Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491 DOI 10.1007/s00210-013-0850-7

ORIGINAL ARTICLE

The in vitro pharmacological profile of TD-1211, a neutral opioid receptor antagonist Pamela R. Tsuruda & Ross G. Vickery & Daniel D. Long & Scott R. Armstrong & David T. Beattie

Received: 24 October 2012 / Accepted: 6 March 2013 / Published online: 4 April 2013 # Springer-Verlag Berlin Heidelberg 2013

Abstract The clinical efficacy of opioid receptor antagonists for the treatment of opioid-induced constipation (OIC) is established. Peripherally selective antagonists are intended to provide OIC symptom relief without compromising the analgesic effects of centrally penetrant opioid agonists. We describe the in vitro profile of a novel opioid receptor antagonist, TD1211, at recombinant (human μ and δ, and guinea pig κ) and rodent native opioid receptors. TD-1211 bound with high affinity to human recombinant μ and δ, and guinea pig κ receptors expressed in CHO-K1 cells (pKd =9.7, 8.6, and 9.9, respectively). The in vitro receptor selectivity of TD-1211 (μ≈κ>δ) is similar to that for the peripherally-selective opioid receptor antagonist methylnaltrexone, but contrasts with the μ selectivity of alvimopan. Functionally, TD-1211 behaved as an antagonist at all three receptor types in both recombinant expression systems (pKb =9.6, 8.8 and 9.5, at μ, δ, and κ, respectively) and rodent native tissue preparations (μ and κ pA2s=10. 1 and 8.8, respectively (guinea pig ileum), and δ pKb =8.4 (hamster vas deferens)). TD-1211 displayed a high degree of selectivity for opioid receptors over a broad panel of cellular targets. These in vitro data justified investigation of the Electronic supplementary material The online version of this article (doi:10.1007/s00210-013-0850-7) contains supplementary material, which is available to authorized users. P. R. Tsuruda (*) : R. G. Vickery Department of Molecular and Cellular Biology, Theravance, Inc., 901 Gateway Boulevard, South San Francisco, CA 94080, USA e-mail: [email protected] D. D. Long Department of Medicinal Chemistry, Theravance, Inc., 901 Gateway Boulevard, South San Francisco, CA 94080, USA S. R. Armstrong : D. T. Beattie Department of Pharmacology, Theravance, Inc., 901 Gateway Boulevard, South San Francisco, CA 94080, USA

preclinical in vivo activity of TD-1211 (Armstrong et al., Naunyn-Schmiedeberg’s Arch Pharm, 2013). Keywords TD-1211 . Opioid receptor . Constipation . Alvimopan . Methylnaltrexone Abbreviations OIC Opioid-induced constipation GI Gastrointestinal CNS Central nervous system HEPES 4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid DMSO Dimethyl sulfoxide DAMGO [D-Ala2, N-MePhe4, Gly-ol]enkephalin DPDPE [D-Pen2,5]enkephalin U69593 N-Methyl-2-phenyl-N-[(5R,7S,8S)7-(pyrrolidin-1-yl)-1-oxaspiro[4.5]dec-8-yl] acetamide EFS Electrical field stimulation BSA Bovine serum albumin CHO Chinese hamster ovary IA Intrinsic activity

Introduction Opioid receptor agonists such as morphine continue to play a critical role in chronic cancer and non-cancer pain control (Walsh 2000; Chou et al. 2009). However, despite their effectiveness, opioid analgesics have significant drawbacks, notably the development of analgesic tolerance and physical dependence, sedation, respiratory depression, and opioidinduced bowel dysfunction (OBD) (Walsh 1990). OBD is characterized by constipation, delayed gastric emptying, abdominal discomfort, and nausea, and affects up to 60 % of patients on chronic opioid analgesic treatment for cancer or non-cancer pain (Brock et al. 2012; De Luca and Coupar

480

1996; Kurz and Sessler 2003; Lucero et al. 2006; Pappagallo 2001; Vanegas et al. 1998). Opioid-induced constipation (OIC), which is the most recognized and meaningful feature of OBD, consists of infrequent, hard stools, difficulty in defecating, abdominal bloating and discomfort, and a sense of incomplete emptying of the bowels (Pappagallo 2001; Walters and Montagnini 2010). OIC can be so severe that patients taper or even discontinue opioid analgesic intake, and as a result, pain control is often suboptimal. In the human gastrointestinal (GI) tract, μ receptors are mainly localized on myenteric and submucosal neurons, and on immune cells in the lamina propria (Sternini et al. 2004). OIC arises from an interaction of opioid agonist with the myenteric and submucosal μ receptors, leading to inhibition of peristalsis, reduced secretion, and sphincter dysfunction (Brock et al. 2012; De Luca and Coupar 1996; Wood and Galligan 2004). Although opioid agonist action in the central nervous system (CNS) is able to influence GI function (Porreca et al. 1986), a direct interaction at the level of the enteric nerves plays the dominant role in influencing motility (Brock et al. 2012; Viscusi et al. 2009). Thus, in humans, systemic doses of morphine that produce constipation are lower than those required for analgesia or other CNSmediated activities. Furthermore, central administration of morphine can provide clinically useful levels of analgesia, with little or no constipation (Lazorthes et al. 1995; Mercadante 1999; Seiwald et al. 1996). Two μ receptor antagonists, naltrexone and naloxone, have been shown to attenuate OIC in clinical studies. However, as these agents readily cross the blood brain barrier, attenuation of opioid-induced analgesia and provocation of an opioid behavioral withdrawal syndrome can occur (Pappagallo 2001; Culpepper-Morgan et al. 1992). The susceptibility of naloxone to first-pass metabolism in humans is claimed to provide some GI selectivity following oral (PO) administration, but careful dose titration is required (Culpepper-Morgan et al. 1992; Meissner et al. 2000). Two more recent μ receptor antagonists, alvimopan (Entereg® ; Cubist Pharmaceuticals) and methylnaltrexone (Relistor®; Salix Pharmaceuticals, Inc.) have reduced CNS penetration (i.e., high peripheral selectivity). Alvimopan is approved for the treatment of post-operative ileus (POI), a condition in which there is stasis of the GI tract for several days or weeks after abdominal surgery. Methylnaltrexone is the only agent currently approved in the US for the treatment of OIC, but may be best suited as a rescue medication due to its subcutaneous route of delivery and relatively short half-life. Data from clinical studies with alvimopan and methylnaltrexone demonstrated that in humans, CNS penetration is not a prerequisite for effective treatment of GI dysfunction (Kraft et al. 2010; Deibert et al. 2010). Here, we describe the in vitro pharmacological properties of the novel opioid receptor antagonist, TD-1211 (3-endo-(8{2-[cyclohexylmethyl-((S)-2,3-dihydroxy-propionyl)amino]

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491

Fig. 1 TD-1211 (3-endo-(8-{2-[cyclohexylmethyl-((S)-2,3-dihydroxypropionyl)amino]ethyl}-8-aza-bicyclo[3.2.1]oct-3-yl) benzamide

ethyl}-8-aza-bicyclo [3.2.1]oct-3-yl) benzamide; Fig. 1). In this study, assays were selected not only to provide a thorough characterization of the pharmacology of TD-1211, and for comparison, other opioid antagonists, but also to afford some insight into its ability to impact GI physiology and OIC pathophysiology. The apparent binding affinity, functional activity, and binding kinetics of TD-1211 were examined at recombinant μ, δ and κ receptors, expressed in cell lines, as described previously (Beattie et al. 2007). The influence of TD-1211 on opioid receptor-mediated responses was studied in two isolated tissue preparations commonly used to investigate opioid receptor pharmacology (Kosterlitz and Watt 1968; McKnight et al. 1985). The guinea pig ileum (Kosterlitz and Watt 1968) was used to evaluate the μ and κ receptor activities of TD-1211, and the hamster vas deferens (McKnight et al. 1985) for δ. The guinea pig ileum is particularly useful as it has been utilized previously to investigate the influence of acute and chronic exposure to opioid receptor agonists (Collier et al. 1981; Lujan and Rodriguez 1981). The majority of OIC patients are likely to commence treatment with a peripherally-selective opioid receptor antagonist on a background of chronic opioid analgesic therapy, and so it was considered of value to determine the activity of TD-1211 using ilea from guinea pigs pretreated chronically with morphine. The off-target activity of TD-1211 was evaluated across a broad panel of receptors, ion channels, and enzymes to explore its selectivity for opioid receptors. The in vivo activity of TD-1211 is described in the accompanying manuscript.

