Contents lists available atScienceDirect
Peptides
journal homepage:www.elsevier.com/locate/peptides
In vitro and in vivo activity of cyclopeptide Dmt-c[
D
-Lys-Phe-Asp]NH
2
, a mu
opioid receptor agonist biased toward
β-arrestin
Katarzyna Gach-Janczak
a,1, Justyna Piekielna-Ciesielska
a,1, Anna Adamska-Bart
łomiejczyk
a,
Karol Wtorek
a, Federica Ferrari
b, Girolamo Calo
’
b, Agata Szymaszkiewicz
c,
Joanna Piasecka-Zelga
d, Anna Janecka
a,⁎aDepartment of Biomolecular Chemistry, Medical University, Lodz, Poland
bDepartment of Medical Sciences, Section of Pharmacology and Italian Institute of Neuroscience, University of Ferrara, 44121 Ferrara, Italy cDepartment of Biochemistry, Faculty of Medicine, Medical University of Lodz, Poland
dInstitute of Occupational Medicine, Research Laboratory for Medicine and Veterinary Products in the GMP Head of Research Laboratory for Medicine and Veterinary
Products, Lodz, Poland
A R T I C L E I N F O Keywords:
Cyclic opioid peptides Binding assay
Calcium mobilization assay
Bioluminescence resonance energy transfer assay
Hot-plate test
Whole gastrointestinal transit test
A B S T R A C T
Morphine and related drugs, which are the most effective analgesics for the relief of severe pain, act through activating opioid receptors. The endogenous ligands of these receptors are opioid peptides which cannot be used as antinociceptive agents due to their low bioactivity and stability in biologicalfluids. The major goal of opioid research is to understand the mechanism of action of opioid receptor agonists in order to improve therapeutic utility of opioids. Analgesic effects of morphine are mediated mostly through activation of the mu opioid re-ceptor. However, in the search for safer and more effective drug candidates, analogs with mixed opioid receptor profile gained a lot of interest. Recently, the concept of biased agonists able to differentially activate GPCR downstream pathways, became a new approach in the design of novel drug candidates. It is hypothesized that compounds promoting G-protein signaling may produce analgesia whileβ-arrestin recruitment may be re-sponsible for opioid side effects. In this report we showed that replacement of the tyrosine residue in the mu-selective ligand Tyr-c[D-Lys-Phe-Asp]NH2with 2′,6′-dimethyltyrosine (Dmt) produced a cyclopeptide Dmt-c[D -Lys-Phe-Asp]NH2with mu/delta opioid receptor agonist profile. This analog showed improved antinociception in the hot-plate test, probably due to the simultaneous activation of mu and delta receptors but also significantly inhibited the gastrointestinal transit. Using the bioluminescence resonance energy transfer (BRET) assay it was shown that this analog was a mu receptor agonist biased towardβ-arrestin. β-Arrestin-dependent signaling is most likely responsible for the observed inhibition of gastrointestinal motility exerted by the novel cyclopeptide.
1. Introduction
Centrally acting opioid agonists, such as morphine, are the most widely used analgesics for the treatment of severe pain[1]. Among the three types of classic opioid receptors, mu, delta and kappa, the mu receptor was identified as the one responsible primarily for the parelieving effects but also for a number of undesired side effects, in-cluding sedation, respiratory depression, inhibition of gastrointestinal transit and also development of tolerance and physical dependence[2]. In the previous decades, extensive structure-activity relationship studies of opioid receptor ligands concentrated on the obtaining analogs with high selectivity for one opioid receptor type. More recently, the development of compounds with mixed opioid profile is gaining a lot of
interest[3,4]. Several opioid analogs with peptide or alkaloid structure that represent this new strategy have already been synthesized and evaluated in vitro and in vivo.
Kappa selective agonists produce analgesia accompanied by some dysphoric effects [5]and this property limited their therapeutic de-velopment. However, mixed mu/kappa agonists (such as ethylk-etazocine) display fewer side-effects than their kappa-selective coun-terparts such as enadoline and spiradoline[6,7]and have been used to treat cocaine addiction[8]. Presumably, the effects produced by mu-agonists help attenuate the dysphoric action associated with kappa agonism.
A combination of buprenorphine (a partial mu agonist/kappa an-tagonist) and naltrexone (a non-selective anan-tagonist) produced
https://doi.org/10.1016/j.peptides.2018.04.014
Received 5 January 2018; Received in revised form 16 April 2018; Accepted 18 April 2018
⁎Corresponding author at: Department of Biomolecular Chemistry, Medical University of Lodz, Mazowiecka 6/8, 92-215 Lodz, Poland. 1Equal contribution.
E-mail address:anna.janecka@umed.lodz.pl(A. Janecka).
Available online 22 April 2018
0196-9781/ © 2018 Published by Elsevier Inc.
antidepressant-like activity in mice and may represent a novel approach in the treatment of depression[9]. The synthesis of mixed mu/kappa receptor antagonists, combining these two activities in one compound, would overcome the abuse liability issue and also the issue of a proper dosing ratio of two separate entities. A new orvinol analog BU10119 with the mu/kappa antagonist affinity profile has recently been ob-tained by Cueva et al.[10]. In vitro BU 10119 showed high affinity for kappa and mu receptors with little efficacy at both these receptors, indicating an antagonist-like profile. The initial characterization of this analog in vivo in mouse models of depression showed the therapeutic potential of BU 10119 for the treatment of depression and other stress-induced conditions[11].
In 1991 Abdelhamid et al.[12]demonstrated for thefirst time that selective delta opioid receptor blockade with naltrindole (delta opioid antagonist) greatly reduced the development of morphine tolerance and dependence in mice. Opioid ligands combining mu agonist/delta an-tagonist activity in one compound were therefore sought for the de-velopment of analgesics with lower propensity to produce tolerance and physical dependence[13].
