Evidence for the Involvement of Cannabinoid CB₁ Receptors in the Bimatoprost-Induced Contractions on the Human Isolated Ciliary Muscle

Abstract

purpose. To evaluate the bimatoprost effects in the isolated human ciliary muscle and to assess how these response can be modulated by AL8810 and SR141716A.

methods. In a myograph system (isometric force measurement), ciliary muscles were exposed cumulatively to PGF_2α, latanoprost, travoprost, bimatoprost, and anandamide (0.1 nM-10 μM). Experiments were also conducted in the presence of AL8810 (FP receptor antagonist; 100 nM) or SR141716A (CB₁ receptor antagonist; 10–100 nM). Contractions were expressed as the percentage of 10 μM carbachol-induced contractions.

results. In quiescent tissues, concentration-response curves for bimatoprost, anandamide, PGF_2α, latanoprost, and travoprost were constructed. Bimatoprost showed an important contractile effect on isolated human ciliary muscle strips (E_max = 125% ± 0.09%); the maximal effect was higher than that obtained with carbachol. Contractions were inhibited by SR141716A (10 and 100 nM) and AL8810 (100 nM).

conclusions. This study showed evidence of direct interaction of bimatoprost with the contractility of the human ciliary muscle through interaction with cannabinoid CB₁ receptor and prostanoid FP receptors.

Glaucoma is a progressive optic neuropathy characterized by the degeneration of retinal ganglion cells and their axons, with resultant visual field defects and loss of vision. Elevated intraocular pressure (IOP) is a well-known risk factor for the development and progression of glaucoma. However, the biochemical basis of elevated IOP and factors contributing to disease progression remain to be fully elucidated. Recent reports show that IOP lowering delays disease progression for patients with normotensive and ocular hypertensive glaucoma, ¹ and the current antiglaucomatous therapy uses ocular hypotensive agents such as the synthetic analogues of prostaglandin F_2α (PGF_2α), including latanoprost and travoprost, and a compound chemically related to prostamide F, bimatoprost. These drugs lower IOP through different mechanisms of action. 

The prostaglandin FP receptor agonists latanoprost and travoprost have been shown to lower IOP by increasing the uveoscleral outflow, ^{2 3 4 5} whereas bimatoprost has been reported to increase aqueous humor flow through uveoscleral and conventional outflow pathways. ⁶ The increased uveoscleral outflow, which appears to be caused by an enlargement of uveoscleral outflow routes and to morphologic changes in the trabecular meshwork causing tissue remodeling in ciliary muscle areas, is thought to be the result of the formation of open spaces in the ciliary muscle. This had first been demonstrated in primate eyes treated with PGF_2α isopropyl ester ⁷ and, recently, with latanoprost. ⁸  

Bimatoprost, a prostamide derivative, is a potent and efficacious ocular hypotensive drug. ^{9 10 11 12} It effectively reduces IOP in patients who do not respond to latanoprost in the form of an isopropyl ester prodrug. ¹³ This clinical evidence suggests that bimatoprost and prostanoid FP receptor agonists stimulate different receptor populations. To date, preclinical pharmacologic evidence demonstrates that bimatoprost and prostanoid FP receptor agonists have different activity profiles. ^{14 15 16} The pharmacology of bimatoprost has been ascribed to interaction with a novel population of receptors distinct from known prostanoid receptors. ¹⁵ Bimatoprost is a unique and potent prostamide agonist ^{14 15} and, in contrast with the free acid forms of latanoprost and travoprost, has only weak interaction with FP receptors. ^{14 15 17 18 19} The prostamides are biologically active compounds derived from anandamide, a fatty acid amide precursor that serves as a substrate for the COX-2 enzyme, and are potent ocular hypotensive agents. 

The principal aim of this study was to investigate the effects of bimatoprost in the isolated human ciliary muscle and to compare them with those of anandamide, ²⁰ an endogenous cannabinoid receptor ligand, and of prostanoid FP receptor agonists such as PGF_2α, latanoprost, and travoprost. Additionally, we have evaluated how these responses can be modulated by AL8810 ²¹ (FP-receptor antagonist) and SR141716A ²² (CB₁-receptor antagonist). 

