Topical Application of Serotonin or the 5-HT X -Agonist 5-CT Intraocular Pressure in Rabbits

Transcription

1 Topical Application of Serotonin or the 5-HT X -Agonist 5-CT Intraocular Pressure in Rabbits Ulrich Meyer-Bothling, Anthony J. Bron, and Neville N. Osborne Purpose. The effect of serotonin (5-hydroxytryptamine: 5-HT) and other agonists on rabbit intraocular pressure (IOP), pupil size, and the breakdown of blood-aqueous barrier were evaluated. Methods. Serotonin and various other agonists were applied topically to the rabbit eye, and intraocular pressure was followed over the next 3 hours using a Digilab 30D pneumatometer. Results. It was demonstrated immunohistochemically that topical 5-HT reached the anterior chamber within 1 hour. Serotonin raised the IOP in a dose-dependent manner over a period of up to 4 hours, with a maximum reached between 30 minutes and 1 hour. A similar effect was observed with the 5-HT,-agonist 5-carboxamidotryptamine (5-CT). Neither tryptamine, 5-hydroxytryptophan, melatonin, gepirone, nor 5,6/5,7-dihdroxytryptamine caused any changes in IOP. Serotonin did not cause a change in pupil size or a breakdown of the blood-aqueous barrier, nor did the aqueous camp change significantly after topical 5-HT administration. Conclusions. The data presented suggest a role for 5-HT in the control of IOP. Previously demonstrated receptors on the iris-ciliary body and the effect of the 5-HT,-agonist 5-CT suggest that the rise in IOP may be caused partly or entirely by an increase in aqueous secretion mediated by 5-HT,-like receptors. Whether or not 5-HT has a role in altering aqueous outflow resistance remains to be seen. An effect of serotonin on other aspects of aqueous dynamics or on the extraocular muscles to cause a change in IOP cannot be excluded. Invest Ophthalmol Vis Sci 1993;34: A body of evidence in the literature suggests that serotonin (5-hydroxytryptamine or 5-HT) has a functional role in the anterior uvea. A transmitter role for the amine was originally suggested by the finding that intraocular pressure (IOP) is lowered in rabbits and dogs after an intravenous injection of serotonin. 1 However, neurones containing serotonin have not been reported to exist either in the iris-ciliary body complex or the trabecular region, although it has been shown that some neurones in the iris-ciliary body take up exogenous 5-HT 2 serotonin. On the other hand, it has been found in the human aqueous humor, 3 and serotonergic agents have been shown to influence IOP although results are conflicting. Thus, Krootila et al 4 reported that topical application of 2% 5-HT caused a decrease in IOP in rabbits whereas an intracameral From Nujfield Uiboratory of Ophthalmology, University of Oxford, Oxford, United Kingdom. Supported by the Wellcome Trust, l^ondon, and supported in part by a grant from A lam Research Uiboratories, Fort Worth, Texas (NNO). Submitted for publication March 27, 1992; accepted July 15, Proprietary interest category: N. Reprint requests: Neville N. Osborne PhD, DSc, Nuffield Uiboratory of Ophthalmology, Walton Street, Oxford, 0X2 6HG, United Kingdom. injection of the drug caused a rise in rabbit IOP. They suggested that the differences between the responses to topical and intracameral 5-HT could be from variations in the concentration of the amine at different sites of action. Intracameral 5-HT might, for example; be of sufficient concentration to cause an inflammatory response, whereas the topically applied amine reaches a level where it has a presynaptic action on adrenergic nerves. Two serotonergic antagonists shown to influence IOP are methysergide and ketanserin. Both substances are thought to have a preferential affinity for 5-HT 2 receptors. 5 Topically applied ketanserin lowers the IOP in rabbits, cats, and monkeys. 6 This was more prominent in eyes with intact sympathetic innervation than in sympathectomized eyes. However, when an acute irritative response in the rabbit eye was induced by topical formaldehyde, an intravenous supply of ketanserin had no effect on the IOP. This was not the case for methysergide, which reduced the IOP, 7 although it is known that methysergide also has effects on 5-HT 1A -type receptors. 8 In contrast, oral administration of ketanserin to glaucomatous patients was Investigative Ophthalmology & Visual Science, September 1993, Vol. 34, No. Copyright Association for Research in Vision and Ophthalmology 3035