Thefirst known compound with mixed mu agonist/delta antagonist profile was the tetrapeptide amide Tyr-Tic-Phe-Phe-NH2 (TIPP-NH2, were Tic represents 1,2,3,4-tetrahydroisoquinoline-3-carboxylic acid) [14]. TIPP-NH2 showed quite high delta antagonist but modest mu agonist potency. Further modifications including replacement of Tyr1 residue by Dmt (Dmt = 2′,6′-dimethyltyrosine) and a reduction of the peptide bond between Tic2 and Phe3 led to a pseudopeptide Dmt-Ticψ[CH2NH]-Phe-Phe-NH2(DIPP-NH2[ψ]) with greatly improved mu potency. As expected, this analog exhibited reduced tolerance com-pared to morphine and no dependence, when administered in-tracerebroventricularly [15]. However, the therapeutic potential of DIPP-NH2[ψ] was compromised by its poor blood-brain barrier (BBB) permeability[16].
Purington et al.[17]reported the synthesis of a cyclic tetrapeptide KSK-103, containing a S‐CH2‐CH2‐S- bridge that showed equal and high affinity for mu and delta receptors but, similar to the previously re-ported ligands, had poor bioavailability. Glycosylation of this cyclo-peptide produced an analog that retained the desirable in vitro profile, while displaying centrally mediated antinociception after in-traperitoneal (i.p.) injection with a potency similar to morphine[18].
It was also shown that potency and efficacy of the mu agonists can be enhanced by delta agonists. Synergistic antinociceptive effects in response to the mu and delta receptor activation were observed in several in vivo studies. For example, enkephalins, DPDPE or deltorphine potentiated antinociceptive effect of morphine[19–22]. These results pointed at a functional interaction between mu and delta opioid re-ceptors and a potential regulatory role of the endogenous delta ligands in controlling pain.
Therefore, the search for single compounds combining mixed mu/ delta agonist activities followed. The best know and the most studied peptide ligand with the dual mu/delta agonist profile is biphalin [Tyr-D -Ala-Gly-Phe-NH-NH-Phe-Gly-D-Ala-Tyr], whose structure is based on
two enkephalin-like fragments connected“tail-to-tail” by a hydrazine bridge [23]. Biphalin showed high affinity for the mu and delta re-ceptors (with Kiabout 1–4 nM), very good antinociceptive activity and duration of action after intrathecal (i.t.) administration[24], induced less physical dependence than morphine[25]and could penetrate the BBB[26]. Several linear analogs of biphalin with amino acid substitu-tions in posisubstitu-tions 2,2′ and 4,4′ were also reported[27,28].
To overcome moderate stability of biphalin in human plasma[29] the group of Mollica developed several cyclic analogs of biphalin with a disulfide linkage[30,31]or a xylene bridge[32]. Such analogs showed increased in vitro affinity for the mu and delta receptors and prolonged in vivo antinociceptive effect[33,34].
In our earlier paper we described a cyclic tetrapeptide, Tyr-c[D
-Lys-Phe-Asp]NH2[35], which was highly mu selective. Here, replacement of Tyr by Dmt resulted in obtaining an analog with mu/delta profile and
very high potency for these two receptors. The pharmacological profile of this cyclopeptide was investigated in detail, both in vitro and in vivo.
2. Materials and methods
2.1. Materials
Peptide synthesis reagents (amino acids, MBHA Rink-Amide resin, TBTU) were purchased from Bachem AG (Bubendorf, Switzerland). Opioid radioligands, [3H]DAMGO, [3H]deltorphin-2 and [3H]U-69593, and cell membranes expressing human recombinant opioid receptors were purchased from PerkinElmer (Krakow, Poland). Loperamide hy-drochloride,β-funaltrexamine hydrochloride (β-FNA) and naltrindole hydrochloride (NLT) were obtained from Tocris Bioscience (Ellisville, MO, USA).
2.2. Peptide synthesis
Peptides were synthesized on solid support, using Fmoc chemistry as described elsewhere [36]. Crude peptides were purified using a Waters semipreparative HPLC (Waters Breeze instrument, Milford, MA, USA) with a Vydac C18column (10μm, 22 × 250 mm), using the sol-vent system of 0.1% TFA in water (A)/80% acetonitrile in water con-taining 0.1% TFA (B). A linear gradient of 0–100% B over 50 min at a flow rate of 1 mL/min was used. The molecular weight of peptides was confirmed by ESI–MS recorded on Bruker micrOTOF-Q mass spectro-meter (Bruker Daltonics, Bremen, Germany). The purity of the cyclo-peptides was verified by analytical HPLC employing a Vydac C18 column (5μm, 4.6 × 250 mm) and the solvent system of 0.1% TFA in water (A)/80% acetonitrile in water containing 0.1% TFA (B). All peptides were at least 96% pure as determined by HPLC monitored at 230 nm.
2.3. Radioligand binding assays
Radioligand binding assays were performed according to the pre-viously described method[37]using commercial membranes of Chi-nese Hamster Ovary (CHO) cells transfected with human opioid re-ceptors. The binding affinities for mu, delta and kappa opioid receptors were determined by radioligand competition analysis using [3H] DAMGO, [3H]deltorphin-2 and [3H]U-69593, respectively, as specific radioligands. Three independent experiments for each assay were car-ried out in duplicate. The data were analyzed by a nonlinear least square regression analysis computer program Graph Pad PRISM 6.0 (Graph Pad Software Inc., San Diego, USA). The IC50values were de-termined from the logarithmic concentration-displacement curves, and the values of the inhibitory constants (Ki) were calculated according to the equation of Cheng and Prusoff[38].
2.4. Calcium mobilization assay
Calcium mobilization assay was performed as reported in detail elsewhere [39], using CHO cells stably co-expressing human re-combinant mu or kappa opioid receptors and the C-terminally modified Gαqi5and CHO cells co-expressing the human recombinant delta opioid receptor and the GαqG66Di5chimeric protein.
Agonist potencies of peptides are given as pEC50that is the negative logarithm of the molar concentration of an agonist that produces 50% of the maximal possible effect. Ligand efficacy was expressed as in-trinsic activity (α).