Methods

Human Ciliary Muscle Preparations

Five human eyes were removed during surgery after trauma in adult living patients and immediately placed at 4°C in a modified Krebs solution gassed with 95% O₂/5% CO₂ to give a pH of 7.4 for transport to the laboratory. Krebs solution had the following composition: 136.8 mM NaCl, 5.4 mM KCl, 2.7 mM CaCl₂, 7 mM MgSO₄ × 0.8 mM H₂O, 5 mM glucose, and 6 mM HEPES. Informed written consent was obtained from all patients according to the Declaration of Helsinki. After the vitreous and lens were removed, the ciliary muscle was quickly isolated under a binocular microscope (Nikon, Tokyo, Japan) and dissected from the scleral spur, lens, and choroids, as previously described, ^{23 24} to be used for functional experiments. Indomethacin (1 μM), a cyclooxygenase inhibitor, was present for the duration of the experiments to prevent the formation of the endogenous prostaglandins. 

Functional Experiment

Ciliary muscle strips approximately 4 mm in width were prepared and suspended in a 20-mL organ bath containing Krebs solution (37°C)continuously aerated with a mixture of 95% O₂ and 5% CO₂. The time from delivery until the tissue was set up in the organ bath was approximately 30 minutes. The preparation was linked with a silk thread to an isometric strain gauge under a constant load of 3 mN. Contractile activity was measured with an appropriate transducer (Fort 10; WPI, Sarasota, FL) connected to a recorder (PowerLab 4/20; ADInstruments, Castle Hill, NSW, Australia). Tissues were allowed to equilibrate for at least 90 minutes, during which the Krebs solution was changed several times. Before any drug addition, tissues were challenged with carbachol (10 μM) at least once to assess the functional state of each preparation. After the carbachol challenge, tissues were washed, and the preload was readjusted just before the onset of the actual study. When more than one concentration of the antagonist was added, a control response to the agonist was recorded to ensure that complete recovery from the block had taken place. Contraction was expressed as a percentage of the reference response elicited by 10 μM carbachol, which was obtained at the end of each concentration-response curve. 

Chemicals

Drugs and reagents used in the present study were obtained as follows: anandamide (in Tocrisolve 100), CP55,940 ((–)-cis-3-[2-hydroxy-4-(1,1-dimethylheptyl) phenyl]-trans- 4-(3-hydroxypropyl) cyclohexanol), AM251 (N-(piperidin-1-yl)-5-(4-iodo phenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrozole-3-carboxamide), and U73122 (1[6-[[(17β)-3-methoxyestra -1,3,5(10)-trien-17-yl] amino]hexyl]-1H-pyrrole-2,5-dione) were supplied by Tocris Bioscience (Bristol, UK); bimatoprost (17-phenyl trinor prostaglandin F_2α ethylamide), latanoprost (17-phenyl-13,14-dihydro trinor prostaglandin F_2α isopropyl ester), travoprost ((+)-fluprostenol isopropyl ester), and AL8810 (9α,15R-dihydroxy-11β-fluoro-15-(2,3-dihydro-1H-inden-2-yl)-16,17,18,19,20-pentanor-prosta-5Z,13E-dien-1-oic acid) were obtained from Cayman (Ann Arbor, MI); SR141716A was a gift from Sanofi Recherche (Montpellier, France); and carbachol hydrochloride and pertussis toxin were from Sigma Aldrich (St. Louis, MO). All other reagents were of analytical grade. Stock solutions (10 mM) of each drug were prepared in dimethyl sulfoxide or ethanol, as appropriate, and subsequent concentrations were diluted in the Krebs solution. The final bath concentration of dimethyl sulfoxide or ethanol was 0.1%. Vehicle control studies indicated that this final concentration of solvent had no effect on the tonus or on the contractility of preparation. 

Data and Statistical Analyses

All data in the text are expressed as mean ± SE, and n refers to the number of preparations. The concentration of contraction, giving a half-maximal response (EC₅₀), was obtained by fitting four-parameter sigmoidal concentration-response curves (Prism version 3.0; GraphPad Software, San Diego, CA) and was reported as its negative logarithm, pEC₅₀. E _max refers to the maximal response achieved. Statistical analysis was performed by analysis of variance (ANOVA) followed by Bonferroni post hoc test (Prism version 3.0; GraphPad Software). Student’s t-test for paired data was used when appropriate. P < 0.05 was considered statistically significant. 