2 3036 Investigative Ophthalmology & Visual Science, September 1993, Vol. 34, No. 10 found to decrease the IOP, though it was found that the systolic blood pressure was also lowered. 9 Evidence for the presence of serotonergic receptors in the rabbit iris-ciliary body is good. Specific binding sites for the amine have been reported and characterized as being 5-HT,-like in nature Serotonin has also been shown to be negatively linked to the metabolism of camp in the rabbit iris-ciliary body, supporting the view that a 5-HT r like receptor is present. 12 Moreover, serotonin has been shown to stimulate phosphoinositide metabolism in the rabbit iris-ciliary body; because this process was attenuated by methysergide and ketanserin, it was concluded that 5-HT 2 receptors are present. 13 The objectives of this study were as follows: first, we wished to provide unequivocal evidence that topically applied serotonin can penetrate the cornea, the conjunctiva, sclera, or both, to reach the iris-ciliary body because it is possible that any observed change in the IOP or pupil size is not the result of serotonin but of a metabolite of the amine. Second, the effect of different concentrations of serotonin on the IOP was measured over a period of 4 hours. Because any reported change in IOP could be from destruction of the blood-aqueous barrier or could involve the second messenger camp, a third aim was to analyze the influence of topical serotonin on aqueous humor protein and camp concentrations. Another objective was to examine the effect of topical serotonin on pupil size. Studies from our laboratory have shown the amine to contract the isolated iris sphincter muscle preparation. 14 MATERIAL AND METHODS Drug Administration and Intraocular Pressure Measurements in Rabbits New Zealand white albino rabbits of both sexes, weighing 2.0 to 4.0 kg, were kept in individual cages in well-defined and standardized conditions (hygrometry- and temperature-controlled room, 12-hour light-dark cycle: hour 0 lights on, hour 12 lights off), and they were fed a dry pellet diet and water ad libitum. The rabbits were entrained regularly to intraocular pressure (IOP) measurements for at least 2 weeks before the onset of the experiments for adaptation purposes. All the experiments were carried out by the same operator using a manometrically calibrated pneumatonometer (Digilab 30D). Each IOP value recorded represents the mean of two separate measurements taken. For eveiy series of experiments, at least six rabbits from the same maternal source were used to ensure a minimum difference in age, v/eight, and external influences. All experiments were carried out at the same time of the day (start: hour 3 circadian time, hour 0 = LIGHTS ON). IOP measurements were performed from the corneal surface after topical application of 30 /A 0.4% benoxinate (oxybuprocaine). IOP was taken at 0, 30, 60, 120, and 180 minutes after the topical application of a drug or vehicle. A single 30-/nl drop of the drug or the vehicle was administered in a masked fashion to the right or left eye at minute 0. The animals were not retested for at least one week after a drug experiment. When aqueous humor was required, the rabbits were first sedated with the neurolept analgesic hypnorm (Janssen [Oxford, UK]) (fentanyl citrate mg/ml, fluanisone 10 mg/ml) intramuscularly, 0.5 ml/kg body weight. Thereafter, 30 /xl of 0.4% benoxinate (oxybuprocaine) was applied topically before removing aqueous humor under sterile conditions with a 30-gauge needle. When rabbits were to be killed, they were sedated and then given an overdose of euthesate (Southern Veterinary [Sussex, UK]) (pentobarbital sodium 200 mg/ml) intravenously. 