Experimental Eye Research 93 (2011) 256e264
Contents lists available at ScienceDirect
Experimental Eye Research journal homepage: www.elsevier.com/locate/yexer
Effect of PF-04217329 a prodrug of a selective prostaglandin EP2 agonist on intraocular pressure in preclinical models of glaucoma Ganesh Prasanna a, *, Samantha Carreiro a, Scott Anderson a, Hovhannes Gukasyan b, Soisurin Sartnurak b, Husam Younis c, David Gale d, Cathie Xiang d, Peter Wells e, Dac Dinh e, Chau Almaden e, Jay Fortner f, Carol Toris g, Michael Niesman a, Jennifer Lafontaine h, Achim Krauss a a
Department of Ocular Biology, Pfizer Global R & D, San Diego, CA 92121, USA Department of Research Formulations, Pfizer Global R & D, San Diego, CA 92121, USA Department of Drug Safety, Pfizer Global R & D, San Diego, CA 92121, USA d Department of Pharmacokinetics and Drug Metabolism, Pfizer Global R & D, San Diego, CA 92121, USA e Department of Biochemical Pharmacology, Pfizer Global R & D, San Diego, CA 92121, USA f Department of Comparative Medicine, Pfizer Global R & D, San Diego, CA 92121, USA g Department of Ophthalmology, University of Nebraska Medical Center, Omaha, NE 68198, USA h Department of Chemistry, Pfizer Global R & D, San Diego, CA 92121, USA b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 November 2010 Accepted in revised form 22 February 2011 Available online 3 March 2011
Better control of intraocular pressure (IOP) is the most effective way to preserve visual field function in glaucomatous patients. While prostaglandin FP analogs are leading the therapeutic intervention for glaucoma, new target classes also are being identified with new lead compounds being developed for IOP reduction. One target class currently being investigated includes the prostaglandin EP receptor agonists. Recently PF-04217329 (Taprenepag isopropyl), a prodrug of CP-544326 (active acid metabolite), a potent and selective EP2 receptor agonist, was successfully evaluated for its ocular hypotensive activity in a clinical study involving patients with primary open angle glaucoma. In the current manuscript, the preclinical attributes of CP-544326 and PF-0421329 have been described. CP-544326 was found to be a potent and selective EP2 agonist (IC50 ¼ 10 nM; EC50 ¼ 2.8 nM) whose corneal permeability and ocular bioavailability were significantly increased when the compound was dosed as the isopropyl ester prodrug, PF-04217329. Topical ocular dosing of PF-04217329 was well tolerated in preclinical species and caused an elevation of cAMP in aqueous humor/iris-ciliary body indicative of in vivo EP2 target receptor activation. Topical ocular dosing of PF-04217329 resulted in ocular exposure of CP-544326 at levels greater than the EC50 for the EP2 receptor. PF-04217329 when dosed once daily caused between 30 and 50% IOP reduction in single day studies in normotensive Dutch-belted rabbits, normotensive dogs, and laser-induced ocular hypertensive cynomolgus monkeys and 20e40% IOP reduction in multiple day studies compared to vehicle-dosed eyes. IOP reduction was sustained from 6 h through 24 h following a single topical dose. In conclusion, preclinical data generated thus far appear to support the clinical development of PF-04217329 as a novel compound for the treatment of glaucoma. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: EP2 agonist glaucoma intraocular pressure pharmacology pharmacokinetics EP2 receptor cynomolgus monkeys dogs rabbits
1. Introduction It is well known based on the AGIS findings (AGIS Investigators, 2000) that low intraocular pressure is associated with reduced progression of visual field defect in glaucomatous patients,
* Corresponding author. Present address. Glaucoma Research e In Vivo Pharmacology (MC: R9-11), Alcon Research Ltd., 6201 South Freeway, Fort Worth, TX 76134, USA. Tel.: þ1 858 603 9551. E-mail address:
[email protected] (G. Prasanna). 0014-4835/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.exer.2011.02.015
indicating that lower IOP (i.e. 85% sequence homology across species (nonhuman primate, rabbit, dog, mouse, and rat) based on nucleotide and protein sequence alignment (Breyer et al., 2001). The homology of human EP2 with human EP1, EP3, and EP4 receptor is 37%, 34%, and 38% respectively (Breyer et al., 2001; Regan, 2003). EP2 receptors are widely expressed in human anterior chamber especially in the iris, ciliary epithelium, ciliary muscle, trabecular meshwork (TM) and in the neural retina (Schlotzer-Schrehardt et al., 2002; Biswas et al., 2004). Activation of EP2 receptor results in the Gs-mediated stimulation of adenylate cyclase resulting in an increase in intracellular cAMP (Regan, 2003). Elevation of cAMP results in the activation of protein kinase A which in turn can stimulate various downstream signaling pathways (Regan, 2003). Involvement of the EP2 receptor in IOP regulation has been demonstrated in several in vivo studies. It has been recently reported that EP2 knock-out mice do not exhibit IOP lowering when dosed with ONO-AE1-259 (Ki of 3 nM), an EP2 agonist, whereas wild-type mice showed a 21% IOP reduction at 2.5 h post-dose (Saeki et al., 2009). Selective EP2 agonists including butaprost and AH13205 which have better adverse effect profiles, lower IOP by 20e30% in normotensive dogs as well as in normotensive and laserinduced ocular hypertensive non-human primates following topical ocular dosing (Woodward et al., 1995; Nilsson et al., 2006). PF-04217329 is an isopropyl ester prodrug of CP-544326, a potent and selective EP2 receptor agonist which was identified as a clinical candidate by Discovery Research at Pfizer Global R&D. PF04217329 was recently evaluated in a clinical study in patients with glaucoma (Schachar et al., 2010). PF-04217329 (Taprenepag isopropyl) at doses 0.0025e0.03% was reported to cause significant IOP lowering (3.0e6.6 mmHg; Least Squares Mean Differences) compared to vehicle-treatment in patients with primary open angle glaucoma (POAG) and ocular hypertension following QD dosing up to 14 days (Schachar et al., 2010). The largest IOP reductions were observed at 8AM (i.e. 24 h post-dose) particularly with 0.01% dose (Schachar et al., 2010). Presently, we have described the preclinical attributes of PF-04217329 which support its clinical candidacy. Specifically the ocular hypotensive effects and ocular bioavailability of PF-04217329 have been determined in preclinical models including rabbits, dogs and laser-induced ocular hypertensive monkeys. 2. Materials and methods 2.1. EP and FP receptor binding assays The receptor binding assay for prostaglandin EP receptors was performed as previously described (Prasanna et al., 2009). Briefly, human EP1, EP2, EP3 and EP4 as well as rat EP2 (rEP2) receptors were
257