Results

Effects of Prostanoids and Prostamides in Quiescent Tissues

In quiescent tissue, bimatoprost, latanoprost, travoprost, PGF_2α, and anandamide evoked concentration-dependent contractions (Fig. 1) . The contraction induced by bimatoprost was significantly (P < 0.001) more pronounced (E_max = 125.3% ± 4.52% of carbachol_max) than those induced by PGF_2α (E_max = 76.9% ± 5.21% of carbachol_max), latanoprost (E_max = 52.1% ± 2.40% of carbachol_max), travoprost (E_max = 53.2% ± 1.76% of carbachol_max), and anandamide (E_max = 56.3% ± 1.20% of carbachol_max). 

Role of Prostaglandin FP Receptors in Ciliary Muscle Contraction Induced by Bimatoprost and FP Receptor Agonists

To evaluate the contribution of prostaglandin FP receptors to the human ciliary muscle contraction in response to bimatoprost, concentration-response curves were examined in the presence of competitive prostaglandin FP receptor antagonist AL8810. ^{18 19 21} Inhibition of prostaglandin FP receptors by AL8810 (100 nM) evoked a small but significant shift to the right of concentration-response curves, giving the pEC₅₀ values of 6.97 ± 0.21 (bimatoprost control: pEC₅₀ = 7.85 ± 0.11; Fig. 2A ). 

The effects of bimatoprost were compared with those of other prostaglandin FP receptor agonists (PGF_2α, latanoprost and travoprost) in isolated human ciliary muscle. PGF_2α produced a concentration-dependent contractile response on human ciliary muscle (pEC₅₀ = 7.06 ± 0.14; Fig. 2B ). This effect was significantly (P < 0.001) inhibited by AL8810 (100 nM; pEC₅₀ = 5.42 ± 0.17) by shifting to the right of concentration-response curves. 

Latanoprost, a PGF_2α analogue, induced a concentration-dependent contraction on human ciliary muscle (pEC₅₀ = 7.08 ± 0.10; Fig. 2C ). Preincubation of ciliary muscle strips with AL8810 (100 nM) revealed an inhibitor effect on the latanoprost-induced contraction by significantly (P < 0.005) shifting to the right of concentration-response curves (pEC₅₀ = 6.43 ± 0.04). 

Travoprost, a PGF_2α analogue, also contracted human ciliary muscle in a concentration-dependent manner (pEC₅₀ = 7.15 ± 0.07; Fig. 2D ). Preincubation of ciliary muscle strips with AL8810 (100 nM) shifted to the of right concentration-response curves to travoprost (pEC₅₀ = 5.71 ± 0.14). 

Contraction of Ciliary Muscle to Bimatoprost

The effects of bimatoprost on the human ciliary muscle strips are illustrated in Figure 1 . The pEC₅₀ value for bimatoprost 7.79 ± 0.08 was similar to its reported potency in the prostamide-sensitive cat iris sphincter ¹² ; the maximal effect (E_max; 125.9% ± 4.46%) was obtained with 10 μM bimatoprost. Interestingly, the bimatoprost-induced contraction was significantly (P < 0.001) inhibited by SR141716A ²² (10 and 100 nM; bimatoprost, pEC₅₀ 7.30 ± 0.07 and 6.81 ± 0.16; Fig. 3 ). The same effect was obtained with an alternative CB₁ receptor-selective antagonist AM251 ²⁵ (100 nM; P < 0.001; Fig. 4 ). The combined effects of AL8810 and SR1417161A on concentration-response curves of bimatoprost were investigated. Preincubation for 20 minutes with AL8810 (100 nM) and SR141716A (100 nM) resulted in complete inhibition of bimatoprost-induced contractions (Fig. 5) . 

Contraction of Ciliary Muscle to Cannabinoid Agonists

Anandamide contracted the human ciliary muscle in a concentration-dependent manner, giving an pEC₅₀ of 7.73 ± 0.06; E_max was produced with 10 μM anandamide (57% of carbachol_max; Fig. 6 ). The contractile effect of anandamide was mimicked by CP55940, a synthetic selective CB₁ receptor agonist. Maximum response was estimated with 10 μM CP55940 (68% of carbachol_max; Fig. 4 ). SR141716A (10 and 100 nM), a selective cannabinoid CB₁ receptor antagonist, evoked a significant rightward shift (P < 0.001) of the anandamide concentration-response curves, and the pEC₅₀ values of 7.17 ± 0.18 and 5.84 ± 0.07 were respectively calculated. AM251 (100 nM), another CB₁ receptor antagonist, inhibited contractions to anandamide and CP55940 (P < 0.01; Fig. 4 ). The prostaglandin FP receptor antagonist AL8810 (100 nM) had no effect on anandamide-evoked contractions (Fig. 7) . 