5-Carboxamidotryptamine was obtained from Glaxo (United Kingdom), and gepirone was supplied by Bristol-Myers (United Kingdom). All other drugs were supplied by Sigma (United Kingdom). All drugs except melatonin were made up in 0.9% NaCl. Melatonin wasfirstdissolved in DSMO (dimethyl sulphoxide), then mixed with 0.9% NaCl. All investigations described conform to the ARVO Resolution on the use of Animals in Research. Immunocytochemistry Thirty microliters of 0.9% saline and 0.1%, 1%, 2%, and 4% 5-HT were topically applied to rabbit eyes (right or left). The rabbits were sedated, and, 1 hour after topical application of the drugs, they were given an intravenous overdose of euthesate (sodium pentobarbital 200 mg/ml). The eyes were removed and washed in normal saline, and the iris-ciliary bodies and retinae dissected. The tissue was fixed for 3 hours with 4% formaldehyde in 0.1 M phosphate buffer, ph 7.4. After fixation, the tissue was washed in 0.1 M phosphate buffer and soaked overnight at 4 C in 0.1 M phosphate buffer containing 30% sucrose. Cryostat sections (10 nm) were mounted on gelatine-coated glass slides and immediately frozen and stored at -20 C. The slides with frozen sections were thawed and washed with 0.2% Triton X-100 in M phosphate buffered saline (PBS-T). The sections were incubated overnight at 4 C with primary monoclonal rat anti-5-ht antibody (Sera-lab, United Kingdom) diluted 1:200 with PBS-T. This and all other incubations were done in humid chambers. The sections were rinsed for 10 minutes in PBS-T and incubated for 40 minutes at 20 C with a secondary polyclonal anti-

3 Increase of Topical Serotonin and 5-CT IOP in Rabbits 3037 tein content was determined using the protein assay described by Bradford.15 The camp content was measured using the competitive camp binding assay described by Brown et al.16 [2,8-3H] Adenosine 3'5'-cyclic monophosphate (50 Ci/mmol) was obtained from Amersham (United Kingdom). All other materials were supplied by Sigma. RESULTS Immunohistochemistry Topical application of saline or different concentrations of 5-HT to rabbit eyes 1 hour before dissection, followed by analysis for the immunohistochemical localization of 5-HT immunoreactivity in the iris/ciliary body and the retina, revealed these results: Serotonin immunoreactivity was absent from both the iris-ciliary body and the retina in saline-treated eyes that is, no endogenous 5-HT could be detected. This was also the case for tissue from eyes treated with 0.1% 5-HT. However, when the eyes were treated with 1%, 2%, or 4% 5-HT, nerve fibers in the iris-ciliary body (Fig. 1) and some amacrine cells in the retina "stained" positively for 5-HT immunoreactivity (Fig. 2). FIGURE 1. Section of a rabbit iris-ciliary body immunohistochemically stained for 5-HT 1 hour after topical application of 30 ii\ saline (control) (top) or.1% 5-HT (bottom). rat IgG FITC-conjugate antibody (fluorescein marker) (Sigma) diluted 1:40 with PBS-T. Slides were rinsed for 10 minutes in PBS-T, and a coverslip was mounted with glycerol-pbs 3:1. The specimens were examined under an epifluorescence microscope (Zeiss [Oberkochen, Germany]) and photographed with Kodak Tri-X 400 ASA film (Eastman Kodak Company, Rochester, NY). Measurements of Pupil Size in Rabbits 5-HT 0.1%, 1%, 2%, 4%, or the vehicle were applied to normal eyes and to eyes treated 15 minutes before with topical carbachol 0.5%, as already described (in a masked manner). Carbachol 0.5% was applied after a 15-minute interval to eyes pretreated with 5-HT 0.1%, 1%, 2%, and 4%. Pupil sizes were then measured with a ruler in constant lighting conditions (800 lux) at different time periods. As for IOP experiments, sets of animals from the same maternal source and of the same ages were used. Aqueous Protein and camp Measurements Aqueous humor was obtained 1 hour after topical application of the drug or the vehicle. The aqueous pro- FIGURE 2. Section of a rabbit retina immunohistochemically stained for 5-HT 1 hour after topical application of 30 /x\ saline (control) (top) or 1% 5-HT (bottom).