recombinantly expressed in human embryonic kidney-293 (HEK293) cells. Membranes were prepared from these cells and used in radioligand binding assays. In the human EP receptor (hEP) radioligand binding assay, CP-544326 competed with [3H]PGE2 ligand (0.0015 mM final concentration) for binding to these receptors (rEP2 ¼ 100 mg/well, hEP1 ¼ 200 mg/well, hEP3 ¼ 2 mg/well, hEP2 and hEP4 ¼ 15 mg/well total protein concentration). Affinity of the compound binding to hEP1, hEP2, hEP3 or hEP4 receptor was measured by displacement of the radiolabeled ligand in the presence of varying doses of CP-544326. Data were collected using the TopCount instrument, a 96/384 well microplate scintillation/luminescence counter (Perkin Elmer) and IC50 were calculated using sigmoidal dose response equation with variable slope (GraphPad Prism). Receptor binding assays were repeated twice and data are reported as an average of the two independent studies. Butaprost (free acid) and PGE2 used as agonists in the cAMP assays were purchased from Cayman Chemical (Ann Arbor, MI). In the FP Radioligand Binding assay, CP-544326 competes with 3 H PGF2a ligand (0.003 mM final concentration) for binding to hFP receptor (210 mg/well total protein amount). Affinity of the compound binding to hFP receptor can be measured by displacement of the radiolabeled ligand. 2.2. Binding assessments of CP-544326 to other G-protein coupled receptors (GPCRs) Binding of CP-544326 to a panel of 37 human GPCR including 15 receptors known to signal via intracellular cAMP elevation was evaluated at CEREP (Redmond, WA). CP-544326 was tested at 10 mM final concentration in these binding assays. 2.3. In vitro cAMP assay The in vitro cAMP assay was performed as previously described (Prasanna et al., 2009). EP2 receptor activation results in an increase in intracellular adenosine 30 ,5-cyclic monophosphate (cAMP) levels. In the cell-based assay, hEP2-HEK293 cells (5000 cells/well) and rEP2-HEK293 cells (4000 cells/well) were harvested and suspended into a 96-well plate (n ¼ 2 independent experiments). CP544326 (0.01, 0.1, 1, 10, 100, and 1000 nM) was incubated for different times with hEP2 or rEP2 cells leading to cAMP production. This cAMP second messenger was then measured using HitHunter cAMP XSþ assay (DiscoveRx, Fremont, CA). In this assay, free cAMP from cell lysate and cAMP conjugated to a fragment of b-galactosidase compete for antibody binding. The cAMP conjugated to the enzyme fragment which is not bound by the antibody, complements b-galactosidase enzyme acceptor and forms a fully active enzyme. A chemiluminescent signal from b-galactosidase substrate hydrolysis detected by TopCount instrument is directly proportional to the amount of free cAMP produced by the cells following exposure to agonist. EC50 data are reported as the average of two independent experiments. 2.4. Intracellular calcium measurement for FP receptor activity The FP Calcium Flux assay uses a florescent dye that binds to intracellular Ca2þ ([Ca2þ]i) released upon activation of FP receptor via Gq signaling to monitor functional activity. Stably expressed hFP receptors in HEK293 cell line was used for assessing the mobilization of [Ca2þ]i following treatment with PGF2a (FP receptor agonist) and CP-544326 (various doses). hFP-HEK cells (60,000 cells/well) were incubated with a dye solution (Calcium 4 Assay Kit from Molecular Devices/MDS) for 1 h at 37 C and 5% CO2. Compound addition and plate reading were both done on the
258
G. Prasanna et al. / Experimental Eye Research 93 (2011) 256e264
FlexStation, which measures changes to [Ca2þ]i via changes in fluorescent dye’s property following binding to intracellular Ca2þ. 2.5. Ocular permeability in rabbit corneal tissues and corneal homogenate stability study Ocular permeability studies using rabbit corneal tissues were performed as previously described (Prasanna et al., 2009). Fresh rabbit corneas were rinsed and equilibrated in BSS solution for 15 min. Corneas (n ¼ 2) were mounted in an using chamber perfusion system (Harvard Apparatus, Holliston, MA) which was placed in the heating block (Harvard Apparatus, Holliston, MA). The exposed surface area of the cornea is approximately 0.2 cm2 per well. One mL of BSS or BSS plus solution was added in the chamber bathing endothelium side (the receiver chamber), while one mL of test compounds (10 mM) in balanced salt solution (BSS) or BSS plus (containing 5 mM glucose) was added in the chamber bathing epithelium side (the donor chamber). Constant mixing of the donor and receiver solutions was achieved with a gentle gas flow of air/ CO2 (95%/5%) through both chambers. The heating block was maintained at 37 C by a circulating water bath. Serial samples of 100 mL were removed every hour during a period of 4 h from the receiver chamber. Each aliquot was replaced immediately with an equal volume of BSS or BSS plus. The samples were analyzed by liquid chromatography (LC) followed by tandem mass spectrometry (MS) (LC/MS/MS). Corneal homogenates were prepared as previously described (Prasanna et al., 2009). Corneal homogenate was prepared in BSS (one cornea/2.5 mL BSS) using a tissue tear homogenizer. The stability study was performed by incubating 10 mM PF-0217329 with corneal homogenate. At designated time points, 100 mL of incubation mixture was sampled and added to 200 mL acetonitrile. The samples were vortexed and centrifuged (3000 rpm for 10 min). The concentration of PF-04217329 was analyzed by liquid chromatography/mass spectroscopy (LC/MS). 2.6. Permeability data analysis Apparent permeability coefficients (Papp) were calculated using the following equation: Papp ¼ (1/A C0) (dM/dt). Where dM/dt is the flux across the cell layers or cornea (nmol/sec), A is the exposed surface area of the rabbit cornea, and C0 is the initial drug concentration (mM) in the donor compartment at t ¼ 0. Flux across the cell layer or cornea (dM/dt) was determined from the slope of the linear portion of the cumulative amount permeated versus time plot. 2.7. Topical formulation of PF-04217329 for in vivo studies PF-04217329 has the physical properties of an oil and is poorly soluble in water. To improve its aqueous solubility, the compound was formulated with a surface active ingredient to achieve concentrations required for preclinical evaluation. A preclinical formulation was developed containing 1% Tween-80, 5% Cremophor-EL and 0.02% benzalkonium chloride in 100 mM phosphate buffer pH 5.5. All the excipients used to formulate PF-04217329 were USP/NF grade, obtained from SigmaeAldrich (St Louis MO). 2.8. In vivo cAMP assay For measuring in vivo ocular levels of cAMP, Dutch-belted rabbits (n ¼ 3 per time point) were topically dosed with 11 mg of PF04217329 as 2 25 mL drops with a 5-min interval between the drops. Rabbits were euthanized at indicated time points and aqueous humor was collected via paracentesis and stored in
centrifuge tubes at 80 C. Iris/ciliary body tissue was rapidly dissected on ice and were homogenized in buffer containing isobutylmethylxanthine (IBMX) and also stored in centrifuge tubes at 80 C. cAMP levels were determined in aqueous humor and iris/ ciliary body using an ELISA per manufacturer’s instructions (Cayman Chemicals; Cat # 581001). Levels of cAMP between untreated and PF-04217329-treated eyes were compared for statistical significance using One-way ANOVA and Tukey’s multiple comparison test (p < 0.05). 2.9. Ocular pharmacokinetic assessments in rabbits Pigmented male Dutch-belted (DB) rabbits (n ¼ 3 rabbits/time point) weighing 1.5e2.0 kg were used. All animals received a 2 25 mL topical dose of PF-04217329 or vehicle in each eye. At indicated times (e.g. 1, 3, 6, and 24 h post-dose), the rabbits were euthanized and enucleated and ocular tissues were dissected. PF04217329 and CP-544326 were isolated from the homogenized rabbit ocular tissue (cornea, iris/ciliary body (ICB), and aqueous humor (AH)) by protein precipitation and quantitated using high performance liquid chromatography/mass spectrometry (LC/MS/ MS). Quantification of samples was performed by adding known amounts of CP-544326 acid to blank matrix for standard curve preparation. The LC/MS/MS system used consisted of two Shimadzu LC-10AD HPLC pumps (Columbia, MD), a CTC HTS PAL autosampler (LEAP, Carbrboro, NC) and a Sciex API 4000 triple quadrupole mass spectrometer (Life Technologies, Foster City, CA). Peak area determination, calculation of the ratio between the analyte to internal standard peak area and determination of sample concentrations was performed using Analyst software (Analyst 1.4.1, Life Technologies, Foster City, CA). 