Materials TD-1211, alvimopan, ADL 08–0011 (active alvimopan metabolite), and methylnaltrexone were synthesized at Theravance, Inc. [3H]TD-1211 was synthesized at Vitrax (Placentia, CA). All other compounds were purchased from Sigma-Aldrich. Each compound was dissolved at 10 mM in DMSO, and then diluted serially in 50 mM HEPES (pH 7.4) with 0.1 % BSA for in vitro recombinant receptor assays or sterile water for studies with animal tissue. Membranes prepared from CHO-K1 cells stably transfected with the human μ receptor were either purchased (Perkin-Elmer, Boston, MA) or prepared in-house (Beattie et al. 2007) from a CHO-K1 cell line expressing the human μ receptor obtained from Dr. George Uhl (NIH). Membranes prepared from CHO-K1 cells stably transfected with the human δ receptor were purchased

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491

from Perkin-Elmer. A CHO-K1 cell line stably transfected with the guinea pig κ receptor (95 % sequence identity to the human κ receptor) was generated in-house and described previously (Beattie et al. 2007).

Methods In vitro binding and functional opioid receptor activity Radioligand binding assays Radioligand binding assays were performed as described previously (Beattie et al. 2007). Briefly, binding assays were conducted in 96-well polypropylene assay plates in a total assay volume of 200 μl containing human μ (4 μg), human δ (2 μg), or guinea pig κ (20–30 μg) receptor membrane protein. In all cases, the final assay buffer was 50 mM HEPES pH 7.4, containing 0.025 % BSA. Displacement assays for determination of pKi values were conducted at room temperature, using a 90-min incubation period; the radioligand ([3H] diprenorphine, [3H]DPN) concentrations were 0.5 nM (human μ receptor), 1.3 nM (human δ receptor), or 0.7 nM (guinea pig κ receptor). Eleven different concentrations (10 pM to 100 μM) of compound were tested. Binding data were analyzed by nonlinear regression analysis with the GraphPad Prism™ Software package (GraphPad Software, Inc., San Diego, CA) using the three-parameter model for one site competition. The Ki values for test compounds were calculated in Prism from the best-fit IC50 values using the Cheng Prusoff equation (Cheng and Prusoff 1973), and the negative decadic logarithm of the Ki value, pKi, is reported. Guanine nucleotide binding assays Membrane-based GTP binding assays were performed using either [35S]GTPγS for δ or κ receptors or in DELFIA GTP Binding Assays (Perkin-Elmer) for μ receptors. Previously, [35S]GTPγS binding data for standard compounds at μ receptors were reported (Beattie et al. 2007), and in general, the potencies and intrinsic activities (IA) are consistent between the two assay formats. Compound dilution and preparation of opioid receptor membranes were conducted as described above. The DELFIA GTP assays were performed in AcroWell 96-well plates in a total volume of 100 μl containing μ receptor membrane protein (5 μg), test compound (10 pM–20 μM), GDP (1 μM), and europiumlabeled GTP (Eu-GTP; 10 nM). The GTPγS binding assays were performed in a total volume of 200 μl in 96-well polypropylene assay plates containing δ or κ receptor membrane protein (10 μg), test compound (10 pM–20 μM), GDP (10 μM), and [35S]GTPγS (0.1 nM). In all cases, final buffer composition was HEPES (50 mM, pH 7.4), MgCl2

481

(10 mM), NaCl (50 mM for μ and κ, or 100 mM for δ), and 0.0125 % BSA. Reactions were incubated at room temperature for 30 min prior to termination by filtration. Concentration–response curves were generated from bound Time Resolved Fluorescence (TRF) or cpm values for both test compounds and receptor-selective agonists (DAMGO, DPDPE, or U69593 for μ, δ, or κ receptors, respectively). pEC50 values and the top and bottom of the curve were determined using a three-parameter sigmoidal dose–response algorithm in GraphPad Prism™. IAs were expressed as a percent of the maximal response evoked by each receptor-selective agonist. Schild regression analysis — TD-1211 Assays were performed in 96-well polypropylene assay plates in a final assay volume of 200 μl containing μ (10 μg), δ (10 μg), or κ (30 μg) receptor membrane protein, GDP (10 μM), [35S]GTPγS (0.1 nM), and 11 different concentrations (10 pM–100 μM) of agonist (DAMGO, DPDPE, or U69593 for μ, δ, or κ receptors, respectively) in the absence or presence of three different concentrations (0.7 to 72 nM) of TD-1211. The final buffer composition was HEPES (50 mM, pH 7.4), MgCl2 (10 mM), NaCl (25 mM for μ receptors or 100 mM for δ and κ receptors), and 0.04 % BSA. The reactions were pre-incubated with TD-1211 (in the absence of agonist) at room temperature for 20 min prior to the addition of the receptor-selective agonist. The assays were incubated for 30 min prior to termination by rapid filtration. Agonist concentration–response curves, in the absence and presence of increasing concentrations of TD-1211, were analyzed using a three-parameter sigmoidal dose–response algorithm in GraphPad Prism™. Values for the curve maximum and minimum were shared across the individual curves. In the presence of antagonist, the concentration of agonist required to produce a response equal to its EC50 in the absence of antagonists was determined (equi-effective concentration). The dose ratios (DR) were then calculated using the equieffective concentration of agonists divided by the EC50 value in the absence of antagonist. A plot of decadic logarithm (DR-1) as a function of antagonist concentration was constructed to determine pA2 or pKb values by Schild regression analysis (Arunlakshana and Schild 1959). If the slope of the Schild regression was found not to be significantly different from unity, it was analyzed with the slope constrained to unity, and the resulting Y-intercept was reported as a pKb value. [3H]TD-1211 binding assays [3H]TD-1211 binding assays were performed in 96-well polypropylene assay plates in a total assay volume of 200 μl containing μ (3 μg) or δ (3 μg) receptor membrane