2.5. Bioluminescence resonance energy transfer (BRET) assay
BRET assay was described in our earlier paper[40]. SH-SY5Y cell lines permanently co-expressing mu-RLuc and one of the transduction protein (Gβ1-RGFP orβ-arrestin 2-RGFP) were used to determine the
interaction of the mu receptor with G-protein andβ-arrestin 2. Methods for cell culturing, retroviral transduction, and BRET assay have been described previously [41,42]. Agonist responses were quantified as stimulated BRET ratio obtained by subtracting the vehicle value from that measured in the presence of the ligand.
For calculation of bias factors EM-2 was used as a standard unbiased ligand.
The concentration response curves of each compound werefitted to the Black-Leff operational model described by Nagi et al.[43]:
= + + response A τ E A τ A K [ ] [ ] ([ ] ) n n m n n An
where [A] is the agonist concentration, the maximal response of the system is given by Em, n is afitting parameter for the slope, the affinity of the agonist is represented by the equilibrium dissociation constant of the agonist-receptor complex (KA), and the efficacy of the agonist is defined by τ. KAandτ are descriptive parameters of intrinsic efficacy and binding affinity and may be directly obtained by fitting experi-mental data to the operational equation and can be expressed as “transduction coefficients” log(τ/KA). The relative efficiency of an agonist producing activation of any pathway can thus be quantified with a “normalized” transduction coefficient, namely Δlog(τ/KA). Fi-nally, the bias factor was calculated as a difference between Δlog(τ/KA) values for a given agonist between the pathways (G protein and β-ar-restin 2):
bias factor =Δlog(τ/KA)Gprotein− Δlog(τ/KA)β-arrestin2
Bias factors are expressed as the mean ± SEM of at least 5 in-dependent experiments.
2.6. In vitro activity in isolated smooth muscle strips
Organ bath studies were performed as described previously[44]. Mice were sacrificed by cervical dislocation. The colon was rapidly removed and 0.5 cm long full-thickness fragments were distincted. Preparations were kept in cold oxygenated Krebs–Ringer Solution. Electricalfield stimulation (EFS; 4 Hz, 24 V, stimulus duration 0.5 ms, train duration 10 s) was applied using a S88X stimulator (Grass Tech-nologies, West Warwick, RI, USA). A tested compound at increasing concentrations (10−12 to 10−6M) was added cumulatively into the organ bath. Tissue was incubated with each concentration for 8 min. As an internal control, the mean amplitude of initial 3 successive con-tractions was used. Changes in contractility (after a tested compound addition) were measured and reported as percentage of internal control.
2.7. Animal tests
2.7.1. Animals
Male Balb/C mice (Institute of Occupational Medicine, Lodz, Poland), weighing 22–24 g, were used for the study. The animals were housed in a room with controlled temperature (22 ± 1 °C), humidity (70 ± 5%) and light/dark cycle conditions (12/12 h), with free access to laboratory chow and tap water. All procedures were approved by the Local Ethical Committee for Animal Research with the following numbers: 29/ŁB662/2013 and 20/ŁB708/2014.
2.7.2. Drugs and pharmacological treatment
The tested compounds were dissolved in DMSO, further diluted with 0.9% NaCl solution to thefinal concentration of 5% DMSO. Animals without treatment (control group) received vehicle alone (5% DMSO in 0.9% NaCl solution). The vehicle given alone had no effects on the observed parameters.
2.7.3. Assessment of antinociception
The antinociceptive effects of peptides were assessed in the
hot-plate test in mice as described earlier[45]. The intracerebroventricular (i.c.v.) injections (10μL/animal) of peptides or vehicle were performed under inhalational isoflurane anesthesia into the left brain ventricle with a Hamilton microsyringe (50μL) connected to a needle (diameter 0.5 mm). Antagonists were administered 15 min prior to the peptide or vehicle injection. A transparent plastic cylinder (14 cm diameter, 20 cm height) was used to confine a mouse on the heated (55 ± 0.5 °C) sur-face of the plate. The animals were placed on the hot-plate 10 min after i.c.v. injection. The latencies to hind paw licking were measured. A cut-off time of 120 s was used to avoid 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 (120 s). The median antinociceptive dose (ED50) was calculated ac-cording to the method of Litchfield and Wilcox[46]. The data were analyzed by a nonlinear least square regression analysis computer program Graph Pad PRISM 6.0 (GraphPad Software Inc., San Diego, USA).
2.7.4. Evaluation of whole gastrointestinal transit
Whole gastrointestinal transit (WGT) test in mice was performed as described[47]. A tested compound was injected i.p. (in thefinal vo-lume of 0.1 mL) 15 min before intragastric (i.g.) administration of a colored marker (0.15 mL of glutinous liquid consisting of 5% Evans blue and 5% Arabic gum). The colored dye was administered with 18-gauge animal feeding tube. Subsequently, mice were placed in the individual cages on a white sheet of paper (in order to facilitate recognition of colored boluses). The whole gastrointestinal transit is calculated as time between i.g. administration of the marker and excretion of thefirst colored bolus.
2.8. Statistical analysis
Statistical analysis was performed using GraphPad Prism 6.0 (GraphPad Software Inc., San Diego, CA, USA). All data are expressed as mean ± SEM Analysis of multiple treatment was performed using one way analysis of variance (ANOVA) followed by Newman–Keuls post-hoc test. P values < 0.05 were considered statistically significant. 3. Results
3.1. In vitro profile of Dmt-c[D-Lys-Phe-Asp]NH2
Our previously reported mu selective cyclopeptide Tyr-c[D
-Lys-Phe-Asp]-NH2(1)[35], was modified by introduction of Dmt into position 1. The obtained analog Dmt-c[D-Lys-Phe-Asp]NH2(2), was examined at the mu, delta and kappa opioid receptors by the competitive binding against [3H]DAMGO, [3H][D-Ala2]deltorphin-2, and U-69,593, respec-tively, and the results are summarized inTable 1. Introduction of Dmt resulted in transforming a mu selective analog1 into a peptide 2 with very high affinity for mu and delta receptors.
The pharmacological profile of the new analog was further eval-uated in vitro at all three opioid receptors in the calcium mobilization
Table 1
Opioid receptor binding affinities of cyclopeptides.