2.10. Ocular tolerability of topical ocular dosing of PF-04217329 in rabbits and dogs Animals (n ¼ 2e3 per treatment/condition) were dosed in one eye with 2 25 mL of a 0.2 mg/mL solution (1% Tween-80, 0.5% Cremophor-EL, 0.02% BAC, pH 5.5) either once (QD) or three times (TID) daily for 4 consecutive days. The total daily doses in the treated eyes were 10 and 30 mg for the QD and TID dose groups, respectively. The other eye received the same volume of vehicle alone. Conjunctival hyperemia, swelling and discharge were assessed post-dosing at various indicated time points post-dose. The ocular adverse effects were performed using the scale as previously described (Prasanna et al., 2009). Briefly, a score of zero indicated bulbar and conjunctival vessels were comparable to controls; 1, minimal (minor hyperemia/redness with no noticeable edema or other side effects.); 2, mild (obvious redness/hyperemia, individual vessels discernible, may have slight edema and eyes may be more sensitive to touch of pneumatonometer or light exposure e photophobia); 3, moderate (all the symptoms of ‘mild’ plus obvious edema or other side effects; individual vessels difficult to identify); 4, severe (diffusely red, individual vessels not discernible). 2.11. IOP measurements in rabbits, dogs, and non-human primates IOP was measured in both eyes of conscious normotensive Dutch-belted (DB) rabbits, normotensive Beagles, and unilaterally lasered ocular hypertensive cynomolgus monkeys following topical ocular dosing of test articles (vehicle and active compound containing formulations) using a Model 30 classic pneumatonometer (Medtronic, Minneapolis, MN). Typically, IOP was measured before administration of test articles (0 h - baseline) and then at 2, 4, 6 and 24 h after test article administration. Animals received 50 mL of topical 0.25% tetracaine HCl (anesthetic, Henry Schein, USA) prior
G. Prasanna et al. / Experimental Eye Research 93 (2011) 256e264
to each IOP measurement session. At least three IOP measurements were taken at each time point for both eyes. After the baseline IOP measurement was taken, dogs and rabbits received 2 25 mL drops (50 mL total) of either vehicle or test compound containing formulation to each eye. For monkey IOP studies, female cynomolgus monkeys between the ages of 4 and 15 years were used in this study (Toris et al., 2005). All had unilateral laser treatment to the trabecular meshwork of the left eye made between 0.2 and 8.3 years earlier to establish a chronic IOP elevation (Toris et al., 2000). In a crossover study design, either 2 25 mL drops or a single 30 mL drop of compound or vehicle were instilled either only in the ocular hypertensive eye or in both eyes with a week of washout between compound or vehicle. In all these and other IOP studies, investigators taking the IOP measurements were masked to the dosing scheme. Both AM and PM dosed IOP studies were performed. For PM dosing, animals were dosed at 8PM and IOP measurements were taken the following day at 12 h and 18 h post-dose (i.e. 8AM and 2PM) whereas that for AM dosing was done at 8AM and IOP was measured at 1 h, 2 h, 4 h, 6 h, and 24 h post-dose. Both single day and multiple day studies were performed with PF-0217329. Six to ten animals were used in each IOP experiment. Statistical significance of drug treatment over that of vehicle-treatment was obtained by Repeated Measures One-way ANOVA with Holm-Sidak multiple comparison test (p < 0.05). IOP reduction (mmHg) of PF04217329 or other active compound versus vehicle (control) was calculated by first subtracting drug or vehicle-treated IOP effects at 2 h, 6 h or 24 h from their respective starting baseline values (delta; set at 100%) and then by subtracting drug effect from vehicle effect for 2 h, 6 h or 24 h time points (delta delta). All animal-related procedures were conducted in accordance with the Association for Research in Vision and Ophthalmology (ARVO) statement for the use of animals in ophthalmic research and were approved by the appropriate animal care and use committees of each institute.
259
Table 1 CP-544326 EP binding affinity, selectivity, and cAMP production. EP
Receptor affinity (IC50)a
cAMP production (EC50)b
hEP2 rEP2 hEP1 hEP3 hEP4
10 nM 15 nM >3200 nM >3200 nM >3200 nM
2.8 nM 1.9 nM NA NA NA
Note: cAMP ¼ cyclic adenosine monophosphate, IC50 ¼ median inhibitory concentration, EC50 ¼ median effective concentration, hEP ¼ human EP, rEP ¼ rat EP, NA ¼ not applicable, PGE2 ¼ prostaglandin E2. IC50 and EC50 values provided herein are the calculated average of n ¼ 2 independent experiments. a IC50 values were obtained from 3H-PGE2 displacement binding assay in human embryonic kidney cells stably transfected with hEP2 and rEP2. b EC50 values were obtained from cAMP assay in stably transfected hEP2 and rEP2 in human embryonic kidney cells.
The acid metabolite, CP-544326 and PF-04217329 prodrug were evaluated for their binding and [Ca2þ]i mobilization activities in cells overexpressing hFP receptors. The IC50 for CP-544326 was calculated to be >100 mM. No binding activity was observed with PF-04217329 and CP-544326 with the hFP receptor. In the calcium flux assay, CP-544326 and butaprost lacked functional activity up to 1 mM whereas PGF2a caused a robust increase in [Ca2þ]i (data not shown). CP-544326 was also tested for binding to 37 human GPCRs including 15 GPCR reported to signal via stimulation of intracellular cAMP (5HT4, 5HT7, Adenosine A1 and A2 receptors, a-2A, a-2B, and b-3 adrenergic receptors, cannabinoid receptors CB1 and CB2, cholecystokinin type A receptor CCK1, dopamine receptor D1, histamine receptors H1 and H2, muscarinic acetylcholine receptor M2, and Mu opioid receptor). CP-544326 did not show appreciable binding to any of the 37 GPCRs when tested at 10 mM final concentration (% inhibition was 30% at 10 mM CP-544326).
3. Results
3.2. Corneal permeability and hydrolysis studies
3.1. In vitro EP receptor binding and cAMP stimulation studies
Corneal permeability studies with PF-04217329 and active acid metabolite, CP-544326 were assessed in the ex vivo rabbit corneal model (n ¼ 2 corneas). The direct permeability of CP-544326 was poor with a Papp value of 1.2 cm6/s (average of 0.91 and 1.58 cm6/s). In comparison, PF-04217329 exhibited high corneal permeability with a Papp value of 7.1 cm6/s (average of 7.62 and 6.66 cm6/s). The ex vivo corneal permeability PF-04217329 is approximately 7-fold higher than that of CP-544326 as assessed in rabbit corneas. PF-04217329 is quickly converted to CP-544326 by esterases in rabbit corneal homogenate with an in vitro T1/2 < 4 min.
PF-04217329 is an isopropyl ester prodrug of the pharmacologically active acid metabolite, CP-544326 (Fig. 1). As shown in Table 1, CP-544326 with an IC50 for human EP2 equal to 10 nM (average of two independent experiments; 9 and 11 nM), is at least 270 times more selective for human EP2 subtype as compared with the other human EP subtypes, 1, 3, and 4. Based on the cell-based efficacy data using rat EP2-HEP293 cells, cAMP levels increased in a dose- and time-dependent manner yielding an average EC50 of 1.9 nM (1.5 and 2.4 nM; from two independent experiments). In human EP2HEK293 cells, CP-544326-mediated cAMP production yielded an EC50 of 2.8 nM (2.5 and 3.1 nM; from two independent experiments), which was similar to that observed for PGE2 (EC50 ¼ 2.6 nM).
