Skip to main content


δ Opioidmimetic Antagonists: Prototypes for Designing a New Generation of Ultraselective Opioid Peptides

Article metrics

  • 132 Accesses

  • 113 Citations



Tyr-Tic (1,2,3,4-tetrahydroisoquinoline-3-carboxylic acid) and Tyr-Tic-Ala were the first peptides with δ opioid antagonist activity lacking Phe, considered essential for opioid activity based on the N-terminal tripeptide sequence (Tyr-d-Xaa-Phe) of amphibian skin opioids. Analogs were then designed to restrain the rotational flexibility of Tyr by the substitution of 2,6-dimethyl-l-tyrosine (Dmt).

Materials and Methods

Tyr and Dmt peptides were synthesized by solid phase and solution methods using Fmoc technology or condensing Boc-Dmt-OH or Boc-Tyr(But)-OH with H-l-Tic-OBut or H-d-Tic-OBut, respectively. Peptides were purified (>99%) by HPLC and characteristics determined by 1H-NMR, FAB-MS, melting point, TLC, and amino acid analyses.


H-Dmt-Tic-OH had high affinity (Kiδ = 0.022 nM) and extraordinary selectivity (Kiμ/Kiδ = 150,000); H-Dmt-Tic-Ala-OH had a Kiδ = 0.29 nM and δ selectivity = 20,000. Affinity and selectivity increased 8700- and 1000-fold relative to H-Tyr-Tic-OH, respectively. H-Dmt-Tic-OH and H-Dmt-Tic-NH2 fitted one-site receptor binding models (η = 0.939−0.987), while H-Dmt-Tic-ol, H-Dmt-Tic-Ala-OH and H-Dmt-Tic-Ala-NH2 best fitted two-site models (η = 0.708−0.801, F 18.9−26.0, p < 0.0001). Amidation increased μ affinity by 10- to 100-fold and acted synergistically with d-Tic2 to reverse selectivity (δ μ). Dmt-Tic di- and tripeptides exhibited δ antagonist bioactivity (Ke = 4−66 nM) with mouse vas deferens and lacked agonist μ activity (> 10 µM) in guinea-pig ileum preparations. Dmt-Tic analogs weakly interacted with κ receptors in the 1 to >20 µM range.


Dmt-Tic opioidmimetic peptides represent a highly potent class of opioid peptide antagonists with greater potency than the nonopioid δ antagonist naltrindole and have potential application as clinical and therapeutic compounds.


Despite the advancements made with the endogenous mammalian enkephalin analogs (1, 2), the amphibian opioid peptides, the dermorphins (μ agonists), and the deltorphins (δ agonists) remain the most selective opioid peptides known (3). Remarkably, sequential ablation of the C-terminal residues of the deltorphins provided abbreviated N-terminal peptides that exhibited reversed selectivity (δμ) (410). These results established that the common N-terminal tripeptide sequence (Tyr-d-Xaa-Phe), featuring a d-amino acid isomer positioned between two aromatic centers, contained the essential residues for interaction with opioid receptors. Structure-activity studies focusing on this portion of the dermorphins/deltorphins suggested pivotal roles for the aromatic side-chains of Tyr1 and Phe3 (1115), leading to the tri- and tetrapeptide antagonists containing 1,2,3,4-tetrahy-dro-3-isoquinoline carboxylate (Tic) at position 2 and Phe at position 3, or 3 and 4 (16, 17).

The first peptides lacking Phe that exhibited opioid activity included the minimal dipeptide H-Tyr-Tic-NH2 and the tripeptides containing Ala3, H-Tyr-Tic-Ala-OH and H-Tyr-Tic-Ala-NH2 (18). This novel dipeptide not only adopted a low energy conformation similar to the nonpeptide N-methyl-naltrindole and shared its δ antagonistic activity (18). To further amplify receptor and antagonist properties of the Tyr-Tic peptides, we introduced 2′,6′-dimethyl-l-tyrosine (Dmt1) to further restrain the Tyr residue that yielded a series of remarkable opioidmimetic compounds with exceptionally high affinity and ultraselectivity for δ opioid receptors. These unusual analogs surpass the δ selectivity of all opioid peptides known to date, including the TIP and TIPP antagonists (16, 17) and other methylated analogs (1922) by orders of magnitude. The results unveil a new model for exceedingly active opioid antagonists with potential clinical and therapeutic applications.

Materials and Methods


H-Dmt-OH, supplied by John H. Dygos, was synthesized by an asymmetric method (23). NαFmoc-Tic and Fmoc-Ala-PEG-PS were from Advanced ChemTech (Lexington, KY, U.S.A.), and Fmoc-PAL-PEG-PS from Millipore (Waltham, MA, U.S.A.), [3H]DPDPE (34.3 Ci/mmol) and [3H]bremazocine (25.4 Ci/mmol) were products of NEN-Dupont (Billerica, MA, U.S.A.), and [3H]DAGO (60.0 Ci/mmol) was from Amersham (Arlington Heights, IL, U.S.A.). The precoated TLC plates of silica gel F254 were from Merck (Darmstadt, Germany). Statistical analyses of binding paradigms used Prism (v. 1.03; Graph-Pad, San Diego, CA, U.S.A.). CytoScint is produced by ICN (Costa Mesa, CA, U.S.A.). Solvents and reagents (e.g., 1-hydroxybenzotriazole, PyBrop, N-methylmorpholine, triethylsilane, isobutylene, trifluoroacetic acid [TFA], acetonitrile) were obtained from Aldrich Chemical Company (Milwaukee, WI, U.S.A.). Diisopropylcarbodiimide (coupling reagent) was from Calbiochem (La Jolla, CA, U.S.A.). All other chemicals were the highest purity available.

