Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Chem Biol Drug Des. Author manuscript; available in PMC 2013 November 1.
Published in final edited form as:
PMCID: PMC3465523

Modulation of Opioid Receptor Ligand Affinity and Efficacy Using Active and Inactive State Receptor Models


Mu opioid receptor (MOR) agonists are widely used for the treatment of pain; however chronic use results in the development of tolerance and dependence. It has been demonstrated that co-administration of a MOR agonist with a delta opioid receptor (DOR) antagonist maintains the analgesia associated with MOR agonists, but with reduced negative side effects. Using our newly refined opioid receptor models for structure-based ligand design, we have synthesized several pentapeptides with tailored affinity and efficacy profiles. In particular, we have obtained pentapeptides 8, Tyr-c(S-S)[DCys-1Nal-Nle-Cys]NH2, and 12, Tyr-c(S-S)[DCys-1Nal-Nle-Cys]OH, which demonstrates high affinity and full agonist behavior at MOR, high affinity but very low efficacy for DOR, and minimal affinity for the kappa opioid receptor (KOR). Functional properties of these peptides as MOR agonists/DOR antagonists lacking undesired KOR activity make them promising candidates for future in vivo studies of MOR/DOR interactions. Subtle structural variation of 12, by substituting D-Cys5 for L-Cys5, generated analog 13 which maintains low nanomolar MOR and DOR affinity, but which displays no efficacy at either receptor. These results demonstrate the power and utility of accurate receptor models for structure-based ligand design, as well as the profound sensitivity of ligand function on its structure.

Keywords: Delta opioid receptor, G protein-coupled receptors, mixed efficacy ligand, mu opioid receptor, opioid, peptide, structure-based design


The recognition that a specific receptor often plays a pivotal role in a disease state shifted the drug discovery paradigm toward a “one disease, one target” approach. The driving force behind this shift was the idea that the more specific a drug, the fewer negative side effects it will elicit. Recently, however, it has been recognized that the simultaneous modulation of multiple targets may generate a more desirable drug profile, in some cases even reducing the development of negative side effects (1, 2). This concept is illustrated in the field of opioid analgesics, where the co-administration of a mu opioid receptor (MOR) agonist with a delta opioid receptor (DOR) antagonist provides all the expected analgesia of a MOR agonist, but with reduced negative side effects, such as constipation and respiratory depression and, more interestingly, reduced tolerance and dependence liabilities (37), features that limit the clinical use of opioid analgesics (8).

Our lab and others (618) have explored the development of mixed efficacy ligands where MOR agonist activity is combined with DOR antagonism in the same molecule using a “merged” pharmacophore (2). Such a multifunctional ligand would possess considerable advantages over the traditional approach of using a combination of selective opioid drugs with possibly differing pharmacokinetic or pharmacodynamic properties. Peptides provide a convenient starting point for the development of multifunctional opioid ligands. Many endogenous and synthetic opioid peptide ligands have been studied and their SAR has been well characterized, providing the foundation for applying rational design to existing structures to explore the MOR and DOR binding pockets.

Previous work in our lab resulted in lead peptide 1 (Tyr-c(S-S)[DCys-Phe-Phe-Cys]NH2) (19), which displays full agonism at MOR and slightly reduced efficacy at DOR as well as high affinity for both receptors (Table 1). Since 1 has appreciable DOR efficacy, we then focused our efforts on designing ligands that retain MOR agonist activity, but with reduced DOR efficacy (6). Using our receptor models for both active and inactive states of MOR and DOR (6, 8, 2024) we predicted that adding bulky aromatic substituents in the third or fourth position of peptide 1 would produce a steric clash in the DOR active state binding site which would not be seen in the DOR inactive site, thus favoring binding to the DOR inactive state and consequently resulting in lower DOR efficacy (6). Docking to corresponding active and inactive state MOR models revealed no analogous receptor state-specific adverse interactions; we therefore predicted that increasing steric bulk at position 3 or 4 would provide the desired MOR agonist/DOR antagonist profile. Incorporation of 1-Nal (1-naphthylalanine) or 2-Nal (2-naphthylalanine) into position 3 or 4 of 1 did indeed yield analogs with high MOR/low DOR efficacy, but all analogs also retained high KOR affinity (6). Such KOR activity is less desirable; while KOR agonists are known to provide some analgesic properties, they are also known to cause dysphoria, which severely limits their usefulness (25). The present study makes further use of our receptor-ligand models to design analogues that maintain high affinity for both MOR and DOR, but not KOR, and display full MOR agonism and DOR antagonism.

Table 1
Binding Affinities and Efficacies of Peptides 1–5

Results and Discussion

To understand the basis for the relatively high KOR binding affinity of 1 and related compounds in the series (19) we docked 1 to our previously developed active state model of KOR (24, 26). The docking suggested an important role of Phe4 of peptide 1 in interacting with KOR. An improved active state model generated from the recently released crystal structure of KOR (27) demonstrates that Phe4 of 1 indeed can form π-stacking interactions with Tyr219 from extracellular loop 2 located at the beginning of helix 5 (Figure 1). These interactions likely contribute to the binding and agonist character of peptide 1. DOR and MOR lack a corresponding aromatic residue in the binding site – DOR has a Ser206 and MOR has a Thr225 in the analogous position. We hypothesized that changing the Phe4 of peptide 1 to a non-aromatic residue would eliminate this favorable aromatic interaction with Tyr219 and thus decrease affinity to KOR. As shown in Table 1, replacing Phe4 of peptide 1 with the aliphatic residues Leu, Ile, or Nle, in peptides 2 through 4 resulted in the predicted decreased binding to KOR (Table 1).