PF-04217329
O
OH
O O
O
N
O
N
O S N O
N N
Esterases
O S N O
N N
CP-544326
Fig. 1. Chemical structure of PF-04217329 (prodrug analog) and CP-544326 (active acid metabolite which is hydrolyzed by the action of esterases in ocular tissues).
3.3. Rabbit ocular pharmacokinetics following topical ocular dosing of PF-04217329 Following topical ocular dosing, PF-04217329, a prodrug, undergoes rapid hydrolysis via esterases in ocular tissues yielding the pharmacologically active acid metabolite, CP-544326. This metabolite is detected in rabbit aqueous humor (AH) and iris/ciliary body (ICB) (Fig. 1). As mentioned above, CP-544326 has a w7-fold higher ex vivo corneal permeability value when tested as the prodrug (PF-04217329) compared to the value obtained for the acid. The higher corneal permeability observed for the isopropyl prodrug (PF-04217329) translates into a significant improvement in ocular exposure following topical administration. Levels for the intact parent compound (PF-04217329) were below the quantitation limit for all these tissues due to rapid hydrolysis in corneal tissue. In rabbits, the exposure of CP-544326 in the ocular tissues after topical administration of PF-04217329 is dose-proportional up to
Aqueous humor (ng/mL)
10000.0 CP-544326 concentration
ICB(ng/g) Cornea (ng/g)
1000.0
100.0
10.0
80.0
1600.0
70.0
1400.0
60.0
1200.0
50.0
1000.0
40.0
800.0
30.0
600.0 400.0
20.0 Aqueous Humor
10.0
200.0
ICB
0.0
0.0 Untreated
1.0 0
4
8
12
16
20
24
Time, hour Fig. 2. Ocular levels of CP-544326 following topical administration of PF-04217329 (11 mg) to Dutch-belted rabbits (n ¼ 3 rabbits/time point). Dosing volume was 25 mL drop 2; dosing concentration ¼ 0.22 mg/mL. CP-544326 was measured as described in materials and methods. ICB ¼ iris/ciliary body.
doses of 8e11 mg per eye. Concentrations of CP-544326 appeared to be highest in the cornea that were 2e3 log units higher than that in AH and ICB tissue (Fig. 2). Furthermore, topical ocular dosing of PF04217329 resulted in ocular exposure of CP-544326 at levels greater than the IC50 for EP2 receptor (i.e. 10 nM). The calculated T1/2 for rabbit ICB was 5.5 h whereas that for cornea was 4.6 h and aqueous humor was 3.7 h respectively (Fig. 2; Table 2). Taken together, these results confirm that dosing of the prodrug improves ocular tissue exposure of CP-544326 following topical application. 3.4. Assessment of ocular cAMP levels in normotensive DB rabbits after topical administration of PF-04217329 As discussed previously, the in vitro cell culture experiments demonstrated that PF-04217329 acted via EP2 to increase cAMP levels (Table 1). Topical administration of PF-04217329 (11 mg) to normotensive DB rabbits also resulted in changes to in vivo cAMP levels in AH and ICB. This experiment was also performed to determine the duration of EP2 target modulation in an in vivo system. PF-04217329 (11 mg) caused a modest increase in cAMP in ICB (2- to 2.5-fold) and in AH (1.5-fold) for up to 3 h after administration compared to that observed in untreated eyes, indicating most likely that EP2 activation was occurring in vivo (Fig. 3). By 24 h, cAMP levels had returned to baseline in both ICB and AH. Experiments with EP2 antagonist to confirm EP2 receptor involvement following PF-04217329 treatment were not performed due to technical challenges including lack of knowledge of actual dose, intrinsic corneal permeability, ocular bioavailability and solubility of the antagonist to be tested. Table 2 Ocular tissue exposure after topical administration of PF-04217329 (11 mg) or CP544326 (10 mg) as well as area-under-the-curve (AUC) and half-life of CP-544326 in Dutch-belted rabbits. Doseda
Cornea AUC0e24 h, ng h/g (T1/2 h)
Aqueous humor AUC0e24 h, ng h/mL (T1/2 h)
Iris-Ciliary Body AUC00-24 h, ng h/g (T1/2 h)
PF-04217329b CP-544326
44122 816 (4.6)
2406 33 (3.7)
630 NDc (5.5)
a
An equal molar of PF-04217329 and CP-544326 was dosed. CP-544326 was monitored since intact parent was undetectable due to rapid hydrolysis. c ND e Not Detected, CP-544326 levels in iris-ciliary body (ICB) were below the detection limit h (hours). b
cAMP (pMol/mg, ICB)
G. Prasanna et al. / Experimental Eye Research 93 (2011) 256e264
cAMP (pMol/ml, aqueous humor)
260
1
3
6
24
Time (hours post-treatment)
Fig. 3. Time-course of cAMP levels after topical administration of PF-04217329 (11 mg; 0.22 mg/mL) to normotensive Dutch-belted rabbits (n ¼ 3 rabbits; 6 eyes). Levels of cAMP in aqueous humor (AH) are indicated in the Y axis while that for iris/ciliary body (ICB) are denoted on the Y 0 axis. cAMP was measured using an ELISA (enzyme-linked immunosorbent assay). * denotes statistical significance of AH cAMP levels and ** denotes statistical significance of cAMP levels in ICB tissues following PF-04217329 treatment at indicated time points compared to the respective AH/ICB levels in untreated eyes (One-way ANOVA and Tukey’s multiple comparison test, p < 0.05).
3.5. Effect of PF-04217329 on rabbit ocular safety and IOP A greater incidence of mild conjunctival hyperemia was observed in rabbit eyes treated once daily with the compound than with vehicle during the first 3 days of the study. In the three times daily and multi-day studies, minimal to moderate conjunctival hyperemia was observed in the compound treated eyes on Days 1e4. No signs of ocular irritation were observed 24 h after the last dose (Day 5). No compound-related effects were observed microscopically in the eyes or eyelids. In a separate rabbit study, PF04217329 was dosed as a 0.22 mg/mL solution at a volume of 2 25 mL. The compound was delivered three times at 5 min intervals (total dose of 33 mg) followed by several hours of intermittent observation. Mild conjunctival hyperemia occurred in 1 of the 2 eyes dosed. The rabbit EP2 receptor has been cloned and its receptor pharmacology has been previously determined by Guan et al. (2002). Therefore, normotensive Dutch-belted (NTDB) rabbits were used to determine the effect of single dose topical administration of PF04217329 on IOP in a time and dose-dependent manner. PF04217329 administered once topically at 1, 2.5, 5, 8, and 10 mg produced a dose- and time-dependent reduction in IOP, while a 0.1mg dose had no effect on the IOP reduction (Fig. 4A; Table 3). As shown in Table 3, doses 8 mg caused a significant reduction in IOP both at 6 h and 24 h. An 8 mg dose of PF-04217329 caused an IOP reduction of 49% at 6 h and 19% at 24 h, while a 10 mg dose caused IOP reduction of 38% both at 6 h and 24 h post-dosing. Statistical significance was observed at 6 h post-dose for all doses whereas that for 24 h was only observed at the highest doses (8 and 10 mg) (Table 3). Another IOP reduction study was performed in NTDB rabbits to determine the IOP effects between 12 h and 24 h after topical administration of PF-04217329 at 5 mg (graph not shown). For that purpose, PF-04217329 was administered in the late evening, and IOP assessments were made the following day at 12 h and 18 h after dosing. From the analysis of IOP data combined from the AM and PM dosing studies, it was evident that peak IOP reduction occurs at 6 h after dosing (34%), and IOP remained stably reduced from 12 h (33%) to 18 h (37%); IOP was still reduced (approximately 9%) at 24 h. A 4-day study of PF-04217329 (5 mg dosed once daily) was performed in NTDB rabbits to assess changes in IOP after multiple days of dosing. As shown in Fig. 4B, the initial IOP reduction observed at
G. Prasanna et al. / Experimental Eye Research 93 (2011) 256e264
A
Table 3 Percent IOP reduction following a single dose of PF-04217329 in normotensive Dutch-belted (NTDB) rabbits and normotensive (NT) dogs.
