<<

Vasoactive Responses of , PGF2␣, , and in Isolated Porcine Ciliary Arteries

Ineta Vysniauskiene, Reto Allemann, Josef Flammer, and Ivan O. Haefliger

PURPOSE. To compare the vasoactive properties of the prosta- these responses can be modulated by SQ 29548 (TP-

noids U46619 ( A2 analogue), F2␣ antagonist) or AL-8810 (FP-receptor antagonist). (PGF2␣), latanoprost free acid, and travoprost free acid in isolated porcine ciliary arteries. METHODS. In a myograph system (isometric force measure- MATERIAL AND METHODS ment), quiescent vessels were exposed (cumulatively) to U46619, PGF , latanoprost, or travoprost (0.1 nM–0.1 mM). 2␣ Vessels Preparation Experiments were also conducted in the presence of SQ 29548 (TP-receptor antagonist; 3–10 ␮M) or AL-8810 (FP-receptor Porcine eyes were obtained from an abattoir immediately after death. antagonist; 3–30 ␮M). Contractions were expressed as the In cold modified Krebs-Ringer solution (NaCl 118 mM, KCl 4.7 mM,

percentage of 100 mM potassium chloride-induced contrac- CaCl2 2.5 mM, K2PO4 1.2 mM, MgSO4 1.2 mM, NaHCO3 25 mM, tions. glucose 11.1 mM, and EDTA 0.026 mM), ciliary arteries were dissected 5 ␮ RESULTS. In quiescent vessels, contractions (concentration–re- and cut into 2-mm segments. In an organ chamber, two 45- m Ϯ tungsten wires were passed through the vessel’s lumen and attached to sponse curves) induced by (0.1 mM) PGF2␣ (87.9% 3.5%), U46619 (66.7% Ϯ 4.1%), and latanoprost (62.9% Ϯ 3.6%) were a force transducer for isometric force measurements (Myo-Interface; JP more pronounced (P Յ 0.001) than those induced by tra- Trading, Aarhus, Denmark). Vessels were left in place for 30 minutes in voprost (23.0% Ϯ 4.4%). Concentration–response curves for modified Krebs-Ringer solution (95% O2,5%CO2, 37°C) before being stretched in a stepwise manner and exposed between stretch steps to PGF2␣, latanoprost, and travoprost were preceded by a smaller contraction peak (0.1 ␮M) that was higher (P Յ 0.05) for 100 mM potassium chloride (KCl) until the optimum passive tension Ϯ Ϯ was reached (i.e., Ͻ10% variation of KCl-induced contractions be- travoprost (24.4% 2.8%) than for PGF2␣ (12.9% 4.6%), but not different (P ϭ 0.58) from latanoprost (22.0% Ϯ 3.0%). The tween two different stretch steps).6 To assess the functional integrity of the endothelium, vessels were precontracted with 0.1 ␮M U46619 50% maximal contraction (PD50: negative log M concentration) of U46619 (Ϫ8.05 Ϯ 0.13) was lower (P Յ 0.001) than those and then exposed to bradykinin. The endothelial function was consid- Ϫ Ϯ Ϫ Ϯ ered to be preserved if bradykinin (100 nM–1 ␮M) evoked at least 80% of latanoprost ( 5.65 0.10), PGF2␣ ( 5.49 0.14), and travoprost (Ϫ5.12 Ϯ 0.52). Contractions were inhibited (P Յ relaxation.7 Vessels were then washed out with modified Krebs-Ringer 0.05–0.001) either by SQ 29548 or AL-8810. solution before any of the following experimental protocols were conducted. CONCLUSIONS. In isolated porcine ciliary arteries, all tested induced contractions. Among them, travoprost ap- peared to be the least potent and U46619 the most efficient. Experimental Protocols (Invest Ophthalmol Vis Sci. 2006;47:295–298) DOI:10.1167/ iovs.05-0760 In a first set of experiments, quiescent ciliary arteries were exposed cumulatively, to increasing concentrations (0.1 nM–0.1 mM) of