No. Sequence Kia[nM]
mu delta kappa
EM-2 Tyr-Pro-Phe-Phe-NH2 2.50 ± 0.05 > 1000 > 1000
1 Tyr- c[D-Lys-Phe-Asp]NH2 0.21 ± 0.02 461 ± 63 684 ± 59
2 Dmt-c[D-Lys-Phe-Asp]NH2 0.24 ± 0.01 1.33 ± 0.06 79 ± 2
All values are expressed as mean ± SEM, n≥ 3.
a Displacement of [3H]DAMGO (mu-selective), [3H]deltorphin-2 (delta-se-lective) and [3H]U-69593 (kappa-selective) from human opioid receptor membrane binding sites.
assay in which CHO cells expressing human recombinant opioid re-ceptors and chimeric G proteins were used to monitor calcium changes. The concentration-response curves of the tested compounds and the standards were obtained and the calculated agonist potencies (pEC50) and efficacies (α) are shown inTable 2. At the mu receptor analog2 displayed even higher potency than the reference mu agonist, en-domorphin-2 (EM-2) and the same maximal effect. At the delta receptor 2 showed the same potency and maximal effect as the selective delta agonist DPDPE. Compared to dynorphin A, the kappa receptor re-ference compound, cyclopeptide2 displayed very low potency and re-duced efficacy. In agreement with the binding assay results, introduc-tion of Dmt transformed analog1 with high mu potency into a non-selective peptide 2 which showed high potency at the mu and delta receptors.
Then, the interaction of the mu opioid receptor with G-protein and β-arrestin 2 were evaluated in the BRET assay, measuring the energy transfer between Renilla Luciferase (RLuc) linked to the mu opioid re-ceptor and Renilla Green Fluorescent Protein (RGFP) linked to the signal transducer proteins (G-protein or β-arrestin 2). In membrane extracts taken from SH-SY5Y cells co-expressing the mu/RLuc and the Gβ1-RGFP fusoproteins, the mu receptor agonist EM-2 promoted re-ceptor/G protein interaction in a concentration dependent manner with potency value of 6.97 and maximal effect (Emax) of 0.80 ± 0.11 sti-mulated BRET ratio (Fig. 1A). Analog1 mimicked the maximal effects of EM-2, being 4 fold more potent (Fig. 1B). In a separate series of experiments, EM-2 promoted mu/G-protein interaction in a con-centration-dependent manner with high potency (pEC50 7.32) and maximal effect of 1.48 ± 0.16 stimulated BRET ratio (Fig. 1C). Under the same experimental conditions cyclopeptide 2 behaved as a full agonist and was 8 fold more potent than EM-2 (Fig. 1D).
In SH-SY5Y cells stably co-expressing the mu/RLuc and the β-ar-restin 2/RGFP fusoproteins, EM-2 promoted receptor/arβ-ar-restin interac-tion in a concentrainterac-tion dependent manner with a potency value of 6.91, and Emaxof 0.35 ± 0.09 stimulated BRET ratio (Fig. 1A). Analog 1 showed higher maximal effects and was 4 fold more potent than EM-2 (Fig. 1B). In the second series of experiments, EM-2 promoted mu/β-arrestin interaction in a concentration dependent manner with a po-tency value of 7.20, and Emaxof 0.38 ± 0.04 stimulated BRET ratio (Fig. 1C). Compound2 was 16 fold more potent and exhibited higher maximal effect than EM-2 (Fig. 1D). These results, which are shown in Fig. 1, were used for calculating bias factors of the ligands: analog1 had a bias factor not statistically different from 0, while analog 2 displayed a modest (14-fold) but significant bias toward β-arrestin 2 (Table 3).
In further in vitro experiments, the effect of analog 2 was evaluated on EFS-induced smooth muscle contractility in the mouse distal colon. Analog2 significantly inhibited EFS-induced colonic contractions in a concentration-dependent manner (10−9to 10−6M) (Fig. 2).
3.2. In vivo profile of cyclopeptide analogs
Assessment of antinociception of cyclopeptide analogs was studied in the mouse hot-plate test (supraspinally mediated analgesia), after i.c.v. administration. Both compounds showed an extremely strong analgesic effect, which was dose-dependent (Fig. 3). Analog2 with the mu/delta profile exerted stronger antinociceptive effect than the parent compound1, which was mu selective. The ED50were 0.6 ng and 2.9 ng for analog2 and analog 1, respectively.
The antinoceptive activity of analog2 was effectively reversed by preemptive i.c.v. injection ofβ-FNA (mu selective antagonist) or NLT (delta selective antagonist) (Fig. 4), indicating that the action of this peptide was mediated by the mu and delta opioid receptors in the brain. To compare the inhibitory effect of analogs 1 and 2 on gastro-intestinal (GI) transit, the mouse WGT test was performed. Tested compounds were administered at three doses (0.1; 0.3; and 1 mg/kg, i.p.) For comparison, loperamide, the well-known antidiarrheal agent, was used. Analog2 with biased profile toward β-arrestin 2/mu exerted the strongest effect. This compound significantly decreased the WGT at all three concentrations. At a dose of 0.1 mg/kg, it delayed excretion of the colored boluses up to 193.7 ± 24.7 min, in comparison to control (90.9 ± 5.0 min) and loperamide (130.0 ± 3.0 min). Analog 1 and loperamide were active only at the highest used dose, which was 1 mg/ kg (Fig. 5).
4. Discussion
Most of the currently available pain-relieving opioids exert their analgesic but also adverse effects primarily through the activation of the mu receptor. However, a large number of biochemical and phar-macological studies provide evidence that there are strong modulatory interactions between mu and delta opioid receptors. Several studies indicate that delta receptor agonists as well as antagonists can sig-nificantly improve the pharmacological effects exerted by mu agonists. In particular, delta agonists can enhance the analgesic potency and efficacy of mu agonists, and delta antagonists can prevent or diminish the development of tolerance and physical dependence induced by mu agonists.
Therefore, the use of agents that simultaneously activate more than one opioid receptor in order to enhance efficacy and/or reduce side effects is a promising approach in the search for innovative analgesics [3,48].