4 3038 Investigative Ophthalmology &: Visual Science, September 1993, Vol. 34, No. 10 The contralateral eyes from animals in which the ipsilateral eyes were treated with normal saline or 5HT (0.1%, 1%, 2%, or 4%) were also analyzed. Serotonin immunoreactivity was not observed in either the iris-ciliary body or retina in these instances, except when 4% 5-HT was used. In this case, a faint positive staining for 5-HT was found in the iris-ciliaiy body and in the retina (amacrine cells) (Fig.3). 7.0 I" j~ o o 5-HT 2* A--A5-HT15K a---d 5-HT 0.1* v vnaci0.9k x i 30 ^2.0 o < Intraocular Pressure Measurements in Rabbits No significant correlation was found between normal IOP and animal age or weight. Normal IOP ranged between 18 mm Hg and 21 mm Hg during the period when experiments were performed. Comparative IOP measurements between the right and left eyes of 24 rabbits revealed a mean difference of 0.42 ±0.13 mm Hg, which did not prove to be statistically significant. When topical drugs were applied to one eye, an IOP effect could often be detected in the contralateral eye, especially with higher concentrations and larger volumes. With the administration of topical 5-HT, a trend to a rise in IOP was found in the contralateral eye. This effect was inconsistent and statistically not FIGURE 3. Section of a rabbit iris-ciliary body (top) and retina (bottom) showing serotonergic immunoreactivity 1 hour after topical application of 30 ^1 4% 5-HT to the contralateral eye time (hours) FIGURE 4. IOP in rabbits after topical application of saline (control) and 0.1%, 1%, or 2% 5-HT. Data are given as changes in IOP (in mm Hg) as means ± SEM (n = 12 for each concentration). *P < 0.05, -{P < compared to IOP at hour 0. significant. If present, it was less marked than in the eye that received the drug. Although the contralateral eye was therefore considered to be a poor control for the effect of a drug, the contralateral eye was still used as a control in determining which eye received the drug in the masked experimental protocol. The actual numeric drug-related changes in IOP values were calculated from the initial baseline IOP value (IOP of the individual rabbit at minute 0 before drug application). Volumes between 20 ju.1 and 100 /xl of a drug of the same concentration topically applied to the eye affected the IOP in the drug-treated eye in the same way. A volume of 30 /A was therefore chosen for routine studies. IOP measurements taken over 24 hours (n = 12) showed relatively stable pressures between hour 2 and hour 9 circadian time (hour 0 = LIGHTS ON). A steady rise in IOP occurred before the lights were turned off (hour 12 = LIGHTS OFF) and continued to do so until a peak was reached at hour 14 circadian time. IOP decreased thereafter until low levels were reached after the lights went on. All experiments with rabbits were conducted between hour 3 and hour 7 circadian time, when the IOP was relatively stable. Topical application of 0.1%, 1%, or 2% 5-HT resulted in an increase in IOP within 30 minutes. All three concentrations used caused a maximum increase in IOP within 30 minutes (0.1% and 2% 5-HT) or 1 hour (1% 5-HT) (Fig. 4). All IOP values within the first 2 hours after drug administration proved to be statistically significant compared to the baseline value. Eyes topically treated with 0.1% 5-HT returned to their original IOP (at hour 0) within 3 hours. Eyes with 1% or 2% 5-HT still had an elevated IOP after 3 hours, although these differences were no longer statistically significant.

5 Increase of Topical Serotonin and 5-CT IOP in Rabbits 3039 A comparison between the effects of 5-HT and the 5-HT r agonist 5-carboxamidotryptamine (5-CT) (both at 1%) on IOP is shown in Fig CT caused a slightly greater increase in IOP than 5-HT. Topical application of tryptamine, 5,6-dihydroxytryptamine, 5,7-dihydroxytryptamine, 5-hydroxytryptophan, melatonin, and gepirone (all at a concentration of 1%) caused no changes in the IOP HT and Pupil Size in Rabbits No difference in pupil size was observed after topical application of different concentrations of 5-HT alone or in combination with carbachol. 5-HT and Aqueous Protein The concentration of protein was measured in samples of aqueous humor and removed from rabbit eyes 1 hour after a topical application of a drug. Eyes topically treated with saline or prostaglandin E 2 were used as controls; prostaglandin E 2 is known to cause a rapid breakdown of the blood-aqueous barrier, with a subsequent increase in aqueous protein content. 