Vehicle (0.2 mg/ml) Vehicle (0.16 mg/ml) Vehicle (0.1 mg/ml) Vehicle (0.05 mg/ml) Vehicle (0.02 mg/ml) Vehicle (0.002 m/ml) PF-04217329 (0.2 mg/ml; 10 ug) 0.16 mg/ml (8 ug) 0.1 mg/ml (5 ug) 0.05 mg/ml (2.5 ug) 0.02 mg/ml (1 ug) 0.002 mg/ml (0.1 ug)
30 28
Species
26
22 20 18
Treatment
Concentration/Dose
% IOP (SEM) % IOP (SEM) reduction reduction at 6 h at 24 h
NTDB PF-04217329 0.002 mg/mL (0.1 mg) 0.02 mg/mL (1 mg) Rabbits 0.05 mg/mL (2.5 mg) 0.1 mg/mL (5 mg) 0.16 mg/mL ((8 mg) 0.2 mg/mL (10 mg)
4 38 45 35 49 38
2% 5%* 4%* 9%* 11%* 6%*
5 0 10 9 19 34
2% 1% 4% 5% 6%* 7%*
(0.1 mg) (1 mg) (2.5 mg) (5 mg) (8 mg) (10 mg)
8 22 37 40 45 37
3% 3%* 3%* 4%* 3%* 4%*
0 15 30 33 37 31
3% 3% 6%* 3%* 4%* 3%*
NT Dogs
24 IOP (mm Hg)
261
PF-04217329 0.002 0.02 0.05 0.1 0.16 0.2
mg/mL mg/mL mg/mL mg/mL mg/mL mg/mL
* Denotes statistical significance versus vehicle-treated eyes and baseline IOP (Repeated Measures ANOVA and Holm-Sidak multiple comparison test; p < 0.05).
16 14 12 0
4
8
12
16
20
24
28
Hours post first dose
B
PF-4217329 (0.1mg/ml; 5 ug/day)
30.0
Vehicle
IOP (mm Hg)
25.0
20.0
15.0
10.0 0
2
4
6
24
30
48
54
72
78
96
Hours post first dose
Fig. 4. Effect of topical ocular dosing of PF-04217329 on intraocular pressure (IOP) responses in normotensive Dutch-Belted (NTDB) rabbits (n ¼ 8 rabbits/group). A) Single dose studies and B) Multiple days of once daily dosing with PF-04217329 (5 mg/ day; dosed once daily for 4 days) on IOP response in rabbits. IOP was measured at indicated time points following a single daily topical ocular administration of PF-04217329 and vehicle after a single day or multiple (four) days of dosing. Baseline IOP was measured prior to dosing with either vehicle or drug. For single day studies, statistical significance was observed at 6 h post-dose for all doses whereas that for 24 h was only observed at highest dose (8 and 10 mg) (Repeated measures ANOVA and Holm-Sidak multiple comparison test; p < 0.05). Symbols denoting statistical significance not included in figure so as to not obscure the graph. For multiple day dosing studies, * denotes statistical significance of PF-04217329-induced IOP reduction versus baseline IOP and vehicle effects determined by Repeated Measures ANOVA and HolmSidak multiple comparison test (p < 0.05).
the first 24-h time point gradually diminished thereafter as observed up to 96 h post-first dose. This finding suggested that NTDB rabbits developed tolerance to multiple days of PF-04217329 of once daily dosing. 3.6. Effect of PF-04217329 on ocular tolerability and IOP in dogs It is known that dogs express EP2 receptors and that they share 88% identity with human EP2 receptors based on protein sequence alignment (Hibbs et al.,1999). Therefore, the IOP lowering effects of selective EP2 agonists (butaprost and PF-04217329) were assessed in ocular normotensive (ONT) dogs (beagles). A similar level of safety and tolerability i.e. mild to moderate effects was observed in dogs as rabbits both following single and multiple days of dosing of PF-04217329.
Butaprost is a very selective and potent EP2 receptor agonist (EC50 ¼ 2.2 nM; hKi of 91 nM) (Breyer et al., 2001), however the binding affinity is 6-fold lower when compared to CP-544326 (hKi of 14 nM), the acid metabolite of PF-04217329. Butaprost-methyl ester (50 mg) was dosed once topically in normotensive dogs and IOP changes were assessed as early as 0.5 h post-dosing. No IOP spike was observed at 0.5 h however, butaprost caused a significant IOP lowering at 2 h (21 1.7%) and 4 h onwards (27 4.3%). Maximum IOP reduction of 32 3.6% was observed at 6 h, while there was no IOP reduction observed at 24 h post-dose (graph not shown). Next, single day IOP studies were performed with PF-04217329 with doses ranging from 0.1 to 10 mg/day (Fig. 5A). As seen in Fig. 5A and Table 3, PF-04217329 caused a dose and timedependent reduction in IOP following once daily dosing in ONT dogs. No IOP lowering effect was observed following a 0.1 mg dose, while a 1 mg dose elicited 6 h and 24 h IOP reduction (Table 3). For higher doses of PF-04217329, it appeared that maximal percent IOP reduction (compared to vehicle) at 6 h was seen between 5 mg and 10 mg doses (40% and 45% respectively) while maximal percent IOP reduction of 37% at 24 h occurred with the 8 mg dose. With a 10 mg dose at 24 h, IOP was reduced by 31%. However, the overall percent of IOP reduction at 24 h post-dose ranged between 30 and 37% for 2.5e10 mg doses (Table 3). The onset of significant IOP reduction with PF-04217329 compared to that seen with vehicle occurred at 2 h with a 1 mg dose. It is interesting to note that 2.5e10 mg doses of PF-04217329 elicited not only robust 6 h IOP responses, which were greater than that seen with butaprost at a substantially higher dose (50 mg), but also exhibited 24 h IOP reductions post-dose that were not seen with butaprost treatment. As noted in Table 3, statistical significance was observed for PF04217329-treated eyes at 2 h, 4 h, and 6 h for 0.1 mg dose whereas 2 h, 4 h, 6 h, and 24 h post-dose was observed for 2.5 mg, 5 mg, 8 mg, and 10 mg (Repeated measures ANOVA and Holm-Sidak multiple comparison test; p < 0.05). The IOP lowering effects of PF-04217329 (1 mg/day) were evaluated in a multi-day once daily dose study in ONT dogs for 5 days. The 1 mg dose was chosen since normotensive DB rabbits showed tolerance at 5 mg/day dose while in ONT dogs, a 1 mg dose showed sub-maximal IOP effect both at 6 h and 24 h. The multi-day dosing at 1 mg/day was also chosen to evaluate the tolerance effect at lower doses in a higher species which is closer to humans than rabbits in terms of anterior chamber development. As seen in Fig. 5B, despite daily dosing of 1 mg PF-04217329, no tolerance was observed in the
262
G. Prasanna et al. / Experimental Eye Research 93 (2011) 256e264
A
3.7. Effects of butaprost and PF-04217329 in laser-induced ocular hypertensive cynomolgus monkeys
Vehicle (0.2 mg/ml) Vehicle (0.16 mg/ml) Vehicle (0.1 mg/ml)
When dosed in the ocular hypertensive eye, the EP2 agonist, butaprost-methyl ester (50 mg) caused IOP lowering of 24 6.2% at 6 h and 12 4.2% at 24 h compared to respective vehicle effects (7 4.6% at 6 h and 0 4% at 24 h) (graph not shown). As shown in Fig. 6A, a single dose of PF-04217329 (11 mg) elicited IOP lowering effects of 23% at 2 h, 35% at 4 h and 33% at 6 h, however in this study the 24 h post-dose time point was not measured. It is important to note that PF-04217329-mediated IOP reduction reached levels that was below the IOP of the normotensive eye. A multi-day dosing study with PF-04217329 at 5 mg/day for 4 days was also performed in laser-induced ocular hypertensive cynomolgus monkeys. As seen in Fig 6B, 5 mg/day dosing of PF04217329 caused an IOP reduction both at 6 h and 24 h post-dose on all days compared to vehicle-treated eye. PF-04217329 caused an acute IOP reduction at 6 h and 24 h time points on days 1 and 2
Vehicle (0.05 mg/ml) Vehicle (0.02 mg/ml) Vehicle (0.002 mg/ml)
28
0.2 mg/ml (10 ug) 0.16 mg/ml (8 ug)
26
0.1 mg/ml (5 ug) 0.05 mg/ml (2.5 ug) 0.02 mg/ml (1 ug)
24
0.002 mg/ml (0.1 ug)