Figure 1
Modeling of peptide 1 (Tyr-c(S-S)[DCys-Phe-Phe-Cys]NH2) docked in the KOR active state model. Highlighted is the π-π interaction between Phe4 of 1 with Tyr219 of KOR (shown by dots). This favorable interaction appears to contribute to ...

As all three analogues with an aliphatic residue in position 4 displayed similar opioid profiles, high affinity and efficacy for MOR and DOR with low KOR activity, for convenience we chose to carry forward with analogs containing the Nle4 substitution. Since cyclic opioid pentapeptides with either D- or L-stereochemistry in residue 5 displayed similar MOR and DOR affinities in the initial examples of this series (19), we also examined the opioid profile of 5, the D-Cys5 diastereomer of 4. This analog displayed a similar binding profile as 4, but with somewhat reduced efficacy at both MOR and DOR (Table 1). Thus, we chose both 4 and 5 as starting points for modifications to reduce DOR efficacy.

To achieve this, we again relied on our previously described receptor models (6, 8, 2024) that suggested that bulkier aromatic side chains replacing Phe3 or Phe4 of the ligands would better fit the large and more open antagonist binding pocket of the receptors in the inactive state, while having steric clashes in the more narrow agonist binding pocket of the active receptor conformation. A greater effect was expected for DOR which has more bulky residues, Lys108, Met199, and Trp284, occluding the ligand binding pocket than MOR, with Asn127, Thr218 and Lys303 at the corresponding positions. Examination of our current, refined models of DOR and MOR in complex with 1 supported our previous suggestions (Figure 2). Of particular importance here is the observation that Phe3 of peptide 1 is in close proximity to Met199 in the DOR agonist binding pocket (Figure 2A). However, in the DOR antagonist binding site (Figure 2B), Met199 is angled away from the ligand, enlarging the binding pocket. Replacing the Phe3 of peptide 1 with a larger residue would be expected to increase the steric clash between the ligand and the active conformation of DOR, disfavoring the binding of the ligand to the active conformation and thus reducing its agonist efficacy, as we have observed before (8). However, due to the presence of the smaller side chain of Thr218 in the MOR binding pocket (Figure 2C) instead of Met199 in the DOR binding pocket, adding steric bulk in the 3 position of peptide 1 should have less of an effect on MOR efficacy. Consequently, we prepared and pharmacologically assessed a series of pentapeptides based on 4 and 5, in which the Phe3 residue was replaced by 1-Nal or 2-Nal (Table 2).

Figure 2
Modeling of peptide 1 (Tyr-c(S-S)[DCys-Phe-Phe-Cys]NH2) docked in the active and inactive state models of DOR (A and B) and MOR (C and D). Docking of 1 to the active state of DOR shows a steric clash between Phe3 of 1 and Met199 of DOR, highlighted by ...
Table 2
Binding Affinities and Efficacies of Peptides 6–15

Peptides 6–9 represent the 1-Nal3 and 2-Nal3 analogs of 4 and 5. As seen in Table 2, replacing the Phe3 of 4 or 5 with 2-Nal (analogues 6 and 7, respectively) greatly reduces efficacy at both MOR and DOR, consistent with our earlier observations (6), while also greatly reducing affinity at DOR. By contrast, the 1-Nal3 analogues 8 and 9 display a more promising profile in which DOR efficacy is more selectively affected and DOR affinity is less drastically reduced. In analogues 6–9 the D-Cys5 diastereomer exhibits a greater ability to reduce efficacy than the corresponding L-Cys5, but this effect is equally expressed at MOR and DOR.

In compounds 10 and 11 we examined the effect of increasing the ring size of the 14-membered disulfide scaffold of 8 and 9 to a 16-membered ethylene dithioether-containing cycle, an approach we have often used to modulate opioid activity (6, 19, 26). In the present case increasing the cycle size had little effect on binding affinity or efficacy, with the exception of a rather large reduction in maximal stimulation at MOR (51% vs. 100% stimulation relative to the standard, DAMGO) and a slight increase in KOR affinity displayed by 10 compared to 8.

Among peptides 6–11, the most promising is 8 which possesses high MOR affinity, full agonist behavior at MOR and greatly reduced DOR efficacy (22% of DOR standard DPDPE). However 8 displays ~18 fold lower affinity at DOR compared to MOR. Furthermore, 8 lacks the undesired KOR activity, which make it a promising ligand for the exploration of functional MOR/DOR interactions.

