Novel Delta Opioid Receptor Agonists with Oxazatricyclodecane Structure

Abstract

We synthesized compounds 4a,c–f,h,i containing the oxazatricyclodecane structure from a novel rearrangement reaction product 2a. All the prepared compounds 4a,c–f,h,i exhibited full agonistic activities for the δ opioid receptor (DOR). Among them, the N-methyl derivative 4c was highly selective, and the most effective DOR agonist in functional assays. Subcutaneous administration of 4c produced dose-dependent and NTI (selective DOR antagonist)-reversible antinociception lacking any convulsive behaviors in the mice acetic acid writhing tests. The N-methyl derivative 4c is expected to be a promising lead compound for selective DOR agonists with a novel chemotype.

The δ opioid receptor (DOR) is one of the three opioid receptor types (μ (MOR), DOR, and κ (KOR)), and activation of this receptor is associated with various pharmacological effects such as antinociceptive, antidepressive, anxiolytic, and cardioprotective effects.¹⁻³ In contrast to the undesirable effects mediated by the MOR such as dependence, constipation, emesis, and respiratory depression or the aversive effects mediated by the KOR,^4,5 the DOR is a promising medical target because it seems to induce neither addictive nor aversive effects. Since the first nonpeptidic DOR agonist TAN-67^6,7 (Figure 1) emerged,³ various nonpeptidic DOR agonists have been reported.¹⁻³ Several investigations revealed that the DOR agonists like BW373U86⁸ and SNC80⁹ (Figure 1) exerted convulsive behaviors.³ However, some DOR agonists such as ADL5747¹⁰ and KNT-127^11,12 (Figure 1) have recently been reported to induce no convulsion. Although SNC80 has been reported to induce the internalization of the DORs and to develop tolerance toward the analgesic, locomotor, and anxiolytic effects, ARM390¹³ (Figure 1) induced hardly any internalization of the DORs and showed tolerance to analgesia but not to locomotor or anxiolytic responses.^14,15 Thus, a distinct DOR agonist interacting with the same DOR sometimes exerted different pharmacological responses. Recently, SNC80, a well-known representative selective DOR agonist, was reported to activate the MOR/DOR heteromer more selectively than the DOR homomer.¹⁶ It is not yet clear why the various DOR agonists mentioned above elicit different pharmacological responses, but the structure of the DOR agonist may account, in part, for their distinct activities. For example, a structural feature of DOR agonists may influence the induction of convulsive behaviors: the DOR agonists that do not cause convulsion had a structure distinct from diarylmethylpiperazine and its related structures such as BW373U86 and SNC80.³ However, diarylmethylpiperazine derivative AZD2327 (Figure 1) reportedly produced no convulsion.¹⁷ The synthesis and pharmacological characterization of DOR agonists with different chemotypes will help to better understand the different pharmacological profiles of distinct DOR agonists. We have recently reported the synthesis and binding affinities for the MOR, DOR, and KOR of an oxazatricyclodecane derivative 2a, which was obtained from endoethanotetrahydrothebaine derivative 1 by a novel rearrangement reaction¹⁸ (Scheme 1). This new compound exhibited moderate affinities for the opioid receptors (K_i (MOR) = 47.7 nM, K_i (DOR) = 174.6 nM, and K_i (KOR) =248.1 nM). The potential opioid ligand 2a was expected to lead to other ligands selective for an opioid receptor type with a unique core structure. Herein, we report the synthesis of novel DOR agonists 4a,c–f,h,i with oxazatricyclodecane structure derived from 2a and their pharmacological properties.

Figure 1.

Structures of DOR agonists, TAN-67, BW373U86, SNC80, ADL5747, KNT-127, ARM390, and AZD2327.

Scheme 1. Potential Opioid Ligand 2a.

The synthesis of the objective compounds 4a,c–f commenced with compound 2a(18) (Scheme 2). The treatment of 2a with 2,2,2-trichloroethyl chloroformate (Troc-Cl) in the presence of K₂CO₃ and the subsequent zinc/AcOH treatment gave norcompound 2b.¹⁹ Various N-substituents were introduced by reductive alkylation of 2b or the alkylation of 2b with an alkyl bromide to provide 2c–f. Compound 2a reacted with CH₂Br₂ in the presence of K₂CO₃ under high dilution conditions (0.0005 M) to provide dioxymethylene compound 3a in 66% yield concomitantly with a dimer in 30% yield in which two 2a units were tethered with a methylene group (see the Supporting Information for details). A portion-wise addition of a solution of 2c–f markedly improved the yields of 3c–f and prevented formation of the dimer. Finally, the O-methyl group in 3a,c–f was removed by a treatment with BBr₃ to give 4a,c–f. Compounds 4h and 4i with respective phenyl and 2-phenethyl groups as the lactam nitrogen substituents were prepared as shown in Scheme 3. After a conversion of ester 5 into 6, the treatment of 6 with t-BuOK in t-BuOH provided an equilibrium mixture of 2g and 7g. An equilibrium mixture of 2h and 7h was prepared from 5 by a previously reported method.¹⁸ The mixture of 2g and 7g or 2h and 7h was reacted with CH₂ClBr in the same manner shown in Scheme 2 to afford dioxymethylene compounds 3g,h. The 2-phenethyl group was introduced on the lactam nitrogen in 3g by alkylation to give 3i.