17 No difference in aqueous protein could be shown after topical application of normal saline, 1%5-HT, 1% isoproterenol, 1% pilocarpine, and 1% timolol. In contrast, prostaglandin E 2 caused a three-fold increase in the aqueous protein content (Fig.6). Cyclic AMP in Aqueous Humor Aqueous humor was withdrawn from rabbit eyes after topical application of 0.9% NaCl (control), 1% 5-HT, 1% isoproterenol, or 1% 5-HT, together with 1% isoproterenol (with a 15-minute interval) for determination of camp content. Analysis of the camp content in aqueous humor from right and left eyes (from ani- FIGURE 6. Aqueous protein concentration 1 hour after topical application of 30 /xl of saline (control), 1% 5-HT, 1% isoproterenol, 1% pilocarpine, 1% timolol, or 0.25% prostaglandin E 2. Data are given as mean concentrations of protein ± SEM (n = 8). *P < compared to control. mals topically treated with 0.9% NaCl) showed that the camp content was statistically identical. Topical application of 5-HT only slightly increased the aqueous camp content, whereas topical isoproterenol treatment elevated the camp markedly. Application of both 5-HT and isoproterenol revealed an increase in camp content compared to eyes treated with 1% isoproterenol only (Fig. 7). To determine whether or not aqueous camp is metabolized by an endogenous phosphodiesterase, aqueous humor samples were withdrawn from six rabbit eyes, and the amount of each sample was divided into halves: one-half was frozen immediately at -70 C, the other half was left at room temperature for 2 hours and frozen afterward. No difference in camp levels was found between the samples immediately frozen and the ones frozen after 2 hours. 5-CT1X A- -A 5-HT1X v v NaCl 0.9X T T time (hours) FIGURE 5. IOP in rabbits after topical application of 1% 5- HT, 1% 5-CT, or saline (control). Data are given as changes in IOP (in mm Hg) as means ± SEM (n= 12 in each case). *P < 0.05, -fp < compared to IOP at hour B 0.5- s Q. O o nvt a: M 0.0 FIGURE 7. Cyclic AMP content in aqueous humor 1 hour after topical application of 0.9% NaCl (control) to both eyes, or 1% 5-HT, 1% isoproterenol, or 1% 5-HT +1% isoproterenol (15-minute intervals) to one eye (n = 8 in each case). *P < 0.05 compared to application of 1% isoproterenol only; \P < compared to control eyes. ontrol ontrol 5-HT nol 1X1

6 3040 Investigative Ophthalmology 8c Visual Science, September 1993, Vol. 34, No. 10 DISCUSSION A major finding of this study is that topical serotonin raises the intraocular pressure in the rabbit. Moreover, it was demonstrated unequivocally that 5-HT reaches the iris-ciliary body within the time during which a rise in IOP occurred. This implies that topically applied 5-HT penetrated the cornea, the conjunctiva sclera, or both, and that it is likely that 5-HT, rather than a metabolite of the amine, is the cause for the change in IOP. No evidence could be found to suggest that the rise in IOP was caused by a breakdown of the blood-aqueous barrier. Because 5-carboxamidotryptamine (5-CT) acts like 5-HT to cause a rise in IOP, the involvement of 5-HTj receptors is implicated. Four mechanisms can be envisaged by which 5-HT might alter IOP: action on outflow resistance; uveoscleral pathway; action on aqueous secredon; action on the extraocular muscles. 29 The action of serotonin in increasing IOP by any of these mechanisms cannot be excluded. As already referred to in the introduction, the actual existence of endogenous serotonin neurones in either the trabecular region or the iris-ciliary body has not been demonstrated. 18 However, when iris-ciliary body tissues are incubated in vitro with exogenous 5- HT, the amine can be shown to be taken up and localized in neurones by use of immunohistochemistry. It is also known that immunohistochemical procedures have failed to localize endogenous serotonin in the mammalian retina, but when the retina is exposed to exogenous amine, serotonin can be localized to occur in certain amacrine cells. 