Synthesis, Biological Activity and Molecular Docking of Chimeric Peptides Targeting Opioid and NOP Receptors

Recently, mixed opioid/NOP agonists came to the spotlight for their favorable functional profiles and promising outcomes in clinical trials as novel analgesics. This study reports on two novel chimeric peptides incorporating the fragment Tyr-c[D-Lys-Phe-Phe]Asp-NH2 (RP-170), a cyclic peptide with high affinity for µ and κ opioid receptors (or MOP and KOP, respectively), conjugated with the peptide Ac-RYYRIK-NH2, a known ligand of the nociceptin/orphanin FQ receptor (NOP), yielding RP-170-RYYRIK-NH2 (KW-495) and RP-170-Gly3-RYYRIK-NH2 (KW-496). In vitro, the chimeric KW-496 gained affinity for KOP, hence becoming a dual KOP/MOP agonist, while KW-495 behaved as a mixed MOP/NOP agonist with low nM affinity. Hence, KW-495 was selected for further in vivo experiments. Intrathecal administration of this peptide in mice elicited antinociceptive effects in the hot-plate test; this action was sensitive to both the universal opioid receptor antagonist naloxone and the selective NOP antagonist SB-612111. The rotarod test revealed that KW-495 administration did not alter the mice motor coordination performance. Computational studies have been conducted on the two chimeras to investigate the structural determinants at the basis of the experimental activities, including any role of the Gly3 spacer.


Introduction
A major challenge in opioid peptide research is the development of novel drugs with analgesic properties similar to morphine but devoid of its severe side effects. Opioid analgesia is mediated by three types of so-called classical opioid receptors designated µ, δ and κ (or MOP, DOP and KOP, respectively), with MOP being a predominant target of morphine-like compounds [1,2]. Unfortunately, activation of MOP is also responsible for most of the adverse effects of opioids [3][4][5]. The fourth member of the opioid family is the nociceptin/orphanin FQ (N/OFQ) receptor (NOP) [6]. The endogenous ligand of NOP is the heptadecapeptide N/OFQ, considered a non-classical opioid peptide [7,8]. In contrast to the classical endogenous opioids (endorphins, enkephalins, dynorphins), N/OFQ lacks Tyr 1 , and this difference could be, at least in part, responsible for the different pharmacological profile of this ligand [9]. Analgesic effects mediated by NOP are more complex than those elicited by any other member of the opioid receptor family. Indeed, depending on the The chimeras were synthesized using a solid-phase procedure on MBHA Rink amide resin, using standard Fmoc/tBu chemistry and TBTU/DIEA as the coupling agents. D-Lys 2 and Asp 5 were introduced as Fmoc-D-Lys(Mtt)-OH and Fmoc-Asp(O-2PhiPr)-OH, respectively ( Figure 1). The synthesis of linear sequences was followed by selective deprotection of the side chains of D-Lys and Asp under mild acidic conditions, and onresin cyclization with TBTU/DIEA.
After cleavage from the resin with 95% TFA and a cocktail of scavengers, both compounds were purified by semipreparative RP HPLC, and their identity was confirmed by high-resolution mass spectrometry (HR ESI-MS). The purity of the compounds was determined to be ≥95% by analytical RP HPLC. The detailed analytical data of the synthesized peptides are provided in the Supporting Information (Table S1, Figures S1 and S2). The chimeras were synthesized using a solid-phase procedure on MBHA Rink amide resin, using standard Fmoc/tBu chemistry and TBTU/DIEA as the coupling agents. D-Lys 2 and Asp 5 were introduced as Fmoc-D-Lys(Mtt)-OH and Fmoc-Asp(O-2PhiPr)-OH, respectively ( Figure 1). The synthesis of linear sequences was followed by selective deprotection of the side chains of D-Lys and Asp under mild acidic conditions, and on-resin cyclization with TBTU/DIEA.
After cleavage from the resin with 95% TFA and a cocktail of scavengers, both compounds were purified by semipreparative RP HPLC, and their identity was confirmed by high-resolution mass spectrometry (HR ESI-MS). The purity of the compounds was determined to be ≥95% by analytical RP HPLC. The detailed analytical data of the synthesized peptides are provided in the Supporting Information (Table S1, Figures S1 and S2).