Peptide Synthesis

Tyr- and Dmt-containing peptides were synthesized by solid phase and solution techniques. Using solid phase methods, the peptide acids and amides were synthesized using Fmoc-PAL-PEG-PS resin (0.18 mmol/g, 0.09 mmol) and Fmoc-Ala-PEG-PS (0.13 mmol/g, 0.065 mmol). NαFmoc-Tic-OH and NαFmoc-Ala-OH were used in the coupling reactions while Dmt was protected as NαBoc-Dmt-OH as reported by Chandrakumar et al. (24). Amino acids were coupled for 1 hr in 4-fold excess diisopropylcarbodiimide in the presence of 4-fold excess 1-hydroxybenzotriazole. In the Dmt-Tic acylation step, double coupling was required; the acylating agent employed, PyBrop, occurred in the presence of 4-fold excess N-methylmorpholine. All peptides were prepared in a Miliigen 9050 synthesizer and crude deprotected peptides (88% TFA, 5% water, and 7% triethylsilane) were purified by preparative HPLC.

In the solution synthesis of peptides, dipeptide acids were prepared by condensing Boc-Dmt-OH or Boc-Tyr-(But)-OH with H-l-Tic-OBut or H-d-Tic-OBut, previously esterified as reported (25) with DCC/HOBt. The dipeptide alcohol (compound 5, Table 1) was prepared by condensing Boc-Dmt-OH and Tic-CH2-OH as given above, while 3-(hydroxymethyl)-3,4-di-hydroisoquinoline (H-Tic-CH2-OH or H-Tic-ol) was synthesized according to Yamaguchi et al. (26). Boc, But, and OBut were removed with 95% TFA at 0°C and crude peptides purified by preparative HPLC. As reported by Marsden et al. (27), Tyr-Tic sequence, depending on the conditions, evolved to diketopiperazine. Dmt-Tic sequence also gave cyclic dipeptide; we could estimate by HPLC (Vidac C18 218 TP 5415, 5 µm particle column [4.5 × 175 mm] using mobile phases A, 10% acetonitrile in 0.1% TFA, and mobile phase B, 60% acetonitrile in 0.1% TFA; linear gradient from 0% B to 100% B at a flow rate of 1 ml/min in 25 min at 220 nm) from the solid material after 1 month at room temperature resulted in 2.5% cyclic dipeptide, while after 36 hr in solution B at room temperature diketopiperazine formation was 55%; after 1 month in DMSO/water (90:10, v/v), less than 5% diketopiperazine was observed.

Table 1 Analytical characterization of the Dmt-Tic and Tyr-Tic peptides
Table 2 Binding affinities and receptor selectivities for Dmt-Tic and Tyr-Tic peptides

Homogeneity of all the purified products was accessed by thin-layer chromatography, analytical HPLC, 1H-NMR, and electrospray-mass spectrometry. Purity was consistently >99%. Melting points were determined on a Kofler apparatus and are uncorrected. Optical rotations used a Perkin-Elmer 241 Polarimeter with a 10-cm water-jacketed cell. Peptides for 1H-NMR were dissolved in 0.5 ml DM-SOd6 to give 1 mM solutions. The 1H-NMR spectra were run at 400 MHz on a Bruker AMX-400 instrument equipped with an Aspect 3000 computer and at 500 MHz on a Bruker AMX-500 coupled to an X-32 computer. One dimensional 1H-NMR spectra were recorded in the Fourier mode with quadrature detection and the water signal suppressed by a low-power selective irradiation in the homogated mode. The determination of mass ions, using a triple stage quadrupole mass spectrometer (TSQ 700; Finnigan MAT, San Jose, CA, U.S.A.), were conducted by Dr. R. Anacardio (Dompé SpA L’Aquila, Italy).

Preparative reverse-phase HPLC used a Waters Delta Prep 3000 Å (30 × 0.3 cm, 15 µm particle size) column. Peptides were eluted with a gradient of 0% to 60% B over 25 min at a flow rate of 30 ml/min using mobile phases A (10%, v/v, acetonitrile in 0.1% TFA) and B (60%, v/v, acetonitrile in 0.1% TFA). Analytical HPLC analyses were performed on a Bruker liquid chromatography LC-21 instrument with Vidac C18 218 TP 541, 5 µm particle column (4.5 × 175 mm) and equipped with a Bruker LC 313 UV variable wavelength detector. Recording and quantification were accomplished with a chromatographic data processor coupled to an Epson computer system (QX-10). Analytical determinations and capacity factor (K′) of the peptides were determined using HPLC conditions in the above solvent systems in a linear gradient from 0% to 100% B in 25 min at a flow rate of 1 ml/min. All analogs showed less than 1% impurities when monitored at 220 nm.