18 In this study, we made use of these observations by immunohistochemically processing the retina and the iris-ciliary body for serotonin localization after topical application of the amine. Because the antibody used recognizes serotonin specifically and not metabolites, this methodology provided us with a precise answer to the question of whether topical application of the amine reached possible sites of actions involved in IOP control. It was found that 1 hour after topical application of 5-HT, an apparent dose-dependent uptake of amine in the iris-ciliary body and retina occurred. Thus, topically applied 5- HT could potentially act on the aqueous humor secreting ciliary epithelium. The 5-HT that can be demonstrated within the retina within 1 hour could be reaching the tissue via diffusion through the vitreous or uveoscleral pathways, or both, as well as through the systemic system. These possibilities are apparent when one compares the amount of "stained" amacrine cells in the retina of the 5-HT treated eye with the retina of the contralateral eye: Numerous cells showed 5-HT immunoreactivity in the 5-HT-treated eye, whereas in the retina of the contralateral eye, only faint staining for 5-HT was revealed in very few cells. Measurements of the IOP over a 24-hour period revealed a circadian rhythm closely related to the lighting conditions. Pressures were lower when the lights were on and higher when they were off. A similar pattern was demonstrated earlier by Gregory, 19 who found a mean difference in IOP of up to 8 mm Hg in the day-night variation. As a consequence of these findings, all subsequent experiments in this study were conducted between hour 3 and hour 7 circadian time (hour 0 = LIGHTS ON) because this was the most stable period, with minimal changes in IOP. It is known that topically applied drugs can reach the contralateral eye. 20 This is also true for 5-HT, which reaches the contralateral eye as demonstrated by immunohistochemistry. The contralateral eye was therefore only used as a control in determining which eye received the drug in the masked experimental protocol. The actual numeric, drug-related changes in IOP were calculated from the initial baseline IOP value at minute 0. Topical application of different concentrations of 5-HT to rabbit eyes revealed a dose-dependent, statistically significant increase in the IOP over at least a 2-hour period. Chiang et al 1 found a decrease in the IOP of dogs and rabbits after intravenous injection of 5-HT. In contrast, Chiou et al 21 reported no change in the IOP after topically or intracamerally administered 5-HT. Similarly, neither did Moro et al 22 find a change in IOP after treatment with 5-HT. However, Krootila et al 4 reported a decrease in the IOP (measured with a pneumatonometer) after topical application of 5-HT and an increase in the IOP (measured electromanometrically with an intracameral needle after pentobarbital anesthesia) after intracameral injection of 5-HT. It has been suggested by Krootila et al 4 that the difference in IOP might be caused by the mode of administration of the amine, and therefore the concentration of 5-HT reaching the site(s) of action might be critical. Intracameral 5-HT might cause an inflammatory response that resembles an irritative response. The decrease in IOP with topical 5-HT, on the other hand, could be caused by a specific receptor-mediated effect. It is worth noting in this context, however, that pentobarbital anesthesia has an effect on IOP. 23 This could explain some of the data reported 4 because some of the experiments were carried out with animals anesthetized with pentobarbital. Topical administration of the 5-HT^agonist 5- carboxamidotryptamine (5-CT) 8 caused an increase in IOP with a time course similar to 5-HT but at a slightly higher level. This demonstrates that 5-CT penetrates the cornea and possibly activates 5-HTj-receptors more effectively than 5-HT itself. This is the first re-

7 Increase of Topical Serotonin and 5-CT IOP in Rabbits 3041 ported evidence of a 5-HTj-agonist influencing IOP. In contrast, studies on the effects of 5-HT antagonists have been reported. Krootila et al 7 used the 5-HT 1A/2 antagonist methysergide and the 5-HT 2 antagonist ketanserin intravenously to attenuate ocular inflammatory effects caused by topical 1% formaldehyde. Pretreatment with methysergide emphatically decreased the IOP (caused by the inflammation). Both methysergide and ketanserin on their own (before the 1% formaldehyde was applied) caused a slight decrease in IOP, methysergide more effectively than ketanserin. Because methysergide has 5-HT 1A and 5-HT 2 antagonist activity as opposed to ketanserin (only 5-HT 2 ), the stronger effect in lowering the IOP is suggestive of a 5-HT 1A effect, possibly involving the secretory ciliary epithelium. The present study shows that topical application of tryptamine, 5,6-dihydroxytryptamine, 5,7-dihydroxytryptamine, and 5-hydroxytryptophan does not cause a change in IOP, suggesting that the 5-HT effect is a specific one. It is likely that all these substances would have a similar ability to penetrate the cornea as serotonin. One further metabolite of serotonin, melatonin, deserves mention here. Melatonin has been extensively studied in relation to the circadian IOP changes. 