Pharmacological Characterization of the Chimeric Peptides In Vitro
In order to evaluate the affinity of novel chimeras at MOP, DOP, KOP and NOP, radioligand binding assays were performed on membrane preparations from CHO cells overexpressing respective recombinant receptors. The results are summarized in Table 1. The parent RP-170, characterized by high MOP and KOP affinity (pKi = 9.21 and 8.53, respectively), was used as a reference opioid ligand. Both chimeras KW-495 and KW-496 retained the binding profile of the parent, RP-170. A small reduction in their MOP affinity was observed as compared to RP-170, and they did not bind to DOP up to the 1 µM concentration. At the KOP, KW-495 had only slightly lower affinity (pKi = 8.40) than the parent; in contrast, KW-496 showed higher KOP affinity (pKi = 9.10). At the NOP, the highest, comparable to N/OFQ affinity, was obtained for Ac-RYYRIK-NH 2 (pKi = 9.29 nM), confirming that this peptide carries a sequence interacting with this receptor. Of the two hybrids, KW-495 efficiently bound to NOP (pKi = 8.65), while KW-496 showed reduced NOP binding affinity (pKi = 7.35 nM). The functional effects of the novel analogs at all three classic opioid receptors and at the NOP receptor were evaluated in the calcium mobilization assay performed in cells co-expressing the respective human recombinant receptors and chimeric G proteins that force Gi receptors to couple with the PLC-IP 3 -Ca 2+ pathway. This assay has been previously set up and validated for the NOP receptor [45] and was later extended to other opioid receptors [46].
The concentration-response curves were obtained for the parent RP-170 and Ac-RYYRIK-NH 2 , and for both chimeras (Supporting Information Figures S3-S5). Agonist potencies (pEC 50 ) and efficacies (α) of the tested ligands are summarized in Table 2. At the MOP, the parent RP-170 showed full efficacy and potency (pEC 50 = 8.93, α = 1.0), even higher than those of the reference MOP agonist EM-2 (pEC 50 = 8.08, α = 1.00). The chimeric KW-495 and KW-496 showed only slightly decreased potency and efficacy. At the DOP, both chimeras were inactive. Consistent with the binding results, both KW-495 and KW-496 stimulated calcium release in KOP-expressing cells with high potency and efficacy.
Summing up, both chimeric peptides displayed slightly reduced but still high binding affinity and agonist activity at MOP and KOP, which could be attributed to the RP-170 portion. At the NOP receptor, only KW-495 mimicked the effects of Ac-RYYRIK-NH 2 , although with a circa 6-fold reduced potency. Therefore, KW-495 was selected for the in vivo experiments.

Antinociceptive Activity
The antinociceptive activity of KW-495 was assessed in the hot-plate test. Mice's response to heat was recorded after intrathecal (i.t.) administration of this analog. The doseresponse experiment results are shown in Figure 2. KW-495 elicited strong, dose-dependent antinociceptive activity with an ED 50  In CHONOP+Gαqi5 cells, the standard agonist N/OFQ evoked a robust concentrationdependent stimulation of calcium release, displaying very high potency and efficacy (pEC50 = 9.26, α = 1.00). Ac-RYYRIK-NH2 and KW-495 mimicked the stimulatory effect elicited by N/OFQ, showing lower potencies (pEC50 = 7.87 and 7.12, respectively) and efficacies (α = 0.80 and 0.69, respectively). KW-496 displayed an incomplete concentration-response curve, being able to stimulate calcium mobilization only at 1 µM. The parent RP-170 was completely inactive ( Table 2).
Summing up, both chimeric peptides displayed slightly reduced but still high binding affinity and agonist activity at MOP and KOP, which could be attributed to the RP-170 portion. At the NOP receptor, only KW-495 mimicked the effects of Ac-RYYRIK-NH2, although with a circa 6-fold reduced potency. Therefore, KW-495 was selected for the in vivo experiments.

Antinociceptive Activity
The antinociceptive activity of KW-495 was assessed in the hot-plate test. Mice's response to heat was recorded after intrathecal (i.t.) administration of this analog. The doseresponse experiment results are shown in Figure 2. KW-495 elicited strong, dose-dependent antinociceptive activity with an ED50 value (jumping response) of 4.59 (1.17-18.09) nmol. In the dose-response experiments with EM-2, the ED50 was 3.15 (1.05-9.51) nmol.

Receptor Antagonist Experiments
To characterize the involvement of opioid and NOP receptors in the antinociceptive action of KW-495, co-administration studies with naloxone (a non-selective opioid receptor antagonist) and SB-612111 (an NOP-selective receptor antagonist) were performed. The antinociceptive effect of KW-495 (6.2 nmol/animal, i.t.) was blocked by both antagonists (1 mg/kg, i.p.), showing the involvement of opioid and NOP receptors in eliciting antinociception ( Figure 3).
To characterize the involvement of opioid and NOP receptors in the antinociceptive action of KW-495, co-administration studies with naloxone (a non-selective opioid receptor antagonist) and SB-612111 (an NOP-selective receptor antagonist) were performed. The antinociceptive effect of KW-495 (6.2 nmol/animal, i.t.) was blocked by both antagonists (1 mg/kg, i.p.), showing the involvement of opioid and NOP receptors in eliciting antinociception ( Figure 3). . n = 10-12 mice for each experimental group. Statistical significance was assessed using one-way ANOVA followed by Dunnett's multiple comparisons test. * is for p < 0.05 and ** is for p < 0.01 as compared with KW-495.