TLC was conducted using the following solvent systems: (A) 1-butanol/HOAc/H20 (3:1:1); and (B) EtOAc/pyridine/HOAc/H20 (6:2:0.6:1.1). Ninhydrin (1%, Merck), fluorescamine (Hoffman-La Roche) and chlorine reagent were used as detection sprays. A compilation of the analytical properties of the peptides are listed in Table 1.

Receptor Binding Analyses

Rat membrane synaptosome preparations (P2 fraction) were prepared as detailed previously, which included a preincubation step to reduce the binding of endogenous ligands and G proteins (14, 28). The δ and μ receptor sites were labeled with [3H]DPDPE (6.0 nM; 34.3 Ci/mmol) and [3H]DAGO (3.5 nM; 60 Ci/mmol), respectively, in assays containing 50 mM HEPES, pH 7.5, 10 mg/mL BSA, 2 μM bestatin, 250 µg bacitracin, 1 mM PMSF, 8% glycerol, and MgCl2 (5 mM for δ binding and 1 mM for μ assays) under equilibrium conditions (2 hr at 22–23°C) (14, 15, 18, 28). On the other hand, κ receptor assays used 1.56 nM [3H]bremazocine, a κ2 agonist (2931), in the HEPES buffer containing protease inhibitors, 10 mM MgCl2 according to Rodriguez et al. (32), and 2 μM each of DAGO and DPDPE to suppress binding to μ and δ receptors, respectively, with incubation at 4°C for 40 min (33). Samples were rapidly filtered through glass fiber filters and washed with 6 ml (3 × 2 ml) ice-cold 50 mM Tris-HCl, pH 7.5, containing either 0.01% chicken egg albumin or bovine serum albumin. The radioactivity of the dry filters (60 min at 70°C) were determined either after overnight incubation in CytoScint for μ and δ receptors or after an hour for κ receptors. Peptide binding was ascertained using three to four synaptosomal preparations (14, 28). The IC50 values were either graphically resolved and converted to affinity constants (Ki) according to Cheng and Prusoff (34) from sample studies in duplicate assays, or conducted in triplicate with 25–40 graded concentrations of peptide and the resulting curves were individually assessed and combined using Prism™ for multisite binding analyses for δ binding (35, 36). The highly stringent statistical criteria to fit a two-site binding model were considered valid only when the following conditions were met: Hill coefficients (η) <0.850, narrow log of the 95% confidence intervals (≤0.1), and p < 0.0001 for the F test between one- and two-site fits (35, 36).


The in vitro pharmacological assays used published methods (15) based on the organ bath technique of Kosterlitz et al. (37). For μ receptors, a 2- to 3-cm segment of guinea-pig ileum mounted in a 20-ml organ bath and for δ receptors, a single mouse vas deferens was obtained from a mature animal. Each peptide in 10–100 μl Krebs solution was tested for μ agonist activity by the inhibition of electrically stimulated contractions in comparison to dermorphin (H-Tyr-d-Ala-Phe-Gly-Tyr-Pro-Ser-NH2) and δ antagonist activity through the inhibitory effects of deltorphin C (H-Tyr-d-Ala-Phe-Asp-Val-Val-Gly-NH2), a δ1 receptor agonist, and compared against naltrindole, a δ receptor antagonist. Dose-response curves were obtained in the absence and presence of multiple concentrations of antagonists and pA2 (the negative logarithm of the concentration to cause 50% inhibition) calculated according to Schild (38).


Opioidmimetic Properties of Dmt-Tic Peptides

The δ selectivities of peptides 1 and 11 (Table 2), greatly surpass that of nonpeptide substances (39) and all other opioid peptides (13, 1117, 36, 4043) by orders of magnitude. The increase is attributed solely to Dmt, since the Tyr-Tic analogs have relatively low affinities and selectivities (Table 2): δ affinity and selectivity of H-Dmt-Tic-OH (peptide 1) was 8500-fold and 1000-fold greater, respectively, than that of H-Tyr-Tic-OH (peptide 6). Even though C-terminally amidated Dmt-peptides 2 and 12 had high δ affinities (although less than peptide 1), the increased μ affinities due to the amide group negated the gain in selectivity of the free carboxylate analogs (1 and 11); analogous results were found with the Tyr-Tic cognates as well (peptides 7 and 16). Reducing the acid function to an alcohol, Dmt-Tic-ol (peptide 5) and Tyr-Tic-ol (peptide 10), moderately affected δ affinity but, like the amide-derivatives, elevated μ binding. Substitution of d-Tic2 simultaneously reduced δ affinity and enhanced μ affinity which led to a reversal of selectivity (δμ) in Dmt-Tic analogs 4, 13, and 14, and Tyr-Tic peptides 9 and 18; an amide group synergistically augmented the expression of μ binding (compare binding data of peptides 4, 14, 9, and 18 with those of peptides 3, 8, 13, and 17).

The interaction to κ2 receptor sites, as defined by the competitive binding assays using [3H]bremazocine, revealed very low affinities (1 to >20 μM) for the Dmt-Tic peptides and, as expected, the δ/κ selectivities were quite similar to the corresponding μ/κ selectivities (Table 3). The Tyr-Tic cognates (peptides 6, 15, and 16) were essentially inactive toward κ receptors.