Involvement of G_i/o and G_q Proteins to Bimatoprost-Induced Contractile Response

Preincubation with the G_i/o protein inhibitor pertussis toxin (500 ng/mL) for 30 minutes abolished the contraction induced by bimatoprost to the high concentrations used (10 μM); subsequently, the preparations responded to latanoprost (10 μM; Fig. 8A ) or travoprost. We explored the involvement of phospholipase C, a pathway typically associated with G_q protein on the bimatoprost-induced contraction of the human ciliary muscle. Bimatoprost (10 μM) was unable to induce a contraction regardless of whether the tissue was pretreated for 30 minutes with U73122 (10 nM), a phospholipase C inhibitor (Fig. 8B) ; the subsequent addition of latanoprost or travoprost (both 10 μM) failed to evoke a contractile response by confirming the involvement of G_q protein. 

Discussion

The topical administration of prostaglandin FP agonists reduced IOP in humans and nonhuman primates by increasing the uveoscleral outflow of aqueous humor. Bimatoprost is a potent ocular hypotensive agent given topically, though it has been suggested that it fundamentally differs from prostaglandin analogues because it lowers IOP through mechanisms that appear to be independent of known receptor signaling. ^{14 15 16}  

The present study aimed to characterize CB₁ receptor involvement in the bimatoprost-induced contractile response on the human ciliary muscle and to compare its effects with those of FP and CB₁ receptor agonists. Bimatoprost elicited potent contractile effects (EC₅₀ = 16.0 ± 0.21 nM) in the human isolated ciliary muscle, results similar to those reported for the cat isolated iris sphincter ¹⁴ (EC₅₀ = 34.0 nM), the cat isolated peripheral lung parenchyma ¹⁵ (EC₅₀ = 35–55 nM), and the rabbit isolated uterus ¹² (EC₅₀ = 28.1 nM). The potent contractile response of bimatoprost was surprisingly antagonized by the selective cannabinoid CB₁ receptor antagonist SR141716A in a concentration-dependent manner, causing a rightward displacement of the concentration-response curves of bimatoprost. In addition, bimatoprost was assayed in the presence of another selective CB₁ receptor antagonist, AM251, ²⁵ which produced an inhibitory effect similar in potency to SR141716A. These findings suggest that the contractile response of bimatoprost has been achieved through activation of CB₁ receptors, though it is at variance with previous studies reporting that bimatoprost has only weak affinity (EC₅₀ values > 10 μM) for CB₁ receptors. ¹⁴ The reason for such controversy is unclear, but it may relate to differences in experimental conditions. Many studies have established the presence of CB₁ receptor on ciliary muscle of different animal species, including humans. ^{26 27 28} Here we have investigated the ability of the cannabinoid agonists anandamide and CP55940 to yield contractions in the human ciliary muscle that were inhibited in presence of the selective CB₁ receptor antagonists SR141716A and AM251. This result provides pharmacological evidence for the presence of the functional CB₁ receptor in the human ciliary muscle, in accordance with our previous study showing evidence of the presence of CB₁ receptors in the bovine ciliary muscle whose activation by cannabinoid agonists contracted the ciliary muscle. ¹⁸ However, cannabinoid agonist contractions were weaker than those of bimatoprost. It is known that cannabinoid CB₁ receptors exert their biological functions by interacting with G_i/G_o proteins to inhibit adenylate cyclase. ^{29 30} For this reason, the contractile effects of bimatoprost have been investigated in presence of pertussis toxin, which blocked the contractile response to bimatoprost without affecting those of other prostanoid agonists such as latanoprost and travoprost, strongly supporting the hypothesis that the contractions evoked by bimatoprost in the human ciliary muscle involved G_i/o protein. 

