The Prostanoid EP₂ Receptor Agonist Butaprost Increases Uveoscleral Outflow in the Cynomolgus Monkey

Abstract

purpose. To investigate the ocular hypotensive effect of the prostanoid EP₂ receptor agonist butaprost and to establish its mechanism of action.

methods. All experiments were performed in cynomolgus monkeys after topical application of butaprost (0.1%). The effects of butaprost on aqueous humor flow were determined by fluorophotometry. Total outflow facility was measured by the two-level, constant-pressure perfusion method, and uveoscleral outflow was determined by perfusion of FITC-labeled dextran through the anterior chamber. Effects on ocular morphology were studied after tissue fixation with transcardial perfusion by paraformaldehyde and immersion fixation of the globe, in animals subjected to long-term treatment with butaprost. Conscious ocular normotensive monkeys and monkeys with unilateral ocular hypertension were used for intraocular pressure (IOP) studies.

results. Butaprost had no significant effect on aqueous humor flow or total outflow facility in ocular normotensive monkeys. Uveoscleral outflow was significantly higher in the butaprost treated eyes than in vehicle treated eyes, 1.03 ± 0.20 vs. 0.53 ± 0.18 μL · min⁻¹. After a 1-year treatment with butaprost, the morphology of the ciliary muscle was changed, showing increased spaces between ciliary muscle bundles and the apparent formation of new outflow channels. In many instances, changes were observed in the trabecular meshwork as well. Butaprost, in a single 0.1% dose, decreased IOP significantly in ocular normotensive monkeys and reduced IOP in laser-induced glaucomatous monkey eyes to the same level as that in the ocular normotensive contralateral eyes.

conclusions. The prostanoid EP₂ receptor agonist butaprost appears to lower IOP by increasing uveoscleral outflow, according to both physiological and morphologic findings. Although the prostanoid EP₂ receptor is structurally and functionally distinct from the FP receptor, the effects of EP₂ and FP receptor stimulation on aqueous humor outflow are similar.

Early studies ^{1 2 3} showed that topical application of prostaglandin (PG) E₂ or PGF_2α to cat or rhesus monkey eyes could cause a long-lasting decrease in intraocular pressure (IOP), without being preceded by an ocular hypertensive phase, which initiated interest in PGs as possible therapeutic agents. Although PGE₂ showed higher efficacy than PGF_2α, PGF_2α was considered to be the best candidate for further development, as PGE₂ was likely to cause more adverse effects and was less stable than PGF_2α in aqueous solution. ⁴ Advances in prostanoid receptor pharmacology enable the reevaluation of PGE₂ analogues for treating glaucoma. 

Presently, the genes encoding eight different prostanoid receptors and their splicing variants have been identified. Prostanoid receptors are designated DP, EP (EP₁–EP₄), FP, IP, and TP, according to the naturally occurring prostanoid that they preferentially recognize: PGD₂, PGE₂, PGF_2α, PGI₂, and TxA₂, respectively. ^{5 6} Of interest, the selective EP₂ receptor agonist AH13205, ⁷ which lacks affinity for the FP receptor, ^{6 7} has been shown to reduce IOP to a moderate extent in the cynomolgus monkey ⁸ and to produce morphologic changes in the aqueous outflow pathways similar to those seen after treatment with latanoprost or bimatoprost, ⁹ indicating that AH13205 exerts its ocular hypotensive effect by increasing uveoscleral outflow. The primary purpose of the present studies was to investigate the ocular hypotensive effect of the other prototypical EP₂ agonist, butaprost. A profound effect on IOP was found in laser-induced ocular hypertensive monkeys, such that IOP in the hypertensive eye was reduced to the level of that in the normotensive eye. Butaprost was, therefore, considered to merit further investigation. Studies were performed to elucidate its physiological mechanism of ocular hypotensive activity and its effects on ocular morphology. 

Methods

Animals and Anesthesia

Experiments described herein adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the animal care and use committees of each institution. 

