Though natural opioid alkaloids such as morphine and codeine (Figure 1) contain a 6-hydroxyl group, synthetic approaches have uncovered that functionalizing position 6 gives rise to a wide range of diverse activities . Thus, position 6 of the morphinan skeleton has been a major target for successful drug developments over the years, leading to various opioid agonists and antagonists that are of importance both for clinical use and research. Oxycodone and oxymorphone (Figure 1), clinically used as opioid analgesics, are two representative examples of structural variation at C-6, where a carbonyl instead of a hydroxyl group is present in position 6. By targeting the chemically highly versatile 6-keto function of morphinan-6-ones as in oxycodone, we have previously reported on a chemically innovative modification giving rise to a novel class of morphinans with acrylonitrile incorporated substructures [29, 30]. The resulted acrylonitrile incorporated 4,5-oxygen bridge-opened N- methylmorphinans (1–3, Figure 1) emerged as high affinity and potent MOP antinociceptive agents, with a pharmacological profile comparable to that of their 6-keto counterparts . The interesting approach to incorporate acrylonitrile substructures into morphinans was further explored by our group and resulted in new derivatives .
In the present study, combining in vitro ligand binding and functional assays and in vivo behavioral approaches, we show that the presence of a cyano group in position 6 in N-methylmorphinans has a strong influence on opioid receptor binding and post-receptor molecular events. In line with our previous findings, having a 6-cyano group in N-methylmorphinans (5 and 6) results in increased MOP receptor activity compared to the lead molecule oxycodone both in vitro and in vivo. In the series of 6-cyanomorphinans, the new derivative 5 was consistently identified to exhibit the highest affinity and selectivity at the MOP receptor and to be the most potent MOP agonist. The design of compound 5 having a 4,14-dimethoxy substitution was attained based on our earlier observations, when a 4-methoxy group and/or a 14-methoxy group, like in compounds 2 and 3, is more favorable for binding and selectivity for the MOP receptor and antinociceptive activity than the corresponding hydroxy counterpart 1
. Herein, we also establish that the presence of a methoxy group in both positions, 4 and 14, has a major impact not only on binding affinities to all three opioid receptor types, and MOP receptor selectivity, but also on agonist potencies and efficacies at this receptor.
We have also examined how the combination of a C-6 cyano functionality together with a closed 4,5-oxygen bridge (compound 6) will affect in vitro and in vivo opioid activities. The two 6-cyanomorphinans 3 and 6 show high and similar affinities at the MOP receptor, and low binding to DOP and KOP receptors. In both functional studies, [35S]GTPγS binding and intracellular calcium mobilization, compounds 3 and 6 acted as potent MOP agonists with comparable EC50 values, and a somewhat reduced efficacy showed by derivative 3. In vivo, the 6-cyanomorphinan 6 with a closed 4,5-oxygen bridge was more potent than its 4,5-oxygen bridge-opened analogue 3 in inducing an antinociceptive effect in mice after s.c. administration (ca. 3 times in the hot-plate, 2 times in the tail-flick and 37 times in the PPQ tests). Closing of the 4,5-oxygen bridge in the 6-acrylonitrile substituted 3 produces no major changes in interaction with the MOP receptor in vitro, but augmented antinociceptive potency. On the other hand, the 14-methoxy-6-cyanomorphinan 6 showed greater MOP receptor affinity and agonist potency than 14-OMC and the 14-hydroxy substituted oxycodone, together with much better antinociceptive properties.
It was of interest to assess the result of the conversion of the 6-acrylonitrile to a 6-amido group on the interaction with opioid receptors, signaling, and the link between antinociceptive efficacy and the mode of action. Since the presence of 4- and 14-methoxy groups was favorable in the case of the 6-cyano substituted N-methylmorphinan 5, the same substitution pattern was applied to the 6-amido analogue 4. It was remarkable to note that the presence of an amido group in position 6 resulted in high affinity at the MOP receptor and also good MOP selectivity. In the [35S]GTPγS functional assay, the 6-amido substituted 4,5-oxygen bridge-opened 4 acted as a highly efficacious agonist at the MOP receptor with several times increased potency than oxycodone, 14-OMC and 4,14-dihydroxy substituted 6-cyanomorphinan 1. The same profile was depicted for compound 4 when stimulating G protein signaling and intracellular calcium release through MOP receptors. When compared to the 6-cyano analogue 5, the 6-amido group in 4 appears to largely affect agonist potency, leading to reduced activity, especially in antinociceptive potency. It is possible that differential metabolism of derivatives 4 and 5 may determine the differences in the in vivo activity. Primary aliphatic amides are known to be rapidly metabolically hydrolyzed , whilst the nitrile group is more stable .