Table 3 Receptor binding characteristics to κ 2 receptor sites

Characteristics of Fits to Receptor Subsite Binding Models

Detailed receptor analyses in Table 4 revealed that the goodness-of-fit to δ receptor subsites by H-Dmt-Tic-OH (Fig. 1) and H-Dmt-Tic-NH2 were distinctly different than those of H-Dmt-Tic-ol, H-Dmt-Tic-Ala-OH (Fig. 1), and H-Dmt-Tic-Ala-NH2. Hill coefficients (η) for H-Dmt-Tic-OH (peptide 1) and H-Dmt-Tic-NH2 (peptide 2) best fitted a simple bimolecular, one-site binding model characterized by a η approximately 1.00. H-Dmt-Tic-ol (peptide 5), H-Dmt-Tic-Ala-OH (peptide 11), and H-Dmt-Tic-Ala-NH2 (peptide 12) better fitted a heterogeneous two-site receptor binding model (η < 0.850 with p < 0.0001 in the F test) (Table 4). The comparative displacement curves for H-Dmt-Tic-OH and H-Dmt-Tic-Ala-OH binding to one- and two-site models are illustrated in Fig. 1.

Table 4 Hill coefficients and Statistical analysis for goodness-of-fit for one- and two-site δ receptor binding models
Fig. 1

Analyses of one- and two-site binding models for H-Dmt-Tic-OH and H-Dmt-Tic-Ala-OH to δ receptors from rat brain membranes

The solid line depicts a one-site binding model while the dashed line is a two-site model determined using Prism™. The open circles represent the mean values and error bars are the standard error of the mean (SEM) for n = 6 experiments; their absence indicates that the SEM falls within the point defined by the circle. The goodness-of-fit of H-Dmt-Tic-NH2 was essentially identical to that of H-Dmt-Tic-OH, while those for H-Dmt-Tic-ol and H-Dmt-Tic-Ala-NH2 were similar to that of H-Dmt-Tic-Ala-OH (c.f., Table 3 for actual values).

Functional Bioassays

Pharmacological bioactivity determined with mouse vas deferens indicated that peptides 1, 2, 5, 11, and 12 exhibited δ antagonist activity as observed with other Tic2-containing peptides (1618). The pA2 values compare favorably to the nonpeptide antagonist natrindole (Table 5); however, in comparison to naltrindole, Dmt-Tic peptides are completely inactive in eliciting an agonist response with guinea-pig ileum, further substantiating the remarkable selectivity observed with these peptides in binding brain membrane receptors (Table 2).

Table 5 Bioactivity of Dmt-Tic and Dmt-Tic-Ala opioidmimetic peptides


Our opioidmimetic peptides represent the ultimate simplification of a substance that potently interacts with the δ opioid receptor with high affinity and an astounding δ selectivity for two compounds (peptides 1 and 11) that surpasses those of previously known opioid peptides and non-peptides by several hundred-fold (14, 1618, 28, 36). The data clearly establishes that 2,6-dimethylation of Tyr in both the di- and tripeptide series (peptide 11) enhanced interaction toward δ receptors for peptides containing a C-terminal carboxylate group in both brain membranes and mouse vas deferens. The presence of Tic in place of Phe3 and/or Phe4 (16) may further stabilize the bioactive conformation or introduce additional bioactive conformations (such as with Tyr-Tic-Phe as the “message”), but is by no means the necessary determinant for antagonism based on the activity of the Dmt-Tic peptides.

Earlier attempts to enhance opioid activity involving modification of the N-terminal Tyr, utilized various conformationally constrained enkephalin analogs or the N-terminal dermorphin/deltorphin sequence Tyr-d-Xaa-Phe; however, the δ selectivities of those peptides were relatively modest (Kiμ/Kiδ = 100−600) with δ affinities in the 5–10 nM range. For instance, alterations that centered around methylation of the aromatic ring, such as [2′-methyl-Tyr1]DPDPE maintained δ affinity and selectivity (44) whereas the activity of [Dmt1]DPDPE was reported to be either weakly δ selective (21) or non-selective (19). On the other hand, while [Tmt1]deltorphin I (Tmt being [2S,3S]-2′,6′-dimethyl-β-methyltyrosine) was active as a δ ligand (22), and [Dmt1]deltorphin II lost δ selectivity due to the concomitant increase in μ affinity (R. Guerrini et al., submitted); however, a Dmt-d-Ala-arylalkylamide analogue was moderately μ selective (24, 45). The increased δ selectivity observed in C-terminal deamidation is consistent with the role of charged groups of the “membrane-assisted receptor selection” hypothesis (46); however, the magnitude of the change is so great that it must reflect constitutional and conformational modifications of the ligand as well.

The low δ/κ and μ/κ selectivities suggest weak affinities for the κ ligand-binding domain. To ensure stringent criteria in the assessment of κ receptors, due to the high affinity towards δ receptors, μM concentrations of DAGO and DPDPE were employed rather than nM quantities (2931, 33) since [3H]bremazocine interacts with multiple receptor binding sites (47).