Peptides 12–15, the C-terminal carboxylic acid counterparts of the carboxamide terminal 8–11, were designed to restore a balance in MOR and DOR affinity, since negatively charged C-terminal groups often interfere with MOR binding (28). As seen in Table 2, the carboxy-terminal analogs displayed the expected decrease in MOR affinity, decreased KOR affinity, but little significant effect on DOR binding. Peptide 13 was an exception, in that DOR affinity improved approximately 13-fold compared with 9 (Ki = 4 nM vs. 56 nM). Both disulfide-containing, C-terminal carboxylic acids, 12 and 13, display the desired binding profile: high affinity for MOR and DOR, low affinity for KOR. The dithioether-containing analogs, 14 and 15 have less desirable binding profiles displaying somewhat lower and/or less balanced MOR and DOR affinity and reduced MOR efficacy.

Examination of the efficacy profiles of 12 and 13 reveals an interesting observation. While 12, like its carboxamide terminal counterpart 8, is a full agonist at MOR, with low partial DOR agonism (~20% maximal stimulation vs. DPDPE), peptide 13, which differs from 12 only in the stereochemistry of the C-terminal Cys, acts as an antagonist at both MOR and DOR. The functional antagonist properties of 13 were confirmed by examining its effect on stimulation of GTPγS binding of MOR by DAMGO and of DOR by DPDPE. In both cases 13 shifted the dose-response curve of the standard ligand with Ke values of 2.13 +/− 0.64 nM and 20.3 +/− 6.3 nM for DAMGO and DPDPE, respectively. Reduced efficacy of D-Cys5 vs L-Cys5-containing analogs is a consistent feature among the compounds shown in Table 2; however the complete elimination of MOR efficacy for 13 was unexpected. An explanation for this behavior can be deduced from examination of 12 and 13 docked to the active and inactive states of MOR. Figure 3A depicts 12 bound to our model of the MOR active receptor. In this model, the C-terminal COO- of 12, while being close to Glu229 from transmembrane helix 5, may also form an ionic bridge with the positively charged side chain of Lys303 from helix 6. This ionic bridge can be formed only in MOR and only in the active conformation, but not in DOR or KOR which have Trp284 (DOR) or Glu297 (KOR) in the corresponding position in helix 6. The favorable ionic interactions of 12 in the agonist binding pocket of MOR may explain its behavior as an efficacious MOR agonist. The unfavorable ionic interaction between the C-terminal carboxylate of 12 and Glu229 from helix 5 in MOR or Glu297 from helix 6 in KOR is consistent with the 6- and 7-fold decreased affinity of 12 to MOR and KOR, respectively, as compared to 8 with a C-terminal amide. For 13, the change in stereochemistry of residue 5 orients the terminal COO- away from Lys303 and toward Glu229, resulting in an unfavorable ionic repulsion. However, in the inactive conformation of MOR (Figure 3B) rotation of Glu229 and Lys233 relieve this repulsion. In the DOR models the C-terminus of 13 is close to the Asp210-Lys214 pair from helix 5 and can form favorable ionic interactions with Lys214 in the inactive receptor conformation. These ligand-receptor interactions help explain the antagonist activity of 13 in MOR and its improved binding and antagonist activity at DOR.

Figure 3
Modeling of peptide 12 (Tyr-c(S-S)[DCys-1Nal-Nle-Cys]OH) docked in the active state model of MOR (A) and 13 (Tyr-c(S-S)[DCys-1Nal-Nle-DCys]OH) docked in the inactive state model of MOR (B). The C-terminal COO- of 12 forms a favorable ionic interaction ...


The studies discussed in this paper were aimed toward developing opioid ligands that display high affinity and efficacy for MOR, high affinity and low efficacy for DOR and low affinity for KOR. Using our validated receptor models of the active and inactive states of all three receptors for structure-based design, we were able to achieve this goal by selectively modulating the affinity and efficacy of our lead peptide 1. First, examination of docking of the lead peptide, 1, to active state KOR suggested the participation of the ligand’s Phe4 residue in an aromatic π–π interaction unavailable in MOR or DOR. Replacement of this Phe4 with an aliphatic residue (peptides 24) achieved the desired result of greatly reducing KOR affinity and efficacy. Next, predicted differences in ligand docking to the DOR active and inactive conformations were exploited by incorporating bulkier residues in the third position of peptide 1 to favor binding to the DOR inactive conformation. As predicted, the use of 1-Nal or 2-Nal in place of Phe3 greatly reduced DOR efficacy. Analogs 8 and 12, in particular, with 1-Nal3, exhibited the desired profile of high MOR/low DOR efficacy. The wide range of affinity and efficacy shown by the closely related analogs in Table 2 reflects both the structural sensitivity of the ligand-receptor interaction and the utility of peptides, whose structures can be easily and subtly manipulated for probing the details of the ligand-receptor interaction. Of particular note is the profound functional difference observed for 12 and 13, which differ only in stereochemistry of the C-terminal residue and which possess similar affinity, but quite different efficacy profiles. The results reported here further validate our receptor models and our approach of using these models for rational design to exploit differences in the opioid receptors highly homologous binding pockets.

Methods and Materials


All reagents and solvents were purchased from commercial sources and used without further purification. All chemicals and biochemicals were purchased from Sigma Aldrich (St. Louis, MO, USA) or Fisher Scientific (Hudson, NH, USA), unless otherwise noted. All tissue culture reagents were purchased from Gibco Life Sciences (Grand Island, NY, USA). Radioactive compounds were purchased from Perkin-Elmer (Waltham, MA, USA). Peptide synthesis reagents, amino acids, and Rink resin were purchased from Advanced Chem Tech (Louisville, KY, USA). Wang resins were purchased from Nova Biochem, EMD (Gibbstown, NJ, USA).