In this study, we described the in vitro functional activities of the previously reported 6-cyanomorphinans 1–3 and oxycodone based on the assessment of MOP receptor-mediated G protein activation and intracellular calcium mobilization. Replacement of the 4-hydroxy group in 6-cyanomorphinan 1 with a 4-methoxy group in analogue 2, or substitution of 14-hydroxyl in compound 1 with a 14-methoxy group in 3 results in 6 to 10 times enhanced agonist potencies and comparable efficacies, upon the test being used. Compared to the 6-ketomorphinans oxycodone and 14-OMC, the 6-cyano substituted N-methylmorphinans 1–3 generally displayed higher agonist activity in vitro, which correlates well with the in vivo results on antinociceptive properties. Among all investigated N-methylmorphinans, derivative 5 is the most potent agonist in terms of G protein coupling and changes in intracellular calcium concentration. This MOP agonist potency enhancement of the new 6-cyanomorphinan 5 compared to the other derivatives established in the two functional assays is in agreement with the outcomes from in vitro binding assays and nociceptive tests, and supports the importance of the presence of both methoxy groups in positions 4 and 14 in this class of opioid morphinans .
The clinically relevant analgesic oxycodone was found as the MOP ligand with the lowest agonist potency in the series of the investigated morphinans. In CHOhMOP cell membranes, oxycodone stimulated [35S]GTPγS binding with a EC50 value of 500 nM, which is lower than the EC50 value of 1.40 μM reported by Thompson et al. in the same cell line . In the same work, a lower relative efficacy as percentage stimulation compared to DAMGO at the human MOP receptor in CHO cells was found for oxycodone (67%), while in our study a higher efficacy, i.e. 92% stimulation relative to DAMGO, was determined (Table 2). Comparable potency (EC50 = 373 nM) and lower relative efficacy (66%) for oxycodone to our data was reported in C6rMOP cells . Similarly, in CHOhMOP cells stably expressing the Gαqi5 chimeric protein, oxycodone exhibited low activity, by producing stimulation of calcium release with an EC50 value of 1,176 nM and an efficacy of 38%. A recent study  reported on changes in intracellular calcium levels produced by oxycodone in human embryonic kidney-293 (HEK293) cells co-expressing the human MOP receptor and Gαqi3 chimeric protein, with low potency (1.74 μM) and high efficacy (100%). Although 14-OMC also displays low agonist potencies at the human MOP receptor in both functional systems, it shows a similar efficacy compared to oxycodone in [35S]GTPγS binding and in calcium mobilization, that is also seen in antinociceptive potency. Mostly, compounds 1–6, oxycodone and 14-OMC were found to be more potent MOP agonists in the terms of G protein activation based on the lower EC50 values by one order of magnitude than the EC50 values for the calcium signaling, and with lesser efficacies measured in the latter. Presumably these differences may be due to variances in receptor reserve in the two cell lines, and/or possibly membranes vs. intact cells. Differences in signaling may also be regulated by the MOP receptor localization within the plasma membrane [38, 39]. Receptor localization within the lipid rafts after agonist binding can promote G protein coupling or recruitment of other intracellular regulatory proteins [40, 41]. Over the past years, increased attention has been drawn to the understanding of intracellular signaling pathways that mediate the therapeutic and/or adverse effects of opioid agonists acting at the MOP receptor [42–44]. In vitro and in vivo studies demonstrate that different opioids can initiate distinct cellular and physiological responses downstream of receptor activation [40, 42]. The nature of MOP receptor signaling and regulation are functions not only of the type and structure of the agonist acting at the receptor but also of the cellular environment in which the receptor is expressed . Moreover, the present understanding of MOP receptor function is persistently increasing, as the crystal structure is now available .