Motor Performance Study (Rotarod Test)
The administration of KW-495 (6.2 nmol/animal, i.t.) did not cause any significant effect on the motor coordination of mice as assessed during the rotarod test when compared to the control group ( Figure 4).  . n = 10-12 mice for each experimental group. Statistical significance was assessed using one-way ANOVA followed by Dunnett's multiple comparisons test. * is for p < 0.05 and ** is for p < 0.01 as compared with KW-495.

Motor Performance Study (Rotarod Test)
The administration of KW-495 (6.2 nmol/animal, i.t.) did not cause any significant effect on the motor coordination of mice as assessed during the rotarod test when compared to the control group ( Figure 4).
To characterize the involvement of opioid and NOP receptors in the antinociceptive action of KW-495, co-administration studies with naloxone (a non-selective opioid receptor antagonist) and SB-612111 (an NOP-selective receptor antagonist) were performed. The antinociceptive effect of KW-495 (6.2 nmol/animal, i.t.) was blocked by both antagonists (1 mg/kg, i.p.), showing the involvement of opioid and NOP receptors in eliciting antinociception ( Figure 3). . n = 10-12 mice for each experimental group. Statistical significance was assessed using one-way ANOVA followed by Dunnett's multiple comparisons test. * is for p < 0.05 and ** is for p < 0.01 as compared with KW-495.

Motor Performance Study (Rotarod Test)
The administration of KW-495 (6.2 nmol/animal, i.t.) did not cause any significant effect on the motor coordination of mice as assessed during the rotarod test when compared to the control group ( Figure 4).