As is well known, opioid agonists interact with GPCRs which upon activation, are phosphorylated by GPCR kinases and subsequently bind β-arrestins, which prevent further coupling of the receptor to G protein [49]. GPCRs can then be internalized and are either recycled to the plasma membrane or degraded. However,β-arrestins are not only ne-gative regulators of the G protein signaling but can promote distinct intracellular signals of their own. Experiments with the β-arrestin 2 knockout (KO) mice showed that signaling mechanism of morphine
Table 2
Effects of the reference agonists and cyclopeptides at human recombinant opioid receptors coupled with calcium signaling via chimeric G proteins.
mu delta kappa
pEC50a(CL95%) α ± SEMb pEC50(CL95%) α ± SEM pEC50(CL95%) α ± SEM
EM-2 7.83 (7.69–7.96) 1.00 Inactive inactive
DPDPE inactive 7.33 (7.20–7.47) 1.00 inactive
Dynorphin A 6.67 (6.17–7.17) 0.83 ± 0.10 7.73 (7.46–8.00) 0.99 ± 0.04 8.81 (8.73–8.90) 1.00
1 8.56 (8.32–8.79) 1.01 ± 0.07 6.62 (6.23–7.01) 1.00 ± 0.13 crc incomplete
2 8.53 (8.25–8.81) 1.03 ± 0.06 7.65 (6.74–8.55) 0.98 ± 0.05 6.11 (5.80–6.42) 0.5 ± 0.02
The crc incomplete means that the maximal effect could not be determined due to the low potency of a compound.
EM-2, DPDPE, and dynorphin A were used as reference agonists for calculating intrinsic activity at the mu, delta and kappa opioid receptors, respectively. a Agonist potency values (pEC
50). b Efficacy values (α).
engage both G protein coupling andβ-arrestin recruitment and that blocking the β-arrestin pathway can reduce unwanted side effects of opioids, such as respiratory depression, inhibition of gastrointestinal motility, and tolerance liability. In theβ-arrestin 2 KO mice morphine induced, via the mu opioid receptor, enhanced and longer-lasting an-tinociceptive effects than in a wild type mice and almost no tolerance [50,51]. On the other hand, the antidiarrheal agent loperamide, which is a peripherally restricted mu receptor agonist, significantly reduced colonic propulsion in wild type mice, and this effect was completely abolished inβ-arrestin 2 KO mice[52]. This suggests that peripherally restrictedβ-arrestin-biased agonists might be useful in the treatment of diarrhea and other hypermotility disorders[53].
The discovery of β-arrestin-mediated signaling pathways opened new possibilities in opioid drug development[54–56]. Agonists biased to either G protein orβ-arrestin may be used to segregate physiological responses downstream of the receptor[57–59].
Molinari et al.[42] measured the possible differential ability of various opioid ligands of peptide and alkaloid structure (including EM-2 and morphine) to induce G-protein andβ-arrestin signaling at the mu and delta receptors. None of the tested ligands showed greater efficacy forβ-arrestin than for the G-protein. The authors suggested that the structure requirements for an agonist to trigger the interaction of the receptor withβ-arrestin are more stringent than those sufficient to in-itiate G-protein coupling.
Fig. 1. Comparison of the effect of EM-2 (panels A and C), Tyr- c[D-Lys-Phe-Asp]NH2(1; panel B), and Dmt-c[D-Lys-Phe-Asp]NH2(2; panel D) at the mu/G protein and mu/β-arrestin 2 interaction. Data are mean ± SEM of at least 5 experiments.
Table 3
Potencies, maximal effects, and bias factors of EM-2, cyclopeptides 1 and 2 in the BRET assay.
Compound mu/G-protein mu/β-arrestin 2 bias factor (CL95%)
pEC50(CL95%) α ± SEM pEC50(CL95%) α ± SEM
EM-2 6.97 (6.73–7.22) 1.00 6.91 (6.76–7.05) 1.00 0.00
1 7.56 (7.15–7.97) 1.08 ± 0.07 7.52 (6.85–8.20) 1.46 ± 0.03 −0.49 (−1.36 to 0.38)
EM-2 7.32 (7.14–7.51) 1.00 7.20 (7.08–7.33) 1.00 0.00
Recently, we have tested several of our opioid peptide analogs to-wards their possible differential mu/G protein and mu/β-arrestin
interactions. The cyclic pentapeptide Dmt-c[D-Lys-Phe-Phe-Asp]NH2 activated both, G protein andβ-arrestin pathways[40]. Reported here tetrapeptide Tyr-c[D-Lys-Phe-Asp]NH2(1) with Tyr1and a shorter se-quence by one Phe residue was also unbiased. However, the Dmt1–containing analog Dmt-c[
D-Lys-Phe-Asp]NH2 (2), activated G protein pathway similarly to EM-2 but promotedβ-arrestin recruitment with a much higher maximal effect than EM-2 (1.46 and 1.0, respec-tively) and therefore showed 14-fold bias towardβ-arrestin. To the best of our knowledge, analog2 is thefirst reported β-arrestin biased opioid peptide.
This ligand activated with high efficacy the mu and delta receptors. Compounds with such profile are known to exert stronger pharmaco-logical effects than the mu selective ligands. Indeed, analog 2 showed a 5-fold more pronounced antinociceptive effect in mice than its mu se-lective parent1. On the other hand, 2 was also found to strongly inhibit the WGT in mice which is consistent with activation of theβ-arrestin pathway.
These results are in accordance with earlier reports indicating that various in vivo activities of opioid agonists arise not only from the ac-tivation of one or more opioid receptors, but also from promoting G-protein orβ-arrestin pathways. Therefore prediction of molecular in-teractions linked to the drug responses in vivo is very complex.
Conflict of interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Author contribution
K. G-J. designed the research study and performed radioligand binding assays, J. P-C. and F.F. performed functional assays, J.P-C. and J P-Z. performed in vivo studies, F.F. and G. C. analyzed calcium and BRET data, A. A-B. and K.W. performed peptide synthesis, A. S. per-formed the ex vivo test, A. J. discussed the results and prepared the manuscript.
