Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Drug Reservoirs in Topical Therapy Joel 5. Mindel,*t Harry Smith,:}: Miriam Jacobs,* Alex B. Kharlamb,* and Alan H. Friedman" The nictitating membrane, corneal epithelium, and corneal stroma were investigated as drug reservoirs. A hydrophilic drug, D,L-epinephrine HC1, or a lipophilic drug, chloramphenicol, was applied topically to rabbit eyes. Tissue levels of radioactive drug-plus-metabolites and unmetabolized epinephrine were assayed up to 24 hrs later. On a per-mg-tissue basis, concentrations of epinephrine-plus-metabolites in the stroma-endothelium were similar to or higher than those in the epithelium. The percentages of radioactivity representing unmetabolized epinephrine in the stroma-endothelium were found to be similar to or higher than those in the epithelium. On a per-mg-tissue basis, concentrations of cltiloramphenicol-plus-metabolites were significantly higher in the epithelium than in the stroma-endothelium during the first 130 min after drug application. While the physical properties of these drugs determined whether a higher concentration was found in the epithelium or stroma-endothelium, the ninefold greater mass of the stroma-endothelium made it the major drug reservoir on a per-whole-tissue basis. The presence or absence of a nictitating membrane had little effect on the level of either drug in the epithelium, stromal-endothelium, or aqueous humor. Invest Ophthalmol Vis Sci 25:346-350, 1984 mans. Salem and Ellison4 found that the greatest concentration of epinephrine bitartrate during the 24 hrs following a single topical application was in the nictitating membrane. Anderson5 reported three times as much epinephrine bitartrate in the nictitating membrane as in the cornea 1 hr after application. The ducts of the Hardarian and nictitans glands empty onto the nictitating membrane. Miller and O'Connor6 pointed out that all three potential reservoirs could be eliminated by amputating the nictitating membrane. The present investigations dealt with the relative importance of the corneal epithelium and stroma-endothelium as drug reservoirs and with the effect produced by amputating the nictitating membrane. The cornea is a barrier that retards the penetration of topically applied drugs. However, there is evidence that the cornea may act as a drug reservoir for those molecules that succeed in penetrating the epithelium. Van Hoose and Leaders1 found that as external pilocarpine concentrations were increased progressively, the cornea appeared to accumulate pilocarpine. Lazare and Horlington2 reported that pilocarpine concentrations in the rabbit cornea remained higher than those in the aqueous humor for at least 4 hrs. Van Hoose and Leaders' believed that the stroma served as the corneal reservoir for pilocarpine. However, Sieg and Robinson3 found that 2 hrs after topical administration, approximately 70% of the corneal pilocarpine was in the epithelial layer. Rabbits possess several potential drug reservoirs that are not found in humans: the nictitans gland, the Harderian gland, and the nictitating membrane. These structures may alter the corneal penetration and distribution of drugs and may invalidate attempts to extrapolate experimental pharmacokinetic data to hu- Materials and Methods Albino rabbits were used. Unilateral or bilateral amputation of the nictitating membrane was performed at least 17 days prior to the experiment. Amputation followed subconjunctival injection of lidocaine 1%. The nictitating membrane was grasped with a forceps and cut off at its base. The corneas of two rabbits with unilateral removal of their nictitating membranes were examined in a masked manner, using electron microscopy, to determine whether the amputation had altered the corneal epithelium. D,L-epinephrine HC1, 50 mM, pH 6 contained D,L[7-l4C]-epinephrine HC1. Chloramphenicol, 7.7 mM, pH 6 contained [dichloroacetyl-l,2-14C]-chloramphenicol. The pH was adjusted using HC1 and NaOH. The rabbits received a 50 ix\ drop of the same drug to both eyes. After application of the drops, the lids were opposed and held away from the globe for several sec- From the Ophthalmology Section (JSM) of the Bronx Veterans Administration Hospital and the Departments of Ophthalmology,* Pharmacology,! and Biomathematical Sciences}: of the Mount Sinai School of Medicine, New York. Supported by the Medical Research Service of the Veterans Administration and National Eye Institute Grants EY 03203 and EY 01867. Dr. Jacobs was supported in part by National Institute of Arthritis, Metabolism and Digestive Diseases Grant 5T35AM07420-04. Submitted for publication: December 23, 1982. Reprint requests: Joel S. Mindel, MD, PhD, Department of Ophthalmology, Annenberg 22-14, the Mount Sinai School of Medicine, One Gustave L. Levy Place, New York, NY 10029. 