Computational Studies
On resuming the above-described experiments, the parent peptide RP-170 was a dual MOP/KOP ligand with some preference for MOP (sub-nM vs. nM, respectively). As it turned out, the chimeric peptides KW-495 and KW-496 showed reduced affinity for MOP, so that KW-495 became a dual MOP/KOP ligand with equal nM affinity. Interestingly, KW-496 gained affinity for KOP, hence becoming a dual ligand with superior affinity for KOP (sub-nM). On the other hand, the chimeric KW-495 was also active towards NOP, hence representing a mixed MOP/KOP/NOP agonist.
The similar MOP affinity of KW-495 and KW-496, having practically the same pKi values in the nM order of magnitude, likely descends from the Tyr-c[D-Lys-Phe-Phe-Asp]-portion inherited from RP-170. The presence of the -Arg-Tyr-Tyr-Arg-Ile-Lys-NH 2 sequence with or without the 3-Gly linker seems to have little relevance. Previous docking simulations showed that RP-170 is in fact able to bind effectively to MOP thanks to a series of favorable interactions, including the ionic bond between the protonated amine of Tyr and Asp147 (3.32) [47]. Therefore, the interactions of KW-495 and KW-496 with MOP have been not investigated further.
On the other hand, other non-obvious experimental pharmacological results aroused some interest, i.e., the increased affinity of KW-496 towards KOP, and the activity of KW-495 towards NOP. In this perspective, we performed computational investigations using the deposited X-ray structures of the receptors. In particular, since the chimeric KW-496 and KW-495 differ only in the 3-Gly sequence, molecular docking computations seemed appropriate to shed light on any role played by this linker.
Since KW-495 is a mixed agonist, we first considered the possibility that this hybrid might represent a bivalent ligand that targets the receptor heterodimer. In other words, each portion of the hybrid peptide might bind the cognate receptor simultaneously, rather than targeting two GPCRs that are not linked. However, this possibility was ruled out due to the lack of a suitable spacer between the two bioactive portions of the sequence. Indeed, it has been well-documented that bivalent ligands targeting GPCR dimers must be bridged by a spacer with a length between 16 and 22 heavy atoms [48,49].
Docking of KW-495/KOP and KW-496/KOP. Hence, the bioactive conformations of KW-495 and KW-496 were analyzed with AutoDock 4.0 [50] in a model of the receptor obtained from the deposited X-ray structure of human KOP in a complex with the potent opioid agonist MP1104 (PDB ID 6B73) [51]. It was postulated that the activity of the two compounds at KOP mainly resided in the cyclopeptide portion. For each ligand, independent docking runs were performed, and the results were scored and clustered. Protein-ligand complexes from selected docked poses were minimized in explicit TIP3P water molecules and equilibrated by molecular dynamics. More details can be found in the literature [52] and in the experimental section. The overlay of KW-495 and KW-496 into KOP is illustrated in Figure S6, and shows that the two peptide conjugates occupy the same cavity of the receptor, albeit with distinct orientations.
The pose of KW-495, Tyr 1 -c[D-Lys-Phe-Phe-Asp]-Arg 6 -Tyr-Tyr-Arg-Ile-Lys 11 -NH 2 , with the best score is shown in Figure 5. As expected, in this structure, the Tyr 1 -c[D-Lys-Phe-Phe-Asp] cyclopentapeptide is deeply buried in the crevice delimited by the TM domains 2, 3 and 5-7, while the C-terminal portion remains external to the TM domain, fully entangled within the bends of the extracellular loop 2 (EL-2), making contact only with the top of the transmembrane helices (TM) 5 and 6.
All details are reported in the Supporting Information ( Figure S6 and S7); for brevity, herein, only the most relevant contacts are discussed. The protonated Tyr 1 of the cyclopeptide adopts a classic disposition in the message-binding region of the receptor, interacting with Asp138 (3.32) carboxylate via an ionic bond; the phenolic side chain makes a hydrogen bond (H-bond) with His291 (6.52) [53]. Other stabilizing contacts include Met142, Ile290 and Ile294. As for the C-terminal sequence Arg 6 -Tyr-Tyr 8 -Arg 9 -Ile-Lys 11 NH 2 , the most relevant interactions involve the charged side chains. Arg 6 makes an ionic bond with Asp223 and H-bonds with Leu212 and Gln213; Tyr 8 phenol is H-bonded to Arg202; Arg 9 is H-bonded to Phe214 and Gln213; Lys 11 NH 2 makes an ionic bond with Asp216 and H-bonds with Ser220 and Leu224 (Supporting Information).
In the best pose of KW-496, Tyr 1 -c[D-Lys-Phe 3 -Phe 4 -Asp]-Gly 3 -Arg-Tyr-Tyr-Arg-Ile-Lys-NH 2 ( Figure 5), the cyclopentapeptide adopts a different orientation within the TM domains as compared to KW-495. The protonated amine of Tyr 1 maintains the salt bridge with Asp138 (3.32) carboxylate, while the phenolic side chain shows interactions with Trp287 (pi-pi), Tyr320 (pi-pi and H-bond Tyr320NH-phenolO) and Ile290 (pi-alkyl). Interestingly, in this pose the message tyramine of the cyclic portion of KW-496 shows some similarities with that of the cyclic peptide alone, RP-170, as calculated in MOP [47]. Albeit not totally unusual for KOP ligands [54], this pose is to some extent an alternative to that of the tetrahydroisoquinoline ring of JDTic and to the docked pose of the ligand dynA(1-8) [55], in which the tyramine phenol groups point towards His291.
The Gly 3 linker is hosted in a vertical position between the two bends of the EL-2. The Arg 9 -Tyr-Tyr-Arg-Ile-Lys 14 -NH 2 portion holds on to the top of the helices 2 and 5-7, wrapped within the EL ribbons. All contacts are discussed in detail in the Supporting Information. Interestingly, the C-terminal peptide contributes to stabilizing the complex thanks to several meaningful contacts (Supporting Information). The guanidinic groups of Arg 9 and Arg 12 show ionic bonds with the carboxylates of Asp223 (top of TM-5) and Asp206 (EL-2), respectively; the C-terminal Lys 14 NH 2 interacts with Ile208(EL-2)C=O (H-bond with NHζ + ), Val118C=O (H-bond with CONH 2 ) and Asn122C=O (H-bond with LysNH).
with Asp138 (3.32) carboxylate, while the phenolic side chain shows interactions with Trp287 (pi-pi), Tyr320 (pi-pi and H-bond Tyr320NH-phenolO) and Ile290 (pi-alkyl). Interestingly, in this pose the message tyramine of the cyclic portion of KW-496 shows some similarities with that of the cyclic peptide alone, RP-170, as calculated in MOP [47]. Albeit not totally unusual for KOP ligands [54], this pose is to some extent an alternative to that of the tetrahydroisoquinoline ring of JDTic and to the docked pose of the ligand dynA(1-8) [55], in which the tyramine phenol groups point towards His291.
The Gly3 linker is hosted in a vertical position between the two bends of the EL-2. The Arg 9 -Tyr-Tyr-Arg-Ile-Lys 14 -NH2 portion holds on to the top of the helices 2 and 5-7, wrapped within the EL ribbons. All contacts are discussed in detail in the Supporting Information. Interestingly, the C-terminal peptide contributes to stabilizing the complex thanks to several meaningful contacts (Supporting Information). The guanidinic groups of Arg 9 and Arg 12 show ionic bonds with the carboxylates of Asp223 (top of TM-5) and Asp206 (EL-2), respectively; the C-terminal Lys 14 NH2 interacts with Ile208(EL-2)C=O (Hbond with NHζ + ), Val118C=O (H-bond with CONH2) and Asn122C=O (H-bond with LysNH). Docking of KW-495/NOP and KW-496/NOP. The docking simulations of KW-495 and KW-496 into NOP were performed as discussed above. Previous studies indicated that the sequence RYYRIK-NH2 competes for the same NOP binding site as N/OFQ [58,59]; therefore, in this study the C-terminal sequence was assumed to be the message portion of the chimeric peptides for NOP. The structure of the receptor was extracted from the crystal structure of NOP in the complex with the ligand C-35, PDB ID 5DHG [60].
In the best-scoring pose of KW-495, Tyr1-c[D-Lys-Phe-Phe-Asp]-Arg 6 -Tyr-Tyr-Arg-Ile-Lys 11 -NH2, the C-terminal portion Tyr 8 -Arg 9 -Ile-Lys 11 NH2 adopts a bent conformation, spanning across the cavity composed by residues of the helices TM-2, 3 and 5-7 ( Figure  6). The complex is nicely stabilized by several strong interactions. The guanidinic group of Arg 9 forms an ionic bond with the carboxylate of Asp138 (3.32) , and a H-bond with Cys200C=O. The branched side chain of Ile 10 nicely fits in a hydrophobic cavity shaped by elements of TM-5 and 6 (Val283, Val279 and Leu301). The C-terminal Lys 11 NH2 is involved Docking of KW-495/NOP and KW-496/NOP. The docking simulations of KW-495 and KW-496 into NOP were performed as discussed above. Previous studies indicated that the sequence RYYRIK-NH 2 competes for the same NOP binding site as N/OFQ [58,59]; therefore, in this study the C-terminal sequence was assumed to be the message portion of the chimeric peptides for NOP. The structure of the receptor was extracted from the crystal structure of NOP in the complex with the ligand C-35, PDB ID 5DHG [60].
In the best-scoring pose of KW-495, Tyr1-c[D-Lys-Phe-Phe-Asp]-Arg 6 -Tyr-Tyr-Arg-Ile-Lys 11 -NH 2 , the C-terminal portion Tyr 8 -Arg 9 -Ile-Lys 11 NH 2 adopts a bent conformation, spanning across the cavity composed by residues of the helices TM-2, 3 and 5-7 ( Figure 6). The complex is nicely stabilized by several strong interactions. The guanidinic group of Arg 9 forms an ionic bond with the carboxylate of Asp138 (3.32) , and a H-bond with Cys200C=O. The branched side chain of Ile 10 nicely fits in a hydrophobic cavity shaped by elements of TM-5 and 6 (Val283, Val279 and Leu301). The C-terminal Lys 11 NH 2 is involved in interactions with Gln107C=O (H-bond), and the carboxylate side chains of both Glu199 and Asp110 (NHζ + , ionic bond). Full details are given in the Supporting Information.