IOP (mm Hg)
22
20
18
16
A
14
PF-04217329 (11 ug)
50
Vehicle
45
12
40 0
4
8
12
16
20
24
Hours post first dose
B
PF-04217329 (0.02 mg/ml; 1 ug/day) Vehicle
25.0
IOP (mmHg)
10
35 30
*
25
IOP (mm Hg)
20
*
*
15
20.0
10 0
2
4
6
Hours Post Treatment
15.0
B
Vehicle
45
PF-04217329 (0.1 mg/ml; 5 ug/day)
10.0
0
6
24
30
48
54
72
78
96
40
Hours post first dose
5 day experiment. The maximum IOP lowering effect of 30% was seen at the first 24 h time point. Thereafter, percent IOP reduction compared to control was stable at 21%, 22% and 19% at 48 h, 72 h, and 96 h post 1st day dose respectively. Intriguingly, there was no cumulative IOP reduction following topical administration of several sub-maximal doses of PF-04217329 in ONT dogs. Similar IOP reduction was observed with 5 mg/day multi-day dosing study (data not shown).
IOP (mmHg)
35 Fig. 5. Effect of topical ocular dosing PF-04217329 on IOP responses in normotensive dogs (n ¼ 8). A) Single dose studies and B) Multiple days of once daily dosing with PF04217329 (1 mg/day; dosed once daily for 4 days) and vehicle on IOP response in dogs. PF-04217329 and vehicle were administered to both eyes of dogs (n ¼ 8) and IOP was measured at indicated time points post-dose. Baseline IOP was measured prior to topical ocular dosing. For single day studies, statistical significance was observed for PF-04217329-treated eyes at 2 h, 4 h, and 6 h for 0.02 mg/mL dose whereas 2 h, 4 h, 6 h, and 24 h post-dose was observed for 0.05 mg/mL, 0.1 mg/mL, 0.16 mg/mL, and 0.2 mg/mL (Repeated measures ANOVA and Holm-Sidak multiple comparison test; p < 0.05). Symbols denoting statistical significance not included in figure so as to not obscure the graph. For multiple day dosing studies, * denotes statistical significance of PF-04217329-induced IOP reduction versus baseline IOP and vehicle effects determined by Repeated Measures ANOVA and Holm-Sidak multiple comparison test (p < 0.05).
30
*
*
25
* *
* *
*
*
*
20 15 10 0
24
48
72
96
120
Hours post first dose Fig. 6. Effect of topical ocular dosing of PF-04217329 on IOP responses in the hypertensive eye in cynomolgus monkeys (n ¼ 8). A) Effect of a single topical ocular dose of PF-04217329 (11 mg) and vehicle and B) effect of once daily dosing of PF-04217329 (5 mg/day) for 4 days and vehicle on IOP in monkeys. Vehicle or PF-04217329 was dosed in the hypertensive eye as part of a crossover study design (n ¼ 8 monkeys) with a week of washout between dosing. PF-04217329 (0.2 mg/mL) was dosed as 2 25 mL drops. IOP readings were taken by pneumatonometry at indicated time points. All these studies were done as a crossover design immediately after the baseline IOP measurement at t ¼ 0. For the single day study, * denotes statistical significance of PF04217329-induced IOP reduction versus baseline IOP and vehicle effects determined by Repeated Measures ANOVA (p < 0.05). For multiple day studies, * denotes statistical significance of PF-04217329-induced IOP reduction versus baseline IOP and vehicle effects were determined by Repeated Measures ANOVA and Holm-Sidak multiple comparison test (p < 0.05).
G. Prasanna et al. / Experimental Eye Research 93 (2011) 256e264
which was stabilized on days 3 and 4ew38% for 6 h and 21% for 24 h. Similar effects were also observed with a 2.5 mg/day dosing of PF-04217329 (Data not shown). In contrast to the findings in rabbits (Fig. 4B) but similar to that in dogs (Fig. 5B) following multi-day dosing with PF-04217329, no apparent loss of efficacy was observed in OHT monkeys (Fig. 6B). 4. Discussion PF-04217329 (Taprenepag isopropyl) is a prodrug of CP-544326, a novel selective and potent EP2 receptor agonist, which was recently evaluated in a clinical study in patients with POAG and ocular hypertension (Schachar et al., 2010). PF-04217329 was shown to significantly lower IOP up to 24 h compared to vehicledosing in patients with POAG following QD dosing for 14 days (Schachar et al., 2010). Key preclinical studies demonstrated that PF-04217329 produced robust IOP efficacy (3 in vivo models), most likely via the activation of EP2 receptors (in vivo and in vitro) with sufficient bioavailability in ocular target tissue and an acceptable safety profile after topical ocular dosing. Hence, these studies support the selection of PF-04217329 for clinical development. PF-04217329 was found to be well tolerated at 10 mg and 30 mg doses in rabbits, and as much as 10 mg in dogs and monkeys following topical ocular dosing. Acute minimal to moderate conjunctival redness was the main adverse effect noted in rabbits and dogs. Putative EP2 receptor activation following CP-544326/PF-04217329 administration was determined via measurement of cAMP elevation both in in vitro assays and in rabbit aqueous humor (AH) and iris/ciliary body (ICB) following topical ocular administration. The absence of detectable levels of PF-04217329 in ocular tissues following topical ocular dosing is a result of rapid hydrolysis in corneal tissue to the active acid metabolite, CP-544326. The greater corneal permeability of PF04217329 compared to CP-544326, results in greater levels of the active metabolite in ocular tissues/compartments. These findings are similar to recently published observations for an EP4 prostaglandin prodrug PF-04475270 (Prasanna et al., 2009). Changes in IOP following topical ocular dosing of PF-04217329 were determined in three in vivo models (in three species) including ocular normotensive Dutch-belted (NTDB) rabbits, ocular normotensive (ONT) dogs and laser-induced ocular hypertensive (OHT) cynomolgus monkeys. In single-day IOP studies, PF-04217329 appears to show similar effects across all three species. Whereas peak levels of CP-544326 and cAMP in rabbit AH/ICB were reached between 1 and 3 h post-topical ocular dosing of PF-04217329, peak efficacy was observed at 6 h and was maintained up to 24 h postdose. Similar to other prostaglandin analogs including latanoprost (FP agonist) and PF-04475270 (prodrug of an EP4 agonist), CP544326 is a potent agonist of the EP2 receptor that can activate several downstream signaling pathways resulting in peak IOP lowering at later time points (i.e. between 6 and 24 h post-dose) most likely due to the activation of second messenger systems including cAMP (Weinreb et al., 1997; Prasanna et al., 2009). In single day once daily topical ocular dosing studies in normotensive dogs, PF-04217329 produced a robust IOP reduction (up to 45%) that was sustained for 24 h post-dose unlike that observed for butaprost. Furthermore, PF-04217329 lowered IOP to a level that was approaching those of episcleral venous pressure in ocular normotensive dogs (10e12 mmHg). This hypotensive effect is perhaps the maximal extent of IOP reduction that can be achieved by pharmacological or surgical intervention (Gelatt et al., 1998). In single day once daily dosing studies in OHT monkeys, AH13205 at 25 mg and 125 mg doses produced a dose-dependent IOP reduction of 15% and 20%, respectively, at 6 h post-dose whereas butaprost (25 mg) produced a robust 30% reduction during the same time period (Woodward et al., 1995; Nilsson et al., 2006).