Several constituents of opioid peptides play significant roles in securing or stabilizing the peptide within the receptor, including the N-terminal amine, C-terminal carboxylate, hydrogen bonding through the Tyr hydroxyl group and the chirality at the α carbon of residue 2. For example, the N-terminal amine appears essential for activity (2) and the C-terminal carboxylate anion apparently interacts with a δ receptor cationic group, simultaneously conferring high-affinity binding to δ the ligand-binding domain and repulsing the ligand from the μ receptor (7). On the other hand, these residues may be involved in extensive internal H bonding which would have a major impact on the solution conformation of the peptide (6, 36, 48). C-terminal amidation in Dmt-Tic opioidmimetic peptides increased association toward μ receptors (Table 2), a phenomenon consistently observed with deltorphin heptapeptide analogs (4, 40, 43): μ binding is enhanced by an overall increase in positive charge as seen in cyclic dermorphin analogs (11) and deltorphins (addition of amine groups or a C-terminal amide) (40, 42). Although the Tyr hydroxyl group assumes an important auxiliary function in flexible opioids, the substitution by Phe1 (2) and nonaromatic ring compounds (13) reduced, but did not eliminate, binding to opioid receptors, consistent with the interaction of the rigid fentanyl nonopioid agonist which lacks hydroxyl groups (49).

The chirality of the α carbon at position 2 (Table 2) is critical for opioid activity (2, 11, 14, 16), which indicates that the ligand-binding domain of δ and μ receptors differ significantly (16). The change in chirality from l- to d-Tic abolished high δ affinity while simultaneously increased μ affinity to yield peptides with either reduced δ selectivity or the acquisition of μ selectivity, similar to changes in affinity and selectivity observed with isomeric modifications in opioid tetrapeptides (16), dermorphin (50, 51) and deltorphin A heptapeptides (14). The fundamental difference between the recognition sites (that can be considered as “message domains” even if this term is generally confined to agonists) between our peptides (Dmt/Tyr-Tic) and all other opioids (Tyr-l/ d-Xaa-Phe or Tyr-l/d-Xaa-Yaa-Phe) emphasizes the importance of the conformation of these molecules. It was shown that the Tyr-Tic peptides reproduce the minimum requirements of volume, conformation, and spatial distribution of the electronic features recently described by Portoghese et al. for several nonpeptidic (and rigid) antagonists for δ and κ receptors (5254). The l chirality of Tic allows the attainment of the proper conformation of Tyr-Tic that appears to mimic the overall shape of naltrindole antagonists (18, 55), whereas the d chirality of Tic in Dmt-Tic-Phe sequences is consistent with the behavior of all Tyr-d-Xaa-Phe peptides and leads to μ agonism. Recently, Bryant et al. (submitted) using Monte Carlo analyses, described a unique solution conformation of Dmt-Tic-OH (peptide 1) consisting of a parallel array between the dimethylated, aromatic side-chain of Dmt and the aromatic ring of Tic.

The incorporation of unnatural heterocyclic amino acids in the N-terminal region of opioid peptides was implemented to reduce conformational flexibility, restrain secondary structure, and determine the influence of Phe3 in deltorphin I on activity (12, 15). Substitution by Tic, among other bicyclic residues (15, 51), yielded peptides with modest receptor binding properties; however, the incorporation of d-Tic into somatostatin, a nonopioid peptide whose analogs evoke an opioid response—based on the initial observations of Terenius (56)—produced a selective and potent μ antagonist (57). The cyclic analogs in the N-terminal tetrapeptide series of deltorphin II (H-Tyr-d-Ala-Phe-Glu), in which Phe3 was replaced by Tic led to an assortment of analogs with distinct properties toward μ receptors (12) comparable to those of [Tic3]dermorphin (50, 51).

Receptor binding analyses with position 4-substituted deltorphin analogs provided evidence of fits to either one- or two-site models that might represent the possible interaction with the pharmacologically defined δ1 and δ2 receptor subtypes, respectively (36). In fact, cDNA cloning experiments have defined a δ receptor distinct from μ and κ receptors and it appears to correspond to the δ2 subtype according to bioassay paradigms (58, 59). A theoretical model to account for the action of agonists and antagonists (60) might also explain this phenomenon; however, neural and peripheral tissues might contain distinct populations of δ receptors, a fact that might explain the differences observed in the Ki and Ke values throughout the literature. Our bioassay data (Table 5) demonstrates that the Dmt-Tic opioidmimetic peptides exhibit antagonistic properties in vitro to a greater extent than the nonopioid compound naltrindole and that peptide 11 also possesses antagonistic activity in vivo (R. Guerrini et al., submitted).

In summary, our opioidmimetic peptides can serve as the pivotal prototypic molecules for designing a new generation of unique opioid antagonists with potential therapeutic applications. As far as the specific message domain requirements are concerned, we conclude that Dmt-Tic, and to a lesser degree Tyr-Tic, represent the ideal universal simulation of the constitutional and conformational requirements of δ antagonists for the recognition site (message ligand-binding domain) of opioid receptors.


  1. 1.

    Hansen PE, Morgan BA. (1984) Structure-activity relationships in enkephalin peptides. In: Udenfriend S, Meienhofer J (eds). The Peptides. Academic Press, Orlando. Vol. 6, pp. 269–321.

  2. 2.

    Hruby VJ, Gehrig CA. (1989) Recent developments in the design of receptor specific opioid peptides. Med. Res. Rev. 9: 343–401.

  3. 3.