fills a deep pocket delimited by TM helices 4 and 5 (Supporting Information).
H-bond), while the phenol ring of Tyr also interacts with Glu197carboxylate (pi-anion).

Discussion
Opioid and NOP receptors are highly expressed in the CNS and play important roles in the direct and indirect control of pain transmission [9,17,61]. The agonists of the classical opioid receptors all produce strong analgesic effects. N/OFQ, when injected intracerebroventricularly (i.c.v.), was shown to display nociceptive activity, suggesting an anti-opioid profile, leading to the assumption that NOP antagonists should have antinociceptive activity [62]. However, other studies reported that N/OFQ potentiated morphine-induced analgesia when injected into the spinal cord, drawing into question the anti-opioid activity of this ligand [63].
With its ambiguous pharmacological profile, N/OFQ appeared as an interesting ligand for use in the design of hybrid molecules. However, due to the long sequence of N/OFQ, a much shorter synthetic peptide, i.e., the NOP ligand Ac-RYYRIK-NH2, was preferred as a fragment that would be convenient to combine with MOP or DOP agonists, yielding potential bivalently active opioid/NOP compounds. Kawano et al. [64] joined this peptide with the MOP agonist, dermorphin (Tyr-D-Ala-Gly-Tyr-Pro-Ser), or its shorter