Scheme 2. Synthesis of 4a,c–f.

Reagents and conditions: (a) Troc-Cl, K₂CO₃, 1,1,2,2-tetrachloroethane, 150 °C; (b) Zn, AcOH, rt, 80% from 2a; (c) aldehyde, AcOH, NaB(OAc)₃H, 1,2-dichloroethane, rt, 74%-quant. (for R = Me, 2-phenethyl); (d) alkyl bromide, NaHCO₃, DMF, rt, 48–92% (for R = allyl, i-Bu); (e) CH₂Br₂, K₂CO₃, DMF (0.0005 M), rt, 66%; (f) CH₂ClBr, K₂CO₃, DMF (0.0004 M), rt, a solution of 2c–f in DMF was added portion-wise. 69–98%; (g) BBr₃, CH₂Cl₂, 0 °C, 76–95%.

Scheme 3. Synthesis of 4h and 4i.

Reagents and conditions: (a) 12 M NH₃aq, EtOH, rt, 73%; (b) t-BuOK, t-BuOH, reflux, quant.; (c) CH₂ClBr, K₂CO₃, DMF (0.0004 M), rt, a solution of 2g and 7g or 2h and 7h in DMF was added portion-wise. 73–93%; (d) 60% NaH, PhCH₂CH₂Br, DMF, rt, 83%; (e) BBr₃, CH₂Cl₂, 0 °C, 61–89%.

The affinities of the prepared compounds 4a,c–f,h,i were evaluated by competitive binding assays (Table 1). All the compounds 4a,c–f,h,i bound to the opioid receptors. The phenolic hydroxy group at the 3-position appeared to play an important role in improving the binding affinities for the opioid receptors compared to the parent compound 2a.²⁰ Except for N-(2-phenethyl)lactam 4i, compounds 4a,c–f,h showed selectivities for the DOR, suggesting that the phenyl group of the substituent on the lactam nitrogen would function as a DOR address such as the phenyl moiety in NTI.^21,22The binding affinities of 4a and 4h for the DOR were better than that of SNC80. Compounds 4c and 4f with respective N-methyl and N-(2-phenethyl) substituents were over 100-fold more selective for the DOR as compared to the KOR. The functional activities of 4a,c–f,h,i were determined by [³⁵S]GTPγS binding and CellKey assays (Tables 2 and 3).²³ The CellKey system utilizes impedance biosensors for detection of cell behaviors and is a functional cell-based assay technology enabling label-free analysis of cell surface receptor activity.^24,25 It is noteworthy that the [³⁵S]GTPγS and CellKey assays differed in the observed output, even though giving similar results. All the compounds 4a,c–f,h,i were full agonists for the DOR. The agonistic activities for the DOR of 4c,f,h were more efficacious than that of SNC80 in both of the functional assays. Compounds 4h and 4i were also potent KOR agonists, whereas compounds 4c and 4f exhibited agonistic activities for the MOR. Although N-methyl derivative 4c had moderate to high efficacy for the MOR and KOR, the potencies for these receptors were poor, which suggested that 4c was highly selective and the most efficacious DOR agonist among the tested compounds. Derivatives 4a,e,f with respective cyclopropylmethyl (CPM), allyl, and 2-phenethyl substituents on the basic nitrogen were more potent agonists for the DOR than N-methyl derivative 4c in both functional assays; however, their functional selectivities for the DOR were lower than that of 4c in [³⁵S]GTPγS binding assays and lower or comparable to that of 4c in CellKey assays. Therefore, the N-methyl substituent on the basic nitrogen appeared to be the optimal group among the tested compounds.

Table 1. Binding Affinities of 4a,c–f,h,i for the Opioid Receptors^a.

	Ki (nM)			selectivity
compd	MORb	DORc	KORd	MOR/DOR	KOR/DOR
SNC80	695	1.04	>1000	668	962
2ae	47.7	175	248	0.27	1.4
4a	3.14	0.313	5.14	10.0	16.4
4c	23.3	1.94	200	12.0	103
4d	186	7.00	119	26.6	17.0
4e	68.4	1.23	56.6	55.8	46.2
4f	45.9	2.59	588	17.7	227
4h	4.61	0.534	1.69	8.6	3.2
4i	1.75	1.16	1.94	1.5	1.7

^aBinding assays were carried out in duplicate using mouse whole brain without cerebellum membranes for MOR and DOR or guinea pig cerebellum membranes for KOR.

^b[³H] DAMGO was used.

^c[³H] DPDPE was used.

^d[³H] U-69,593 was used.

^eRef (18).

Table 2. Functional Activities of 4a,c–f,h,i for the Opioid Receptors Assessed by [³⁵S]GTPγS Binding Assays^a.