All experiments were performed on cynomolgus monkeys (Macaca fascicularis). Aqueous humor flow measurements were made in ocular normotensive monkeys lightly and briefly sedated with 5 to 15 mg ketamine (kg body weight)⁻¹. For total outflow facility and uveoscleral outflow studies, anesthesia consisted of an initial intramuscular dose of 15 mg (kg body weight)⁻¹ of ketamine and 1.5 mg (kg body weight)⁻¹) of diazepam. Supplemental doses of ketamine (7.5 mg (kg body weight)⁻¹) were given every 30 minutes and of diazepam (0.75 mg (kg body weight)⁻¹) every 60 minutes. Measurements of IOP were made by pneumatonometry in conscious, trained monkeys. Before any measurement involving touching the cornea, 1 drop of proparacaine (Ophthetic; Allergan, Irvine, CA) was applied topically. 

Administration of Butaprost

In aqueous humor flow experiments, either 25 μL of 0.1% butaprost or vehicle was applied topically to both eyes of sedated monkeys. For all other pharmacological studies butaprost was administered to one eye and vehicle to the contralateral eye. Measurements of IOP, aqueous humor flow, and total outflow facility were made after a single drop. Uveoscleral outflow was determined after 5 days of once-daily administration. Histologic studies were performed after 1 year of once-daily application to the right eye. The animals were fully conscious during application but were restrained in custom-designed chairs. 

Determination of Aqueous Humor Flow

All measurements were made in ketamine-sedated animals. At 8:00 AM on each of the two measurement days, corneal thickness and anterior chamber depth were measured by ultrasound pachymetry (Sonomed, Inc., Lake Success, NY), and corneal diameter was measured with a ruler. Cornea and anterior chamber volumes were calculated from these values. ¹⁰ One 20-μL drop of fluorescein (10%) was then applied to the cornea. In a randomized masked fashion, vehicle or butaprost was administered to both eyes at 8:30 AM. 

Between 12:30 and 3:30 PM, with the animal placed prone, the fluorescence of the cornea and anterior chamber was measured with a scanning ocular fluorophotometer (Fluorotron; OcuMetrics, Palo Alto, CA). Scans were taken in duplicate at 60-minute intervals for four sets. These data were used to determine aqueous humor flow (F_a). ¹¹ Measurements were repeated 6 days later, 4 hours after a topical drop of fluorescein and an alternate drop of either butaprost or vehicle. 

Determination of Total Outflow Facility

Total outflow facility was determined according to the two-level constant pressure-perfusion method. ¹² The anterior chambers were cannulated by a single needle, connected in series by polyethylene tubing to a pressure transducer and a reservoir, filled with mock aqueous humor. ¹³ The connection to the reservoir was clamped during cannulation and the initial part of the experiment, allowing determination of spontaneous IOP. Determination of total outflow facility was started 4 to 6 hours after a single dose of butaprost and was made by alternating between two pressure levels, approximately 5 and 10 mm Hg above spontaneous IOP. The inflow from the reservoir was determined during five 10-minute periods (three low and two high levels), allowing four values of outflow facility to be determined and the mean value was used for statistical analysis. At the end of the experiment, the connection to the reservoir was clamped, to allow determination of spontaneous IOP once again. 

After the experiment, the needles were gently removed, and the monkey received bilateral treatment with 3 drops of ofloxacin (Ocuflox; Allergan, Irvine, CA), and intramuscular injection of 2 mg/kg of flunixin meglumine (Banamine; Schering-Plough, Kenilworth, NJ) and 0.1 mg/kg of atropine. This combination of drugs were used to minimize the risk of inflammation and development of synechiae. Baseline values, without drug treatment, were obtained in the same animals 11 months after drug treatment. 

Determination of Uveoscleral Outflow

One month after the baseline measurements of total outflow facility, the same animals were used for determination of uveoscleral outflow by the anterior chamber perfusion technique ¹⁴ using FITC-labeled dextran (70 kDa) as tracer ¹⁰ at a concentration of 0.7% in mock aqueous humor. 