Solid-Phase Peptide Synthesis

Peptides were synthesized using standard solid phase Fmoc (fluorenylmethyloxycarbonyl) chemistry on a CS Bio CS336X Peptide Synthesizer (CS Bio Company, Menlo Park, CA, USA), using previously described protocols (29). C-terminal amide peptides were synthesized using Rink resin, C-terminal acid peptides were synthesized using Fmoc-Wang resin preloaded with the C-terminal amino acid. A 20% solution of piperidine in N-methyl-2-pyrrolidone (NMP) was used to remove the first Fmoc protecting group before synthesis and again to remove the Fmoc-protecting group after each coupling cycle. Coupling was performed using a four-fold excess of amino acid and a solution of 0.4 M hydroxybenzotriazole (HOBt) and O-benzotriazole- N, N, N′, N′-tetramethyl-uroniumhexafluoro- phosphate (HBTU) in dimethylformamide (DMF), in the presence of diisopropylethylamine (DIEA). After the synthesis was complete, the resin was washed with NMP, then with dichloromethane, and dried under vacuum. The peptides were cleaved from the resin and side-chain-protecting groups removed by treatment at room temperature for 2 h with a cleavage cocktail consisting of 9.5 mL trifluoroacetic (TFA) acid, 0.25 mL triisopropylsilane (TIS) and 0.25 mL H2O. The solution was concentrated in vacuo, and peptides were precipitated using cold, fresh diethylether. The filtered crude material was then purified using a Waters semipreparative HPLC (Waters Corporation, Milford, MA, USA) with a Vydac Protein and Peptide C18 column, using a linear gradient 10% Solvent B (0.1% TFA acid in acetonitrile) in Solvent A (0.1% TFA acid in water) to 60% Solvent B in Solvent A, at a rate of 1% per minute. The identity all peptides were determined ESI-MS performed on an Agilent Technologies LC/MS system using a 1200 Series LC and 6130 Quadrupole LC/MS (Agilent Technologies, Santa Clara, CA, USA) in positive mode with 50–100 μL injection volume and a linear gradient of 0% Solvent D (0.02% TFA and 0.1% acetic acid (AcOH) in acetonitrile) in Solvent C (0.02% TFA and 0.1% AcOH in water) to 60% Solvent D in Solvent C in 15 min. The purity of all peptides was determined using a Waters Alliance 2690 Analytical HPLC (Waters Corporation, Milford, MA, USA) and Vydac Protein and Peptide C18 reverse phase column, using a linear gradient of 0–70% Solvent B in Solvent A at a rate of 1% per minute. Linear peptides were purified to ≥ 95% purity by UV absorbance at 230 nm.

Disulfide Cyclization of Linear Peptides

Pure linear disulfhydryl-containing peptide was dissolved at 1mg/mL in argon saturated solution of 1% (v/v) AcOH in H2O at 4C. The pH of the peptide solution was raised to 8.5 using NH4OH, followed by the addition of 4 molar equivalents of K3Fe(CN)6. The reaction mixture was stirred on ice, under argon for two minutes and quenched by addition of glacial acetic acid to pH 3.5. The reaction mixture was incubated with 100–200 mesh anion exchange resin AG-3-X4 (Biorad, Hercules, CA, USA) and swirled occasionally at room temperature until the solution was colorless. The crude mixture was then filtered, concentrated in vacuo, and purified to ≥98% purity as determined by UV absorbance at 230 nm as described above to yield the disulfide linked cyclized peptides. The identity of cyclic peptides was determined by ESI-MS as described above.

Dithioether Cyclization of Linear Peptides

A DMF solution of the linear peptide (15 mg/40 mL) containing 5 molar equiv of 1,2-dibromoethane was added dropwise to a round-bottom flask containing 10 molar equiv of potassium tert-butoxide in 100 mL of anhydrous DMF saturated with argon, on ice. The reaction was stirred for 2 h under argon, on ice, and then quenched with to pH 3.5 with glacial acetic acid. Solvents were removed in vacuo, and the residue was purified to ≥ 98% purity as determined by UV absorbance at 230 nm as described above to afford the alkyl dithioether cyclized peptide. The identity of cyclic peptides was determined by ESI-MS as described above.

