Investigative Ophthalmology & Visual Science, Vol. 30, No. 7, July 1989 Copyright © Association for Research in Vision and Ophthalmology The Ocular Hypotensive Effects of Demeclocycline, Tetracycline and Other Tetracycline Derivatives Ira Wallace, Theodore Krupin, Richard A. Srone, and Jay Moolchandani Demeclocycline, tetracycline and other tetracycline derivatives lowered intraocular pressure (IOP) in rabbits following intravitreal injection, but the onset of this effect was not evident until 1 or more days after drug administration. Of the drugs tested, demeclocycline was the most active ocular hypotensive agent. Demeclocycline caused a dose-dependent decrease in IOP. The maximum IOP decrease of approximately 12 mm Hg occurred 5 days after intravitreal administration of 0.5 mg, with the effect persisting for over a week. Demeclocycline did not alter tonographically measured aqueous humor outflow facility or episcleral venous pressure. Based on calculated aqueous humor flow rates following 0.2 mg demeclocycline, a 62% decrease in aqueous humor formation occurred 7 days after intravitreal injection. The flow-to-diffusion ratio for ascorbate was reduced 54% 6 days after the intravitreal administration of demeclocycline, a change also consistent with suppression of aqueous humor forma- tion. Anterior chamber aqueous humor protein concentration was increased 6 days after demeclocy- cline administration. No histologic changes were present in the treated eyes by light microscopy. Intravitreal demeclocycline similarly lowered IOP in cats, with the duration of effect lasting up to 20 days. Invest Ophthalmol Vis Sci 30:1594-1598,1989 Possible roles in the regulation of intraocular pres- sopressin.6 The most active tetracycline derivative in sure (IOP) for biologically active peptides, either this regard is demeclocycline, which may even induce those present locally within peptide containing nerve nephrogenic diabetes insipidus.7 Demeclocycline is fibers or those reaching the eye through the blood, an effective therapy for the syndrome of inappropri- remain poorly defined.1 Intravenous administration ate ADH secretion because of this effect.8 of the nanopeptide vasopressin, or antidiuretic hor- In the eye, vasopressin does not stimulate cyclic mone (ADH), stimulates aqueous humor formation AMP production in isolated ciliary processes,9 and and elevates IOP in rabbits,2'3 apparently as the result desmopressin does not elevate aqueous humor cyclic of a direct effect on active sodium transport across the AMP.4 Paradoxically, however, we have found that ciliary epithelium.2 Intravenous desmopressin, a syn- demeclocycline inhibits the IOP rise caused by des- thetic analog of vasopressin, similarly increases mopressin.4 Demeclocycline alone shows no acute aqueous humor formation and raises IOP.4 effect on IOP; but during the course of our studies, we In its antidiuretic action on the kidney, both vaso- observed that demeclocycline has an ocular hypoten- pressin and desmopressin stimulate the production of sive effect developing days after local administration the intracellular second messenger hormone, cyclic to the eye. We therefore studied the direct effects of adenosine 3'5' monophosphate (cyclic AMP), which demeclocycline, tetracycline and other tetracycline initiates a series of events that ultimately decrease derivatives on IOP. urine flow.5 The tetracycline drugs inhibit the vaso- pressin-induced stimulation of cyclic AMP in the kidney and thus block the antidiuretic action of va- Materials and Methods Awake, 2- to 3-kg albino rabbits and 3- to 4-kg cats were restrained in cloth wrappers or in plexiglass From the Glaucoma Service, Department of Ophthalmology, boxes specially designed for each species. Systemic University of Pennsylvania School of Medicine, Scheie Eye Insti- tute, Philadelphia, Pennsylvania. anesthetic agents were not used during any of the Supported by NIH Grants EY-05267, EY-05454, a National experiments. IOP was measured after 0.5% topical Glaucoma Research Grant from the American Health Assistance proparacaine HC1 anesthesia using a Digilab Model Foundation (Dr. Krupin), and the Archie E. Cruthirds Research 30R pneumatonometer (Cambridge, MA) manomet- Fund of the Scheie Eye Institute. rically calibrated for rabbit and cat eyes. Drugs for Submitted for publication: March 24, 1988; accepted December 12, 1988. intravitreal injection were dissolved in filtered (0.22 Reprint requests: Theodore Krupin, MD, Scheie Eye Institute, ^m) distilled water or normal saline. The pH was 51 N. 39th Street, Philadelphia, PA 19104. adjusted with NaOH as close to neutrality as possible 1594 Downloaded from iovs.arvojournals.org on 09/25/2021 No. 7 TETRACYCLINES AND INTRAOCULAR PRESSURE / Wallace er al 1595 without drug precipitation. An acidic solution was required for the 0.2 mg (pH 6.6) and 0.5 and 1.0 mg (pH 6.3) injections. Osmolarity was adjusted with NaCl to between 350 and 360 mOsm. +) >- O) Intravitreal injections were performed after topical 2 I -4 0.5% proparacaine HC1 anesthesia using a 30-gauge o" IE needle on a 