    Erspamer V. (1992) The opioid peptides of the amphibian skin. Int. J. Dev. Neurosci. 10: 3–30.

  4. 4.

    Sagan S, Amiche M, Delfour A, Camus A, Mor A, Nicolas P. (1989) Differential contribution of C-terminal regions of dermorphin and dermenkephalin to opioid-sites selection and binding potency. Biochem. Biophys. Res. Commun. 163: 726–732.

  5. 5.

    Sagan S, Amiche M, Delfour A, Mor A, Camus A, Nicolas P. (1989) Molecular determinants of receptor affinity and selectivity of the natural δ-opioid agonist, dermenkephalin. J. Biol. Chem. 264: 17100–17106.

  6. 6.

    Balboni G, Marastoni M, Picone D, et al. (1990) New features of the δ opioid receptor: conformational properties of deltorphin I analogues. Biochem. Biophys. Res. Commun. 169: 617–622.

  7. 7.

    Lazarus LH, Salvadori S, Tomatis R, Wilson WE. (1991) Opioid receptor selectivity reversal in deltorphin tetrapeptide analogues. Biochem. Biophys. Res. Commun. 178: 110–115.

  8. 8.

    Melchiorri P, Negri L, Falconieri Erspamer G, et al. (1991) Structure-activity relationships of the δ-opioid-selective agonists, deltorphins. Eur. J. Pharmacol. 195: 201–207.

  9. 9.

    Marastoni M, Tomatis R, Lazarus LH, Salvadori S. (1991) On the degradation of deltorphin peptides by plasma and brain homogenate. Farmaco 46: 1273–1279.

  10. 10.

    Sagan S, Charpentier S, Delfour A, Amiche M, Nicolas P. (1992) The aspartic acid in deltorphin I and dermenkephalin promotes targeting to δ-opioid receptor independently of receptor binding. Biochem. Biophys. Res. Commun. 187: 1203–1210.

  11. 11.

    Schiller PW, Nguyen TM-D, Chung NN, Lemieux C. (1989) Dermorphin analogues carrying an increased positive net charge in their “message” domain display extremely high μ opioid receptor selectivity. J. Med. Chem. 32: 698–703.

  12. 12.

    Schiller PW, Weltrowska G, Nguyen TM-D, et al. (1991) Conformational restriction of the phenylalanine residue in a cyclic opioid peptides analogue: Effects on receptor selectivity and stereospecificity. J. Med. Chem. 34: 3128–3132.

  13. 13.

    Mosberg HI, Kroona HB. (1992) Incorporation of a novel conformationally restricted tyrosine analog into a cyclic, δ opioid receptor selective tetrapeptide (JOM-13) enhances δ receptor binding affinity and selectivity. J. Med. Chem. 35: 4498–4500.

  14. 14.

    Lazarus LH, Salvadori S, Balboni G, Tomatis R, Wilson WE. (1992) Stereospecificity of the amino acid side chains in deltorphin defines binding to opioid receptors. J. Med. Chem. 35:1222–1227.

  15. 15.

    Salvadori S, Bryant SD, Bianchi C, Balboni G, Attila M, Lazarus LH. (1993) Phe3-substituted analogues of deltorphin C. Spatial conformation and topography of the aromatic ring in peptide recognition by δ opioid receptors. J. Med. Chem. 36: 3748–3756.

  16. 16.

    Schiller PW, Nguyen TM-D, Weltrowska G, et al. (1992) Differential stereochemical requirements of μ vs. δ opioid receptors for ligand binding and signal transduction: Development of a class of potent and highly δ-selective peptide antagonists. Proc. Natl. Acad. Sci. 89: 11871–11875.

  17. 17.

    Schiller PW, Weltroska G, Nguyen TM-D, Wilkes BC, Chung NN, Lemieux C. (1993) TIPP [ψ]: A highly potent and stable pseudopeptide δ opioid receptor antagonist with extraordinary δ selectivity. J. Med. Chem. 36: 3182–3187.

  18. 18.

    Temussi PA, Salvadori S, Amodeo P, et al. (1994) Selective opioid dipeptides. Biochem. Biophys. Res. Commun. 198: 933–939.

  19. 19.

    Chandrakumar NS, Stapelfeld A, Beardsley PM, et al. (1992) Analogs of the δ opioid receptor selective cyclic peptide [2-d-penicillamine, 5-d-penicillamine]-enkephalin: 2′,6′-dimethylty rosine and Gly3-Phe4 amide bond isostere substitutions. J. Med. Chem. 35: 2928–2938.

  20. 20.

    Pizele BS, Hamilton RW, Kudla KD, et al. (1992) Enkephalin analogs as systemically active antinociceptive agents: O- and N-alkylated derivatives of the dipeptide amide 2,6-dimethyl-l-tyrosyl-N-(3-phenylpropyl)-d-alaninamide. J. Med. Chem. 37: 888–896.

  21. 21.

    Hansen Jr DW, Stapelfeld A, Savage MA, et al. (1992) Systemic analgesic activity and δ-opioid selectivity in [2,6-dimethyl-Tyr1,D-Pen2,D-Pen5]enkephalin. J. Med. Chem. 35: 684–687.

  22. 22.