U46619, PGF2␣, latanoprost, or travoprost. In a second set of experi- he analogue U46619 and prostaglandin ments, quiescent vessels were first incubated for 30 minutes with 1 ␮ TF2␣ (PGF2␣) belong to the family. U46619 is different concentrations (3–10 M) of SQ 29548 (TP-receptor antago- regarded as a thromboxane A-receptor (TP) agonist and PGF2␣ nist) and then exposed to increasing concentrations (0.1 nM–0.1 mM) 1 as a prostaglandin F-receptor (FP) agonist. U46619 and PGF2␣ of U46619, PGF2␣, latanoprost, or travoprost. In a third set of experi- are known to be strong vasoconstrictors, particularly in the ments, quiescent vessels were first incubated for 30 minutes with ocular circulation.2,3 different concentrations (3–30 ␮M) of AL-8810 (FP-receptor antago- Latanoprost and travoprost are potent ocular hypotensive nist) and then exposed to increasing concentrations (0.1 nM–0.1 mM)

drugs that share some chemical structural similarities with of U46619, PGF2␣, latanoprost, or travoprost. Timed control experi- 4 PGF2␣. In the present study, we compared the vasoactive ments were conducted in parallel with only the vehicle solution. responses of U46619, PGF2␣, latanoprost, and travoprost in isolated porcine ciliary arteries and additionally assessed how Drugs and Statistical Analysis

U46619, PGF2␣, latanoprost free acid, and travoprost free acid were From the Laboratory of Pharmacology and Physiology, University obtained from Cayman Chemicals (Ann Arbor, MI) and dissolved in Eye Clinic, Basel, Switzerland. pure ethanol (maximal final concentration in the organ chamber, Supported in part by Alcon Laboratories, Fort Worth, Texas. Յ1%). Bradykinin and SQ 29548 were purchased from Sigma-Aldrich Submitted for publication June 16, 2005; revised July 15, 2005; (Buchs, Switzerland). AL-8810 was provided by Alcon Laboratories accepted November 21, 2005. (Fort Worth, TX). Concentrations are expressed as final molar concen- Disclosure: I. Vysniauskiene, Alcon Laboratories (F); R. Alle- trations in the organ chambers. Contractions are given as percent of mann, Alcon Laboratories (F); J. Flammer, Alcon Laboratories (F); 100 mM KCl-induced maximal contraction. Results are provided as I.O. Haefliger, Alcon Laboratories (F) mean Ϯ SEM with n representing the number of vessels studied (one The publication costs of this article were defrayed in part by page vessel per eye). The concentration causing 50% of the maximal con- charge payment. This article must therefore be marked “advertise- ment” in accordance with 18 U.S.C. §1734 solely to indicate this fact. traction is expressed as a negative log M concentration (PD50). Statis- Corresponding author: Ivan O. Haefliger, Laboratory of Ocular tical comparisons were conducted with a one-way ANOVA multiple Pharmacology and Physiology, University Eye Clinic, Mittlere Strasse comparison followed by the Bonferroni test, with a two-tailed P Յ 0.05 91, PO Box, CH-4012 Basel, Switzerland. considered to be statistically significant.

Investigative Ophthalmology & Visual Science, January 2006, Vol. 47, No. 1 Copyright © Association for Research in Vision and Ophthalmology 295

Downloaded from tvst.arvojournals.org on 09/28/2021 296 Vysniauskiene et al. IOVS, January 2006, Vol. 47, No. 1

RESULTS Effect of Prostanoids in Quiescent Vessels In quiescent vessels (Fig. 1), all prostanoids induced contrac- tions (concentration–response curve). At a 0.1 mM concentra-

tion, U46619, PGF2␣, latanoprost free acid, and travoprost free acid evoked contractions that were significantly different from those in timed control experiments. Contractions induced by Ϯ Ϯ PGF2␣ (87.9% 3.5%), U46619 (66.7% 4.1%), and latano- prost (62.9% Ϯ 3.6%) were significantly (P Յ 0.001) more pronounced than those induced by travoprost (23.0% Ϯ 4.4%). Ϫ Ϯ Յ The PD50 of U46619 ( 8.05 0.13) was significantly (P Ϫ Ϯ 0.001) lower than that of latanoprost ( 5.65 0.10), PGF2␣ (Ϫ5.49 Ϯ 0.14), and travoprost (Ϫ5.12 Ϯ 0.52). It has to be noted that the concentration–response curves evoked by

PGF2␣, latanoprost free acid, and travoprost free acid were preceded by a small peak contraction that occurred at a con- centration of 0.1 ␮M and that was significantly different from that in the timed control. This first peak contraction was Յ significantly (P 0.05) different for travoprost than for PGF2␣ but not significantly (P ϭ 0.58) different for latanoprost.