0.025 ml Hamilton syringe with a Chaney adaptor. The final volume injected was 0.01 ml, except for the 1.0 mg dose of demeclocycline -8 which required 0.02 ml. In all rabbits, one eye re- 0.01 0.05 0.1 0.2 0.5 1.0 ceived intravitreal drug and the fellow control eye Intravitreal demeclocycline (mg) received an equal volume of intravitreal filtered ster- ile saline with an adjusted pH similar to the drug Fig. 1. Dose-dependent effect of intravitreal demeclocycline on intraocular pressure in rabbits. The change in intraocular pressure solution. In the cats, each animal received a single (IOP) 8 days after intravitreal injection is illustrated. The values uniocular injection of either drug or saline. Except represent [(IOPcontrO| eye) - (IOP,reated eye)] for 8-14 rabbits at each where indicated, the rabbits but not the cats also re- demeclocycline dose. Using t-statistics on the paired differences, ceived 10 mg/kg intraperitoneal indomethacin (40 IOP is significantly reduced following the 0.1 mg (P < 0.05) and the higher (P < 0.001) concentrations of demeclocycline. mg/ml in 1 M Na2CO3) 2 to 3 hr before intravitreal injection. A stable baseline IOP was established both before indomethacin pretreatment and again before chamber (Ch) aqueous humor and in the plasma (Cp). intravitreal drug injection. After drug administration, Assuming that the rate of diffusion remains constant, IOP generally was measured hourly for 6 hr and then this relationship has been employed to measure the every 24 or 48 hr until it returned to baseline. To effects of various agents on the flow rate.12 The pro- control for possible diurnal fluctuations, daily IOP tein concentration of these anterior chamber aqueous readings were recorded between 1 and 3 PM. samples also was measured using Biuret and Folin Aqueous humor dynamics were studied in rabbits phenol reagents (Total Protein Kit No. 690, Sigma following intravitreal injection of 0.2 mg demeclocy- Chemical Co., St. Louis, MO). cline. Seven days after demeclocycline administra- For histopathologic study, three rabbits were sacri- tion in six animals, aqueous humor outflow facility ficed with an overdose of sodium pentobarbital 6 was measured on awake and restrained animals by days after intravitreal injection of 0.2 mg demeclocy- tonography (4 min recordings) using an electronic cline. Both the drug and saline injected eyes were Schiotz tonometer connected to a graph recorder. fixed in formalin, embedded in paraffin, serial-sec- Aqueous humor flow rate was calculated using the tioned in four quadrants, and stained with hematox- Goldmann equation,10 F = (Po - Pv) C; where F ylin/eosin or with Periodic Acid Schiff. = aqueous humor flow rate in /itl/min, Po = intraoc- The effects on IOP of six additional tetracycline ular pressure in mm Hg, Pv = episcleral venous pres- compounds were tested after intravitreal administra- sure in mm Hg, and C = aqueous humor outflow tion of 0.5 mg in rabbits using the same protocol, facility in /xl/min/mm Hg. Episcleral venous pressure except that no indomethacin pretreatment was given. was measured with an episcleral venomanometer in IOP was measured daily in these rabbits for 14 days. six rabbits before and 7 days following intravitreal Intravenous demeclocycline, 40 mg/kg, was ad- demeclocycline. ministered via the marginal ear vein in eleven rabbits. Six days after intravitreal injection of 0.2 mg dem- IOP was measured at 3 hr and 6 hr following injec- eclocycline in another 15 awake rabbits, aqueous tion and then daily for a total of 8 days. humor samples were obtained by posterior and ante- Statistical analysis used the student t-test on paired rior chamber paracentesis after topical 0.5% propara- differences to compare drug- and vehicle-treated eyes caine HC1 anesthesia. These aqueous humor samples within each experimental group and an unpaired t- were placed immediately in 4.0% metaphosphoric test for intergroup analysis. Drug effects are reported acid and titrated with dichlorphenol-indophenol for as mean ± SEM, with P < 0.05 considered signifi- estimation of ascorbate concentration. The ascorbate cant. This study conformed to the ARVO Resolution ratio of the flow coefficient (kfa) to the diffusion coef- on the Use of Animals in Research. ficient (kdpa) was calculated from the Kinsey-Palm formula," where under steady state conditions: Results = (d X C " C )/(C - C ) a p h a (0 Intravitreal injection of demeclocycline in rabbits where d is the Donnan factor, C the ascorbate con- resulted in a dose-dependent decrease in IOP (Fig. 1). centration in the anterior chamber (Ca) or posterior At the 0.1 mg dose, a statistically significant decrease Downloaded from iovs.arvojournals.org on 09/25/2021 1596 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / July 1989 Vol. 30 mg%) compared to control eyes (anterior chamber ascorbate = 23.2 ± 1.8 mg%, posterior chamber = 31.8 ± 2.3 mg%). The significant (P < 0.005) de- crease in the ratio and the increase in the concentra- tion of posterior chamber ascorbate are consistent with a decreased entry of water into the eye.