    Qian X, Kōvér KE, Shenderovich MD, et al. (1994) Newly discovered stereochemical requirements in the side-chain conformation of δ opioid agonists for recognizing opioid δ receptors. J. Med. Chem. 37: 1746–1757.

  23. 23.

    Dygos JH, Yonan EE, Scaros MG, et al. (1992) A convenient asymmetric synthesis of the unnatural amino acid 2,6-dimethyl-l-tyrosine. Synthesis 8: 741–743.

  24. 24.

    Chandrakumar NS, Yonan PK, Stapelfeld A, et al. (1992) Preparation and opioid activity of analogues of the analgesic dipeptide, 2,6-dimethyl-l-tyrosyl-N-(3-phenylpropyl)-d-alaninamide. J. Med. Chem. 35: 223–233.

  25. 25.

    Castopanagiotis AA, Preston J, Weinstein B. (1966) Amino acids and peptides, V. Synthesis of the C-terminal tripeptide sequence (A27–A29) of glucagon. J. Org. Chem. 31: 3398–3400.

  26. 26.

    Yamaguchi R, Hamasaki T, Sasaki T, et al. (1993) A highly effective one-pot bicycloannulation methodology for the synthesis of berban and yohimban systems based on organation-mediated three-component coupling (N-acylative pentadienylation of C = N bonds). J. Org. Chem. 58: 1136–1143.

  27. 27.

    Marsden BJ, Nguyen TM-D, Schiller PW. (1993) Spontaneous degradation via diketopiperazine formation of peptides containing a tetrahydroisoquinoline-3-carboxylic acid residue in the 2-position of the peptide sequence. Int. J. Pept. Prot. Res. 41: 313–316.

  28. 28.

    Lazarus LH, Wilson WE, de Castiglione R, Guglietta A. (1989) Highly selective δ-opioid receptor peptide from preprodermorphin gene sequence. J. Biol. Chem. 264: 3047–3050.

  29. 29.

    Corbett AD, Paterson SJ, McKnight AT, Magnan J, Kosterlitz HW. (1982) Dynorphin1,8 and dynorphin1–9 are ligands for the κ subtype of opiate receptor. Nature 299: 79–81.

  30. 30.

    Standifer KM, Cheng J, Brooks AI, et al. (1994) Biochemical and pharmacological characterization of mu, delta and kappa3 opioid receptors expressed in BE(2)-C neuroblastoma cells. J. Pharmacol. Exp. Ther. 270: 1246–1255.

  31. 31.

    Webster JL, Polgar WE, Brandt SR, et al. (1993) Comparison of κ2-opioid receptors in guinea pig brain and guinea pig ileum membranes. Eur. J. Pharmacol. 231: 251–258.

  32. 32.

    Rodriguez FD, Bardaji E, Traynor JR. (1992) Differential effects of Mg2+ and other divalent cations on the binding of trititated opioid ligands. J. Neurochem. 59: 467–472.

  33. 33.

    Synder KR, Story SC, Heidt ME, et al. (1992) Effect of modification of the basic residues of dynorphin A-(1-13) amide on κ opioid receptor selectivity and opioid activity. J. Med. Chem. 35: 4330–4333.

  34. 34.

    Cheng YC, Prusoff WH. (1973) Relationships between the inhibition constant (Ki) and the concentration of inhibition which cause 50 per cent inhibition (I50) of an enzymatic reaction. Biochem. Pharmacol. 22: 3099–3108.

  35. 35.

    Attila M, Salvadori S, Balboni G, Bryant SD, Lazarus LH. (1993) Synthesis and receptor binding analysis of dermorphin hepta, hexa- and pentapeptide analogues. Evidence for one- and two-site binding models for the μ-opioid receptor. Int. J. Peptide Prot. Res. 42: 550–559.

  36. 36.

    Bryant SD, Attila M, Salvadori S, Guerrini R, Lazarus LH. (1994) Molecular dynamics conformations of deltorphin analogues advocate δ opioid binding site models. Peptide Res. 7: 175–184.

  37. 37.

    Kosterlitz HW, Lees GM, Wallis DI, Watt AJ. (1968) Non-specific inhibitory effects of morphine-like drugs on transmission in the superior cervical ganglion and guinea-pig isolated ileum. Br. J. Pharmacol. 34: 691P–692P.

  38. 38.

    Schild HO. (1947) pA, a new scale for the measurement of drug action. Br. J. Pharmacol. 2: 189–206.

  39. 39.

    Collin E, Cesselin F. (1991) Neurobiological mechanism of opioid tolerance and dependence. Clin. Neuropharmacol. 14: 465–488.

  40. 40.

    Lazarus LH, Salvadori S, Santagada V, Tornatis R, Wilson WE. (1991) Function of negative charge in the “address domain” of deltorphins. J. Med. Chem. 34: 1350–1359.

  41. 41.

    Charpentier S, Sagan S, Delfour A, Nicolas P. (1991) Dermenkephalin and deltorphin I reveal similarities within ligand-binding domains of μ- and δ-opioid receptors and an additional address subsite on the δ-receptors. Biochem. Biophys. Res. Commun. 179: 1161–1168.

  42. 42.

    Lazarus LH, Salvadori S, Grieco P, Wilson WE, Tomatis R. (1992) Unique sequence in deltorphin C confers structural requirements for δ opioid receptor selectivity. Eur. J. Med. Chem. 27: 791–797.