Each eye was cannulated with three 25-gauge needles. Two of the needles were connected by polyethylene tubing to 5 mL push–pull coupled syringes, driven by a pump (Harvard Apparatus, South Natick, MA). The third needle was coupled to a reservoir with an in-line pressure transducer, to measure IOP and maintain it at a predetermined level of 15 mm Hg. During the cannulation and the initial part of the experiment, the connection to the reservoir was clamped to measure spontaneous IOP. A 30-minute perfusion of the anterior chamber commenced 4 to 6 hours after the last dose of butaprost. During the initial 5 minutes, a high pump speed (0.2 mL/min) was used to achieve a quick exchange of anterior chamber contents. The pumps were then stopped briefly, and the pull syringes were exchanged with empty ones. The perfusion was continued at low speed (0.05 mL/min) for 25 minutes, with IOP held at 15 mm Hg. After 30 minutes, the animal was euthanatized by intravenous injection of 2 mL Eutha-6, and the perfusates in the pull syringes were collected to determine the mean concentration of tracer. Both eyes were perfused with mock aqueous without tracer at a high speed (0.2 mL/min) for 10 minutes, to rinse the tracer from the anterior chambers and the episcleral–extraocular vascular system. The eyes were then immediately enucleated and dissected into anterior and posterior sclera, extraocular tissues, ciliary body, choroid, retina, vitreous, and fluids (aqueous humor + 10 mL of phosphate-buffered saline [PBS], used to wash the dish and instruments used for dissection). The cornea, lens, and iris were excluded, as these tissues are not thought to contribute to uveoscleral outflow. The tissue samples were then stored at −20°C until assayed. Thawed tissue samples were homogenized in PBS, and placed on ice for 60 minutes. The samples were centrifuged for 20 minutes at 3500g (4°C). The supernatants were transferred to new tubes and centrifuged at 20,000g (10°C) for 60 minutes. The resultant supernatants were stored refrigerated until analyzed. 

The concentration of tracer in the perfusates and tissue supernatants were determined by the use of a spectrofluorometer (LS-50B; Perkin-Elmer, Wellesley, MA). Total uveoscleral outflow (F_u) was calculated:  

\[\mathrm{F}_{\mathrm{u}}{=}\ {{\sum}_{{\bullet}}}\ T_{\mathrm{tissue}}{\times}T_{\mathrm{a}}^{\mathrm{{-}}1}{\times}t^{\mathrm{{-}}1}\]

where T _tissue is the amount of tracer in the tissue, T _a is the mean concentration of tracer in the anterior chamber during the perfusion, and t is the perfusion time (30 minutes). 

Morphology

For 1 year, the right eyes of five monkeys were treated once daily with 25 μL 0.1% butaprost. Five different animals were similarly treated with vehicle (5% Poloxamer 407 dissolved in distilled water). Contralateral eyes were untreated. Two untreated animals served as age-matched control subjects. 

Tissue fixation was achieved by transcardial perfusion with 4% paraformaldehyde for 10 minutes, in animals that were deeply anesthetized with pentobarbital sodium. After the temporal quadrant was marked, the eyes were enucleated, and small, wedge-shaped pieces were removed from the four quadrants of the sclera. The cornea was removed, and the globe was then immersion fixed in Ito’s solution. This method allows adequate penetration of fixative into the globe, whereas the ciliary muscle remains attached to the insertion of the posterior and anterior tendons and thereby retains its configuration. 

The eyes, fixed in Ito’s solution, were rinsed in cacodylate buffer. The specimens were postfixed in 1% OsO₄, dehydrated in an ascending series of alcohols, and embedded in Epon, according to standard methods. Semithin sections were cut on a microtome (Ultracut OmU3; Reichert, Vienna, Austria) and stained with toluidine blue. Ultrathin sections were stained with uranylacetate and lead citrate and viewed in an electron microscope (EM 902; Carl Zeiss Meditec, Oberkochen, Germany). 