Cell Lines and Membrane Preparations

C6-rat glioma cells stably transfected with a rat μ (C6-MOR) or rat δ (C6-DOR) opioid receptor (30) and Chinese hamster ovary (CHO) cells stably expressing a human κ (CHO-KOR) opioid receptor (31) were used for all in vitro assays. Cells were grown to confluence at 37°C in 5% CO2 in Dulbecco’s Modified Eagle’s Medium containing 10% fetal bovine serum and 5% penicillin/streptomycin. Membranes were prepared by washing confluent cells three times with ice cold phosphate-buffered saline (0.9% NaCl, 0.61 mM Na2HPO4, 0.38 mM KH2PO4, pH 7.4). Cells were detached from the plates by incubation in warm harvesting buffer (20 mM HEPES, 150 mM NaCl, 0.68 mM EDTA, pH 7.4) and pelleted by centrifugation at 200xg for 3 min. The cell pellet was suspended in ice-cold 50 mM Tris-HCl buffer, pH 7.4 and homogenized with a Tissue Tearor (Biospec Products, Inc, Bartlesville, OK, USA) for 20 s at setting 4. The homogenate was centrifuged at 20,000xg for 20 min at 4 C, and the pellet was rehomogenized in 50 mM Tris-HCl with a Tissue Tearor for 10 s at setting 2, followed by recentrifugation. The final pellet was resuspended in 50mM Tris-HCl and frozen in aliquots at −80°C. Protein concentration was determined via Bradford assay using bovine serum albumin as the standard.

Radioligand Binding Assays

Opioid ligand-binding assays were performed using competitive displacement of 0.2 nM [3H]diprenorphine (250 μCi, 1.85TBq/mmol) by the test compound from membrane preparations containing opioid receptors. The assay mixture, containing membrane suspension (20 μg protein/tube) in 50 mM Tris-HCl buffer (pH 7.4), [3H]diprenorphine, and various concentrations of test peptide, was incubated at room temperature for 1 h to allow binding to reach equilibrium. The samples were rapidly filtered through Whatman GF/C filters using a Brandel harvester (Brandel, Gaithersburg, MD, USA) and washed three times with 50 mM Tris-HCl buffer. The radioactivity retained on dried filters was determined by liquid scintillation counting after saturation with EcoLume liquid scintillation cocktail in a Wallac 1450 MicroBeta (Perkin-Elmer, Waltham MA, USA). Nonspecific binding was determined using 10 μM naloxone. Ki values were calculated using nonlinear regression analysis to fit a logistic equation to the competition data using GraphPad Prism version 5.01 for Windows. The results presented are the mean ± standard error from at least three separate assays performed in duplicate.

Stimulation of [35S]GTPγS Binding

Agonist stimulation of [35S] guanosine 5′-O-[gamma-thio]triphosphate ([35S]GTPγS, 1250 Ci, 46.2TBq/mmol) binding was measured as described previously (32). Briefly, membranes (10–20 μg of protein/tube) were incubated 1 h at room temperature in GTPγS buffer (50 mM Tris-HCl, 100 mM NaCl, 5 mM MgCl2, pH 7.4) containing 0.1 nM [35S]GTPγS, 100 μM guanosine diphosphate (GDP), and varying concentrations of test peptides. Peptide stimulation of [35S]GTPγS was compared with 10 μM standard compounds [D-Ala2, N-MePhe4, Gly-ol]-enkephalin (DAMGO) at MOR, D-Pen2,5- enkephalin (DPDPE) at DOR, or U69,593 at KOR. The reaction was terminated by rapidly filtering through GF/C filters and washing ten times with GTPγS buffer, and retained radioactivity was measured as described above. The results presented are the mean ± standard error from at least three separate assays performed in duplicate; maximal stimulation was determined using nonlinear regression analysis with GraphPad Prism.

Determination of Ke through stimulation of [35S]GTPγS Binding

Agonist stimulation of [35S] guanosine 5′-O-[gamma-thio]triphosphate ([35S]GTPγS, 1250 Ci, 46.2TBq/mmol) binding was measured for known agonists, [D-Ala2, N-MePhe4, Gly-ol]-enkephalin (DAMGO) at MOR and D-Pen2,5- enkephalin (DPDPE) at DOR, as described above. Control wells contained only membranes (10–20 μg of protein/tube) GTPγS buffer (50 mM Tris-HCl, 100 mM NaCl, 5 mM MgCl2, pH 7.4) containing 0.1 nM [35S]GTPγS, 100 μM guanosine diphosphate (GDP), and varying concentrations of known agonists. Test wells contained all of the same components as the control wells, as well as a constant concentration of the test antagonist. The assay mixture was incubated for 1 hr at room temperature and was terminated by rapidly filtering through GF/C filters and washing ten times with GTPγS buffer. Retained radioactivity was measured as described above. The results presented are the mean ± standard error from at least three separate assays performed in duplicate; EC50 values were determined using nonlinear regression analysis with GraphPad Prism and Ke values calculated based of these EC50 values.

Receptor Modeling

The homology modeling of opioid receptors in complexes with peptide ligands was performed as previously described (6, 8, 19, 33). The procedure included the following steps: 1) residue substitution in corresponding structural template(s); 2) rigid body helix movement to reproduce structural rearrangement during receptor activation observed in crystal structures of rhodopsin and adrenergic receptor (34); 3) peptide ligand docking in accordance with mutagenesis-derived constraints; and 4) refinement of receptor-ligand complex using distance geometry and energy minimization with CHARMm. The validity of this modeling procedure has been assessed in blind prediction experiments of structural modeling of MOR (6, 8), A2a-adenosine receptor (35), CXCR4, and D3 dopamine receptor (33) performed before the release of the corresponding crystal structures. The following comparison with experimental structures showed relatively high accuracy of our homology models: rmsd were between 1.5 and 2.5 A for seven transmembrane helices (33, 35). A comparison of our previously developed opioid receptor models (6, 8, 24, 26) and recently released crystal structures of the mouse MOR (36) and the human KOR (27) demonstrated the high reliability in prediction of ligand-receptor interactions in the more conserved “message” region located deeply in the ligand binding pocket, and less precise modeling in the “address” region of flexible extracellular loops which are responsible for ligand selectivity. Despite some inaccuracies, the previous models suggested the important role of interactions between Met199 and Trp284 of DOR and pentapeptide Phe3 and Phe4 side chains, respectively, and aromatic interactions between pentapeptide Phe4 side chain and residues from the extracellular loop 2 (6, 7, 24, 26).

