Novel 6β-acylaminomorphinans with analgesic activity

Abstract

Aminomorphinans are a relatively young class of opioid drugs among which substances of high in vitro efficacy and favorable in vivo action are found. We report the synthesis and pharmacological evaluation of novel 6β-acylaminomorphinans. 6β-Morphinamine and 6β-codeinamine were stereoselectively synthesized by Mitsunobu reaction. The aminomorphinans were subsequently acylated with diversely substituted cinnamic acids. In vitro binding studies on cinnamoyl morphinamines showed moderate affinity for all opiate receptors with some selectivity for mu opioid receptors, while cinnamoyl codeinamines only showed affinity for mu opioid receptors. In vivo analgesia studies showed significant analgesic activity of 6β-cinnamoylmorphinamine mediated by mu and delta receptors. The lead compound was found to be roughly equipotent to morphine (ED₅₀ 3.13 ± 1.09 mg/kg) but devoid of the dangerous side-effect respiratory depression, a major issue associated with traditional opioid therapy.

Morphine, the principal drug of the opioid family is among the most important agents used for the treatment of severe pain. [1] Clinically relevant effects, including various dangerous adverse effects are predominantly mediated by mu-opioid receptors (MOR-1). The structural features of the morphine skeleton have long been the basis of successful drug development and the development of novel morphine-like drugs with improved efficacy, receptor selectivity and side-effects. Aminomorphinans and their derivatives containing a nitrogen atom instead of oxygen in position C-6 of the morphinan ring system have been used in the development of a number of new opioids. Naloxamine and naltrexamine were the first two representatives of this family followed by an array of compounds generated by acylation and alkylation of the C-6 amino moiety, especially that of β-naltrexamine (β-NTA). [2] This group of compounds includes the novel kappa-opioid receptor (KOR-1) agonist nalfurafine (TRK-820) (Remitch® in Japan) used as a treatment for hemodialysis-related uremic pruritus [3,4] and β-funaltrexamine (β-FNA), a covalent MOR-1 antagonist affinity label with partial KOR-1 agonist activity has been widely used to investigate opioid receptor mechanisms. [5,6]

More recently iodobenzoylnaltrexamide (IBNtxA), a very potent analgesic with a novel pharmacological profile devoid of most side-effects characteristic of traditional opioid agonists due to its activity at a recently identified target (6TM/E11 sites) has been reported. [7,8] The naloxone analog, IBNalA, is more selective for these sites and has a similar side-effect profile. [9] The equatorial β-acylamino moiety is a common structural feature of these molecules. The C-6 amino group has been acylated by a diverse range of substituted aliphatic and aromatic carboxylic acids, including furanylacrylic acid (nalfurafine), 3-iodobenzoic acid (IBNtxA and IBNalA) and methylfumaric acid (β-FNA).

The cinnamoyl amides of β-NTA possess KOR-1 agonist and MOR-1 antagonist activity, similar to β-FNA. [10–12] C-14 cinnamoyl amides (e.g. clocinnamox) and esters of various morphinans have received significant attention for their varying MOR-1 activity (Figure 1). [13–15] 6-arylamido morphinans have been proposed as analogues of morphine-6-O-glucuronide. [16] Similarly, aryl naloxamides have been considered as alcohol cessation agents [17,18] and as potential MOR-1 antagonists. [19–22] N-Naphthoyl-naltrexamide (NNTA) reportedly targets MOR-KOR opioid dimers. [23]

Figure 1. Structures of pharmacologically important aminomorphinans.

To the best of our knowledge, C-6 acylamines of morphine and its congeners carrying Δ7–8 double bond and a methyl group on N-17 have not been reported. Herein, we present the synthesis of novel C-6 acylamines of morphine and codeine analogues acylated by a series of cinnamic acids. The synthesized compounds were evaluated for their binding affinities to mu (MOR-1), kappa₁ (KOR-1) and delta (DOR-1) opioid receptors and analgesia. Based on in vivo results, 6β-cinnamoylmorphinamine was subjected to more detailed studies.

The 6β-amino derivatives of morphine and codeine were synthesized using the Mitsunobureaction. [24] Morphine (1) was selectively acetylated in the C-3 position and subsequently treated with phthalimide and diisopropyl azodicarboxylate (DIAD) in the presence of Ph₃P. Codeine (5) was reacted directly under the same conditions. The 6β-phthalimidoyl intermediates (3,6) were isolated as HCl salts and then treated with hydrazine hydrate in ethanol to yield the desired 6β-amino derivatives (4, 7). For amidation reactions, the appropriate carboxylic acids were treated with thionyl chloride and reacted with the 6β-amino compounds in dichloromethane in the presence of Et₃N. In case of 6β-morphinamine (4) formation of 3-O-esters could also be detected; the amidation reaction mixture was evaporated to dryness and dissolved in methanolic aqueous sodium carbonate solution and left to stand overnight to ensure cleavage of C-3 phenolic esters. The crude products were purified by column chromatography using chloroform-methanol 9:1 isocratic eluent and crystallized from hexane (Scheme 1, Figure 2).