Determination of IOP

IOP measurements (model 30R pneumotonometer; Digilab, Boston, MA) were made in ocular normotensive monkeys and in monkeys rendered unilaterally ocular hypertensive by argon laser photocoagulation of the trabecular meshwork. The measurements were made 1 hour before and 6 hours after treatment, in trained, conscious monkeys seated in custom chairs. 

Statistical Analysis

Student’s paired two-tailed t-test was used for all statistical comparisons. All data are the mean ± SEM. 

Results

Aqueous Humor Flow

On vehicle and butaprost treatment days, there was no significant difference in aqueous flow between left and right eyes. Butaprost did not affect aqueous humor flow, when compared with the same eyes on vehicle-treatment day (Table 1) . 

Total Outflow Facility

A single dose of butaprost (0.1%) had no significant effect on total outflow facility in ocular normotensive monkeys. Neither was there a significant difference in outflow facility, when compared with the baseline, determined 11 months later (Fig. 1) . 

Uveoscleral Outflow

The highest amount of tracer was recovered from the ciliary body and the anterior and posterior sclera. For all tissues, the amount of tracer was higher in the butaprost-treated eye than in the control eye, but the difference was statistically significant only for extraocular tissues and the wash fluids (Fig. 2) . Total uveoscleral outflow was 1.03 ± 0.29 μL · min⁻¹ in the butaprost-treated eyes and 0.53 ± 0.18 μL · min⁻¹ in the control eyes (P ≤ 0.01). 

Morphology

After 1 year of daily treatment with 0.1% of the EP₂ receptor agonist butaprost, morphologic changes in the anterior eye segment were restricted to the anterior ciliary muscle and the trabecular meshwork. The ciliary processes including the epithelial cells and the adjacent capillaries appeared normal. There were no signs of inflammatory reactions in the anterior eye segment. 

Ciliary Muscle.

The general morphology of the ciliary muscle was not changed by chronic butaprost treatment compared with controls. The tips were at the level of the scleral spur, and the circular portion formed an inner edge giving the muscle the typical triangular appearance (Fig. 3) . By light microscopy, the muscle fibers appeared normal in all quadrants of the eye. There were, however, changes at the tips of the ciliary muscle, where the muscle bundles were separated by enlarged intermuscular spaces. Enlarged spaces in the butaprost-treated eyes were only present in the anterior third of the muscle and were restricted to the longitudinal and reticular portion (Figs. 3B 3C) . Spaces described as + in Table 2were short and restricted mainly to the transition zone between the longitudinal and reticular portion of the anterior ciliary muscle (Fig. 3B) . If enlarged intermuscular spaces were present between the longitudinal and reticular muscle portions and extended further posteriorly, they were described as ++ in Table 2(Fig. 3C) . Such ++ spaces were found only in the right butaprost-treated eyes (Table 2) . 

Changes described as + were also found in the left, nontreated, contralateral control eyes (Table 2) . A contralateral treatment effect has also been found after 1 year of treatment with PGF_2α analogues. ⁹ We assume that a contralateral effect cannot be avoided in conscious monkeys treated unilaterally for 1 year. 

At the ultrastructural level, both ++ and + intermuscular spaces appeared as straight channels, incompletely lined by elongated endothelial-like cells (Fig. 4) . These cells were separated from the ciliary muscle cells by their basement membrane and small amounts of fibrillar and amorphous material. Within the intermuscular spaces, capillaries were often in contact with processes of the endothelial-like cells lining the optically empty spaces (Fig. 4) . 

In both treated and contralateral nontreated eyes, the intermuscular spaces were never distributed evenly throughout the circumference of the eye. In most eyes, they were restricted to the lower quadrant, whereas the upper quadrants of the same eyes appeared normal (Table 2) . 

Trabecular Meshwork.