263
While butaprost’s effect on IOP reduction in monkeys (by 34%) in our study was similar to previous findings by Nilsson et al. (2006), PF-04217329 (11 mg) produced an even greater reduction of 45% in IOP for up to 6 h post-dose in OHT monkeys. In multi-day IOP studies, data appear to be most consistent in terms of sustained and stable IOP reduction at 24 h over days 2e5 in ONT dogs and OHT primates compared to data from rabbits, which exhibited apparent tachyphylaxis following 4 days of topical ocular dosing. In a previous report in ocular normotensive monkeys, EP2 agonist, AH13205 (25 mg) dosed twice daily for 5 days caused a 2e3 mmHg reduction in IOP on days 3, 4, and 5 (Woodward et al., 1995). No tolerance was seen in this study and this was similar to our findings in OHT monkeys. A possible explanation for the observed tachyphylactic response to multiple day dosing of PF04217329 in rabbits but not in dogs or monkeys is that rabbits have a poorly developed ciliary body, the predominant site for EP2mediated IOP reduction. Alternatively, differences in EP2 receptor distribution and cellular signaling in rabbits could lead to receptor uncoupling and/or activation of robust counteracting mechanisms to EP2-induced IOP reduction. In any case, no tachyphylaxis was observed in the recently completed clinical study of once daily dosing up to 14 days with PF-04217329 indicating that multi-day IOP studies in dogs and monkeys were more translatable to humans (Schachar et al., 2010). In the human anterior chamber, immunolabeling indicates the presence of EP2 receptors in ciliary body and ciliary process (pars plana and pars plicata) (Schlotzer-Schrehardt et al., 2002; Biswas et al., 2006). Additionally, EP2 receptors line the inner wall of the Schlemm’s canal and are present in the TM cells lining the beams (Kamphuis et al., 2004). Interestingly, EP2 immuno-colocalization was seen with asmooth muscle actin in posterior TM tissue of human anterior chamber, suggestive of EP2 activation and TM relaxation (Kamphuis et al., 2004). Previously, it has been demonstrated using quantitative PCR that EP2 receptor mRNA expression is most abundant in human TM tissues than other PG receptors (Kamphuis et al., 2001). For instance, EP2 receptor mRNA was 18-times higher than EP1 mRNA, 228 times higher than EP3 mRNA and 15 times greater than EP4 mRNA expression (Kamphuis et al., 2001). The EP2 receptor distribution in dog and monkey anterior chamber is presently unknown making species comparisons of receptor expression and pharmacodynamic effects on IOP much more difficult. It is hypothesized that EP2 agonists may increase aqueous humor outflow via relaxing TM tissues thereby lowering IOP. EP2 activation by butaprost or AH13205 has been shown to cause relaxation of carbachol-precontracted bovine TM strips as well as ciliary muscle (CM) strips (Krauss et al., 1997). Relaxation induced by both EP2 agonists was stronger in TM strips than in CM strips. The signaling pathways that are activated by EP2 agonists result in promoting the relaxation of bovine TM strips and can include activation of maxi-K high conductance outward potassium channels (Stumpff et al., 2005). Additionally, EP2 agonist-induced cAMP production and subsequent IOP lowering in the anterior chamber could also involve other downstream signaling pathways in TM and CM that affect cell adhesion and/or cause cytoskeletal changes including Rho kinase inhibition, activation of A-kinase anchoring protein (AKAP), induction of COX-2 and PGE2 synthesis, phosphorylation of GSK-3 and Tcf/ lymphoid enhancer factor transcriptional activation, and b-catenin signaling (Stumpff et al., 2005; Fujino et al., 2002; Rosch et al., 2005). It is of interest to mention that butaprost has been shown to up-regulate Cyr61 transcription in a dose and time-dependent manner in EP2 transfected HEK cells and human CM cells (Liang et al., 2003). Cyr61 is a cysteine-rich protein that is a ligand for integrin receptors, utilizes Rho signaling and may be involved in remodeling of extracellular matrix which is particularly relevant in CM and TM tissues and of consequence for IOP regulation. It is likely
264
G. Prasanna et al. / Experimental Eye Research 93 (2011) 256e264
that the aforementioned signaling pathways may be activated by CP-544326, however confirmation is still required. Despite the EP2 agonist mediated signal transduction effects on TM, previous data from aqueous humor outflow studies in normal cynomolgus monkeys suggest that butaprost produces a w2-fold increase in uveoscleral outflow implicating the involvement of CM tissues in IOP reduction rather than an increase in outflow facility (Nilsson et al., 2006). Furthermore, morphological changes in the anterior segment, particularly in CM tissues, have been observed following long term treatment of normal monkeys with 0.1% AH13205 (Richter et al., 2003). Specifically in CM tissues of AH13205-treated monkeys, there was an enlargement of space between muscle bundles in the anterior longitudinal and reticular portion of the muscle compared to that seen in controls. This increase in CM intertissue spaces may be indicative of increased matrix metalloproteinase (MMP) activity, consequently increasing the uveoscleral outflow route for aqueous humor as shown previously with prolonged treatment with EP2 agonists (Richter et al., 2003). The increase in MMP expression is known to occur following topical prostanoid treatment including latanoprost, PGF2a and a PGE2 analog, 11-deoxy PGE1, which may be responsible for the increase in uveoscleral outflow (Lindsey et al., 1996; Weinreb et al., 1997; Gaton et al., 2001). Therefore, it is highly likely that PF04217329 could also lower IOP mainly via increasing uveoscleral outflow and studies are underway to determine this possibility. In conclusion, PF-04217329, a prodrug of CP-544326 (active acid metabolite), is a selective and potent EP2 agonist that significantly lowers IOP for up to 24 h following a single topical ocular dose and was well tolerated in 3 different preclinical species including rabbits, dogs and monkeys in all 3 species at clinically relevant doses. Such a robust IOP reduction was correlated with levels of CP544326 albeit in a phase-shifted manner, such that peak IOP levels occurred 6e24 h post-treatment whereas peak levels of CP-544326 were found 2-6 h post-treatment of PF-04217329. PF-04217329 appears to be a promising compound targeting a novel mechanism, and has successfully demonstrated clinical efficacy in patients with primary open angle glaucoma. Acknowledgments The authors wish to thank Joanne Dean, Robert Alger, Joseph Doro, Robert Lebel, Jim Levis, and Stan LeBlanc (Canine facility in Comparative Medicine, PGRD Groton, CT) and Shan Fan and Lisa Stapp (Department of Ophthalmology, UNMC), for their immense technical and collaborative help with the IOP studies as well as Kimberly Cameron and Vishwas Paralkar (CVMED, PGRD, Groton, CT) for their scientific advice on the EP2 program. The efforts of Walter Collette III (Pfizer Drug Safety and Toxicology) are also appreciated. Primate studies were supported via a Pfizer contract to Dr. Carol Toris (Department of Ophthalmology, UNMC). This study was sponsored by Pfizer Inc. References The advanced glaucoma intervention study (AGIS): 7. The relationship between control of intraocular pressure and visual field deterioration. The AGIS Investigators. Am. J. Ophthalmol. 130 (4), 2000, 429e440.