FIGURE 1. Contracting effect of the prostanoids U46619, PGF2␣,la- tanoprost free acid (latanoprost), and travoprost free acid (travoprost) Effect of SQ 29548 on in quiescent isolated porcine ciliary arteries. Statistical comparisons Prostanoid-Induced Contractions were made versus timed control (control) and U46619 (§), PGF2␣ (‡), latanoprost (†), or travoprost (*), with a single, double, or triple symbol In the presence of different concentrations of SQ 29548 (TP- corresponding to P Յ 0.05, P Յ 0.01, or P Յ 0.001, respectively. receptor antagonist), the concentration-response curve of U46619 (TP-receptor agonist) was significantly (P Յ 0.001) shifted to the right, without any decrease in the maximal contraction observed, indicating that SQ 29548 is likely to act as a competitive inhibitor for U46619 (Fig. 2). In the presence

FIGURE 2. Effect of the TP-receptor antagonist SQ 29548 on contractions

induced by U46619, PGF2␣, latano- prost free acid (latanoprost), and travoprost free acid (travoprost) in quiescent isolated porcine ciliary ar- teries. Statistical comparisons were made between timed control (con- trol) and the effect of 3 ␮M (*) and 10 ␮M (†) SQ 29548 on the contraction induced by the prostanoids, with a single, double, or triple symbol cor- responding to P Յ 0.05, P Յ 0.01, or P Յ 0.001, respectively.

Downloaded from tvst.arvojournals.org on 09/28/2021 IOVS, January 2006, Vol. 47, No. 1 Prostanoid-Induced 297

FIGURE 3. Effect of the FP-receptor antagonist AL-8810 on contractions

induced by U46619, PGF2␣, latano- prost free acid (latanoprost), and travoprost free acid (travoprost) in quiescent isolated porcine ciliary ar- teries. Statistical comparisons were made between timed control (con- trol) and the effect of 3 ␮M(*), 10 ␮M (†), 20 ␮M (‡), and 30 ␮M (§) AL-8810 on the contraction induced by the prostanoids, with a single, double, or triple symbol correspond- ing to P Յ 0.05, P Յ 0.01, or P Յ 0.001, respectively.

of 10 ␮M SQ 29548, the concentration–response curve evoked tions. It also has to be noted that the vasoconstrictive response

by PGF2␣ also tended to be shifted to the right, whereas those of U46619 differed from the three other prostanoids in the evoked by latanoprost and travoprost approached zero. sense that the contractions evoked by PGF2␣, latanoprost free acid, and travoprost free acid were preceded by a small peak Effect of AL-8810 on contraction. This peak contraction was observed at a concen- Prostanoid-Induced Contractions tration as low as 0.1 ␮M for the three drugs. The peak-contrac- In the presence of increasing concentrations of AL-8810 (FP- tion induced by latanoprost and travoprost at a concentration ␮ receptor antagonist), the concentration-response curves of of 0.1 M did not differ significantly. In contrast, at higher concentrations (0.1 mM), the contractions evoked by PGF2␣ PGF2␣ (FP-receptor agonist), but also of U46619 and latano- prost tended to be shifted to the right in a concentration- and latanoprost free acid were significantly higher than those dependent manner (Fig. 3). In the presence of AL-8810, the induced by travoprost. The present study further demonstrates concentration–response curve of travoprost tended to be abol- that when the vasoconstrictive effects of latanoprost and tra- ished. voprost are compared, it clearly appears that latanoprost is a stronger vasoconstrictor than travoprost in isolated porcine ciliary arteries, at least at high concentrations that might not DISCUSSION necessarily be clinically relevant. The results of the present study indicate that U46619 (TP- The results of the present study are in agreement with different publications indicating that U46619 and PGF2␣ are receptor agonist), PGF2␣ (FP-receptor agonist), latanoprost free 2,3 acid, and travoprost free acid have vasoconstrictive properties strong vasoconstrictors in vitro. Such studies have shown in isolated porcine ciliary arteries. They also show that SQ that high concentrations of PGF2␣ can for example cause 29548 (TP-receptor antagonist) and AL-8810 (FP-receptor an- vasoconstriction in rabbit posterior ciliary arteries, bovine ret- 8–10 tagonist) had an inhibitory effect on the contractions evoked inal arteries, and monkey ophthalmic arteries. The present by these drugs. study is also in agreement with reports indicating that latano- Among the four vasoconstrictors tested, U46619 (throm- prost can act as a vasoconstrictor.2,11 Indeed, Astin2 reported boxane A agonist) appeared to be the most efficient vasocon- similar contractions in bovine ciliary artery preparations with 2 ␮ strictor, with a PD50 that was more than twofold lower than latanoprost at concentrations above 1 M. These contractions the PD50 of PGF2␣. In addition, U46619 was able to produce were inhibited by the TP-receptor antagonist GR 32191B. highly potent contractions that were almost equivalent to It has to be mentioned that, in humans, about 2 hours after