In most of the butaprost-treated eyes, there were also morphologic changes in the TM. These changes were either restricted to the cribriform region (+ in Table 2 ) or were seen in both the cribriform region and the lamellated TM (++ in Table 2 ; Fig. 5B ). The entire cribriform region was enlarged in the inward–outward direction (Figs. 5A 5B) . In places, the endothelium of Schlemm’s canal appeared pressed against the outer wall (Fig. 5A) . In these regions, the collector channels and especially their connections to Schlemm’s canal were often enlarged (Fig. 5B) . In areas with + changes, the trabecular lamellae appeared almost normal (Fig. 5A) . Places with ++ changes, revealed an expanded cribriform region. In addition, in ++ areas, the connective tissue of the central core of most of the trabecular lamellae were largely reduced and only small, short remnants of the lamellae were seen (Fig. 5B) . 

Electron microscopic investigation of the affected TM revealed that, in the cribriform region, some TM cells were disconnected from the endothelial lining of Schlemm’s canal. They also appeared disconnected from the cribriform elastic network (Fig. 6) . The TM cells of both the cribriform and corneoscleral TM appeared activated with increased amounts of rough endoplasmic reticulum and Golgi membranes. In places, TM cells bridged spaces between trabecular lamellae and cribriform cells (Fig. 6) . Within some of the trabecular lamellae, in areas with ++ changes, the elastic fibers were still present, but there was loss of collagen material. Remnants of fibrillar material were seen and the endothelial lining was incomplete. 

In general the TM changes were found in the same eye and regions as the muscle changes. TM changes assigned ++ were also restricted to the treated right eyes, but were present in more quadrants of each eye than the muscle changes (Table 2) . 

Intraocular Pressure

Butaprost (0.1%) significantly lowered normal monkey IOP (Fig. 7) . In glaucomatous monkeys a single application of butaprost (0.1%) to the hypertensive eye reduced the IOP to that of the contralateral normotensive eye (Fig. 8) . 

Discussion

The prostanoid EP₂ receptor agonist butaprost ^{5 15 16 17} reduced IOP in monkeys by a mechanism that appears to involve increased uveoscleral outflow, which was supported by examining the morphology of the ciliary body. The observed morphologic changes in the ciliary muscle were similar to those reported for a pharmacologically diverse series of prostanoid analogues. ⁹  

Morphologic changes were also observed in the trabecular meshwork after 1 year of treatment with butaprost. The extracellular matrix in the TM appeared diminished. The TM changes produced by butaprost were more pronounced than those observed for the prostanoid analogues previously investigated. ⁹ Given the extensive changes produced by butaprost at the TM level, an increase in conventional outflow may have been expected. It should be noted, however, that the morphologic changes were evaluated after 1 year of butaprost treatment, whereas total outflow facility was measured 4 to 6 hours after a single dose of butaprost. In parallel experiments, the EP₂ agonist AH13205 had no effect on outflow facility, after 5 days of twice-daily treatment (Nilsson S, et al. IOVS 2004;45:ARVO E-Abstract 4663). It therefore seems unlikely that the more selective and potent EP₂ agonist butaprost ⁶ should have affected outflow facility after 5 days of treatment. Thus, the effects on TM morphology and lack of effect on total outflow facility may be viewed as either the result of TM changes occurring as a more delayed response to continued butaprost treatment or that such changes have no net resultant effect on total outflow facility. 

The morphologic changes in the contralateral nontreated eye may indicate an increased drainage via the uveoscleral outflow, also in the fellow eye. However, the values on uveoscleral outflow in the control eyes were similar to those obtained in another study, in which the investigated drug had no effect on uveoscleral outflow (Kharlamb A, et al. IOVS 2004;45:ARVO E-Abstract 1035). This indicates that the morphologic changes in the contralateral eyes are the results of the longer drug treatment. 