Breyer, R.M., Bagdassarian, C.K., Myers, S.A., Breyer, M.D., 2001. Prostanoid receptors: subtypes and signaling. Annu. Rev. Pharmacol. Toxicol. 41, 661e690. Review. Fujino, H., West, K.A., Regan, J.W., 2002. Phosphorylation of glycogen synthase kinase-3 and stimulation of T-cell factor signaling following activation of EP2 and EP4 p.ostanoid receptors by prostaglandin E2. J. Biol. Chem. 277 (4), 2614e2619. Gaton, D.D., Sagara, T., Lindsey, J.D., et al., 2001. Increased matrix metalloproteinases 1, 2, and 3 in the monkey uveoscleral outflow pathway after topical prostaglandin F(2 alpha)-isopropyl ester treatment. Arch. Ophthalmol. 119, 1165e1170. Gelatt, K.N., Brooks, D.E., Samuelson, D.A., 1998. Comparative glaucomatology. I: the spontaneous glaucomas. J. Glaucoma 7 (3), 187e201. Review. Guan, Y., Stillman, B.A., Zhang, Y., et al., 2002. Cloning and expression of the rabbit prostaglandin EP2 receptor. BMC Pharmacol. 2, 14. Hibbs, T.A., Lu, B., Smock, S.L., et al., 1999. Molecular cloning and characterization of the canine prostaglandin E receptor EP2 subtype. Prostaglandins Other Lipid Mediat 57 (2e3), 133e147. Inatani, M., Honjo, M., Tokushige, H., Azuma, J., Araie, M., 2008. Intraocular pressure-lowering effects and safety of topical administration of a selective ROCK inhibitor, SNJ-1656, in healthy volunteers. Arch. Ophthalmol. 126, 309e315. Kamphuis, W., Schneemann, A., van Beek, L.M., et al., 2001. Prostanoid receptor gene expression profile in human trabecular meshwork: a quantitative realtime PCR approach. Invest. Ophthalmol. Vis. Sci. 42 (13), 3209e3215. Kamphuis, W., Schneemann, A., Shichi, H., et al., 2004. Immunolocalization of prostanoid EP receptor isotypes in human trabecular meshwork. Curr. Eye Res. 29 (1), 17e26. Kim, N., Crosson, C., Lam, T., Christian, B., Busse, C., Cantone, G., Baumgartner, R., McCauley, T., McVicar, W., 2010. INO-8875, an Adenosine A1 Agonist, Lowers Intraocular Pressure Through the Conventional Outflow Pathway. ARVO. poster # 3238/D780. Krauss, A.H., Wiederholt, M., Sturm, A., et al., 1997. Prostaglandin effects on the contractility of bovine trabecular meshwork and ciliary muscle. Exp. Eye Res. 64 (3), 447e453. Liang, Y., Li, C., Guzman, V.M., et al., 2003. Comparison of prostaglandin F2alpha, bimatoprost (prostamide), and butaprost (EP2 agonist) on Cyr61 and connective tissue growth factor gene expression. J. Biol. Chem. 278 (29), 27267e27277. Lindsey, J.D., Kashiwagi, K., Boyle, D., 1996. Prostaglandins increase proMMP-1 and proMMP-3 secretion by human ciliary smooth muscle cells. Curr. Eye Res. 15 (8), 869e875. Nilsson, S., Drecoll, E., Lutjen-Drecoll, E., et al., 2006. The prostanoid EP2 receptor agonist butaprost increases uveoscleral outflow in the cynomologous monkey. Invest. Ophthalmol. Vis. Sci. 47 (9), 4042e4049. Prasanna, G., Fortner, J., Xiang, C., Zhang, E., Carreiro, S., Anderson, S., Sartnurak, S., Wu, G., Gukasyan, H., Niesman, M., Nair, S., Rui, E., Lafontaine, J., Almaden, C.D., Wells, P., Krauss, A., 2009. Ocular pharmacokinetics and hypotensive activity of PF-04475270, an EP4 p.ostaglandin agonist in preclinical models. Exp. Eye Res. 89 (5), 608e617. Regan, J.W., 2003. EP2 and EP4 receptor signaling. Life Sci. 74, 143e153. Richter, M., Krauss, A.H., Woodward, D.F., 2003. Morphological changes in the anterior eye segment after long-term treatment with different receptor selective prostaglandin agonists and a prostamide. Invest. Ophthalmol. Vis. Sci. 44 (10), 4419e4426. Rosch, S., Ramer, R., Brune, K., et al., 2005 Dec 16. Prostaglandin E2 induces cyclooxygenase-2 expression in human non-pigmented ciliary epithelial cells through activation of p38 and p42/44 mitogen-activated protein kinases. Biochem. Biophys. Res. Commun. 338 (2), 1171e1178. Schachar, R.A., Raber, S., Courtney, R., Zhang, M., Bosworth, C., 2010. Dose-Escalating, Double-Masked, Vehicle-Controlled Trial of the IOP-Reducing Effect of EP2 Agonist, Taprenepag Isopropyl (PF-04217329). ARVO. poster #175/A398. Schlotzer-Schrehardt, U., Zenkel, M., Nusing, R.M., 2002. Expression and localization of FP and EP prostanoid receptor subtypes in human ocular tissues. Invest. Ophthalmol. Vis. Sci. 43, 1475e1487. Stumpff, F., Boxberger, M., Krauss, A., et al., 2005. Stimulation of cannabinoid (CB1) and prostanoid (EP2) receptors opens BKCa channels and relaxes ocular trabecular meshwork. Exp. Eye Res. 80 (5), 697e708. Toris, C.B., Zhan, G.-L., Wang, Y.-L., Zhao, J., McLaughlin, M.A., Camras, C.B., Yablonski, M.E., 2000. Aqueous humor dynamics in monkeys with laserinduced glaucoma. J. Ocul. Pharmacol. Ther. 16, 19e27. Weinreb, R.N., Kashiwagi, K., Kashiwagi, F., 1997. Prostaglandin increase matrix metalloproteinase release from human ciliary smooth muscle cells. Invest. Ophthalmol. Vis. Sci. 38 (13), 2772e2780. Woodward, D.F., Bogardus, A.M., Donello, J.E., et al., 1995. Molecular characterization and ocular hypetensive properties of prostanoid EP2 receptor. J. Ocul. Pharmacol. Ther. 11 (3), 447e454.