21 ' 24 ' 25 Yet, we could not demonstrate a change in IOP after topical application. In this case, it is more likely that melatonin did not penetrate the cornea; Chiou et al 21 found an increase in IOP when melatonin was given intracamerally. Krootila et al 4 noted an increase in the aqueous protein content after intracameral infusion of 5-HT and, similarly, after intracameral ketanserin infusion without changes in IOP. This is not explained by the known mode of action of this 5-HT 2 -antagonist. Activation of 5-HT 2 receptors can potentially trigger a prostaglandin response, 26 although that has not been demonstrated in the front of the eye. It is possible that the study was confounded by the measurement of aqueous humor protein 2 hours after manometric IOP recording, because the corneal puncture involved could itself induce a protein increase in the rabbit aqueous humor. 17 In another study, Krootila et al 7 report that ketanserin and methysergide fail to inhibit the breakdown of the blood-aqueous barrier induced by topical pretreatment of 1% formaldehyde. This could indicate that 5-HT is not involved in a bloodaqueous barrier breakdown. In the present study, aqueous humor was sampled 1 hour after topical application of 1% 5-HT. At this time, it was shown that a peak in the increase in IOP had been reached. No difference in aqueous protein content was noted between these eyes and eyes treated with the vehicle only. In comparison, eyes topically treated with 0.25% prostaglandin E 2 (which is known to cause an inflammatory response) showed a threefold increase in the aqueous protein content, as previously reported by Eakins. 17 Pilocarpine, isoproterenol, and timolol had, as expected, no effect on the protein content in the aqueous humor. One can conclude that the rise in IOP that occurs within 1 hour of topical application of 5-HT is not the result of a breakdown in the blood-aqueous barrier. Whether or not 5-HT has a direct effect on the vascular system when applied in other ways (e.g., intravenously or intracamerally) does not contradict the findings of this study. Studies from our laboratory on isolated rabbit sphincter smooth muscle have shown that 5-HT has no effect on the resting tension of the muscle but causes a relaxation of the carbachol-contracted sphincter muscle. 14 This suggests a possible functional role for 5-HT in the control of pupil diameter via an inhibitory effect on cholinergic induced miosis. This hypothesis was tested in this study. Topical administration of 5-HT to the normal rabbit eye had no effect on pupil size. In rabbit eyes pretreated with carbachol, topical 5-HT did not modify the degree of miosis. A possible reason for the lack of correlation between in vitro and in vivo data is that the methodology used to measure pupil size is not sensitive enough to record minute changes induced by serotonin. In the in vitro studies, an isotonic transducer mechanism is used to record changes in muscle length. Based on results previously presented, 12 a decrease in aqueous camp could have been expected after topical administration of 5-HT because in the rabbit iris ciliary body, serotonin receptors are negatively coupled to adenylate cyclase. We therefore measured camp content of the aqueous humor 1 hour after topical application of isoproterenol, 5-HT, or both drugs together. Interestingly, no evidence of phosphodiesterase enzyme activity was found in the aqueous humor because the camp content did not decay with time in isolated samples. Topical application 1% isoproterenol raised the camp level more than two-fold, whereas 1% 5-HT raised the camp only slightly. The normal camp values (0.57 ± 0.02 pmol/ 50 fxl aqueous) compare favorably with data from Yashitomi et al, 27 where values between 0.6 and 1.5 pmol/50 /x\ aqueous humor have been reported, depending on the time during the light-dark cycle. Neufeld et al 28 found a 150% camp increase in the aqueous humor after topical administration of 2% isoproterenol, and a more than 300% rise in camp with topical 1% adrenaline. The source of the measured camp is not known. Because there are usually no cells present in the aqueous humor, camp probably originates from tissues surrounding the anterior chamber. And because