Discussion
Opioid and NOP receptors are highly expressed in the CNS and play important roles in the direct and indirect control of pain transmission [9,17,61]. The agonists of the classical opioid receptors all produce strong analgesic effects. N/OFQ, when injected intracerebroventricularly (i.c.v.), was shown to display nociceptive activity, suggesting an anti-opioid profile, leading to the assumption that NOP antagonists should have antinociceptive activity [62]. However, other studies reported that N/OFQ potentiated morphine-induced analgesia when injected into the spinal cord, drawing into question the anti-opioid activity of this ligand [63].
With its ambiguous pharmacological profile, N/OFQ appeared as an interesting ligand for use in the design of hybrid molecules. However, due to the long sequence of N/OFQ, a much shorter synthetic peptide, i.e., the NOP ligand Ac-RYYRIK-NH 2 , was preferred as a fragment that would be convenient to combine with MOP or DOP agonists, yielding potential bivalently active opioid/NOP compounds. Kawano et al. [64] joined this peptide with the MOP agonist, dermorphin (Tyr-D-Ala-Gly-Tyr-Pro-Ser), or its shorter analog Tyr-D-Ala-Phe-β-Ala. MOP and NOP ligands were connected tail-to-tail by various spacers. The so-obtained dermorphin-RYYRIK-NH 2 hybrids retained high binding affinities for both receptors, which increased with the length of the spacer. The authors concluded that longer spacers prevented the interference of the closely located pharmacophores with their respective binding sites. In vivo, the hybrids showed potent and dose-dependent antinociceptive activities in the tail-flick test in mice following i.t. administration [58].
Guillemyn et al. [31] linked Ac-RYYRIK-NH 2 to the opioid ligand Dmt-D-Arg-Abaβ-Ala-NH 2 (Dmt: 2 ,6 -dimethyl-L-tyrosine; Aba = tetrahydro-4-amino-2-benzazepinone). The resulting hybrid, Dmt-D-Arg-Aba-β-Ala-Arg-Tyr-Tyr-Arg-Ile-Lys-NH 2 (KGNOP1), showed decreased affinity to opioid and NOP receptors, behaving as an agonist at MOP and an antagonist at NOP. In rats, KGNOP1 produced a strong analgesic effect after i.t. administration, but repeated daily injections led to the development of tolerance [65]. Further in vivo studies showed that in the formalin test, KGNOP1 was highly potent and efficacious in rodent models of acute nociceptive, acute and chronic inflammatory pain and chronic neuropathic pain after systemic (i.v. or s.c.) administration. Antinociceptive effects of KGNOP1 were reversed by naltrexone and SB-612111, indicating the involvement of both MOP and NOP receptor agonism. This is in contrast to the in vitro data indicating antagonism at NOP [66].
Recently, Erdei et al. [59,67] described three hybrid ligands, combining Ac-RYYRIK-NH 2 with N-terminal tetra-or pentapeptide sequences of DOP and KOP agonists, i.e., Leu-enkephalin and dynorphin. Both fragments were linked either directly or through a 3-Gly spacer. The YGGF-RYYRIK-NH 2 analog deprived of a spacer lost most of the DOPmediated activity but exhibited the most efficacious agonist action in the MOP-mediated G-protein stimulation assay.
In this context, we decided to design chimeric peptides composed of RP-170, a mixed MOP/KOP cyclic peptide ligand with a preference for MOP, and the NOP active Ac-RYYRIK-NH 2 (Figure 1) . Herein, we report the synthesis and the initial pharmacological testing of two chimeras in which RP-170 and -RYYRIK-NH 2 were linked directly (KW-495) or through a 3-Gly spacer (KW-496).
In vitro binding and functional assays showed that KW-496 became a mixed ligand with inverted KOP > MOP affinity, without activity at NOP. On the other hand, KW-495 could simultaneously and potently bind and activate opioid and NOP receptors. In vivo, KW-495 produced a dose-dependent antinociceptive effect when given i.t., and this effect was antagonized by naloxone and SB-612111. This finding demonstrates the involvement of both classical opioid and NOP receptors in the antinociceptive action of KW-495. Finally, this peptide did not alter the locomotor activity, and there was no change in the motor coordination and balance of mice.
Taken together, these findings demonstrate that the cyclic opioid RP-170 and the NOP ligand Ac-RYYRIC-NH 2 can be successfully joined to generate a chimeric peptide able to simultaneously target NOP and classical opioid receptors.
Computational simulations of the chimeric peptides docked in the receptors may help us in understanding the structural determinants at the basis of the comparatively higher KOP activity of KW-496, and the significant NOP activity of KW-495. In KW-495, the NOP-active sequence RYYRIK-NH 2 is directly connected to the MOP/KOP ligand RP-170, while KW-496 possesses a Gly 3 sequence in between.
In KW-495/KOP, the protonated Tyr 1 of the cyclopeptide is predicted to adopt the classic disposition within the message-binding region of the receptor, forming the expected ionic bond with Asp138 (3.32) carboxylate and a hydrogen bond (H-bond) with His291 (6.52) ; the C-terminal RYYRIK-NH 2 portion remains external to the TM domain, fully entangled within the bends of EL-2.
In contrast, the presence of a Gly 3 spacer in KW-496/KOP is predicted to allow a different position of the cyclopeptide and of the C-terminal RYYRIK-NH 2 . The computations suggest that the spacer spans vertically across the transmembrane domain. This thin and flexible connection seems to allow the N-terminal cyclopentapeptide to adopt an alternative orientation ( Figure 5 and Supporting Information). In this structure, the tyramine message portion of the cyclic peptide of KW-496 maintains in KOP a similar disposition to that adopted by the tyramine of the cyclic peptide RP-170 alone as calculated in MOP [47]. This significant difference might in part explain the higher experimental affinity of KW-496 as compared to KW-495. In addition, the residues of the C-terminal section can attain further, highly favorable interactions with the residues at the outer side of TM2 and the EC loops ( Figure 5).
The significant NOP activity of KW-495, deprived of any linkers, was somewhat unexpected. Kawano et al. [64] joined the MOP agonist dermorphin to RYYRIK-NH 2 with various spacers, and observed that binding to both receptors increased with the length of the spacer. The authors proposed that longer spacers allowed avoiding any interference of the closely located pharmacophores with their respective binding sites.
The structure of KW-495/NOP is characterized by a transversal, bent conformation of the C-terminal tetrapeptide ( Figure 6 and Supporting Information). In 2012, Daga and Zaveri used the homology model of NOP to analyze the binding of Phe-Gly-Gly-Phe-Thr-Gly-Ala-Arg-Lys-Ser-Ala-Arg-Lys, N/OFQ (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13). These authors proposed that the N-terminal tetrapeptide Phe-Gly-Gly-Phe of N/OFQ penetrates deeply into the binding pocket in a straight conformation, and the N-terminal amine makes an essential charge interaction with Asp130 [68]. Very recently, Dumitrascuta et al. simulated NOP binding of the chimeric peptide Dmt 1 -D-Arg-Aba-β-Ala-Arg-Tyr-Tyr-Arg 8 -Ile-Lys 10 NH 2 (KGNOP1), a full agonist for both MOP and NOP [66]. Similar to KW-496, the computations performed by Dumitrascuta et al. showed an extended conformation of the C-terminal tetrapeptide, with Arg 8 and Lys 10 forming salt bridges with Asp130 3:32 , while the linear N-terminal tetrapeptide was excluded from the receptor, being fully exposed to the solvent. In contrast to KGNOP1, the N-terminal portion of KW-495 external to the receptor is represented by a large cyclopeptide scaffold. The rigidity and the noteworthy impediment exerted by the latter seem to prevent the C-terminal sequence from fully plunging into the receptor's depths. Nevertheless, the computations predict that the flexible C-terminal sequence is still capable of establishing strong interactions with NOP, while the cyclopeptide is tightly pressed against the outside of the receptor, and significantly contributes to overall binding with several strong interactions (Supporting Information).
Crude peptides were purified by preparative reversed-phase HPLC on a Vydac C18 column (10 µm, 22 mm × 250 mm), flow rate 2 mL/min, 20 min linear gradient from water/0.1% (v/v) TFA to 80% acetonitrile/20% water/0.1% (v/v) TFA. The purity of the final peptides was verified by analytical HPLC employing a Vydac C18 column (5 µm, 4.6 mm × 250 mm), flow rate 1 mL/min and the same solvent system over 50 min. The purity of the obtained peptides was >95%. Calculated values for protonated molecular ions were in agreement with those determined by high-resolution mass spectroscopy with electrospray ionization (ESI-MS) (Table S1).