  43. 43.

    Lazarus LH, Salvadori S, Attila M, et al. (1993) Interaction of deltorphin with opioid receptors: molecular determinants for affinity and selectivity. Peptides 14: 21–28.

  44. 44.

    Portoghese PS. (1989) Bivalent ligands and the message-address concept in the design of selective opioid receptor antagonists. Trends Pharmacol. Sci. 10: 230–235.

  45. 45.

    Hammond DL, Stapelfeld A, Drower EJ, Savage MA, Tam L, Mazur RH. (1994) Antinociception produced by oral, subcutaneous or intrathecal administration of SC-39566, an opioid dipeptide. J. Pharmacol. Exp. Ther. 268: 607–615.

  46. 46.

    Schwyzer R. (1986) Molecular mechanism of opioid receptor selection. Biochemistry 25: 6335–6342.

  47. 47.

    Wollemann M, Benyhe S, Simon J. (1993) The kappa-opioid receptor: Evidence for the different subtypes. Life Sci. 52: 599–611.

  48. 48.

    Amodeo P, Motta A, Tancredi T, et al. (1992) Solution structure of deltorphin I at 265 K: A quantitative NMR study. Peptide Res. 4: 48–55.

  49. 49.

    Castiglione-Morelli MA, Lelj F, Pastore A, et al. (1987) A 500-MHz proton nuclear magnetic resonance study of mu opioid peptides in a simulated receptor environment. J. Med. Chem. 37: 2067–2073.

  50. 50.

    Tourwé D, Verschueren D, Van Binst G, Davis P, Porreca F, Hruby VJ. (1992) Dermorphin sequence with high δ-affinity by fixing the Phe side chain to trans at χ1. Bioorg. Med. Chem. Lett. 2: 1305–1308.

  51. 51.

    Schiller PW, Weltrowska G, Nguyen T M-D, Wilkes BC, Chung NN, Lemieux C. (1992) Conformationally restricted deltorphin analogues. J. Med. Chem. 35: 3956–3961.

  52. 52.

    Portoghese PS. (1992) The role of concepts in structure-activity relationship studies of opioid ligands. J. Med. Chem. 35: 1927–1937.

  53. 53.

    Portoghese PS, Sultana M, Takemori AE. (1988) Naltrindole, a highly selective and potent non-peptide delta opioid receptor antagonist. Eur. J. Pharmacol. 146: 185–186.

  54. 54.

    Takemori AE, Sultana M, Nagase H, Portoghese PS. (1992) Agonist and antagonist activities of ligands derived from naltrexone and oxymorphone. Life Sci. 50: 1491–1495.

  55. 55.

    Tancredi T, Salvadori S, Amodeo P, et al. Conversion of enkephalin and dermorphin into δ-selective opioid antagonists by single-residue substitution. (1994) Eur. J. Biochem. 224: 241–247.

  56. 56.

    Terenius L. (1976) Somatostatin and ACTH are peptides with partial antagonist-like selectivity for opiate receptors. Eur. J. Pharmacol. 38: 211–213.

  57. 57.

    Kazmierski W, Wire SW, Lui GK, et al. (1988) Design and synthesis of somatostatin analogues with topographical properties that lead to highly potent and specific mu opioid receptor antagonists with greatly reduced binding at somatostatin receptors. J. Med. Chem. 31: 249–253.

  58. 58.

    Standifer KM, Chien C-C, Wahlestedt C, Brown GP, Pasternak GW. (1994) Selective loss of δ opioid analgesia and binding by antisense oligodeoxynucleotides to a δ opioid receptor. Neuron 12: 805–810.

  59. 59.

    Bilsky EJ, Bernstein RN, Pasternak GW, et al. (1994) Selective inhibition of [d-Ala2, Glu4] deltorphin antinociception by supraspinal, but not spinal, administration of an antisense oligodeoxynucleotide to an opioid delta receptor. Life Sci. 55: PL37–PL43.

  60. 60.

    Szabadi E. (1977) A model of two functionally antagonistic receptor populations activated by the same agent. J. Theor. Biol. 69: 101–112.

Download references


We are indebted to Dr. J. H. Dygos (G. D. Searle and Co.) for generously providing us with 2′,6′-dimethyl-l-tyrosine. We thank Dr. R. Anacardio, Dompé SpA L’Aquila-Italy for the FAB-mass spectroscopic determinations. We appreciate the part-time assistance of K. Jefferies and T. McElwee, and the consistent library support by S. Fuller, R. J. Hester, E. M. Leadem, and F. T. Lyndon. We are grateful to Dr. C. Gruppi at Morgan & Finnegan, Attorneys at Law, New York, NY, for studiously preparing and submitting our data for a patent application, Serial Number 08/347.531, through the aegis of the National Institutes of Health Office of Technology Transfer. S. Salvadori was supported in part by grants from CNR Progetto Finalizzato Chimica Fine e Secondaria II and Murst.

Author information

Correspondence to Lawrence H. Lazarus.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Salvadori, S., Attila, M., Balboni, G. et al. δ Opioidmimetic Antagonists: Prototypes for Designing a New Generation of Ultraselective Opioid Peptides. Mol Med 1, 678–689 (1995) doi:10.1007/BF03401608

Download citation