From the pharmacological standpoint, it may be regarded as surprising that EP₂ receptor stimulation results in ocular effects virtually indistinguishable from those produced by FP receptor stimulation, ⁹ because these are quite distinct receptor entities. These unexpected similarities are not adequately explained by current information on ocular prostanoid receptor pharmacology. Both prostanoid receptors are widely represented in human ocular tissues and are present in cells and tissues involved in aqueous humor dynamics. ¹⁸ The endogenous agonists, PGE₂ and PGF_2α, are not very selective, ⁶ indicating that part of the effect of PGF_2α on uveoscleral outflow could be via activation of EP₂ receptors and vice versa. However, it seems very unlikely that selective analogues, such as butaprost (EP₂) and latanoprost (FP), ⁶ should interfere with each other’s receptors under in vivo conditions. Thus, a common pathway in the postreceptor signaling seems to be the most plausible explanation, and there are several possible sites of interaction. 

Some insight is being provided by gene regulation studies that may begin to explain the similarities between EP₂ and FP receptor effects on uveoscleral outflow. The gene Cyr 61, belonging to the CCN (connective tissue growth factor/cysteine-rich 61/nephroblastoma overexpressed) gene family, has been implicated in tissue remodeling ^{19 20} and could be regarded as a candidate for the initiation of PG-induced remodeling of the ciliary body. It seems more than coincidental that both EP₂ and FP agonists cause upregulation of Cyr 61, in cultured human ciliary muscle cells ²¹ and produce similar remodeling of the ciliary body in living monkeys. ⁹ This provides a potential explanation for upstream EP₂ agonist effects on uveoscleral outflow and their similarity to FP agonist effects. It is also important to note that both FP and EP₂ receptor agonists cause matrix metalloproteinase (MMP) secretion from ciliary smooth muscle cells, providing further association with events resulting in the morphologic changes typical of increased uveoscleral outflow. ^{22 23} At the second-messenger level, both FP and EP₂ receptor stimulation upregulates the orphan nuclear receptor Nur 77 in human ciliary smooth muscle and TM cells by a protein kinase C (PKC)–dependent pathway. ²⁴ Thus, to some extent, both FP and EP₂ agonists effects on uveoscleral outflow may be reconciled at the gene and second-messenger level in human ciliary smooth muscle cells. 

Another intriguing similarity between EP₂ agonists and FP agonists is their potential to stimulate the endogenous formation of prostanoids in the anterior segment. FP agonists have been shown to stimulate endogenous formation of PGE₂ in ocular tissues, ^{25 26} cultured melanocytes, ²⁷ and nonpigmented ciliary epithelial (NPE) cells. ^{28 29} In NPE cells, the formation of PGE₂ is also stimulated by the EP₂ agonist butaprost. ²⁹ The increased PGE₂ formation appears to be caused by an upregulation of cyclooxygenase-2 (COX-2), ^{27 28 29} which is accompanied by phosphorylation of mitogen-activated protein kinases (MAPK) p38 and p42/44, ^{28 29} indicating a common signaling pathway for FP and EP₂ agonists. Furthermore, both latanoprost and PGE₂ upregulated the expression of MMP-1 in NPE cells, with PGE₂ being effective already at nanomolar concentrations. ²⁸ These findings prompted the authors to suggest that latanoprost, via Ca²⁺, PKC and MAPKs, upregulate the expression of COX-2 and hence the formation of PGE₂, which subsequently increases the MMP-1 expression. MMP-1 could then be released from the nonpigmented ciliary epithelium, transported by the aqueous humor to the ciliary muscle and trabecular meshwork, to induce tissue remodeling and facilitation of outflow. ²⁸ However, PGE₂ is not the only prostaglandin released by FP receptor stimulation ²⁵ and COX-2 is also responsible for the formation of prostaglandin glycerol esters and ethanolamides (prostamides) from the endocannabinoids 2-arachidonylglycerol and anandamide, respectively. ^{30 31} Thus, several prostanoids with potential effects on uveoscleral outflow could be formed endogenously after exogenous application of prostanoid analogues. Furthermore, prostamides, and possibly prostaglandin glycerol esters, may have been misidentified as prostaglandins; commercially available antibodies directed against PGE₂ and PGF_2α cross-reacted with PGE₂ ethanolamide and PGF_2α ethanolamide, respectively. ³²  