	MOR		DOR		KOR
compd	EC50 (nM)	Emax (%)b	EC50 (nM)	Emax (%)c	EC50 (nM)	Emax (%)d
SNC80	NTe	NTe	1.9	100	NTe	NTe
4a	2.8	13.7	1.1	92.8	80.5	69.1
4c	113	110	11	112	478	83.6
4d	223	8.4	15.6	96.4	760	65.6
4e	2.7	5.4	6.5	94.6	231	74.0
4f	2.3	83.0	9.2	115	NDf	NDf
4h	9.0	25.6	0.98	118	6.5	42.9
4i	2.1	19.7	0.41	103	3.9	51.2

^a[³⁵S]GTPγS binding assays were carried out in duplicate using human MOR, DOR, or KOR expressed CHO cells.

^bE_max was calculated as the % of the response obtained with DAMDO.

^cE_max was calculated as the % of the response obtained with SNC80.

^dE_max was calculated as the % of the response obtained with U-69,593.

^eNot tested.

^fNot determined.

Table 3. Functional Activities of 4a,c–f,h,i for the Opioid Receptors Assessed by CellKey Assays^a.

	MOR		DOR		KOR
compd	EC50 (nM)	Emax (%)b	EC50 (nM)	Emax (%)c	EC50 (nM)	Emax (%)d
SNC80	0.14	6.8	1.7	100	5264	5.9
4a	1.8	9.6	1.54	88.7	39.8	50.5
4c	1350	36.1	141	138	307	79.2
4d	400	12.2	140	130	333	57.9
4e	5.1	11.3	2.0	77.2	NDe	NDe
4f	639	40.2	20.5	108	12530	22.6
4h	1.2	8.8	0.39	123	1.2	80.4
4i	5.2	4.8	0.62	90.6	2.4	75.8

^aCellKey assays were carried out in duplicate using human MOR, DOR, or KOR expressed HEK293 cells.

^bE_max was calculated as the % of the response obtained with DAMGO.

^cE_max was calculated as the % of the response obtained with SNC80.

^dE_max was calculated as the % of the response obtained with (−)-U-50,488H.

^eNot determined.

We next assessed the antinociceptive effects of 4c in mice by acetic acid writhing tests. Subcutaneously administered 4c significantly exhibited antinociception in a dose-dependent manner and its EC₅₀ value was 5.26 mg/kg (Figure 2a). No convulsive behaviors were observed. The antinociceptive effects induced by 4c were attenuated by the selective DOR antagonist NTI but not by the selective MOR antagonist β-FNA or the selective KOR antagonist nor-BNI (Figure 2b). Taken together, these results indicate that compound 4c could be a promising lead compound for selective DOR agonists with a novel chemotype, the oxazatricyclodecane structure

Figure 2.

(a) Antinociceptive effect of 4c administered subcutaneously in the mice acetic acid writhing tests. The statistical significance of differences between the groups was assessed with one-way ANOVA followed by Bonferroni’s test. *p < 0.05 and ***p < 0.001 versus saline treated mice. (b) Effects of opioid receptor antagonists on the antinociception induced by subcutaneous treatment of 4c in the mice acetic acid writhing tests. The statistical significance of differences between the groups was assessed with one-way ANOVA followed by Bonferroni’s test. ***p < 0.001 versus saline treated mice. #p < 0.05 versus 4c treated mice.

In conclusion, we synthesized novel DOR agonists 4a,c–f,h,i with oxazatricyclodecane structure. Among the synthesized compounds, N-methyl derivative 4c was highly selective and the most effective DOR agonist. Subcutaneous administration of 4c produced dose-dependent and NTI-reversible antinociception without any convulsive behaviors. N-Methyl derivative 4c is expected to be a promising lead compound for selective DOR agonists containing the novel oxazatricyclodecane structure.

Glossary

ABBREVIATIONS

Bn

benzyl

CHO

chinese hamster ovary

CPM

cyclopropylmethyl

DAMGO

[d-Ala², N-Me-Phe⁴, Gly-ol⁵]-enkephalin

DOR

δ opioid receptor

DPDPE

[d-Pen², d-Pen⁵]-enkephalin

β-FNA

β-funaltrexamine

HEK

human embryonic kidney

KOR

κ opioid receptor

MOR

μ opioid receptor

nor-BNI

nor-binaltorphimine

NTI

naltrindole

Troc

2,2,2-trichloroethoxycarbonyl

Supporting Information Available

Experimental procedures for the synthesis and characterization of the compounds, the in vitro activity assay, the in vivo mice acetic acid writhing assay, and the spectral data of the reported compounds. This material is available free of charge via the Internet at http://pubs.acs.org.

Author Present Address

^# (H.N.) International Institute for Integrative Sleep Medicine, University of Tsukuba, 1–1–1 Tennodai, Tsukuba, Ibaraki 305–8577, Japan.

Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

This work was partly supported by JSPS KAKENHI (23590133) [Grant-in-Aid for Scientific Research (C)] (H.F) and Adaptable and Seamless Technology Transfer Program through target-driven R&D, JST (AS2315040G) (to H.F, K.H, S.H, T.I, E.N., T.N., and H.N.).

The authors declare no competing financial interest.