482

protein, or a 1-ml assay volume containing κ (20–60 μg) receptor membrane protein. Saturation binding studies for determination of pKd values were performed using 12 different concentrations of [3H]TD-1211 (3 pM–4.7 nM for μ, 15 pM–26 nM for δ receptors, and 2 pM–2 nM for κ receptors). The association rate constants were determined by incubating increasing concentrations of [3H]TD-1211 with membranes expressing μ, δ, or κ receptors, for varying lengths of time. Bound radioactivity was plotted against time and analyzed using a one-phase association kinetic equation in GraphPad Prism™. The observed association rate constants (kobs) were obtained and plotted as a function of [3H]TD-1211 concentration. These data were fit to a linear function where the slope of the line is the radioligand association rate constant (kon). To determine the dissociation rate constants for [3H]TD-1211, membranes expressing μ, δ or κ receptors were incubated with [3H]TD-1211 for 1 h at room temperature. Naloxone (10 μM) was added at various times and the binding reactions then terminated by rapid filtration. Bound radioactivity data were plotted as a function of time and analyzed using a one-phase exponential decay model in GraphPad Prism to determine the radioligand dissociation rate constant (koff). Opioid receptor selectivity Broad panel screening The off-target activity of TD-1211 was examined in binding and functional enzyme assays performed at Cerep (Paris, France), a contract research organization (see catalog for assay conditions). Radioligand binding or enzyme inhibition studies were conducted using cell lines (recombinant or endogenous target expression), animal tissue or purified enzymes. The percent inhibition of specific radioligand binding or enzymatic activity was measured at a single concentration of TD-1211 (1 μM). hERG and sodium channel activity The investigation of off-target activity of TD-1211 included an analysis of the ability of TD-1211 to affect human ether-àgo-go-related gene (hERG) potassium and sodium channel currents. Characterization of activity on hERG channel currents is of particular importance (mandated by regulatory agencies); hERG channel inhibition is associated with life threatening cardiac dysrhythmias, such as Torsades de Pointe (Lagrutta et al. 2008). Whole-cell electrophysiological recordings from CHO-K1 cells stably transfected with hERG were conducted as described previously (Smith et al. 2008). Cisapride (20 nM) served as a positive control. Whole-cell electrophysiological protocols for the measurement of sodium channel activity were similar to previous descriptions (Smith

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491

et al. 2006). Mexiletine (300 μM) and lidocaine (300 μM) were used as positive controls for Nav1.2a and Nav1.5, respectively. Ex vivo functional opioid receptor activity All animal experiments were conducted in accordance with the principles of laboratory animal care provided by the Institutional Animal Care and Usage Committee of Theravance, Inc. Ilea and vasa deferentia were dissected from adult, male, Dunkin Hartley guinea pigs (200–300 g, Harlan) and Golden Syrian hamsters (85–100 g; Charles River), respectively. In the majority of experiments, the effects of test agents were studied in tissue from untreated animals. In some experiments, however, ilea were dissected from guinea pigs pretreated chronically with morphine, as previously described (Beattie et al. 2007). Segments of ileum (20 cm in length, 10–30 cm caudal to the ileocecal junction) or the vasa deferentia were removed, and placed in Krebs–Henseleit physiological buffer, containing (in mM): KCl (4.7), KH2PO4 (1.2), MgSO4 anhydrous (1.2), NaCl (118.1), D-glucose (11.1), NaHCO3 (25.0), CaCl2 (2.6), and indomethacin (0.001; to inhibit prostaglandin synthesis). Segments of ileum (approximately 4 cm in length) and the vasa deferentia were mounted on stimulation electrode assemblies, under a tension of 1 and 0. 2 g, respectively, in a 10-ml tissue bath filled with Krebs– Henseleit buffer (37 °C). Changes in isometric force were recorded (Beattie et al. 2007). Guinea pig ileum Tissues were subjected to one of three different treatment protocols (Beattie et al. 2007): (a) Electrical field stimulation (EFS) was initiated (10 V, a just maximal voltage), 1 ms pulse width, 0.1 Hz frequency), and cumulative concentration response curves were constructed to test compounds (concentrations increasing at half log increments from 0.001 to 3 μM). The potency, with respect to augmentation or attenuation of the EFS evoked contractions, was analyzed by iterative curve fitting to a sigmoidal dose– response algorithm using GraphPad Prism™: Y = Bottom + (Top − Bottom)/(1 + 10(LogEC50 − X) * Slope), where X is the log [compound (M)], and Y is the response (starting at the Bottom of the curve and going to the Top with a sigmoid shape). The mean potency (pEC50 or pIC50, i.e., the negative log of the molar concentration associated with 50 % of the maximum response) and IA (i.e., maximum % change of the EFS evoked contraction) were determined. (b) EFS was initiated, and a cumulative concentration response curve was constructed to either endomorphin-1 (0.0001–3 μM) or U69593 (0.0001–30 nM), selective μ or κ receptor agonists, respectively. EFS was then terminated,

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491

and tissues were washed. EFS was re-initiated, followed by the addition of a single concentration of test compound or vehicle. After 15 min, the endomorphin-1 or U69593 concentration response curve was repeated, to maximal effect. The concentration ratio (with respect to the agonist concentration effect curves in the absence and presence of test compound) was calculated, and a pA2 (Arunlakshana and Schild 1959) or pKb (from a single antagonist concentration) value was calculated to quantify antagonist potency. (c) To evaluate the effects of test compounds on the spontaneous activity of the ileum following chronic treatment of guinea pigs with morphine, tissues were exposed to a single concentration of test compound (0.001–1 μM) or vehicle. The spontaneous mechanical activity was integrated (Acknowledge™ Waveform Acquisition and Analysis software), and the contractile area during the 15 min after challenge with the test compound or its vehicle was expressed as a percentage change from that recorded in the 15-min period immediately preceding the application.

483

TD-1211 displaced [3H]DPN binding to human recombinant μ and δ, and guinea pig recombinant κ receptors in a concentration-dependent manner, with equilibrium pKi values of 9.8±0.2, 8.8±0.1, and 9.9±0.2, respectively (Fig. 2 and Table 1). When compared to other reported peripherally selective compounds, the rank order of apparent binding affinities was TD-1211 > alvimopan = ADL 08–0011 > methylnaltrexone at both μ and δ, and TD-1211 > alvimopan > ADL 08–0011= methylnaltrexone at κ receptors. The equilibrium dissociation constants (pKd) of [3H]TD-1211, determined by saturation binding analysis, were 9.7±0.1, 8.6±0.1, and 9.9±0.2 at μ, δ and κ

Hamster vas deferens EFS was initiated (ten pulses at 10 Hz every 30 s, 10 V, 1 ms), and a cumulative concentration response curve (concentrations increasing at log increments, from 0.001 to 10 μM) was constructed to [D-Pen2,5] enkephalin (DPDPE), a selective δ receptor agonist (Calderon et al. 1994). Upon completion of the concentration response curve, EFS was terminated, and tissues were washed five times every 2 min and twice more at 10-min intervals. After another 30 min, EFS was reinitiated followed by the addition of a single concentration of test compound or vehicle, after 10 additional min. Following a 15-min equilibration period, the DPDPE concentration response curve was repeated. The potency of DPDPE, with respect to attenuation of the EFS evoked contractions, in the absence and presence of test drug, was analyzed by iterative curve fitting to a logistic equation using GraphPad Prism™ as described above. The ratio of the DPDPE EC50 values in the absence and presence of test compound was calculated, and a pKb value was calculated to quantify antagonist potency.

Results Opioid receptor binding and functional potency The apparent in vitro binding affinity (pKi) and potency (pEC50) of TD-1211 were determined using membrane preparations from CHO-K1 cells expressing recombinant μ, δ and κ receptors. Other potent standard opioid receptor ligands were examined for comparison, including alvimopan and its metabolite (ADL 08–0011), naltrexone, methylnaltrexone, loperamide, morphine, DAMGO, DPDPE, and U69593.

Fig. 2 Determination of TD-1211 binding inhibition (pKi values) at human recombinant μ (a), human recombinant δ (b), or guinea pig recombinant κ receptors (c) (each n=3). Standard opioid agonists DAMGO, DPDPE, and U69593 and antagonists naloxone, ADL 08–0011, and methylnaltrexone are shown for reference

484

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491

Table 1 Apparent binding affinities (pKi values, n ≥ 17) and functional activities (pEC50 and IA, as a percentage of the respective receptor selective agonists, DAMGO, DPDPE, U69593, n ≥ 3) at human recombinant μ and δ, or guinea pig recombinant κ receptors Compound

TD-1211 Alvimopan ADL 08-0011 Naloxone Naltrexone Methylnaltrexone Loperamide Morphine DAMGO DPDPE U69593

μ receptor

δ receptor

κ receptor

pKi

pEC50

IA

pKi

pEC50

IA

pKi

pEC50

IA

9.8±0.2 9.6±0.1 9.6±0.1 9.2±0.1 9.9±0.1 8.2±0.1 9.0±0.1 7.9±0.2 7.9±0.2 5.3±0.2 <5.2

n.d. n.d. n.d. n.d. n.d. n.d. 9.0±0.2 8.0±0.2 8.3±0.2 5.9±0.3 n.d.

−3 (−6, −1) −7 (−10, −5) −17 (−20, −15) 7 (4, 10) 11 (8, 13) 19 (16, 21) 119 (112, 126) 112 (106, 117) 97 (93, 102) 98 (92, 104) n.d.

8.8±0.1 8.3±0.2 7.8±0.2 7.8±0.1 8.2±0.1 6.3±0.1 6.7±0.1 6.3±0.1 5.9±0.1 7.9±0.2 <5.0

n.d. n.d. n.d. n.d. n.d. n.d. 6.0±<0.1 5.8±0.1 5.6±0.2 8.1±0.2 5.2±0.7

−16 (−18, −13) −15 (−30, 1) −24 (−32, −16) −9 (−19, 1) −8 (−15, 0) 3 (−4, 10) 119 (100, 138) 88 (73, 103) 84 (69, 99) 109 (103, 115) 42 (28, 56)

9.9±0.2 8.5±0.2 7.7±0.1 8.7±0.1 9.4±0.1 7.7±0.1 6.6±0.2 6.8±0.2 5.8±0.2 <5.1 8.1±0.2

n.d. n.d. n.d. n.d. n.d. n.d. Not tested 5.6±0.5 n.d. n.d. 7.0±0.2

8 (4, 13) 10 (2, 19) 6 (3, 9)a 10 (5, 16) 18 (7, 30) 7 (0, 14) 53 (0, 105) 9 (2, 16) 5 (0, 10) 108 (97, 119)

pEC50 values are reported as the mean ± SEM, and mean IA values are reported with the 95 % confidence interval a

IA value for ADL 08–0011 at κ receptor reported by Beattie et al. (2007)

receptors, respectively (Fig. S1), in good agreement with the TD1211 pKi values at each receptor. The functional activity of TD-1211 was determined in guanine nucleotide binding assays, in which the ability of a test compound to evoke opioid receptor-mediated GTP exchange was evaluated. TD-1211 did not evoke receptormediated increases in bound nucleotide (Eu-GTP or [35S] GTPγS) in membranes expressing the human recombinant μ and δ, or guinea pig recombinant κ receptor. The IA (with 95 % confidence interval) of TD-1211 at the human μ receptor, relative to the full agonist DAMGO, was −3 (−6, −1), consistent with neutral antagonism. Similar observations were made with alvimopan whereas ADL 08–0011 displayed some inverse agonist activity, and methylnaltrexone showed some partial agonist activity. At the human δ and guinea pig κ receptor subtypes, the IA values (with 95 % confidence interval) for TD-1211 were −16 (−18, −13) relative to DPDPE, and 8 (4, 13) relative to U69593, respectively (Fig. 3 and Table 1). Similar profiles were observed with ADL 08–0011 and alvimopan at both δ and κ receptors. By contrast, methylnaltrexone did not show significant activity at either δ or κ receptors. The functional activities of these standard compounds are consistent with results published previously (Beattie et al. 2007). TD-1211 antagonized DAMGO-, DPDPE-, and U69593-stimulated GTPγS binding by μ, δ, and κ receptors, respectively. Increasing concentrations of TD1211 produced parallel rightward-shifts in the agonist concentration response curves, and Schild regression analyses yielded pKb values of 9.6±0.1, 8.8±<0.1, and 9.5± 0.1 for TD-1211 against DAMGO, DPDPE, and

U69593, respectively (Fig. 4). These pKb values were consistent with both the equilibrium binding pKi and pKd values at each receptor subtype. Rate constants for [3H]TD-1211 binding to μ, δ and κ receptors were determined in association and dissociation kinetic studies (Fig. 5; Table 2). The association rate constant for [3H]TD-1211 binding to μ receptors was 0.18 nM−1 min−1 and the dissociation rate constant was 0.038 min−1 (t1/2 =18 min). Dissociation of [3H]TD1211 was faster from both δ and κ receptors (koff =0.37 and 0.20 min−1, respectively), and the corresponding association rate constants were 0.088 nM−1 min−1 at δ and 1.5 nM −1 min−1 at κ receptors. The calculated kinetic pKd values (pKd =9.7 at μ, 8.4 at δ, and 9.9 at κ) for [3H]TD-1211 were well matched to the equilibrium pKd and pKi values. Opioid receptor selectivity The selectivity of TD-1211 for opioid receptors was examined using binding and functional assays at 80 different receptors, ion channels, and enzymes (data not shown). When tested at 1 μM (i.e., a concentration 6,300-fold higher than its μ Ki value), TD-1211 did not inhibit radioligand binding or functional activity at any of the tested targets. The magnitude of hERG potassium tail currents was unaffected by the presence of TD-1211 (3 μM). At 1 μM, TD-1211 had no effect on the magnitude of Na+ currents in cell lines heterologously expressing the human brain (Nav1.2) or cardiac (Nav1.5) voltage-gated sodium channels.

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491

485

and ADL 08–0011 (each at 1 nM–3 μM) produced concentration dependent increases in amplitude (Table 3). Alvimopan and ADL 08–0011 were more potent than methylnaltrexone with respect to the change in EFS induced contractile amplitude (mean pEC50 or pIC50 values of 8.0, 7.5, or 6.4, respectively; Table 3). The mean IA values of methylnaltrexone, alvimopan, and ADL 08–0011, with respect to attenuation or augmentation of the EFS mediated contractile amplitude, were 37 %, 25 %, and 27 %, respectively (Table 3). Antagonism of agonist mediated attenuation of EFS evoked responses

Fig. 3 Functional potency and intrinsic activity of TD-1211 at human recombinant μ (a), human recombinant δ (b), or guinea pig recombinant κ receptors (c) (n = 3). Standard opioid agonists morphine, DPDPE, and U69593 and antagonists naloxone, ADL 08–0011, and methylnaltrexone are shown for reference

μ, δ and κ receptor activity in isolated tissue Influence of test agents on EFS evoked responses Cumulative addition of drug vehicle to the guinea pig ileum had no effect on the amplitude of the EFS evoked contractions. Endomorphin-1 (1 nM–3 μM) and U69593 (0.1 nM– 1 μM), μ and κ opioid agonists, respectively, produced a concentration dependent inhibition of the EFS evoked responses (Table 3). TD-1211 (1 nM–3 μM) had no effect on the EFS evoked responses, while methylnaltrexone (1 nM– 3 μM) was associated with a concentration dependent reduction in the EFS mediated contractions, and alvimopan

The μ receptor antagonist potencies of TD-1211, alvimopan, ADL 08–0011 and methylnaltrexone were determined by investigating their ability to inhibit the endomorphin-1mediated inhibition of EFS evoked contractions. Concentration-dependent rightward shifts in the agonist concentration response curve were produced by TD-1211, alvimopan, ADL 08–0011 and methylnaltrexone, relative to vehicle (Fig. 6). The rank order of antagonist potency (Table 4) was TD-1211 (pA 2 = 10.1) > alvimopan (pA 2 = 9.6) ≥ ADL 08–0011 (pA2 =9.4) > methylnaltrexone (pA2 =7.6). For each compound, the slope of the Schild plot was not significantly different from unity (data not shown). TD-1211 and ADL 08–0011 produced concentration dependent, parallel shifts in the endomorphin-1 concentration–response curve consistent with competitive antagonism (Fig. 6). The κ receptor antagonist potency of TD-1211, and for comparison, alvimopan, ADL 08-0011, and methylnaltrexone, was determined in the guinea pig ileum with respect to their ability to attenuate U69593-induced inhibition of EFS evoked contractions. Parallel rightward shifts in the agonist concentration response curve were produced by TD-1211, alvimopan, ADL 08–0011, and methylnaltrexone, but not by vehicle. The rank order of κ receptor antagonist potency (Table 4) was TD-1211 (pKb =8.8) > alvimopan (pA2 =8.4) > ADL 08–0011 (pA2 =7.2) > methylnaltrexone (pA2 =6.6). The functional selectivities of TD-1211, alvimopan, ADL 08–0011, and methylnaltrexone for the μ versus κ receptor were 20-, 16-, 158-, and 10-fold, respectively, in the guinea pig ileum (Table 4). The δ receptor agonist, DPDPE (1 nM–10 μM), inhibited, in a concentration dependent manner, the EFS evoked contractions of the hamster vas deferens (mean (±SEM) pIC50 =7. 2±0.1; n=30). TD-1211 (0.03 μM; n=7), alvimopan (0.1 μM; n=5), ADL 08–0011 (0.3 μM; n=5), and methylnaltrexone (10 μM; n=4) produced rightward shifts in the DPDPE concentration response curve, with mean pKb values of 8.4, 8.9, 7. 6, and 6.4, respectively (Table 4). TD-1211 was therefore 3fold less potent than alvimopan, and 6- or 100-fold more potent than ADL 08–0011 or methylnaltrexone, respectively,

486

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491

Fig. 4 TD-1211-mediated antagonism of opioid agoniststimulated GTPγS exchange: DAMGO activation of human recombinant μ (a), DPDPE activation of human recombinant δ (b), or U69593 activation of guinea pig recombinant κ receptors (c). Agonist concentration– response curves in the absence and presence of TD-1211 (left panels) were used to calculate dose ratios and fit pKb values by Schild regression analysis (right panels)

with regard to its δ receptor antagonist potency in this tissue. The functional selectivities of TD-1211, alvimopan, ADL 08– 0011, and methylnaltrexone for the guinea pig ileal μ versus hamster δ receptor were 50-, 5-, 63-, and 16-fold, respectively (Table 4). Effect of test agents in morphine-dependent tissue In ilea isolated from guinea pigs treated subcutaneously for 7 days with vehicle (10 μl/h), application of vehicle or TD1211 (1 nM–10 μM) to the tissue bath had no effect on the spontaneous contractile activity of the ileum (Fig. 7a). In contrast, methylnaltrexone (1 nM–10 μM) reduced, and alvimopan and ADL 08–0011 (both at 1 nM–10 μM) increased, in a concentration-dependent manner, the spontaneous contractile activity (Fig. 7a). In ilea isolated from

morphine-treated guinea pigs (45 mg/kg/day for 7 days), addition of vehicle had no effect on spontaneous contractile activity (Fig. 7b). TD-1211, alvimopan, ADL 08–0011, and methylnaltrexone each produced statistically significant increases in activity (Fig. 7b). TD-1211 was less potent than alvimopan, ADL 08–0011, or methylnaltrexone in this regard (mean (±SEM) pEC50 values of 6.1±0.4, 7.9±0.3, 8.2 ±0.4, and 6.7±0.5, respectively). Methylnaltrexone had a lower IA than the other test compounds.

Discussion TD-1211, a novel, opioid receptor antagonist, displays high binding affinity (pKi and pKd values) at recombinant μ, δ, and κ receptors, and modest selectivity (<10-fold) for μ and

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491

487

Fig. 5 Determination of binding kinetic constants for [3H]TD-1211 at human recombinant μ (a), δ (b), or guinea pig recombinant κ receptors (c). Fitted kobs values from individual association binding studies are plotted against radioligand concentration to determine association constants (left panels). A representative binding dissociation curve for each receptor is shown (right panels)

κ (over δ) receptors. This selectivity profile is similar to that of naloxone, naltrexone, and methylnaltrexone, but contrasts with that of alvimopan and ADL 08–0011 (>10-fold selectivity for μ over both δ and κ receptors). In the guinea pig isolated ileum preparation, TD-1211 antagonized opioid agonist-induced attenuation of EFS-evoked contractile activity, an effect consistent with inhibition of myenteric neurotransmission and acetylcholine release (Nakayama et al. 1990; Nishiwaki et al. 1998a). Inhibition of enteric acetylcholine release is an important mechanism underlying opioid agonist-induced reductions in motility (Nishiwaki et al. 1998b; Wood and Galligan 2004). TD-1211 was a potent μ and κ receptor antagonist in the guinea pig ileum, as indicated by the inhibition of endomorphin-1 and U69593induced responses against EFS. The ability of μ and κ activation to inhibit EFS-evoked acetylcholine release from myenteric neurons, and resultant contractile activity is established (Kromer et al. 1982; Nakayama et al. 1990). In terms of μ receptor antagonism, TD-1211 was approximately

3-, 5-, and 300-fold more potent than alvimopan, ADL 08– 0011, and methylnaltrexone, respectively, and was 2.5-, 40-, and 158-fold more potent with regards to κ receptor antagonism, respectively. TD-1211 was also a potent δ receptor antagonist in the hamster vas deferens (3-fold less potent than alvimopan, and 6- and 100-fold more potent than ADL 08– 0011 and methylnaltrexone, respectively). In general, the pKb values of TD-1211, alvimopan, ADL 08–0011 and methylnaltrexone in the hamster vas deferens were similar to the pKi values at the human recombinant δ receptor. The μ receptor antagonist potencies of TD-1211, alvimopan, and ADL 08–0011 in this study correlated well with the μ receptor binding affinities of the compounds. This correlation is consistent with the acknowledged role of the μ receptor in the guinea pig ileum and with the high μ receptor selectivity of endomorphin-1. The 20-fold selectivity of TD-1211 for μ receptors over κ receptors in the guinea pig ileum contrasts with its similar affinity for the receptors in the cell based studies. Like TD-1211, methylnaltrexone had a lower κ

488

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491

Table 2 Kinetic constants for [3H]TD-1211 binding to human recombinant μ and δ, or guinea pig recombinant κ receptors Receptor

kon (nM−1 min−1)

koff (min−1)

t1/2 (min)

pKd

μ δ κ

0.18 0.088 1.5

0.038 0.37 0.20

18 2 3

9.7 8.4 9.9

receptor antagonist potency (pA2 =6.6) in the isolated tissue preparation than would have been anticipated from its binding affinity at this receptor (pKi =7.6). The reason for these discrepancies is unclear, although in the membrane-based functional GTPγS binding-assay the pKb for TD-1211 at the κ receptor was consistent with its binding affinity. The significance, if any, of the difference in the μ/δ and μ/κ selectivities of TD-1211, alvimopan and ADL 08–0011 remains to be determined. Morphine and other commonly used opioid analgesics are μ receptor-selective agonists, and OIC is accepted to occur via an interaction with μ, rather than κ or δ receptors on enteric neurons (Brock et al. 2012; De Luca and Coupar 1996; Wood and Galligan 2004). It is possible, however, that κ or δ receptor activation by endogenous opioid ligands may regulate GI function in some circumstances. In human and guinea pig colonic circular smooth muscle, δ receptor activation reduces nonadrenergic, noncholinergic inhibitory responses (Kojima et al. 1994; Mako et al. 2000; Zagorodnyuk and Maggi 1994). In guinea pigs and mice, κ or δ receptor agonist-induced reductions in GI transit are inhibited by receptor-specific antagonists (Culpepper-Morgan et al. 1995; Pol et al. 1994). Conceivably, antagonism of κ or δ receptors by TD-1211 could confer an efficacy advantage over antagonists with higher μ receptor selectivity, although if only μ receptor antagonism is necessary for optimal clinical efficacy, interaction at κ or δ would be irrelevant. It has been demonstrated recently that antagonism of δ receptors can inhibit μ agonist-induced reductions in GI transit in mice (Wade et al. 2012). Based on its μ receptor antagonist properties in vitro, and appropriate in vivo characteristics (see

Table 3 Potencies (pEC50 or pIC50 values [mean ± SEM] for augmentation or attenuation, respectively) and IA values (mean and 95 % confidence intervals, negative: attenuation, positive: augmentation) of compounds in the electrically stimulated ileum Compound

pEC50 or pIC50 value

IA

Endomorphin-1 U69593 TD-1211 Methylnaltrexone Alvimopan ADL 08-0011

8.1±0.2 (n=12) 8.7±0.4 (n=4) No activity 6.4±0.3 (n=5) 8.0±0.6 (n=6) 7.5±0.5 (n=6)

−73 (−66, −80; n=12) −75 (−70, −80; n=4) −37 (−24, −49; n=5) 25 (5, 45; n=6) 27 (9, 45; n=6)

Fig. 6 Antagonism of μ receptor agonist (endomorphin-1)-induced attenuation of EFS evoked contractions by a TD-1211 (n=3 or 4), b alvimopan (n=4), c ADL 08–0011 (n=4 or 6), d methylnaltrexone (n= 3–7), and vehicle (n=6–8) in the guinea pig ileum

Armstrong et al. 2013), TD-1211 would be expected to prevent opioid agonist-induced interference of GI peristalsis, secretion and sphincter dysfunction via an interaction at the level of the myenteric and submucosal plexuses. Receptor kinetics can influence the duration of drug action. Slow dissociation of alvimopan (t1/2 =30 min) and buprenorphine (t1/2 =44 min) from μ receptors has been proposed to explain their extended duration of action, relative to their plasma half lives (Cassel et al. 2005; Yassen et

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491

Fig. 7 Differentiation of TD-1211 (n=3–7), alvimopan (n=4–8), ADL 08–0011 (n=4–7), methylnaltrexone (n=4–6), and vehicle (n=23 or 27) with respect to their influence on the spontaneous mechanical activity of the ileum isolated from guinea pigs treated chronically with a vehicle and b morphine (*p<0.05, ANOVA, followed by a Dunnett’s post-hoc test vs. vehicle)

al. 2005). The kinetics of [3H]TD-1211 binding to opioid receptors (at room temperature) were examined in association and dissociation binding studies. The dissociation half life of [3H]TD-1211 was fast at both δ and κ (t1/2 =2 min at δ and 3 min at κ, respectively), but slower at μ receptors (t1/2 =18 min). Dissociation of [3H]TD-1211 at μ receptors was slower than that of either naloxone or methylnaltrexone, but similar to values reported for alvimopan binding (in-house data not shown; Cassel et al. 2005). The influence of TD1211 binding kinetics on its activity in vivo remains to be determined.

489

It has been proposed that a compound with μ receptor neutral antagonist properties may not precipitate withdrawal to the same degree as an inverse-agonist when tested in preclinical animal models of morphine dependence (Wang et al. 2001). It follows that a neutral antagonist, or a weak partial agonist, may provide advantages with respect to the symptoms of GI withdrawal in opioid agonist-experienced patients. Whether or not a neutral antagonist and a partial agonist are able to prevent, to the same degree, GI dysmotility evoked by an opioid receptor full agonist in OIC patients, remains to be determined. Functionally, TD-1211 did not display agonist activity at human μ, human δ, or guinea pig κ receptors in nucleotide binding assays, but antagonized the activity of the receptor selective agonists DAMGO, DPDPE, and U69593. The absence of functional activity of TD-1211 at μ and κ receptors suggests that it is a neutral antagonist at these receptors, whereas — in contrast — it exhibited moderate inverse agonist activity at δ receptors. The μ receptor profile of TD1211 contrasted with that of ADL 08–0011 which showed inverse agonism at μ receptors, and methylnaltrexone, which behaved as a partial μ receptor agonist. Use of the guinea pig isolated ileum preparation permitted further assessment of the different IAs of TD-1211 and comparator agents, noted with the recombinant receptor functional assays. TD-1211 had no effect on EFS evoked contractions of the morphine-naïve ileum, consistent with a neutral antagonist profile. Methylnaltrexone produced a concentration dependent attenuation of the electrically evoked contractions of the ileum, activity consistent with μ receptor partial agonist activity. Alvimopan and ADL 08–0011 were associated with modest increases in the amplitude of EFS evoked contractions, actions indicative of opioid receptor inverse agonism or a non-opioid receptor interaction. These observations were consistent with studies reported previously (Beattie et al. 2007) and those in the experiments evaluating the effects of the test agents on the resting tension of the ileum. In morphine-naïve guinea pig ileum, TD-1211 had no effect, while methylnaltrexone decreased, and alvimopan and ADL 08–0011 increased spontaneous activity. In agreement with data from studies using naloxone (Collier et al. 1981; Lujan and Rodriguez 1981), chronic (7-day) morphine treatment of guinea pigs influenced the activity profile of TD-1211, alvimopan, ADL 08–0011 and methylnaltrexone. After

Table 4 Antagonist potencies (pKb or pA2 values; n ≥ 4) at the guinea pig ileal μ and κ, and hamster vas deferens δ receptors, and the μ/κ and μ/δ receptor selectivities, of TD-1211, alvimopan, ADL 08–0011, and methylnaltrexone Compound

μ receptor pA2

κ receptor pKb or pA2

δ receptor pKb

μ/κ selectivity

μ/δ selectivity

TD-1211 Alvimopan ADL 08-0011 Methylnaltrexone

10.1 9.6 9.4 7.6

8.8 8.4 7.2 6.6

8.4 8.9 7.6 6.4

20 16 158 10

50 5 63 16

490

morphine exposure, each compound was associated with increased spontaneous contractile activity, although alvimopan and ADL 08-0011 had a higher potency in this regard than TD-1211 or methylnaltrexone. Naloxone and naltrexone, which have no IA in morphine-naïve test systems, potentiate cyclic AMP levels in recombinant cell lines and contract the guinea pig ileum after morphine exposure (Collier et al. 1981; Lujan and Rodriguez 1981; Sadée et al. 2005; Wang et al. 1994). Data from several studies implicate a role of μ or κ receptors in the increased contractility that occurs in opioid agonist-dependent ileum (Beattie et al. 2007; Brent et al. 1993; Johnson et al. 1987; Valeri et al. 1995). It is evident from other studies that opioid receptors may be constitutively active, to a degree that is regulated by agonist exposure (Sadée et al. 2005; Wang et al. 2007). Agonist pretreatment not only increases constitutive activity of μ receptors, but also can convert neutral antagonist activity of naloxone and naltrexone into inverse agonist activity via suppression of basal signaling (Sadée et al. 2005; Wang et al. 2007). The excitatory effects noted for each compound in this study with morphine-dependent guinea pig ileum may represent such inverse agonist activity. It remains to be determined whether the apparent differences in the IAs of TD-1211, methylnaltrexone and alvimopan/ADL 08–0011 (i.e., neutral antagonism, partial agonism and inverse agonism, respectively) have any clinical relevance in OIC patients. Additionally, while activation of peripheral opioid receptors by endogenous opioid peptides, released, for example, from immune cells, may contribute to visceral analgesia (Kapitzke et al. 2005), its importance is unclear. Compounds with distinct opioid receptor IAs may modulate differentially these anti-nociceptive effects. A detailed comparison of clinical data, when available, for different opioid receptor antagonists, and further preclinical characterization of their pharmacological properties, are required. In this report, we have demonstrated that TD-1211 is a novel, potent opioid receptor antagonist in vitro. Several clinical studies with TD-1211 in healthy subjects and patients with OIC have recently been completed. TD-1211 was generally well tolerated and demonstrated robust efficacy (based on pre-defined increases in spontaneous, and complete and spontaneous bowel movements) in Phase 2b studies (Vickery et al. 2012), supporting its further clinical development. Acknowledgments The authors would like to thank Shanti Amagasu, Madhavi Ravindran, Courtney Gee, Ngoc Mai, Joey Yung, and Shana Johnson Rabidoux for technical support, and Uwe Klein for a thoughtful review of the manuscript.

References Armstrong SR, Campbell CB, Richardson CL, Vickery RG, Tsuruda PR, Long DD, Hegde SS, Beattie DT (2013) The in vivo pharmacology of TD-1211, a peripherally-selective opioid receptor

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491 antagonist in development for the treatment of opioid-induced constipation. Naunyn-Schmiedeberg’s Arch Pharm. doi:10.1007/ s00210-013-0844-5 Arunlakshana O, Schild HO (1959) Some quantitative uses of drug antagonists. Br J Pharmacol Chemotherap 14:48–58 Beattie DT, Cheruvu M, Mai N, O’Keefe M, Johnson-Rabidoux S, Peterson C, Kaufman E, Vickery R (2007) The in vitro pharmacology of the peripherally restricted opioid receptor antagonists, alvimopan, ADL 08–0011 and methylnaltrexone. Naunyn Schmiedeberg’s Arch Pharmacol 375:205–220 Brent PJ, Chahl LA, Cantarella PA, Kavanagh C (1993) The kappaopioid receptor agonist U50,488H induces acute physical dependence in guinea-pigs. Eur J Pharmacol 241:149–156 Brock C, Olesen SS, Olesen AE, Frokjaer JB, Andresen T, Drewes AM (2012) Opioid-induced bowel dysfunction. Drugs 72:1847–1865 Calderon SN, Rothman RB, Porreca F, Flippen-Anderson JL, McNutt RW, Xu H, Smith LE, Bilsky EJ, Davis P, Rice KC (1994) Probes for narcotic receptor mediated phenomena. 19. Synthesis of (+)-4[(alpha R)-alpha-((2S,5R)-4-allyl-2,5-dimethyl-1-piperazinyl)-3methoxybenzyl]-N,N-diethylbenzamide (SNC 80): a highly selective, nonpeptide delta opioid receptor agonist. J Med Chem 37:2125–3128 Cassel JA, Daubert JD, DeHaven RN (2005) [(3)H]Alvimopan binding to the micro opioid receptor: comparative binding kinetics of opioid antagonists. Eur J Pharmacol 520:29–36 Cheng Y, Prusoff WH (1973) Relationship between the inhibition constant (Ki) and the concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic reaction. Biochem Pharmacol 22:3099–3108 Chou R, Fanciullo GJ, Fine PG, Adler JA, Ballantyne JC, Davies P, Donovan MI, Fishbain DA, Foley KM, Fudin J, Gilson AM, Kelter A, Mauskop A, O’Connor PG, Passik SD, Pasternak GW, Portenoy RK, Rich BA, Roberts RG, Todd KH, Miaskowski C (2009) Clinical guidelines for the use of chronic opioid therapy in chronic noncancer pain. J Pain 10:113–130 Collier HOJ, Cuthbert NJ, Francis DL (1981) Model of opiate dependence in the guinea-pig isolated ileum. Br J Pharmacol 73:921– 932 Culpepper-Morgan JA, Holt PR, LaRoche D, Kreek MJ (1995) Orally administered opioid antagonists reverse both mu and kappa opioid agonist delay of gastrointestinal transit in the guinea pig. Life Sci 56:1187–1192 Culpepper-Morgan JA, Inurrisi CE, Portenoy RK, Foley K, Houde RW, Marsh F, Kreek MJ (1992) Treatment of opioid-induced constipation with oral naloxone: a pilot study. Clin Pharmacol Therap 52:90–95 Deibert P, Xander C, Blum HE, Becker G (2010) Methylnaltrexone: the evidence for its use in the management of opioid-induced constipation. Core Evid 4:247–258 De Luca A, Coupar IM (1996) Insights into opioid action in the intestinal tract. Pharmacol Ther 69:103–115 Johnson SM, Williams JT, Costa M, Furness JB (1987) Naloxoneinduced depolarization and synaptic activation of myenteric neurons in morphine-dependent guinea pig ileum. Neurosci 21:595– 602 Kapitzke D, Vetter I, Cabot PJ (2005) Endogenous opioid analgesia in peripheral tissues and the clinical implications for pain control. Ther Clin Risk Manag 1(4):279–297 Kojima Y, Takahashi T, Fujina M, Owyang C (1994) Inhibition of cholinergic transmission by opiates in ileal myenteric plexus is mediated by kappa receptor. Involvement of regulatory inhibitory G protein and calcium N-channels. J Pharmacol Exp Therap 268:965–970 Kosterlitz HW, Watt AJ (1968) Kinetic parameters of narcotic agonists and antagonists, with particular reference to N-allylnoroxymorphone (naloxone). Br J Pharmacol 33:266–277