8 3042 Investigative Ophthalmology & Visual Science, September 1993, Vol. 34, No. 10 camp is a universal secondary messenger involved in many functions, whether aqueous camp content reflects events in the secretory process remains a question. In conclusion, the present results show that topically applied serotonin can reach the iris-ciliary body and is associated with an elevation of IOP. Because 5-CT also causes this effect, it is argued that activation of 5-HT, receptors, possibly in the ciliary process, elevates IOP. No evidence could be found that 5-HT causes a breakdown of the blood-aqueous barrier, which would explain the change in IOP. An effect on other aspects of aqueous dynamics or on the extraocular muscles has not been excluded. Key Words intraocular pressure, rabbit, serotonin, 5-carboxamidotryptamine, ciliary body References 1. Chiang TS. Effects of intravenous infusions of histamine, 5-hydroxytryptamine, bradykinin and prostaglandins on intraocular pressure. Arch hit Pharmacodyn. 1974; 207: Osborne NN, Tobin AB. Serotonin-accumulating cells in the iris-ciliary body and cornea of various species. ExpEyeRes. 1987;44: Martin XD, Brennan MC, Lichter PR. Serotonin in human aqueous humor. Ophthalmology. 1988;95: Krootila K, Palkama A, Uusitalo H. Effect of serotonin and its antagonist (ketanserin) on intraocular pressure in the rabbit./ Ocular Pharmacol. 1987; 3: Leysen JE, Awouters F, Kennis L, et al. Receptor binding profile of R 41468, a novel antagonist at 5-HT 2 receptors. Life Sci. 1981;28: Chang FW, Burke JA, Potter DE. Mechanism of the ocular hypotensive action of ketanserin. J Ocular Pharmacol. 1985; 1: Krootila K, Uusitalo H, Palkama A. Effect of alphaadrenergic and serotonergic irritative response in the rabbit eye. Exp Eye Res. 1987;45: Peroutka SJ. 5-Hydroxytryptamine receptor subtypes. Pharmacol & Toxicol. 1990;67: Costagliola C, Scibelli G, Fasano ML, Ferrara LA, Mastropasqua L. Effect of oral ketanserin on intraocular pressure in glaucomatous patients. Exp Eye Res. 1991;52: Tormay A, Severin C, Mittag T. Role of ciliary process 5-HT receptors in ocular pressure. ARVO Supplement. Invest Ophthalmol Vis Sci. 1985; 26: Mallorga P, Sugrue MF. Characteristics, of serotonin receptors in the iris and ciliary body of the albino rabbit. Curr Eye Res. 1987;6: Tobin AB, Osborne NN. Evidence for the presence of serotonin receptors negatively coupled to adenylate cyclase in the rabbit iris-ciliary body, f Neurochem. 1989;53: Tobin AB, Unger W, Osborne NN. Evidence for the presence of serotonergic nerves and receptors in the iris-ciliary body complex of the rabbit. / Neuroscience. 1988;8: Barnett NL, Osborne NN. The effect of serotonin on the rabbit isolated iris sphincter muscle. Curr Eye Res. 1993:(in press). 15. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analyt Biochem. 1976; 72: Brown BL, Ekins RP, Albano JDM. Saturation assay for cyclic AMP using endogenous binding protein. Adv Cyclic Nucleotide Res. 1972; 2: Eakins KE. Prostaglandin and non-prostaglandin mediated breakdown of the blood-aqueous barrier. Exp Eye Res. 1977:483. Supplement. 18. Osborne NN, Nesselhut T, Nicholas DA, Patel S, Cuello AC. Serotonin containing neurones in vertebra retina. J Neurochem. 1982;39: Gregory DS. Timolol reduces IOP in normal NZW rabbits during the dark only. Invest Ophthalmol Vis Sci. 1990;31: Zimmerman TJ, Kaufman HE. Timolol, dose-response and duration of action. Arch Ophthalmol. 1977;95: Chiou GCY, Aimoto T, Chiou LY. Melatonergic involvement in diurnal changes of intraocular pressure in rabbit eyes. Ophthalmic Res. 1985; 17: Moro F, Scapagnini U, Scaletta S, Drago F. Serotonin nerve endings and regulation of pupillary diameter. Ann Ophthalmol. 1981: Van Bijesterveld OP, Van Loenen AC, Ten Ham M. The effect of hypotensive drugs on the intraocular pressure after waterloading in rabbits. Documenta Opthalmologica. 1981; 52: Reppert SM, Sagar SM. Characterization of the daynight variation of retinal melatonin content in the chick. Invest Ophthalmol Vis Sci. 1983;24: Samples JR, Krause G, Lewy AJ. Effect of melatonin on intraocular pressure. Current Eye Res. 1988; 7: Sekar MC, Hokin LE. The role of phosphoinositides in signal transduction./ Membrane Biol. 1986; 89: Yashitomi T, Horio B, Gregory DS. Changes in aqueous norepinephrine and cyclic adenosine monophosphate during the circadian cycle in rabbits. Invest Ophthalmol Vis Sci. 1991;32: Neufeld AH, Jampol LM, Sears ML. Cyclic-AMP in the aqueous humor: The effects of adrenergic agents. ExpEyeRes. 1972; 14: Potter DH, Crossen CE, Heath AR, Ogidigben MJ. Review: Alpha 2 and DA 2 agonists as antiglaucoma agents: Comparative pharmacology and clinical potential./ OCM/ Pharmacol. 1990;6: