laboratory. or with the drugs indicated above followed by isolation of 3H-inositol-1 -phos- 3 3 phate (IP), H-lnositol-1,4-bisphosphate (IP2), and H-inositol-1,4,5-trisphosphate (IP3). The data indicate the mean ± SEM of nine assays. The Student's t-test was used to ascertain significant differences in IP3 levels between control and methoxamine-treated cultures (P < 0.01) and between methoxamine- Key words: cornea, epithelium, alpha ,-adrenoceptors, al- treated and methoxamine + prazosin-treated cultures (P < 0.05). pha ,-adrenergic, receptors, human, inositol phosphate

Downloaded from iovs.arvojournals.org on 09/28/2021 3072 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / November 1991 Vol. 32

Acknowledgments adrenergic receptors in the regulation of ion transport in the frog cornea. Am J Physiol 230:1487, 1976. The authors thank Cheryl Dake and Nels Holmberg for 13. Fischer FH, Schnitz L, Hoff W, Schartl S, Liegl O, and Wieder- their expert assistance in the execution of these experi- holt M: Sodium and chloride transport in the isolated human ments. cornea. Pflugers Arch 373:179, 1978. 14. Klyce SD and Wong RKS: Site and mode of adrenaline action on chloride transport across the rabbit corneal epithelium. J References Physiol 266:77, 1977. 15. Friedenwald JS and Buschke W: The effects of excitement, of 1. Ehinger B: Distribution of adrenergic nerves to orbital struc- epinephrine and of sympathectomy on the mitotic activity of tures. Acta Physiol Scand 62:291, 1964. the corneal epithelium in rats. Am J Physiol 141:689, 1944. 2. Laties A and Jacobowitz D: A histochemical study of the 16. Neufeld AH, Olsen JS, and Zawistowski KA: Autonomic re- adrenergic and cholinergic innervation of the anterior segment ceptors in the rabbit cornea. In ARVO Convention Abstracts. of the rabbit eye. Invest Ophthalmol 3:592, 1964. Chicago, IL, Association for Research in Vision and Ophthal- 3. Tervo T and Palkaraa A: Adrenergic innervation of the rat mology, 1978, p. 189. corneal epithelium. Invest Ophthalmol 15:147, 1976. 17. Walkenbach RJ, Bylund DB, Chao WT, and Gibbs SR: Adren- 4. Neufeld AH, Zawistowski KA, Page ED, and Bromberg BB: ergic and cholinergic receptors in the cornea. Invest Ophthal- Influences on the density of beta-adrenergic receptors in the mol Vis Sci 24(Suppl):198, 1983. cornea and iris-ciliary body of the rabbit. Invest Ophthalmol 18. Akhtar RA: Effects of norepinephrine and 5-hydroxytrypta- VisSci 17:1068, 1978. mine on phosphoinositide-P04 turnover in rabbit cornea. Exp 5. Colley AM and Cavanagh HD: Binding of [3H]di- Eye Res 44:849, 1987. hydroalprenolol and [3H]quinuclidinyl benzilate to intact cells 19. Walkenbach RJ and Ye GS: Muscarinic cholinoceptor regula- of cultured corneal epithelium. Metab Pediatr Syst Ophthal- tion of cyclic GMP in human corneal epithelium. Invest Oph- mol 6:75, 1982. thalmol VisSci 32:610, 1991. 6. Walkenbach RJ, Gibbs SR, Bylund DB, and Chao WT: Char- 20. Sunder-Raj CV, Freeman IL, and Brown SI: Selective growth acteristics of beta-adrenergic receptors in bovine corneal epithe- of rabbit corneal epithelial cells in culture and basement mem- lium: Comparison of fresh tissue and cultured cells. Biochem brane collagen synthesis. Invest Ophthalmol Vis Sci 19:1222, Biophys Res Commun 121:664, 1984. 1980. 7. Walkenbach RJ, LeGrand RD, and Barr RE: Characterization 21. Martin TF: Thyrotropin-releasing hormone rapidly activates of adenylate cyclase activity in bovine and human corneal epi- the phosphodiester hydrolysis of polyphosphoinositides in thelium. Invest Ophthalmol Vis Sci 19:1080, 1980. GH3 pituitary cells. J Biol Chem 258:14816, 1983. 8. Walkenbach RJ and LeGrand RD: Regulation of cyclic AMP- 22. Berridge MJ, Downes PC, and Hanley MR: Lithium amplifies dependent protein kinase and glycogen synthase by cyclic agonist-dependent phosphatidylinositol responses in brain and AMP in the bovine cornea. Exp Eye Res 33:111, 1981. salivary glands. Biochem J 206:587, 1982. 9. Reinach PS and Kirchberger MA: Evidence for catecholamine- 23. Lowry OH, Rosebrough NJ, Farr AL, and Randall R: Protein stimulated adenylate cyclase activity in frog and rabbit corneal measurement with the Folin phenol reagent. J Biol Chem epithelium and cyclic AMP-dependent protein kinase and its 193:265, 1951. protein substrates in frog corneal epithelium. Exp Eye Res 24. Rosenthal HE: Graphic method for the determination and pre- 37:327, 1983. sentation of binding parameters in a complex system. Anal 10. Chalfie M, Neufeld AH, and Zadunaisky JA: Action of epi- Biochem 20:525, 1967. nephrine and other cyclic AMP-mediated agents on the chlo- 25. Minneman KP: Alpha,-adrenergic receptor subtypes, inositol ride transport of the frog cornea. Invest Ophthalmol 11:644, phosphates, and sources of cell Ca2+. Pharmacol Rev 40:87, 1972. 1988. 11. Klyce SD, Neufeld AH, and Zadunaisky JA: The activation of 26. Bylund DB: Biochemistry and pharmacology of the alpha-1 chloride transport by epinephrine and DB cyclic-AMP in the adrenergic receptor. In The Alpha-1 Adrenergic Receptors, cornea of the rabbit. Invest Ophthalmol 12:127, 1973. Ruffolo RR, Jr, editor. Clifton, New Jersey, The Humana 12. Montoreano R, Candia OA, and Cook P: Alpha and beta- Press, 1987, pp. 19-69.

Downloaded from iovs.arvojournals.org on 09/28/2021