The effects of a single 0.1% dose of butaprost on IOP in the “glaucomatous” monkey model were profound. Butaprost essentially normalized IOP in the “glaucomatous” monkey model, with IOP of the ocular hypertensive eyes being reduced to that of the ocular normotensive, contralateral eyes within 2-hours of posttreatment. In this model, butaprost appears to have greater efficacy than all classes of ocular hypotensive drugs currently used in clinical practice, including prostanoid analogues. ^{33 34 35 36 37 38} This high efficacy could be regarded as unexpected because the nature and degree of the morphologic changes in the ciliary body produced by chronic butaprost treatment are not obviously different from those produced by clinical doses of latanoprost and bimatoprost. ⁹ As it seems unlikely that altered gene expression and subsequent remodeling of the outflow pathways should have fully developed within 2 hours, one may suspect that the rapid decrease in IOP may be caused by increased uveoscleral outflow due to relaxation of the ciliary muscle. A recent study shows that PGE₁ and PGE₂ inhibit contraction of the monkey ciliary muscle in vitro, whereas FP agonists is without effect. ³⁹ In the cat ciliary muscle, prostanoid-induced relaxation is mediated by EP₂ and DP receptors. ⁴⁰  

It may be concluded at this juncture that the EP₂ agonist butaprost rapidly produces profound ocular hypotension. The underlying mechanism appears to involve increased uveoscleral outflow. 

 Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, April 2004, Fort Lauderdale, Florida.

 Supported by Allergan, Inc.

 Submitted for publication December 20, 2005; revised March 20 and April 28, 2006; accepted July 19, 2006.

 Disclosure: S.F.E. Nilsson, Allergan, Inc. (F); E. Drecoll, None; E. Lütjen-Drecoll, None; C.B. Toris, Allergan, Inc. (F); A.H.-P. Krauss, Allergan, Inc. (E); A. Kharlamb, Allergan, Inc. (E); A. Nieves, Allergan, Inc. (E); T. Guerra, Allergan, Inc. (E); D.F. Woodward, Allergan, Inc. (E)

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Aqueous Humor Flow

	Vehicle Treatment			Butaprost Treatment
		OD	OS	OD	OS
Fa (μL · min−1)	1.88 ± 0.29	2.07 ± 0.28	2.17 ± 0.34	2.38 ± 0.34

 Measurements were obtained under baseline conditions and after bilateral treatment with a single dose of 0.1%, 25 μL butaprost. Data are the mean ± SEM; n = 8, except for left-eye butaprost treatment, where n = 7.

Figure 1.

 

Total outflow facility on baseline day and treatment day, 4 to 6 hours after treatment with vehicle or butaprost (0.1%, 25 μL), respectively (n = 6). Mean ± SEM.

Figure 2.

 

Effect of butaprost on uveoscleral outflow. Butaprost (0.1%; 25 μL) was given once daily for 5 days, and the determination of uveoscleral outflow was made 4 to 6 hours after the last dose. Flow to different tissues, calculated from the amount of tracer recovered from the tissue, and total uveoscleral outflow (F_u) are shown. Mean ± SEM. *P ≤ 0.05 and **P ≤ 0.01 (paired, two-tailed Student’s t-test; n = 6).

Figure 3.

 

Sagittal semithin sections through the anterior ciliary body of a butaprost-treated eye. A normal appearing part of the ciliary muscle (A), a part of the ciliary muscle with short enlarged spaces in the transition zone between longitudinal and circular muscle portion (arrows, + in Table 2 ) (B), and an area of the muscle tip with changes described as ++ in Table 2(C). The enlarged intermuscular spaces (arrows) are present between bundles of the longitudinal and reticular portion of the muscle and extend farther posteriorly than in areas described as + in Table 2 .

Table 2.