346 Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933112/ on 05/05/2017 347 DRUG RESERVOIRS IN TOPICAL THERAPY / Mindel er ol. No. 3 Table 1. Cornea and aqueous humor drug levels, mean ± standard error without nictitating membrane pmol Drug + metabolites (number of eyes) Per mg Corneal epithelium Minutes* 10 40 70 130 190 370 1450 Epinephrine 37.7 ± 19.4 ± 20.8 ± 4.2 ± 16.8 ± 4.5 ± 1.9 ± 8.2 (5) 3.4(4)$ 4.9 (5)$ 1.1 (6) 8.8 (8) 2.4 (4) 0.4 (6) Chloramphenicol 214.3 ±65.1 (6)f 126.7 ± 36.2 (6)t 70.8 ± 16.2 (6)t 30.2 ± 9.9 (8)t 8.8 ± 4.0 (4) 3.5 ± 0.6 (4)$ 2.3 ± 0.8 (4) Per mg Stroma-endothelium Epinephrine 86.2 ± 47.2 ± 38.6 ± 3.0 ± 9.5 ± 1.8 ± 1.4 ± 30.5 (5) 11.9(4)$ 8.5 (5)$ 1.3(6) 2.1 (8) 0.6(4) 0.3(6) Per nl Aqueous humor Chloramphenicol Epinephrine Chloramphenicol 68.5 ± 39.3 ± 23.1 ± 11.2 ± 8.8 ± 6.2 ± 3.9 ± 2.2 ± 0.7 (6) 2.8 ± 0.9 (4) 2.2 ± 0.6 (6) 0.6 ± 0.2 (6) 1.3 ±0.4(8) 0.4 ± 0.2 (4) 0.1 ±0.0(6) 3.2 ± 0.7 (6) 6.7 ± 2.1 (6) 5.1 ±0.5 (6) 2.5 ± 0.5 (8) 1.3 ±0.5 (4) 0.2 ± 0.0 (4) 0.0 ± 0.0 (4) 8.4 (6)t 6.7 (6)t 1.8 (6)t 2.2 (8)t 3.0 (4) 1.1 (4)$ 0.3 (4) * After topical application of 50 /tl: D,L-epinephrine HO, 50 mM, pH 6 or chloramphenicol 7.7 mM, pH 6. t Epithelial drug level significantly (P < 0.05) higher than stromal-endothelial drug level at this point in time. % Stromal-endothelial drug level significantly (P< 0.05) higher than epithelial drug level at this point in time. onds to prevent overflow. The rabbits were killed 10, 40, 70, 130, 190, 370 and 1,450 min later with an intravenous injection of pentobarbitol. Each eye quickly was enucleated and dipped once in a fresh solution of NaCl, 0.9% to remove any drug remaining in the tear film. The aqueous humor was removed as completely as possible using a 26-gauge needle inserted at the corneal limbus. The volume of aqueous humor aspirate was recorded, and the fluid was placed in a scintillation vial. The anterior chamber was refilled with air to restore the shape of the anterior segment of the eye and facilitate dissection. The cornea was removed at its limbus, weighed, and placed epitheliumside down for 1 min on an n-heptanol-soaked filter paper disc approximately 15 mm in diameter.7 The filter paper disc then was placed in a scintillation vial. The remainder of the cornea, ie, the stroma-endothelium, was reweighed and placed in a scintillation vial. Tissue solubilizer was added to all vials; after the tissues were dissolved, their radioactivity in Bray's solution was determined with the use of a scintillation counter. In 20 additional rabbits, the tissues were handled similarly, except that unmetabolized epinephrine was assayed after being separated from epinephrine metabolites.8'9 To estimate the amount of drug transferred from epithelium to stroma-endothelium during the 1-min period on the filter paper disc, 10 ^1 of the epinephrine (seven determinations) or chloramphenicol (five determinations) solution were placed on the surface of an n-heptanol-soaked paper disc immediately before an untreated cornea was placed epthelium-side down on the disc. After 1 min of contact the stroma-endothelium was lifted from the filter paper disc. The relative amounts of radioactivity remaining in the filter paper and transferred to the stroma-endothelium were determined by dissolving the samples and assaying them as previously described. Four corneas were examined histologically to confirm that 1 min of contact with n-hepatanol removed the corneal epithelium. Results Rabbits tolerated the absence of their nictitating membranes for the 2.5-5-week period prior to drug application without objective evidence of discomfort. The corneas remained clear and lustrous. Masked evaluation of electron micrographs at 7500 X magnification failed to indicate any consistant differences