Membrane Preparation
CHO cells (a kind gift from Dr F. Marshall and Mrs N. Bevan, GSK, Stevenage, UK) stably expressing the human nociceptin/orphanin FQ peptide receptor (CHO hNOP ) were cultured in Dulbecco's Modified Eagles Medium/Nutrient Mixture F12 Ham (50/50) supplemented with 5% fetal calf serum, penicillin (50 IU/mL), streptomycin (100 µg/mL), fungizone (2.5 µg/mL) and L-glutamine. Stock cell culture media were further supplemented with geneticin (G418) (2 µg/mL), hygromycin-B (2 µg/mL) and cells grown at 37 • C in 5% carbon dioxide humidified air. Cells were used at confluence and membrane fragments prepared by homogenization of cells followed by centrifugation at 13,500 rpm for 10 min at 4 • C. This was repeated for a total of three times [69]. Protein concentration was determined using the method described by Lowry [70].

Binding Data Analysis
The data were analyzed by a nonlinear least square regression analysis computer program Graph Pad PRISM 6.0 (Graph Pad Software Inc., San Diego, CA, USA). The IC 50 values were determined from the logarithmic concentration-displacement curves, and the values of the inhibitory constants (K i ) were calculated according to the equation of Cheng and Prusoff [71].

Calcium Mobilization Functional Assay
For the experiments, CHO cells (a kind gift from Prof. T. Costa, Rome, Italy) stably co-expressing the human MOP, DOP, KOP or NOP and the C-terminally modified G-protein were generated as described [46]. Briefly, cells incubated for 24 h in 96-well black, clearbottom plates were loaded with medium supplemented with probenecid (2.5 mmol/L), calcium-sensitive fluorescent dye Fluo-4 AM (3 µmol/L) and pluronic acid (0.01%) and kept for 30 min at 37 • C. Following aspiration of the loading solution and a washing step, serial dilutions of peptide stock solutions were added. After placing cell culture and compound plates into the FlexStation II (Molecular Devices, Sunnyvale, CA, USA), changes in fluorescence of the cell-loaded calcium-sensitive dye Fluor-4 AM were measured. On-line additions were carried out in a volume of 50 µL/well.

Functional Assay Data Analysis and Terminology
All data were analyzed using Graph Pad Prism 6.0 (La Jolla, CA, USA). Concentrationresponse curves were fitted using the four-parameter log-logistic equation. Data are expressed as mean ± SEM of at least four experiments performed in duplicate. Agonist effects were expressed as maximum change in percent over the baseline fluorescence. Baseline fluorescence was measured in wells treated with vehicle. Agonist potency was expressed as pEC 50 , which is the negative logarithm to base 10 of the agonist molar concentration that produces 50% of the maximal possible effect of that agonist. Concentration-response curves to agonists were fitted with the four-parameter logistic nonlinear regression model:

Animals
Male CD-1 mice (Charles River, Italy), weighing 25-30 g, were used for the study. The animals were housed in a room with controlled temperature (21 ± 1 • C), humidity (70 ± 5%) and light/dark cycle conditions (12/12 h), with free access to laboratory chow and tap water. Mice were housed for at least 1 week before the experimental sessions in sawdust-lined plastic cages (6 mice in each cage). Animal studies and research protocols were approved by the Service for Biotechnology and Animal Welfare of the Istituto Superiore di Sanità and authorized by the Italian Ministry of Health, according to Legislative Decree 26/14 (protocol number: 198/2013-B), which implemented the European Directive 2010/63/EEC on laboratory animal protection. Animal welfare was routinely checked by veterinarians from the Service for Biotechnology and Animal Welfare.