Acknowledgments
This work was supported by the Polpharma Scientific Foundation grant (11/XIII/2014 to KG) and grants from the Medical University of Lodz (No 503/1-156-02/503-11-002 and 502-03/1-156-02/502-14-301) and from the University of Ferrara (FAR grant to G.C.). KG is re-cipients of the Polish L’Oréal UNESCO Awards for Women in Science.
Fig. 2. The inhibitory effect of analog 2 (10−12to 10−6M) on EFS induced longitudinal smooth muscle contractions in mouse colon. Data represent mean ± SEM (n = 7-8). Statistical significance was assessed using one-way ANOVA and a post hoc multiple comparison Student–Newman–Keuls test. *p < 0.05, **p < 0.01, ***p < 0.001, as compared with control (vehicle alone).
Fig. 3. The effect of different doses of the tested analogs in the mouse hot-plate test. Results are expressed as percentage (mean ± SEM) of the maximal pos-sible effect (%MPE) for the inhibition of hind paw licking induced by i.c.v. injection of a peptide. n = 6–8 mice for each experimental group.
Fig. 4. Antagonist effect of β-FNA and NLT (both at 1 μg/animal, i.c.v.) on the inhibition of hind paw licking induced by administration of analog2 (1 ng/ animal, i.c.v.) in the mouse hot-plate test. Results are expressed as percentage (mean ± SEM) of the maximal possible effect (%MPE). n = 6–8 mice for each experimental group. Statistical significance was assessed using one-way ANOVA and a post hoc multiple comparison Student–Newman–Keuls test. ***p < 0.001, as compared with analog2.
Fig. 5. Dose-response inhibitory effect of loperamide and peptides 1 and 2 on the mouse WGT. A peptide or vehicle were injected i.p. 15 min before the in-tragastric (i.g.) administration of a colored marker. Data represent mean ± SEM of n = 6–8 mice for each experimental group. Statistical sig-nificance was assessed using one-way ANOVA and a post hoc multiple com-parison Student-Newman-Keuls test. ***p < 0.001, *p < 0.05, as compared with control (vehicle treated mice).
References
[1] H.S. Smith, J.F. Peppin, Toward a systematic approach to opioid rotation, J. Pain Res. 7 (2014) 589–608.
[2] R. Benyamin, A.M. Trescot, S. Datta, R. Buenaventura, R. Adlaka, N. Sehgal, S.E. Glaser, R. Vallejo, Opioid complications and side effects, Pain Phys. 11 (2008) 105–120.
[3] R. Morphy, Z. Rankovic, Designed multiple ligands. An emerging drug discovery paradigm, J. Med. Chem. 48 (2005) 6523–6543.
[4] R. Morphy, Z. Rankovic, Designing multiple ligands– medicinal chemistry strate-gies and challenges, Curr. Pharm. Des. 15 (2009) 587–600.
[5] A. Pfeiffer, V. Brantl, A. Herz, H.M. Emrich, Psychotomimesis mediated by kappa opiate receptors, Science 233 (1986) 774–776.
[6] S.S. Negus, N.K. Mello, Effects of kappa opioid agonists on cocaine self-adminis-tration under a progressive ratio schedule in rhesus monkeys, Drug Alcohol Depend. 63 (2001) S113.
[7] N.K. Mello, S.S. Negus, Interactions between kappa opioid agonists and cocaine. Preclinical studies, Ann. NY Acad. Sci. 909 (2000) 104–132.
[8] B.M. Greedy, F. Bradbury, M.P. Thomas, K. Grivas, G. Cami-Kobeci, A. Archambeau, K. Bosse, M.J. Clark, M. Aceto, J.W. Lewis, J.R. Traynor, S.M. Husbands, Orvinols with mixed kappa/mu opioid receptor agonist activity, J. Med. Chem. 56 (2013) 3207–3216.
[9] A. Almatroudi, S.M. Husbands, C.P. Bailey, S.J. Bailey, Combined administration of buprenorphine and naltrexone produces antidepressant-like effects in mice, J. Psychopharmacol. 29 (2015) 812–821.
[10] J.P. Cueva, C. Roche, M. Ostovar, V. Kumar, M.J. Clark, T.M. Hillhouse, J.W. Lewis, J.R. Traynor, S.M. Husbands, C7β-methyl analogues of the orvinols: the discovery of kappa opioid antagonists with nociceptin/orphanin FQ peptide (NOP) receptor partial agonism and low, or zero, efficacy at mu opioid receptors, J. Med. Chem. 58 (2015) 4242–4249.
[11] A. Almatroudi, M. Ostovar, C.P. Bailey, S.M. Husbands, S.J. Bailey, Antidepressant-like effects of BU10119, a novel buprenorphine analogue with mixed κ/μ receptor antagonist properties, in mice, Br. J. Pharmacol. (2017),http://dx.doi.org/10. 1111/bph.14060Epub ahead of print.
[12] E.E. Abdelhamid, M. Sultana, P.S. Portoghese, A.E. Takemori, Selective blockage of delta opioid receptors prevents the development of morphine tolerance and de-pendence in mice, J. Pharmacol. Exp. Ther. 258 (1991) 299–303.
[13] S. Ananthan, Opioid ligands with mixed mu/delta opioid receptor interactions: an emerging approach to novel analgesics, AAPS J. 8 (2006) 118–125.
[14] P.W. Schiller, T.M. Nguyen, G. Weltrowska, B.C. Wilkes, B.J. Marsden, C. Lemieux, N.N. Chung, Differential stereochemical requirements of mu vs. delta opioid re-ceptors for ligand binding and signal transduction: development of a class of potent and highly delta-selective peptide antagonists, Proc. Natl. Acad. Sci. U. S. A. 89 (1992) 11871–11875.
[15] P.W. Schiller, M.E. Fundytus, L. Merovitz, G. Weltrowska, T.M. Nguyen, C. Lemieux, N.N. Chung, T.J. Coderre, The opioid mu agonist/delta antagonist DIPP-NH(2)[Psi] produces a potent analgesic effect no physical dependence, and less tolerance than morphine in rats, J. Med. Chem. 42 (1999) 3520–3526.