Scheme 1. Synthesis of 6β-cinnamoyl morphinamines and codeinamines.

(i) acetic anhydride, NaHCO₃, H₂O, r.t., 1 h. (ii) phthalimide, DIAD, Ph₃P, benzene, r.t., 2h. (iii) hydrazine hydrate, EtOH. (iv) 1.1 eq. acyl chloride, Et₃N, CH₂Cl₂, r.t., 2 h. (v) 1.1 eq. acyl chloride, Et₃N, CH₂Cl₂, r.t., 2 h. then Na₂CO₃, H₂O, MeOH, r.t., 12h.

Figure 2. Acyl side chains of the synthesized 6β-cinnamoyl morphinamines and codeinamines.

The compounds showed moderate to high affinity for MOR-1 with lower affinities for KOR-1 and DOR-1 (Table 1). Not surprisingly, affinities of codeinamine-derived amides (9a–e) were significantly lower than those of the amides of morphinamine (8a–e). Derivatives 8a–e had very similar subnanomolar affinitites for MOR-1 irrespective of the aryl ring substituent of the acylamino moiety, while KOR-1 binding was approximately 10-fold lower, with highest KOR-1/MOR-1 K_i ratio for 8c (55.4) and the lowest for 8e (6.8). DOR-1 receptor affinity was roughly two magnitudes lower than MOR-1 affinity. DOR-1/MOR-1 K_i ratio for 8a was 102.6. Cinnamoyl morphinamines in our series lacked the marked KOR-1 affinity seen with the previously reported acylated aminomorphinans. [10]

Table 1.

Receptor binding and in vivo analgesia data of selected 6β-acylaminomorphinans^a

Compound	Affinity (Ki, nM)			Analgesia (ED50, mg/kg, s.c.)
	MOR-1	KOR-1	DOR-1
8a	0.10 ± 0.02	2.90 ± 0.66	10.26 ± 6.76	3.13 ± 1.09
8b	0.15 ± 0.03	1.97 ± 0.01	9.38 ± 1.53	>10
8c	0.19 ± 0.09	10.52 ± 0.90	14.52 ± 6.82	>10
8d	0.74 ± 0.12	5.43 ± 1.38	15.26 ± 1.74	>10
8e	0.12 ± 0.006	0.81 ± 0.09	5.15 ± 0.75	>10
9a	4.73 ± 1.64	>100	>100	>10
9b	3.81 ± 0.15	>100	>100	>10
9c	5.44 ± 0.93	>100	>100	>10
9d	25.16 ± 13.02	>100	>100	>10
9e	3.74 ± 0.83	>100	>100	>10
morphine	4.60 ± 1.81b			4.96 ± 0.96c

^aCompetition studies were performed with the indicated compounds against ¹²⁵I-BNtxA (0.1 nM) in membranes from CHO cells stably expressing the indicated cloned mouse opioid receptors. K_i values were calculated from the IC₅₀ values [26] and represent the means ± SEM of at least three independent replications. ¹²⁵IBNtxA K_D values for MOR-1, KOR-1, DOR-1 sites were 0.11, 0.03 and 0.24, respectively.

^bValues from the literature. [7]

^cValues from the literature. [25]

Next, the substances were administered to mice subcutaneously to determine their in vivo analgesic activity (Table 1, Figure 3 A). Somewhat surprisingly only the unsubstituted cinnamoyl morphamine analog 8a (ED₅₀ 3.13 ± 1.09 mg/kg) was analgesic with potency similar to that of morphine (4.96 ± 0.97 mg/kg s.c.). [25] Despite similar affinities against MOR-1, KOR-1 and DOR-1 binding, substitution on the aromatic ring of the cinnamoyl acid eliminated the analgesic activity of analogs (8b–e) at the highest dose tested, i.e. 10 mg/kg. Thus, the in vivo active compound 8a was selected as lead compound and further characterized. The analgesic activity of 8a was partially reversed by both the MOR-1 selective antagonist β-FNA and the DOR-1 selective antagonist NTI but, interestingly, was insensitive to the KOR-1 selective antagonist norBNI (Table 1, Figure 3 B).

Figure 3. Pharmacology of 6β-cinnamoylmorphinamine 8a.