To explain the potent and efficacious contractile action of bimatoprost, we have further investigated the role of FP receptors in the modulation of contraction. Indeed, the pharmacological actions of bimatoprost through the FP prostaglandin receptor have been the subject of some debate. Woodward et al. ^{14 15} report that bimatoprost has no observed significant interaction with prostanoid receptor, whereas Sharif et al. ^{17 18 19} suggest the opposing conclusion that the agonist actions of bimatoprost were mediated by prostanoid FP receptor. Recently, it has shown that the bimatoprost-induced effects were antagonized by AGN 204396. ³¹ This compound is not yet available and in our experiments has not been tested. Our findings clearly demonstrated that AL8810 effectively inhibited the bimatoprost-evoked contraction, behaving as a competitive inhibitor and shifting to the right the concentration-response curves and supporting the idea that bimatoprost activity may also be mediated through FP receptors. When FP receptor agonists PGF_2α, latanoprost, and travoprost were evaluated in our experiment, the constrictor potency obtained was lower than that of bimatoprost for all compounds used, and efficacy was submaximal (approximately 75%, 52%, and 53% of the maximal response of carbachol, respectively). As expected, these responses were effectively antagonized by AL8810. Furthermore, because it has been thought that bimatoprost induces also a prostanoid FP-dependent contraction in this preparation, we investigated the action of bimatoprost in the presence of the phospholipase C inhibitor U73122. Both bimatoprost and latanoprost/travoprost failed to produce contractions, suggesting that the stimulation of phospholipase C plays a key role to trigger the contractile effect. Intriguingly, when the selective CB₁ and FP receptor antagonists SR141716A and AL8810, respectively, were preincubated together, bimatoprost-induced contractions on the human ciliary muscle were completely inhibited, demonstrating a synergic action. 

Taken together, the data in this study showed that bimatoprost is a stronger agonist than prostaglandin analogues or cannabinoid agonists in stimulating contractile activity of the human ciliary muscle; we think that this effect could be attributed to the stimulation of different receptor systems on the ciliary smooth muscle. This evidence further contributes to our understanding of the functional and pharmacological roles of bimatoprost in ocular tissues involved in IOP modulation. 

In conclusion, this investigation demonstrates that bimatoprost has a strong contractile action on the human ciliary muscle, stimulating functional responses through cannabinoid CB₁ and prostanoid FP receptors. In particular, we have demonstrated that bimatoprost-induced contractions were inhibited by selective competitive CB₁ and FP receptor antagonists and that, when the two antagonists were used together, the response was completely abolished. 

 Supported by “Ministero dell’Università e della ricerca scientifica e tecnologica” Cofinanziamento 2005.

 Submitted for publication August 1, 2006; revised February 9 and April 16, 2007; accepted June 8, 2007.

 Disclosure: M.R. Romano, None; M.D. Lograno, None

 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.

 

Figure 1.

 

Comparison of the effects of bimatoprost (□), PGF_2α (▴), latanoprost (▾), travoprost (⋄), and anandamide (AEA) (○) on the human isolated ciliary muscle. Values are mean ± SE of percentage of response to 10 μM carbachol; n = 4.

Figure 2.

 

Effects of the prostanoid FP receptor antagonist AL8810 100 nM on contractile responses to bimatoprost, PGF2α, latanoprost, and travoprost in the human ciliary muscle. Data are given as mean ± SE; n = 4.

Figure 3.

 

Effects of the cannabinoid CB₁ receptor antagonist SR141716A (10 and 100 nM) on contractions evoked by bimatoprost in the human ciliary muscle. Data are given as mean ± SE; n = 4.

Figure 4.

 

Effects of the cannabinoid CB₁ receptor antagonist AM 251 (100 nM) on contraction to bimatoprost, anandamide (AEA), and CP55940 in the human ciliary muscle. Data are given as mean ± SE; n = 3. *P < 0.01; **P < 0.001.

Figure 5.

 

Effect of coadministration of AL8810 (100 nM) and SR141716A (100 nM) on contraction to bimatoprost in human ciliary muscle. Data are given as mean ± SE; n = 3.

Figure 6.

 

Effects of the cannabinoid CB₁ receptor antagonist SR141716A (10 and 100 nM) on contractions evoked by anandamide in the human ciliary muscle. Data are given as mean ± SE; n = 4.

Figure 7.

 

Effect of AL8810 (100 nM) on contractions induced by anandamide in the human ciliary muscle. Data are given as mean ± SE; n = 4.

Figure 8.

 

Traces show one experiment in which pretreatment (30 minutes) with pertussis toxin (400 ng/mL) blocked the bimatoprost (10 μM) response that subsequently respond to latanoprost (10 μM) (A). Pretreatment (30 minutes) with U73122 blocked the bimatoprost (10 μM) and latanoprost (10 μM) responses (B).