[16] P.W. Schiller, Bi- or multifunctional opioid peptide drugs, Life Sci. 86 (2010) 598–603.
[17] L.C. Purington, K. Sobczyk-Kojiro, I.D. Pogozheva, J.R. Traynor, H.I. Mosberg, Development and in vitro characterization of a novel bifunctional μ-agonist/δ-an-tagonist opioid tetrapeptide, ACS Chem. Biol. 6 (2011) 1375–1381.
[18] H.I. Mosberg, L. Yeomans, J.P. Anand, V. Porter, K. Sobczyk-Kojiro, J.R. Traynor, E.M. Jutkiewicz, Development of a bioavailableμ opioid receptor (MOPr) agonist, δ opioid receptor (DOPr) antagonist peptide that evokes antinociception without development of acute tolerance, J. Med. Chem. 57 (2014) 3148–3153. [19] P. Horan, R.J. Tallarida, R.C. Haaseth, T.O. Matsunaga, V.J. Hruby, F. Porreca,
Antinociceptive interactions of opioid delta receptor agonists with morphine in mice: supra- and sub-additivity, Life Sci. 50 (1992) 1535–1541.
[20] L. He, N.M. Lee, Delta opioid receptor enhancement of mu opioid receptor-induced antinociception in spinal cord, J. Pharmacol. Exp. Ther. 285 (1998) 1181–1186. [21] F. Porreca, A.E. Takemori, M. Sultana, P.S. Portoghese, W.D. Bowen, H.I. Mosberg,
Modulation of mu-mediated antinociception in the mouse involves opioid delta-2 receptors, J. Pharmacol. Exp. Ther. 263 (1992) 147–152.
[22] N.A. Martin, P.L. Prather, Interaction of co-expressed mu- and delta-opioid re-ceptors in transfected rat pituitary GH(3) cells, Mol. Pharmacol. 59 (2001) 774–783.
[23] A.W. Lipkowski, A.M. Konecka, I. Sroczyńska, Double-enkephalins-synthesis, ac-tivity on guinea-pig ileum, and analgesic effect, Peptides 3 (1982) 697–700. [24] D. Kosson, A. Klinowiecka, P. Kosson, I. Bonney, D.B. Carr, E. Mayzner-Zawadzka,
A.W. Lipkowski, Intrathecal antinociceptive interaction between the NMDA an-tagonist ketamine and the opioids, morphine and biphalin, Eur. J. Pain. 12 (2008) 611–616.
[25] M. Yamazaki, T. Suzuki, M. Narita, A.W. Lipkowski, The opioid peptide analogue biphalin induces less physical dependence than morphine, Life Sci. 69 (2001) 1023–1028.
[26] A. Kosson, I. Krizbai, A. Lesniak, M. Beresewicz, M. Sacharczuk, P. Kosson, P. Nagyoszi, I. Wilhelm, P. Kleczkowska, A.W. Lipkowski, Role of the blood-brain barrier in differential response to opioid peptides and morphine in mouse lines divergently bred for high and low swim stress-induced analgesia, Acta. Neurobiol. Exp. 74 (2014) 26–32.
[27] O. Frączak, A. Lasota, A. Leśniak, A.W. Lipkowski, A. Olma, The biological con-sequences of replacing D-Ala in biphalin with amphiphilicα-alkylserines, Chem. Biol. Drug Des. 84 (2014) 199–205.
[28] O. Frączak, A. Lasota, P. Kosson, A. Leśniak, A. Muchowska, A.W. Lipkowski, A. Olma, Biphalin analogs containingβ(3)-homo-amino acids at the 4, 4' positions: synthesis and opioid activity profiles, Peptides 66 (2015) 13–18.
[29] S.M. Cowell, Y.S. Lee, Biphalin the foundation of bivalent ligands, Curr. Med. Chem. 23 (2016) 3267–3284.
[30] A. Mollica, P. Davis, S.W. Ma, F. Porreca, J. Lai, V.J. Hruby, Synthesis and biolo-gical activity of thefirst cyclic biphalin analogues, Bioorg. Med. Chem. Lett. 16 (2006) 367–372.
[31] A. Mollica, A. Carotenuto, E. Novellino, A. Limatola, R. Costante, F. Pinnen, A. Stefanucci, S. Pieretti, A. Borsodi, R. Samavati, F. Zador, S. Benyhe, P. Davis, F. Porreca, V.J. Hruby, Novel cyclic biphalin analogue with improved anti-nociceptive properties, ACS Med. Chem. Lett. 5 (2014) 1032–1036.
[32] A. Stefanucci, A. Carotenuto, G. Macedonio, E. Novellino, S. Pieretti, F. Marzoli, E. Szűcs, A.I. Erdei, F. Zádor, S. Benyhe, A. Mollica, Cyclic biphalin analogues in-corporating a xylene bridge: synthesis, characterization, and biological profile, ACS Med. Chem. Lett. 8 (2017) 858–863.
[33] R. Costante, F. Pinnen, A. Stefanucci, A. Mollica, Potent biphalin analogs withμ/δ mixed opioid activity: in vivo and in vitro biological evaluation, Arch. Pharm. 347 (2014) 305–312.
[34] A. Mollica, F. Pinnen, R. Costante, M. Locatelli, A. Stefanucci, S. Pieretti, P. Davis, J. Lai, D. Rankin, F. Porreca, V.J. Hruby, Biological active analogues of the opioid peptide biphalin: mixedα/β(3)-peptides, J. Med. Chem. 56 (2013) 3419–3423. [35] J. Piekielna, A. Kluczyk, L. Gentilucci, M.C. Cerlesi, G. Calo', C. Tomböly,
K.Łapiński, T. Janecki, A. Janecka, Ring size in cyclic endomorphin-2 analogs modulates receptor binding affinity and selectivity, Org. Biomol. Chem. 13 (2015) 6039–6046.
[36] R. Perlikowska, D. Malfacini, M.C. Cerlesi, G. Calo', J. Piekielna, L. Floriot, T. Henry, J.C. Do-Rego, C. Tömböly, A. Kluczyk, A. Janecka, Pharmacological characterization of endomorphin-2-based cyclic pentapeptides with methylated phenylalanine residues, Peptides 55 (2014) 145–150.