(A) Analgesia: Cumulative dose-response curves were carried out on groups of mice (n = 10) with 8a at the indicated doses (s.c.) and analgesia tested 30 min later at peak effect. The ED₅₀ value was 3.13 ± 1.09 mg/kg in CD1 mice by using the radiant heat tail-flick assay. Results were evaluated as %MPE [(observed latency − baseline latency)/(maximal latency − baseline latency)] and shown as the average of each group (n = 10). (B) Sensitivity of 8a to opioid antagonists: Groups of mice (n = 6) received a fixed dose of 8a (5 mg/kg, s.c.) alone or with NorBNI (10 mg/kg, s.c.), β-FNA (40 mg/kg, s.c.), NTI (20 mg/kg, s.c.). β-FNA and NorBNI were given 24h before 8a while NTI was given 15 min before 8a. Tail flick analgesia was measured 30 min after 8a. Similar results were observed in two independent replications. 8a analgesia is insensitive to NorBNI while analgesia is partially antagonized by both β-FNA and NTI (ANOVA followed by Bonferroni multiple comparison test (p < 0.05). (C) Respiratory rate. Animals were randomly assigned to receive saline (n = 3), 8a (24 mg/kg, n = 3), or morphine (20 mg/kg, n = 3). Each animal’s baseline average breath rate was measured every 5 min for 25 min before drug injection, and breath rates after drug injection are expressed as a percent of baseline. 8a did not depress respiratory rate and was not significantly different from saline at any time point, whereas morphine decreased respiratory depression in comparison with both saline and 8a (p < 0.05) as determined by repeated-measures ANOVA followed by Bonferroni multiple-comparison test.

The [³⁵S]GTPγS-binding assay was used to characterize the functional activity of 8a on in vitro opioid transfected cell lines (Table 2). 8a produced near maximum stimulation in all cell lines when compared with the MOR-1 agonist DAMGO, DOR-1 agonist DPDPE and KOR-1 agonist U50,488H suggesting that it acts as a full agonist at all three opioid receptors. 8a showed a 25-fold and 4-fold selective potency for MOR-1 and DOR-1 over the KOR-1 in this assay which partially explains why in spite of high, nanomolar affinity for KOR-1 it does not exert KOR-mediated agonism in vivo, as shown by the in vivo behavior assay in which 8a analgesia could be antagonized by selective MOR-1 and DOR-1 antagonists but not by KOR-1 antagonist. Since in vivo behavior is a better predictor of functional activity, it is safe to say that analgesia of 8a is mediated by agonism at both MOR-1 and DOR-1.

Table 2.

Opioid receptor efficacy of 8a^a

Compound	EC50 (nM)			% stimulation
	MOR-1	KOR-1	DOR-1	MOR-1	KOR-1	DOR-1
8a	1.38 ± 0.8	36.5 ± 19.2	9.1 ± 3.0	95.3 ± 0.8	106.3 ± 8.5	101.7 ± 2.8
DPDPE	ndb	ndb	2.05 ± 0.6
DAMGO	5.6 ± 3	ndb	ndb
U50,488H	ndb	21.7 ± 4.6	ndb

^aEfficacy data were obtained using agonist induced stimulation of [35S]GTPγS binding assay. Efficacy is represented as EC50 (nM) and percent maximal stimulation relative to standard agonist DAMGO (MOR-1), DPDPE (DOR-1), or U50,488H (KOR-1) at 100 nM. All values are expressed as the mean ± SEM of three separate assays performed in triplicate.

^bNot determined

Perhaps the most serious, potentially life-threatening side-effect of opiate therapy is respiratory depression. We tested the effect of 8a on the respiratory rate of CD1 mice and found that at doses approximately four times the ED₅₀ morphine (20 mg/kg) caused the expected marked respiratory depression, while after treatment with even a high, 8-times the ED₅₀ dose of 8a (24 mg/kg) no significant difference was found in respiratory rate compared with saline (Figure 3 C). Significantly reduced risk of respiratory depression has been reported in the literature for MOR-1/DOR-1 mixed opioid agonists, including DPI-3290 and the peptide MMP-2200. [27, 28]

In conclusion we have synthesized a series of novel 6β-cinnamoyl morphinamines carrying various cinnamoyl side chains. Characterization of compounds in vitro and in vivo revealed high affinity for MOR-1 receptors. Analgesic activity of 6β-cinnamoylmorphinamine was found to be comparable to morphine, but without causing respiratory depression, a major side-effect of commonly used opioids. Functional assays and reversal of analgesia by selective antagonists revealed an interesting receptor activity profile. The results suggest that new 6β-acylamino derivatives of morphine and its congeners are promising new opioids worthy of further investigation.

• We designed and synthesized the 6β-acylamino derivatives of morphine and codeine

• The synthesis of 10 novel potentially analgesic compounds is reported

• The compounds were tested for opioid receptor binding and in vivo analgesia

• 6β-Cinnamoylmorphinamine were shown to possess MOR-1/DOR-1 analgesia

• 6β-Cinnamoylmorphinamine did not cause respiratory depression