Naunyn-Schmiedeberg's Arch Pharmacol (2013) 386:479–491 Kraft M, Maclaren R, Du W, Owens G (2010) Alvimopan (entereg) for the management of postoperative ileus in patients undergoing bowel resection. Pharm Therap 35:44–49 Kromer W, Steigemann N, Shearman GT (1982) Differential effects of SKF 10,047 (N-allyl-normetazocine) on peristalsis and longitudinal muscle contractions of the isolated guinea-pig ileum. NaunynSchmiedeberg’s Arch Pharmacol 321:218–222 Kurz A, Sessler DI (2003) Opioid-induced bowel dysfunction: pathophysiology and potential new therapies. Drugs 63:649–671 Lagrutta AA, Trepakova ES, Salata JJ (2008) The hERG channel and risk of dug-acquired cardiac arrhythmia: an overview. Curr Topics Med Chem 8:1102–1112 Lazorthes YR, Sallerin BA-M, Verdie J-CP (1995) Intracerebroventricular administration of morphine for control of irreducible cancer pain. Neurosurg 37:422–428 Lucero M, Von Scheele B, Blackard R, Milanova T, Bell T (2006) Amer Pain Soc Meeting, poster #959. Lujan M, Rodriguez R (1981) Pharmacological characterization of opiate physical dependence in the isolated ileum of the guineapig. Br J Pharmacol 73:859–866 Mako E, Ronai AZ, Adam G, Juhasz G, Ritter L, Lestar B, Crunelli V (2000) Modulation by GABA(B) and delta opioid receptors of neurally induced responses in isolated guinea-pig taenia coli and human colonic circular muscle. J Physiol Paris 94:135–138 McKnight AT, Corbett AD, Marcoli M, Kosterlitz HW (1985) The opioid receptors in the hamster vas deferens are of the delta-type. Neuropharmacol 24:1011–1017 Meissner W, Schmidt U, Hartmann M, Kath R, Reinhart K (2000) Oral naloxone reverses opioid-associated constipation. Pain 84:105–109 Mercadante S (1999) Problems of long-term spinal opioid treatment in advanced cancer patients. Pain 79:1–13 Nakayama S, Taniyama K, Matsuyama S, Ohgushi N, Tsunekawa K, Tanaka C (1990) Regulatory role of enteric mu and kappa opioid receptors in the release of acetylcholine and norepinephrine from guinea pig ileum. J Pharmacol Exp Therap 254:792–798 Nishiwaki H, Saitoh N, Nishio H, Takeuchi T, Hata F (1998a) Inhibitory effect of endomorphin-1 and −2 on acetylcholine release from myenteric plexus of guinea pig ileum. Jap J Pharmacol 78:83–86 Nishiwaki H, Saitoh N, Nishio H, Takeuchi T, Hata F (1998b) Relationship between inhibitory effect of endogenous opioid via mu-receptors and muscarinic autoinhibition in acetylcholine release from myenteric plexus of guinea pig ileum. Jap J Pharmacol 77:279–286 Pappagallo M (2001) Incidence, prevalence, and management of opioid bowel dysfunction. Am J Surgery 182:11S–18S Pol O, Ferrer I, Puig MM (1994) Diarrhea associated with intestinal inflammation increases the potency of mu and delta opioids on the inhibition of gastrointestinal transit in mice. J Pharmacol Exp Ther 270(1):386–391 Porreca F, Galligan JJ, Burks TF (1986) Central opioid receptor involvement in gastrointestinal motility. Trends Pharmacol Sci 7:104–107 Sadée W, Wang D, Bilsky EJ (2005) Basal opioid receptor activity, neutral antagonists, and therapeutic opportunities. Life Sci 76:1427–1437 Seiwald M, Alesch F, Kofler A (1996) Intraventricular morphine administration as a treatment possibility for patients with intractable pain. Wien Klin Wochenschr 108:5–8 Smith JA, Amagasu SM, Hembrador J, Axt S, Chang R, Church T, Gee C, Jacobsen JR, Jenkins T, Kaufman E, Mai N, Vickery RG

491 (2006) Evidence for a multivalent interaction of symmetrical, Nlinked, lidocaine dimmers with voltage-gated Na+ channels. Mol Pharmacol 69:921–931 Smith JAM, Beattie DT, Marquess D, Shaw J-P, Vickery RG, Humphrey PPA (2008) The in vitro pharmacological profile of TD-5108, a selective 5-HT4 receptor agonist with high intrinsic activity. Naunyn Schmiedebergs Arch Pharmacol 378:125–137 Sternini C, Patierno S, Selmer IS, Kirchgessner A (2004) The opioid system in the gastrointestinal tract. Neurogastroenterol Motil 16(suppl 2):3–16 Valeri P, Morrone LA, Romanelli L, Amico MC (1995) Acute withdrawal after bremazocine and the interaction between mu- and kappa-opioid receptors in isolated gut tissues. Br J Pharmacol 114:1206–1210 Vanegas G, Ripamonti C, Sbanotto A, De Conno F (1998) Side effects of morphine administration in cancer patients. Cancer Nursing 21:289–297 Vickery R, Li Y-P, Schwertschlag U, Singla N, Webster L, Canafax D (2012) TD-1211 demonstrates improvement in bowel movement frequency without impacting analgesia in a Phase 2b study of patients with opioid-induced constipation. Pain Week, Las Vegas, NV, September 5–8. Poster #121. http://www.painweek.org/ scientificposters/2012-accepted-abstracts Viscusi ER, Gan TJ, Leslie JB, Foss JF, Talon MD, Du W, Owens G (2009) Peripherally acting mu-opioid receptor antagonists and postoperative ileus: mechanisms of action and clinical applicability. Anaesth Analg 108:1811–1822 Wade PR, Palmer JM, McKenney S, Kenigs V, Chevalier K, Moore BA, Mabus JR, Saunders PR, Wallace NH, Schneider CR, Kimball ES, Breslin HJ, He W, Hornby PJ (2012) Modulation of gastrointestinal function by MuDelta, a mixed μ opioid receptor agonist/μ opioid receptor antagonist. Br J Pharmacol 167(5):1111–1125 Walsh TD (1990) Prevention of opioid side effects. J Pain Symptom Manag 5:362–367 Walsh TD (2000) Pharmacological management of cancer pain. Seminars in Oncology 27:45–63 Walters JB, Montagnini M (2010) Current concepts in the management of opioid-induced constipation. J Opioid Management 66:435– 444 Wang D, Bilsky EJ, Porreca F, Sadée W (1994) Constitutive mu opioid receptor activation as a regulatory mechanism underlying narcotic tolerance and dependence. Life Sci 54:339–350 Wang D, Raehal KM, Bilsky EJ, Sadée W (2001) Inverse agonists and neutral antagonists at mu opioid receptor (MOR): possible role of basal receptor signaling in narcotic dependence. J Neurochem 77:1590–1600 Wang D, Sun X, Sadée W (2007) Different effects of opioid antagonists on mu-, delta-, and kappa-opioid receptors with and without agonist pretreatment. J Pharmacol Exp Therap 321:544–552 Wood JD, Galligan JJ (2004) Function of opioids in the enteric nervous system. Neurogastroenterol Motil 16(suppl 2):17–28 Yassen A, Olofsen E, Dahan A, Danhof M (2005) Pharmacokinetic– pharmacodynamic modeling of the antinociceptive effect of buprenorphine and fentanyl in rats: role of receptor equilibration kinetics. J Pharmacol Exp Ther 313:1136–1149 Zagorodnyuk V, Maggi CA (1994) Electrophysiological evidence for different release mechanism of ATP and NO as inhibitory NANC transmitters in guinea-pig colon. Br J Pharmacol 112:1077–1082