Drugs and Pharmacological Treatment
The tested compounds were dissolved in DMSO and further diluted with 0.9% NaCl solution to the final concentration of 0.1% DMSO. Animals without treatment (control group) received the vehicle alone (0.1% DMSO in 0.9% NaCl solution). The vehicle given alone had no effects on the observed parameters.

Assessment of Antinociception
The antinociceptive effects of peptides were assessed in the hot-plate test in mice as described earlier [72]. The intrathecal (i.t.) injections (5 µL/animal) of peptide or vehicle were performed under inhalational isoflurane anesthesia using the method described by Hylden and Wilcox [73]. Injections were performed with disposable, 30-gauge, 1 2 -inch needles mated to a 10 µL syringe. Mice were held firmly by the pelvic girdle, while the needle was introduced at an angle of about 20 • above the vertebral column. The needle was inserted between the L5 and L6 spinous processes and moved carefully forward to the intervertebral space as the angle of the syringe was decreased to about 10 • . The tip of the needle was inserted approximately 0.5 cm into the vertebral column. Antagonists (naloxone hydrochloride or SB-612111 hydrochloride) were intraperitoneally (i.p.) administered 15 min prior to the peptide or vehicle injection at the dose of 1 mg/kg. A transparent plastic cylinder (14 cm diameter, 20 cm height) was used to confine a mouse on the heated (55 ± 0.5 • C) surface of the plate (Ugo Basile, No. 7280, Gemonio, Italy). The animals were placed on the hot plate 5 min after the i.t. injection of saline (control) or peptide, and the latencies to jumping were measured at 5, 10, 20, 30, 45 and 60 min after administration of peptide. A cut-off time of 120 s was used to prevent tissue injury. The percentage of the maximal possible effect (%MPE) was calculated as: %MPE = (t1 − t0)/(t2 − t0) × 100, where t0: control latency, t1: test latency, and t2: cut-off time. The median antinociceptive dose (ED50) was calculated according to the method of Litchfield and Wilcox [74]. The data were analyzed by a nonlinear least-square regression analysis computer program, Graph Pad PRISM 6.0 (GraphPad Software Inc., San Diego, CA, USA).

Assessment of Motor Coordination
Rotarod apparatus (Ugo Basile, No. 8466, Italy) was used to estimate the probable influence of peptide on motor coordination of animals [75]. In the training period, 3 trials were conducted for 3 consecutive days. Animals that were able to stay on the rotating (16 revolutions/min) bar more than 180 s were chosen for the tests. In the experimental session, each animal was tested on the rotarod, and duration to fall from the rotating bar was noted.

Molecular Docking
The KOP model was obtained from the deposited X-ray structure PDB ID 6B73, while the NOP model was obtained from the crystal structure PDB ID 5DHG. The initial structures of the ligands were optimized using the program ORCA [76] with a Hartree-Fock method and a 6-31G* basis set.
The simulations were performed with AutoDock 4.0, using the Lamarckian genetic algorithm, which combines global (genetic algorithm) and local (Solis and Wets algorithm) search. Ligands and receptors were further processed using the AutoDock Tool Kit (ADT). Gasteiger-Marsili charges were loaded on the ligands in ADT and solvation parameters were added to the final structure using the Addsol utility. The docking runs were applied on an initial population of 100 random structures, with a maximum number of energy evaluations = 150, mutation rate = 0.02, a crossover rate = 0.80, elitism = 1.25. The socalled pseudo-Solis and Wets algorithm was applied to local search, with the maximum of iterations per local search per ligand = 200. The grid maps representing the system in the actual docking process were calculated with Autogrid. Grid dimensions were 80 × 80 × 80, and the spacing between the grid points and the center of the cavity left by the ligand after its removal was set to 0.1 Å. Structures differing < 1.0 Å in positional root mean square deviation (RMSD) were clustered together. For each cluster, the structure with the most favorable free energy of binding was selected for subsequent minimization of the protein-ligand complexes.
The selected ligand-protein complex structures were prepared by assigning AM1-BCC charges [77] to the ligands and AMBER 14SB force field [78] parameters to the proteins. Moreover, the complexes were solvated with TIP3P water molecules. Additional Na + and Clions were placed in the water box to obtain a physiological ion strength (200 mM). Systems were minimized and equilibrated at 300 K and 1 atm by performing 1 ns MD simulations with position restraints on the protein backbone.