For this study we used X-ray structures of the mouse MOR (PDB ID: 4dkl) and the human KOR (PDB ID: 4djh) as inactive receptor conformations for docking peptide ligands in a mode similar to that of co-crystalized antagonists. To minimize steric hindrances, manual docking of peptides in low-energy conformations was followed by the automated rigid docking implemented in QUANTA (Accelrys Inc). The MOR crystal structure was also used to generate the homology model of the inactive conformation of the human DOR (UniProt ID: P41143, residues 46–333).

The comparison of this DOR model with the experimental structure of the mouse DOR (PDB ID: 4ej4) (38) released after manuscript submission demonstrates their close resemblance (rmsd 1.16 Å for 249 Cα-atoms) in all regions except some parts of extracellular loops 2 and 3 (residues 192–195, 288–295). In the experimental DOR structure residues Asp290–Arg291 of the extracellular loop 3 appear to be located deeper in the binding pocket than we anticipated, thus forming additional contacts with the variable “address” region of peptides. However, the shift of this loop does not alter essential interactions between the common “message” region of cyclic peptides and receptor residues, which has been previously suggested based on our and other mutagenesis data (18, 19). These interactions included hydrogen-bonding between peptide N+ and Asp from TM3 (Asp128 in DOR), water-mediated hydrogen-bonding between OH-group of peptide Tyr1 and His from TM6 (His278 in DOR), aromatic interactions between Tyr1 of peptide and Tyr from TM3 (Tyr129 in DOR), and interactions of the disulfide bridge of the peptide with TM6 (Val281 in DOR).

Further, we used the crystal structure of the human KOR (PDB ID: 4djh) together with our previous active state models (6, 8) for modeling of the active conformations of the mouse MOR using the procedure described above. The KOR structure was chosen as a modeling template because it demonstrates some inward movement of the extracellular ends of TMs 2, 5, 6 and 7 and the outward shift of TM3, as compared to other opioid receptor structures (PDB IDs: 4dkl, 4ej4). Similar helix movements were attributed to the activated receptor conformation (34). However, although the experimental KOR structure shows some active-like rearrangements in the extracellular region, it does not reproduce the main characteristic feature of the activated receptors in the intracellular part -- the TM6 rotation and the large shift of its N-terminus away from TM3 and closer to TM5 along with the shift of the end of TM7 toward TM2 (34). To reproduce these large relocations of intracellular ends of TMs 5, 6, and 7 associated with receptor activation, we used constraints from our previous models of the active receptor conformations (6, 8). The thus obtained model of the activated MOR after energy refinement was used as a template for modeling of the active states of KOR and DOR using residue substitution followed by energy minimization. These models of the opioid receptor active conformations were used for docking high efficacy peptide ligands. Low-energy conformations of cyclic pentapeptides were generated using previously developed pharmacophore models of tetrapeptides (37) with additional conformational search in the region of the fifth residue and disulfide bridge. Coordinates of MOR (active and inactive states), DOR (inactive state), and KOR (active state) with peptide 1 can be downloaded from our web site (


We would like to thank Katarzyna Sobczyk-Kojiro for her assistance with the synthesis of precursors to peptides 13 and 15. This work was funded by NIH grants DA04087 (J.R.T.) and DA03910 (H.I.M., I.D.P). J.P.A. was supported by NIH training grants DA007281 and GM007767. L.C.P. was supported by NIH training grant DA007267.