[37] A. Misicka, A.W. Lipkowski, R. Horvath, P. Davis, T.H. Kramer, H.I. Yamamura, V.J. Hruby, Topographical requirements for delta opioid ligands: common struc-tural features of dermenkephalin and deltorphin, Life Sci. 51 (1992) 1025–1032. [38] Y. Cheng, W.H. Prusoff, Relationship between the inhibition constant (K1) and the
concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic reaction, Biochem. Pharmacol. 22 (1973) 3099–3108.
[39] V. Camarda, G. Calo', Chimeric G proteins influorimetric calcium assays: experi-ence with opioid receptors, Methods Mol. Biol. 937 (2013) 293–306.
[40] J. Piekielna-Ciesielska, F. Ferrari, G. Calo’, A. Janecka, Cyclopeptide Dmt-[D
-Lys-p-CF3-Phe-Phe-Asp]NH2; a novel G protein-biased agonist of the mu opioid receptor,
Peptides 101 (2018) 227–233,http://dx.doi.org/10.1016/j.peptides.2017.11.020. [41] D. Malfacini, C. Ambrosio, M.C. Gro', M. Sbraccia, C. Trapella, R. Guerrini,
M. Bonora, P. Pinton, T. Costa, G. Calo', Pharmacological profile of nociceptin/ orphanin FQ receptors interacting with G-proteins andβ-arrestins 2, PLoS One 10 (2015) e0132865.
[42] P. Molinari, V. Vezzi, M. Sbraccia, C. Grò, D. Riitano, C. Ambrosio, I. Casella, T. Costa, Morphine-like opiates selectively antagonize receptor-arrestin interac-tions, J. Biol. Chem. 285 (2010) 12522–12535.
[43] K. Nagi, G. Pineyro, Practical guide for calculating and representing biased sig-naling by GPCR ligands: a stepwise approach, Methods 92 (2016) 78–86. [44] M. Zielińska, A. Jarmuż, A. Wasilewski, G. Cami-Kobeci, S. Husbands, J. Fichna,
Methyl-orvinol-Dual activity opioid receptor ligand inhibits gastrointestinal transit and alleviates abdominal pain in the mouse models mimicking diarrhea-pre-dominant irritable bowel syndrome, Pharmacol. Rep. 69 (2017) 350–357. [45] R. Perlikowska, J. Piekielna, L. Gentilucci, R. De Marco, M.C. Cerlesi, G. Calo,
R. Artali, C. Tömböly, A. Kluczyk, A. Janecka, Synthesis of mixed MOR/KOR effi-cacy cyclic opioid peptide analogs with antinociceptive activity after systemic ad-ministration, Eur. J. Med. Chem. 109 (2016) 276–286.
[46] R.J. Tallarida, R.B. Murray, Manual of Pharmacologic Calculations: With Computer Programs, Springer Science & Business Media, 2012.
[47] K. Gach, J.C. Do-Rego, J. Fichna, M. Storr, D. Delbro, G. Toth, A. Janecka, Synthesis and biological evaluation of novel peripherally active morphiceptin analogs, Peptides 31 (2010) 1617–1624.
[48] W. Fujita, I. Gomes, L.A. Devi, Revolution in GPCR signalling: opioid receptor heteromers as novel therapeutic targets: IUPHAR review 10, Br. J. Pharmacol. 171 (2014) 4155–4176.
[49] L.M. Luttrell, R.J. Lefkowitz, The role of beta-arrestins in the termination and transduction of G-protein-coupled receptor signals, J. Cell Sci. 115 (2002) 455–465. [50] L.M. Bohn, R.J. Lefkowitz, M.G. Caron, Differential mechanisms of morphine
an-tinociceptive tolerance revealed in (beta)arrestin-2 knock-out mice, J. Neurosci. 22 (2002) 10494–10500.
[51] J.T. Lamberts, J.R. Traynor, Opioid receptor interacting proteins and the control of opioid signaling, Curr. Pharm. Des. 19 (2013) 7333–7347.
[52] K.M. Raehal, J.K. Walker, L.M. Bohn, Morphine side effects in beta-arrestin 2 knockout mice, J. Pharmacol. Exp. Ther. 314 (2005) 1195–1201.
[53] L.M. Bohn, K.M. Raehal, Opioid receptor signaling: relevance for gastrointestinal therapy, Curr. Opin. Pharmacol. 6 (2006) 559–563.
[54] S. Rajagopal, K. Rajagopal, R.J. Lefkowitz, Teaching old receptors new tricks: biasing seven-transmembrane receptors, Nat. Rev. Drug Discov. 9 (2010) 373–386. [55] J.D. Violin, R.J. Lefkowitz, Beta-arrestin-biased ligands at seven-transmembrane
receptors, Trends Pharmacol. Sci. 28 (2007) 416–422.
[56] T. Kenakin, L.J. Miller, Seven transmembrane receptors as shapeshifting proteins: the impact of allosteric modulation and functional selectivity on new drug dis-covery, Pharmacol. Rev. 62 (2010) 265–304.
[57] J.D. Violin, A.L. Crombie, D.G. Soergel, M.W. Lark, Biased ligands at G-protein-coupled receptors: promise and progress, Trends Pharmacol. Sci. 35 (2014) 308–316.
[58] A. Madariaga-Mazón, A.F. Marmolejo-Valencia, Y. Li, L. Toll, R.A. Houghten, K. Martinez-Mayorga, Mu-Opioid receptor biased ligands: a safer and painless dis-covery of analgesics? Drug. Discov. Today 22 (2017) 1719–1729.
[59] X. Liu, L. Zhao, Y. Wang, J. Zhou, D. Wang, Y. Zhang, X. Zhang, Z. Wang, D. Yang, L. Mou, R. Wang, MEL-N16. a series of novel endomorphin analogs with good an-algesic activity and a favorable side effect profile, ACS Chem. Neurosci. 8 (2017) 2180–2193.