those observed with PGF2␣, which was, among the four pro- a single topical ocular instillation of latanoprost eye drop solu- stanoids tested, the one that induced the most potent contrac- tion, the maximal concentration of latanoprost measured in the

Downloaded from tvst.arvojournals.org on 09/28/2021 298 Vysniauskiene et al. IOVS, January 2006, Vol. 47, No. 1

aqueous humor was ϳ0.1 ␮M. At this concentration, as shown Acknowledgments in the present study, this drug was able to induce significant contractions in isolated porcine ciliary arteries (first peak con- The authors thank Julie Crider (Alcon Laboratories) for providing traction).12 However, the use of radioactively labeled micro- critical, fair, and constructive comments regarding the manuscript. spheres to assess ocular blood flow showed no major circula- tory changes in the posterior pole of monkeys, feline, and References rabbit eyes after a single topical administration of 6 to 10 ␮g latanoprost (corresponding to four times the topical dose ap- 1. Coleman RA, Smith WL, Narumiya S. VIII. International Union of plied in humans).13 Nevertheless, this observation does not Pharmacology Classification of prostanoid receptors: properties, exclude that this type of drug could lead, in some predisposed distribution, and structure of the receptors and their subtypes. Pharmacol Rew. 1994;46:205–229. patients, to a vasoconstriction. Indeed, it has been reported 14 2. Astin M. Effects of , F2alpha, and latanoprost acid that latanoprost could exacerbate angina pectoris. on isolated ocular blood vessels in vitro. J Ocul Pharmacol Ther. Because PGF2␣, which is considered to be a potent FP- 1998;14:119–128. 1 receptor agonist, shares some close chemical structural simi- 3. Pellanda N, Flammer J, Haefliger IO. L-NAME and U 46619- induced larities with latanoprost and travoprost, one would expect that contractions in isolated porcine ciliary arteries versus vortex veins. the contractions induced by these drugs would be specifically Klin Monatsbl Augenheilkd. 2001;218:366–369. inhibited by an FP-receptor antagonist such as AL-8810 and not 4. Hellberg MR, Salley VL, Mclaughlin MA, et al. Preclinical efficacy of by a TP-receptor antagonist such as SQ 29548. Similarly, travoprost, a potent and selective FP ago- 1 nist. J Ocul Pharmacol Ther. 2001;17:421–432. U46619, considered to be a potent TP-receptor agonist, 5. Haefliger IO, Flammer J, Lu¨tscher T. Nitric oxide and endothelin-1 should be inhibited by SQ 29548 (TP-receptor antagonist) and are important regulators of human ophthalmic artery. Invest Oph- not by AL-8810 (FP-receptor antagonist). The present study thalmol Vis Sci. 1992;33:2340–2343. illustrated that this is definitely not the case regarding the 6. Dettmann E, Vysniauskiene I, Wu R, Flammer J, Haefliger IO.