1. Morphy R, Kay C, Rankovic Z. From Magic Bullets to Designed Multiple Ligands. Research Focus Reviews. 2004;9:641–52. [PubMed]
2. Morphy R, Rankovic Z. Designing Multiple Ligands - Medicinal Chemistry Strategies and Challenges. Current Pharmaceutical Design. 2009;15:587–600. [PubMed]
3. Abdelhamid EE, Sultana M, Portoghese PS, Takemori AE. Selective Blockage of the Delta Opioid Receptors Prevents the Development of Morphine Tolerance and Dependence in Mice. The Journal of Pharmacology and Experimental Therapeutics. 1991;258:299–303. [PubMed]
4. Fundytus ME, Schiller PW, Shapiro M, Weltrowska H, Coderre TJ. Attenuation of Morphine Tolerance and Dependence with the Highly Selective Delta Opioid Receptor Antagonist TIPP(psi) European Journal of Pharmacology. 1995;286:105–8. [PubMed]
5. Hepburn MJ, Little PJ, Gringas J, Khun CM. Differential Effects of Naltrindole on Morphine-Induced Tolerance and Physical Dependence in Rate. The Journal of Pharmacology and Experimental Therapeutics. 1997;281:1350–6. [PubMed]
6. Purington LC, Pohozheva ID, Traynor JR, Mosberg HI. Pentapeptides Displaying Mu Opioid Receptor Agonist and Delta Opioid Receptor Partial Agonist/Antagonist Properties. Journal of Medicinal Chemistry. 2009;52:7724–2231. [PMC free article] [PubMed]
7. Schiller PW. Bi- or Multifunctional Opioid Peptide Drugs. Life Sciences. 2009;86:598–603. [PMC free article] [PubMed]
8. Purington LC, Sobczyk-Kojiro K, Pogozheva ID, Traynor JR, Mosberg HI. Development and in Vitro Characterization of a Novel Bifunctional Mu-Agonist/Delta-Antagonist Opioid Tetrapeptide. Journal of Chemical Biology. 2011;6:1375–81. [PMC free article] [PubMed]
9. Anathan S, Kezar HS, Carter RL, Saini SK, Rice KC, Wells JL, et al. Synthesis, Opioid Receptor Binding, and Biological Activities of Naltrexone-derived Pyrido- and Pyrimidomorphans. Journal of Medicinal Chemistry. 1999;42:3527–38. [PubMed]
10. Anathan S, Khare NK, Saini SK, Seitz LE, Bartlett JL, Davis P, et al. Identification of Opioid Ligands Possessing Mixed Mu Agonist/Delta Antagonist Activity Among Pyridomorphans Derived from Naloxone, Oxymorphone, and Hydromorphone. Journal of Medicinal Chemistry. 2004;47:1400–12. [PubMed]
11. Balboni G, Guerrini R, Salvadori S, Bianchi C, Rizzi D, Bryant SD, et al. Evaluation of the Dmt-Tic Pharmacophore: Conversion of a Potent Delta-opioid Receptor Antagonist into a Potent Delta Agonist and Ligands with Mixed Properties. Journal of Medicinal Chemistry. 2002;45:713–20. [PubMed]
12. Balboni G, Salvadori S, Guerrini R, Negri L, Giannini E, Jinsmaa Y, et al. Potent Delta-opioid Receptor Agonists Containing the Dmt-Tic Pharmacophore. Journal of Medicinal Chemistry. 2002;45:5556–63. [PubMed]
13. Cheng K, Kim IJ, Lee MJ, Adah SA, Raymond TJ, Bilsky EJ, et al. Opioid Ligands with Mixed Properties from Substituted Enatiomeric N-phenethyl-5-phenylmorphans: Synthesis of Mu-Agonist Delta-Antagonist and Delta-Inverse Agonists. Organic and Biomolecular Chemistry. 2007;5:1177–90. [PubMed]
14. Heibel AC, Lee YS, Bilsky EJ, Guivelis D, Deschamps JR, Parrish D, et al. Synthesis and Structure-Activity Relationships of a Potent Mu-Agonist Delta-Antagonist and an Exceedingly Potent Antinociceptive in the Enantiomeric C9-substituted 5-(3-hydroxyphenyl)-N-phenylethylmorphan Series. Journal of Medicinal Chemistry. 2005;50:3765–76. [PubMed]
15. Salvadori S, Guerrini R, Balboni G, Bianchi C, Bryant SD, Cooper PS, et al. Further Studies on the Dmt-Tic Pharmacophore: Hydrophobic Substituents at the C-terminus Endow Delta Antagonists to Manifest Mu Agonism of Mu Antagonism. Journal of Medicinal Chemistry. 1999;42:5010–9. [PubMed]
16. Yamamoto T, Nair P, Vagner J, Largent-Milnes T, Davis P, Ma SW, et al. A Structure-Activity Relationship Study and Combinatorial Synthetic Approach to C-Terminal Modified Bifunctional Peptides that are Delta/Mu Opioid Receptor Agonists and Neurokinin 1 Receptor Antagonists. Journal of Medicinal Chemistry. 2008;51:1369–76. [PMC free article] [PubMed]
17. Schiller PW, Fundytus ME, Merovitz L, Weltrowska G, Nguten TM-D, Lemieux C, et al. The Opioid Mu Agonist/Delta Antagonist DIPP-NH2(psi) Produces a Potent Analgesic Effect, no Physical Dependence and less Tolerance than Morphine in Rate. Journal of Medicinal Chemistry. 1999;42:3520–6. [PubMed]
18. Schmidt R, Vogel D, Mrestani-Klaus C, Brandt W, Neubert K, Chung NN, et al. Cyclic beta-Casomorphin Analgoues with Mixed Mu Agonist/Delta Antagonist Properties: Synthesis, Pharmacological Characterization and Conformational Aspects. Journal of Medicinal Chemistry. 1994;37:1136–44. [PubMed]