between corneas with nictitating membranes and those without. Light-microscopic examination showed that 1 min of contact with the n-hepatanol-soaked filter paper disc removed the corneal epithelium, leaving the basement membrane and stroma intact. The mean ± SE percent of drug transferred to the stroma-endothelium during the 1-min period of epithelial dissolution was 5.1 ± 1.4% using the epinephrine solution and 6.0 ± 2.9% using the chloramphenicol solution. The mean ± SE weights of the corneal epithelium and stroma-endothelium, determined by weighing 70 corneas before and after exposure to n-heptanol, were 6.07 ± 0.32 mg and 56.32 ± 1.82 mg, respectively. The average agreement of the weights of the two corneas of a given rabbit was 92 ± 0.5%. Tables 1 and 2 give the mean ± SE drug-plus-metabolite concentrations per mg corneal epithelium, per mg stroma-endothelium, and per n\ aqueous humor in rabbits with and without nictitating membranes. Table 3 gives the percent of radioactivity representing unmetabolized epinephrine in these three tissues. The role of the nictitating membrane as a reservoir was determined by comparing the drug-plus-metabolite levels in corneal epithelia of animals with and without Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933112/ on 05/05/2017 348 Vol. 25 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / March 1984 Table 2. Cornea and aqueous humor drug levels, mean ± standard error* with nictitating membrane pmol Drug + metabolites Per mg Corneal epithelium Minutes •f 10 40 70 130 190 Epinephrine 62.9 140.0 22.5 17.9 7.2 ±21.4 ± 54.0 ± 5.6 ± 5.8 ± 1.0 Per mg Stroma-endothelium Chloramphenicol 375.3 209.0 48.5 23.0 19.0 ± 25.4$ ± 51.0$ ± 10.6$ ± 3.5$ ± 1.8$ Epinephrine 42.7 129.8 19.8 20.4 5.8 ± 62 ± 55.2 ± 5.4 ± 9.2 ± 1.3 Chloramphenicol 86.5 51.6 18.2 8.9 7.0 ± 13.8$ ± 10.0$ ± 5.2$ ± 2.3$ ± 2.1$ Per nl Aqueous humor Epinephrine 0.9 ± 0.2 14.4 ± 6.8 2.1 ±0.4 4.2 ± 2.4 1.2 ±0.3 Chloramphenicol 4.0 10.4 4.2 2.2 1.9 ± 0.4 ± 2.0 ± 1.3 ± 0.2 ±0.4 * Each value determined from four eyes. t After topical application of 50 pi: D,L-epinephrine HC1, 50 mM, pH 6 or chloramphenicol 7.7 mM, pH 6. $ Epithelial drug level significantly (P «£ 0.05) higher than stromal-endothelial drug level at this point in time. nictitating membranes and stroma-endothelia of animals with and without nictitating membranes. Each point in time was tested using a student's /-test. Values were assumed significant if P < 0.05. For a given drug, there were no significant differences at any point in time between eyes with and without nictitating membranes. The roles of the epithelium and stroma-endothelium as reservoirs were determined by comparing their drug-plus-metabolite levels at each point in time on a per-mg-tissue basis. In rabbits with nictitating membranes, the per mg epithelial and stromal-endothelial levels were not significantly different after epinephrine application. However, the chloramphenicol-plus-metabolite levels were significantly higher per mg epithelium than per mg stroma-endothelium at all points in time for rabbits with nictitating membranes. In rabbits without nictitating membranes, the epinephrine-plus-metabolite levels per mg stroma-endothelium were significantly higher than those per mg epithelium only at 40 and 70 min after drug application (Table 2). After chloramphenicol administration to rabbits without nictitating membranes, the per-mgepithelium drug-plus-metabolite levels were significantly higher for at least 130 min after application, but not from 190 min to 1,450 min after application. The percentages of unmetabolized epinephrine (Table 3) were compared in eyes without nictitating membranes to contralateral eyes with nictitating membranes using a paired t-test. The percents of unmetabolized epinephrine in the epithelium, stroma-endothelium, and aqueous humor were not altered significantly by the presence or absence of a nictitating membrane, with one exception: 10 min after epinephrine application, corneal epithelia of eyes without nictitating membranes had a slightly higher percent of unmetabolized epinephrine than epithelia of eyes with nictitating membranes (40 ± 6% vs 36 ± 4%). The percents of unmetabolized epinephrine in the stroma-endothelium were not significantly different from those in the corresponding