vasoconstrictive response of PGF2␣ and U46619 in isolated Adrenomedullin-induced endothelium-dependent relaxation in porcine ciliary arteries. Indeed, both AL-8810 (FP-receptor an- porcine ciliary arteries. Invest Ophthalmol Vis Sci. 2003;44:3961– tagonist) and SQ 29548 (TP-receptor antagonist) were able to 3966. 7. Zhu P, Beny JL, Flammer J, Lu¨tscher T, Haefliger IO. Relaxation by inhibit the contraction induced by the FP-receptor agonist ϩ bradykinin in porcine ciliary artery: role of nitric oxide and K - PGF2␣ as well as by the TP-receptor agonist U46619. channels. Invest Ophthalmol Vis Sci. 1997;38:1761–1767. Different conditions could lead to such observations. For 8. Hayashi E, Yoshitomi T, Ishikawa H, Hayashi R, Shimizu K. Effects example, although some agents may activate one receptor of isopropyl on rabbit ciliary artery. Jpn J Ophthal- preferentially, they may exhibit some modest affinity for other mol. 2000;44:214–220. prostaglandin-receptor subtypes. In the case of the FP-receptor 9. Nielsen P, Nyborg N. Calcium antagonist-induced relaxation of the antagonist AL-8810 it has been shown that it can produce a prostaglandin-F2␣ response of isolated calf retinal resistance arter- slight inhibition of TP-receptor-mediated phosphoinositide ies. Exp Eye Res. 1989;48:329–335. turnover in human nonpigmented ciliary epithelial cells at a 10. Ohkubo B, Chiba S. Vascular reactivities of simian ophthalmic and ciliary arteries. Curr Eye Res. 1987;6:1197–1204. concentration of 100 ␮M.15 Another condition that may affect 11. Brogiolo G, Flammer J, Haefliger IO. Latanoprost is a vasoconstric- the vasoactive response of these ocular blood vessels is the tor in isolated porcine ciliary arteries. Klin Monatsbl Augenheilkd. possible existence of a heterogeneous population of TP recep- 2001;218:373–375. tors.16 For example, it is known that the human TP receptor 12. Sjoquist B, Stjernschantz J. Ocular and systemic pharmacokinetics isoforms TP␣ and TP␤ show different sensitivities to inhibition of latanoprost in humans. Surv Ophthalmol. 2002;47(suppl 1):S6– by endogenous and exogenous mediators.17 In any event, one S12. has to keep in mind that, since the specificity of the TP- and 13. Stjernschantz J, Selen G, Astin M, Resul B. Microvascular effects of FP-receptor antagonists tested can always be questioned, no selective prostaglandin analogues in the eye with special reference to latanoprost and glaucoma treatment. Prog Retin Eye Res. 2000; definite conclusion can be drawn regarding the possible stim- 19:459–496. ulation of different receptors by the prostanoids tested in the 14. Mitra M, Chang B, James T. Drug points: exacerbation of angina present study. Nevertheless, it clearly appears that SQ 29548 associated with latanoprost. BMJ. 2001;323:783. and AL-8810 could strongly inhibit the contractions induced by 15. Griffin BW, Klimko P, Crider JY, Sharif NA. AL-8810: a novel all prostanoids studied in the present trial. prostaglandin F2␣ analog with selective antagonist effects at the In conclusion, in isolated porcine ciliary arteries and at the prostaglandin F2␣ (FP) receptor. J Pharm Exp Ther. 1999;290: concentrations tested, the maximal contractions evoked by 1278–1284. U46619, PGF , and latanoprost free acid were much higher 16. Krauss AH, Woodward DF, Burk RM, et al. Pharmacological evi- 2␣ dence for heterogeneity: implication for than those evoked by travoprost free acid, and the contractions the eye. J Ocul Pharmacol Ther. 1997;13:303–312. evoked by any of the four drugs were inhibited either by the 17. Walsh M-T, Kinsella BT. Regulation of the human prostanoid TP␣

TP-receptor antagonist SQ 29548 or the FP-receptor antagonist and TP␤ receptor isoforms mediated through activation of the EP1 AL-8810. and IP receptors. Br J Pharmacol. 2000;131:601–609.

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