19. Przydzial MJ, Pogozheva ID, Ho JC, Bosse KE, Sawyer E, Traynor JR, et al. Design of High Affinity Cyclic Pentapeptide Ligands for the Kappa Opioid Receptors. Journal of Peptide Research. 2005;66:255–62. [PubMed]
20. Fowler CB, Pogozheva ID, Lomize AL, LeVine H, Mosberg HI. Refinement of a Homology Model of the Mu-opioid Receptor Using Distance Constraints from Intrinsic and Engineered Zinc-binding Sites. Biochemistry. 2004;43:8700–10. [PubMed]
21. Fowler CB, Pogozheva ID, Lomize AL, LeVine H, Mosberg HI. Complex of an Active Mu-opioid Receptor with a Cyclic Peptide Agonist Modeled from Experimental Constraints. Biochemistry. 2004;43:15796–810. [PubMed]
22. Pogozheva ID, Lomize AL, Mosberg HI. The Transmembrane 7-alpha-bundle of Rhodopsin: Distance Geometry Calculations with Hydrogen Bonding Constraints. Biophysics. 1997;72:1963–85. [PubMed]
23. Pogozheva ID, Lomize AL, Mosberg HI. Opioid Receptor Three-Dimensional Structures from Distance Geometry Calculations with Hydrogen Bonding Constraints. Biophysics. 1998;75:612–34. [PubMed]
24. Pogozheva ID, Przydzial MJ, Mosberg HI. Homology Modeling of Opioid Receptor-Ligand Complexes Using Experimental Constraints. AAPS Journal. 2005;7:43–57. [PMC free article] [PubMed]
25. Pfeiffer A, Brantl V, Herz A, Emrich HM. Psychotomimesis Mediated by Kappa Opioid Receptors. Science. 1986;233:774–6. [PubMed]
26. Mosberg HI, Omnaas JR, Smith CB, Medzihradsky F. Cyclic Disulfide and Dithioether-Containing Opioid Tetrapeptides: Development of a Ligand with Enhanced Delta Opioid Receptor Selectivity and Potency. Life Sciences. 1988;43:1013–20. [PubMed]
27. Wu H, Wacker D, Mileni M, Katritch V, Han GW, Vardy E, et al. Structure of the Human Kappa Opioid Receptor in Complex with JDTic. Nature. 2012 doi: 10.1038/nature10939. [PMC free article] [PubMed] [Cross Ref]
28. Mosberg HI, Fowler CB. Development and Validation of Opioid Ligand-Receptor Interaction Models: The Structural Basis of Mu vs. Delta Selectivity. Journal of Peptide Research. 2002;60:329–35. [PubMed]
29. Przydzial MJ, Pogozheva ID, Andrews SM, Tharp TA, Traynor JR, Mosberg HI. Roles of Residues 3 and 4 in Cyclic Tetrapeptide Ligand Recognition by the Kappa Opioid Receptor. Journal of Peptide Research. 2005;26:333–42. [PubMed]
30. Lee KO, Akil H, Woods JH, Traynor JR. Differential Binding Properties of Oripavines at Cloned Mu- and Delta-Opioid Receptors. European Journal of Pharmacology; 1999:323–30. [PubMed]
31. Husbands SM, Neilan CL, Broadbear J, Grundt P, Breeden S, Aceto MD, et al. BU74, A Complex Oripavine Derivative with Potent Kappa Opioid Receptor Agonism and Delayed Opioid Antagonism. European Journal of Pharmacology; 2005:117–35. [PubMed]
32. Traynor JR, Nahorski SR. Modulation by Mu-Opioid Agonists of Guanosine-5′-O(3-[35S]thio)triphosphate Binding to Membranes from Human Neuroblastoma SHY5Y Cells. Molecular Pharmacology; 1995:848–54. [PubMed]
33. Kufareva I, Rueda M, Katritch V, Stevens RC, Abagyan R. participants GD. Status of GPCR Modeling and Docking as Reflected by Community-wide GPCR Dock 2010 Assesment. Structure. 2011;19:1108–26. [PMC free article] [PubMed]
34. Congreve M, Langmead CJ, Mason JS, Marshall FH. Progress in Structure Based Drug Design for G Protein-Coupled Receptors. Journal of Medicinal Chemistry. 2011;54:4283–311. [PMC free article] [PubMed]
35. Michino M, Abola E, Participants GD, Charles L, Brooks I, Dixon JS, Moult J, et al. Community-wide Assesment of GPCR Structure Modeling and Ligand Docking. Nature Reviews Drug Discovery; 2008;8:455–63. [PMC free article] [PubMed]
36. Manglik A, Kruse AC, Kobilka TS, Thian FS, Mathiesin JM, Sunahara RK, et al. Crystal Structure of the Mu Opioid Receptor Bound to a Morphinan Antagonist. Nature; 2012 doi: 10.1038/nature10954. [PMC free article] [PubMed] [Cross Ref]
37. Lomize AL, Flippen-Anderson JL, George C, Mosberg HI. Conformational Analysis of the Delta Receptor Selective Cyclic Opioid Peptide Tyr-c[DCys-Phe-DPen]OH (JOM13). Comparison of X-ray Cystallographic Structures, Molecular Mechanics Simulations, and 1H NMR Data. Journal of the American Chemical Society. 1994;116:429–36.
38. Granier S, Manglik A, Kruse AC, Kobilka TS, Thian FS, Weis WI, Kobilka BK. Structure of the δ-opioid receptor bound to naltrindole. Nature. 2012;485(7398):400–404. [PMC free article] [PubMed]