 

Semiquantitative Evaluation of the Ciliary Muscle Tips and the TM in the Study Groups

Animal	Right Eye: Butaprost					Left Eye: Untreated
		Quadrant	Muscle Clefts	TM Changes	Quadrant	Muscle Clefts	TM Changes
Butaprost-treated monkeys
R45	1	−	+	1	+	−
	2	−	−	2	−	−
	3	−	−	3	−	−
	4	−	−	4	−	−
R46	1	−	+	1	+	+
	2	−	−	2	−	+
	3	−	+	3	−	−
	4	++	+	4	−	+
R47	1	−	−	1	−	−
	2	−	−	2	−	−
	3	−	−	3	+	−
	4	++	+	4	+	−
R48	3	++	++	3	−	+
	4	++	++	4	+	+
R49	3	−	−	3	+	−
	4	−	+	4	−	−
Vehicle-treated monkeys
R50	1	−	−	1	−	−
	2	−	−	2	−	−
	3	−	+	3	−	−
	4	−	−	4	−	−
R51	1	−	−	1	−	−
	2	+	+	2	+	−
	3	+	−	3	+	+
	4	+	−	4	+	−
R52	1	−	−	1	−	−
	2	−	−	2	−	−
	3	−	−	3	+	−
	4	−	−	4	−	−
R53	3	−	−	3	−	−
	4	−	−	4	−	−
R54	3	−	+	3	−	+
	4	+	+	4	−	−
Untreated control subjects
R55	2	−	−	1	−	−
	4	−	−	3	−	−
				4	−	−
R56	1	−	−	1	−	−
	2	−	−	2	−	−
	4	−	−	3	−	−

 Quadrants: 1, upper nasal; 2, upper temporal; 3, lower nasal; 4, lower temporal. −, no change; ciliary muscle clefts +, short enlarged spaces at the transition zone between reticular and longitudinal portion (see Fig. 3B ); ++, long, enlarged spaces between anterior longitudinal and reticular muscle fibers (Fig. 3C) . TM changes: +, changes restricted to the cribriform TM (Fig. 5A) ; ++, changes present in the cribriform and lamellated TM (Fig. 5B) .

Figure 4.

 

Electron micrograph of an enlarged intermuscular space. The space is lined by endothelial-like cells (arrowhead) forming contacts with the vessel wall of the capillary (C).

Figure 5.

 

Semithin sagittal sections through the trabecular meshwork. In an area described as + changes in Table 2(A), note that the cribriform region was widened and at places in contact with septa of the outer wall. Most of the trabecular lamellae appeared normal. In an area described as (++) in Table 2(B), the inner wall endothelium at several places was disconnected from the cribriform region (arrows). Several TM lamellae were reduced in thickness, some appeared nearly destroyed (arrowheads). SC, Schlemm’s canal; CC, collector channel.

Figure 6.

 

Electron micrograph of the inner wall of Schlemm’s canal (SC) and the trabecular meshwork in an area with ++ changes (Table 2) . At places the cribriform cells are disconnected from the canal endothelium (arrowheads) and seem to bridge enlarged cribriform spaces with adjacent partly destroyed corneoscleral lamellae. Arrows: destroyed lamellae with loss of collagen and incomplete TM cell lining.

Figure 7.

 

Effect of butaprost (0.1%, single dose) on IOP in normotensive monkey eyes. Butaprost-treated eyes, vehicle-treated eyes. (Mean ± SEM, **P < 0.01; paired, two-tailed Student’s t-test; n = 6).

Figure 8.

 

Effect of butaprost (0.1%; single dose) on IOP, in unilaterally glaucomatous monkeys. Butaprost was applied to the ocular hypertensive eye and an equal volume of vehicle was applied to the contralateral normotensive eye at time 0. Mean ± SEM **P < 0.01 (change in IOP in treated eyes compared with change in IOP in control eyes; paired two-tailed Student’s t-test; n = 6).

The authors thank the Animal Sciences Group at Allergan, Inc. for treating and caring for the animals and Lisa Rubin for preparing the manuscript.