epithelium, with one exception: 10 min after epinephrine application to eyes without nictitating membranes, the percent of unrnetabolized epinephrine was significantly higher in the stroma-endothelium than in the epithelium. Discussion Epinephrine and chloramphenicol were chosen for this study because both are in clinical use and because their pharmacokinetics have been studied in rabbits.2'4'5 The physical properties of the two drugs differ: epi- Table 3. Percent radioactivity (mean ± SE) as unmetabolized epinephrine Time after D,L-epinephrine HCl applied* Tissue Nictitating membrane absent (OD) Corneal epithelium Corneal stroma-endothelium Aqueous humor Nictitating membrane present (OS) Corneal epithelium Corneal stroma-endothelium Aqueous humor 10 min 70 min 190 min 1,450 min 40 ± 6% (4)f$ 68 ± 0% (4)$ 54 ± 2% (4) 23 ± 7% (5) 27 ± 8% (5) 32 ± 4% (4) 24 ± 6% (5) 24 ± 1% (4) 27 ± 6% (5) 18 ± 2% (4) 19 ± 1% (4) 16 ± 1% (4) 36 ± 6% (4)f 65 ± 5% (4) 53 ± 3% (4) 17 ± 1% (5) 20 ± 4% (5) 27 ± 10% (4) 19 ± 1% (5) 24 ± 2% (4) 30 ± 12% (5) 16 ± 1% (4) 18 ± 0% (4) 17 ± 2% (4) * Numbers in parentheses indicate number of eyes assayed. t Significant difference between eyes with and without nictitating membranes (P < 0.02). X Significant difference between epithelial and stromal levels (P < 0.02). Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933112/ on 05/05/2017 No. 3 DRUG RESERVOIRS IN TOPICAL THERAPY / Mindel er ol. nephrine HC1 is readily soluble in water, while chloramphenicol is more soluble in lipids, eg, chloramphenicol is 150 times more soluble in ethanol than in water.l0 Thus, at pH 6, there was no difficulty in making a solution of 50 mM, D,L-epinephrine HC1, while the maximum chloramphenicol concentration was 7.7 mM. These concentrations are used clinically, ie, 50 mM epinephrine HC1 is a 1% concentration and 7.7 mM chloramphenicol is a 0.25% concentration. The corneal penetration of drugs increases in proportion to lipid solubility," presumably because the major barrier to drug penetration is the lipids in the epithelial cell membrane. The stroma is relatively acellular and consists primarily of collagen.12 On the basis of the physical properties of the drugs and the cornea, we predicted that chloramphenicol would penetrate into the epithelium more readily than epinephrine. This expectation was fulfilled. Ten minutes after application, the chloramphenicol-plus-metabolite levels in the corneal epithelium were six times higher than the epinephrine-plus-metabolite levels, despite that the concentration of the epinephrine drop was six times greater. The presence or absence of a nictitating membrane did not alter this ratio significantly (Tables 1 and 2). We also predicted that with the passage of time, the corneal epithelium would remain the primary reservoir for chloramphenicol, while the corneal stroma would become the primary reservoir for the more hydrophilic epinephrine molecules. To a large extent, this prediction was also correct. On a per-mg-tissue basis, the epithelial levels of chloramphenicol-plusmetabolites remained significantly higher than those in the stroma for at least 130 min. This was true for rabbits with and without nictitating membranes. The stromal levels of epinephrine-plus-metabolites tended to be higher than those in the epithelium, but these differences achieved significance only at 40 and 70 min after application and only in rabbits without nictitating membranes. The proportions of radioactivity representing unmetabolized epinephrine also tended to be higher in the stroma-endothelium than in the epithelium (Table 3), but this difference was significant only at 10 min after epinephrine application and only in rabbits without nictitating membranes. Although the presence of the nictitating membrane increased the relative importance of the corneal epithelium as a depot for epinephrine, this effect was not especially marked, because epinephrine-plus-metabolite levels in the epithelia of rabbits with nictitating membranes were not significantly different from those in rabbits without nictitating membranes. Nor were significant differences found for stromal-endothelium and aqueous humor epinephrine-plus-metabolite levels when rabbits with and without nictitating membranes were compared. When corresponding corneal epithe- 349 lial, stromal-endothelial, and aqueous humor levels of chloramphenicol-plus-metabolites were compared between rabbits with and without nictitating membranes, these too failed to show significant differences. Thus, the results of the present study do not support the theory that the nictitating membrane acts as a major drug depot. Which was the more important drug depot, the corneal epithelium or the corneal stroma? When data were analyzed on a per-mg-tissue basis, the answer varied, depending on the drug and the time after application. However, the mean stromal weight, 56.32 mg, was more than nine times greater than the mean epithelial weight, 6.07 mg. As a result, on a ptv-wholetissue basis, the corneal stroma was the major drug depot. This was so for epinephrine and chloramphenicol at all points in time. Thus, a distinction should be made when discussing drug reservoirs between total tissue content and tissue concentration. When epinephrine was separated from its metabolites, the importance of the stroma as a reservoir for unmetabolized molecules was increased only minimally. Presumably, the greater metabolic activity of the epithelium13 diminished its proportion of unmetabolized epinephrine, but this disadvantage was largely offset by the hydrophilic nature of epinephrine metabolites. These metabolites would tend to accumulate in the stroma-endothelium. No attempt was made to separate chloramphenicol from its metabolites. There is evidence that chloramphenicol is metabolized within the cornea. For example, Green and MacKeen14 reported that 2 hrs after topical chloramphenicol application, about 45% of rabbit aqueous humor radioactivity was due to chloramphenicol metabolites. Therefore, the present data may not accurately reflect the relationship of unmetabolized chloramphenicol in the epithelium to that in the stroma-endothelium. Lippert and co-workers15 reported results similar to ours. Two and three hours after the topical administration of tritiated dexamethasone to rabbits, the epithelial radioactivity was higher, on a per-mg-tissue basis, than that in the stroma. But the stroma remained the major depot on a per-whole-tissue basis. Sieg and Robinson316 reported that the highly lipophilic drug fluorometholone was similarly distributed to the lipophilic chloramphenicol used in the present study. Ten minutes after application of fluorometholone, the major corneal reservoir was the epithelium, even when calculations were on a per-whole-tissue basis. Thereafter, from 20 to 120 min following application, the stroma was the major drug reservoir. Surprisingly, for pilocarpine, which was less lipid-soluble than fluorometholone, the corneal epithelium was the primary reservoir during the entire 2 hrs of their study. Perhaps pilocarpine exerted pharmacologic effects that altered Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933112/ on 05/05/2017 350 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / March 1984 its distribution. Sieg and Robinson removed the corneal epithelium by scraping with a scalpel blade. The present authors used n-heptanol.7 Both techniques suffered from a common defect: drug was released from disrupted epithelial cells into the stroma. Estimates of the amounts of drug released into the stroma during dissolution with n-heptanol yielded mean values well under 10%. These errors were considered acceptable. An assumption in the preceding analysis was that epinephrine and its metabolites distributed by passive diffusion. This may have been partially incorrect. Epinephrine has multiple pharmacologic effects. Among the known actions of adrenergic agonists are: increasing lacrimal gland tear production, 17 activating a cyclicAMP sensitive corneal epithelial transport system,18 being accumulated by corneal and nictitating membrane nerve endings (especially of the adrenergic type19), and increasing aqueous humor outflow facility.20 A second assumption concerned the use of a racemic mixture of D- and L-epinephrine. We believed that the corneal distributions of both epinephrine isomers and their mebabolites were similar because both isomers are substrates for the degradative enzymes monoamine oxidase and catechol-O-methyl transferase.21'22 A third assumption was that an insignificant amount of corneal and aqueous humor radioactivity resulted from systemically absorbed drug molecules. Systemic absorption does occur, eg, Anderson5 found 8 picograms epinephrine per ml blood 1 hr after topical application of 0.1 % epinephrine bitartrate. In summary, on a per-mg-tissue basis, the corneal epithelium appeared to be the major drug reservoir for highly lipophilic drugs, such as chloramphenicol, while the corneal stroma appeared to be the major drug reservoir for more hydrophilic drugs, such as epinephrine. On a per-whole-tissue basis, because of its greater mass, the corneal stroma appeared to be the major drug reservoir for many lipophilic and possibly all hydrophilic drugs. The presence or absence of a nictitating membrane did not significantly alter drug pharmacokinetics. Key words: epinephrine, chloramphenicol, pharmacokinetics, cornea, nictitating membrane, eye, drug References 1. Van Hoose MD and Leaders FE: Role of the cornea in the biologic response to pilocarpine. Invest Ophthalmol 13:377, 1974. 2. Lazare R and Horlington M: Pilocarpine levels in the eyes of rabbits following topical application. Exp Eye Res 21:281, 1975. Vol. 25 3. Sieg JW and Robinson JR: Mechanistic studies on transcorneal permeation of pilocarpine. J Pharm Sci 65:1816, 1976. 4. Salem H and Ellison T: Pharmacologic action and tissue distribution of epinephrine bitartrate in the rabbit eye. Ann Ophthalmol 5:417, 1973. 5. Anderson JA: Systemic absorption of topical ocularly applied epinephrine and dipivefrin. Arch Ophthalmol 98:350, 1980. 6. Miller D and O'Connor P: The influence of the nictitating membrane on steroid inhibition of limbal wound healing. Acta Ophthalmol 55:586, 1977. 7. Cintron C, Hassinger L, Kublin CL, and Friend J: A simple method for the removal of rabbit corneal epithelium utilizing n-heptanol. Ophthalmic Res 11:90, 1979. 8. Locke S, Cohen G and Dembiec D: Uptake and accumulation of 3H-6,7-dihydroxytetrahydroisoquinoline by peripheral sympathetic nerves in vivo. J Pharmacol Exp Ther 187:56, 1973. 9. Cohen G and Katz S: 6,7-Dihydroxytetrahydroisoquinoline: evidence for in vivo inhibition of intraneuronal monoamine oxidase. J Neurochem 25:719, 1975. 10. Clarke EGC: Isolation and Identification of Drugs, London, William Clowes and Sons, Ltd, 1969, p. 246. 11. Leibowitz HM and Kupferman A: Bioavailability and therapeutic effectiveness of topically administered corticosteroids. Trans Am Acad Ophthalmol Otolaryngol 79:78, 1975. 12. Rodrigues MM, Waring GO, Hackett J, and Donohoo P: Cornea, Chapter 8. In Biomedical Foundations of Ophthalmology, Vol. 1, Duane TD and Jaeger EA, editors. Philadelphia, Harper and Row, 1982, pp. 6-8. 13. Anderson JA, Davis WL, and Wei C-P: Site of ocular hydrolysis of a prodrug dipivefrin and a comparison of its ocular metabolism with that of the parent compound epinephrine. Invest Ophthalmol Vis Sci 19:817, 1980. 14. Green K and MacKeen DL: Chloramphenicol retention on and penetration into the rabbit eye. Invest Ophthalmol 15:220, 1976. 15. Lippert U, Mosebach KO, and Dardenne MV: Untersuchungen uber die Verteilung von Dexamethason im Kaninchenauge nach lokaler Applikation. Klin Monatsbl Augenheilkd 164:225, 1974. 16. Sieg JW and Robinson JR: Mechanistic studies on transcorneal permeation of fluoromethalone. J Pharm Sci 70:1026, 1981. 17. Aberg G, Adler G, and Wikberg J: Inhibition and facilitation of lacrimal flow by /8-adrenergic drugs. Acta Ophthalmol 57:225, 1979. 18. KJyce SD, Neufeld AH, and Zadunaisky JA: The activation of chloride transport by epinephrine and Db cyclic-AMP in the cornea of the rabbit. Invest Ophthalmol 12:127, 1973. 19. Laties A and Jacobowitz D: A histochemical study of the adrenergic and cholinergic innervation of the anterior segment of the rabbit eye. Invest Ophthalmol 3:592, 1964. 20. Chubak GS, Weiss MB, Yablonski ME, and Podos SM: Effects of topical epinephrine on rabbit aqueous humor dynamics. ARVO Abstracts. Invest Ophthalmol Vis Sci 19(suppl):194, 1980. 21. Nagatsu T: Biochemistry of Catacholamines, Baltimore, University Park Press, 1973. pp. 106-107, 128. 22. Leeper LC, Weissbach H, and Udenfriend S: Studies of the metabolism of norepinephrine, epinephrine and their O-methyl analogs by partially purified enzyme preparations. Arch Biochem Biophys 77:417, 1958. Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933112/ on 05/05/2017