Discovery of Peripheral κ-Opioid Receptor Agonists as Novel Analgesics

Abstract

κ-Opioid receptor agonists with high selectivity over the μ-opioid receptor and peripheral selectivity are attractive targets in the development of drugs for pain. We have previously attempted to create novel analgesics with peripheral selective κ-opioid receptor agonist on the basis of TRK-820. In this study, we elucidated the biological properties of 17-hydroxy-cyclopropylmethyl and 10α-hydroxy derivatives. These compounds were found to have better κ-opioid receptor selectivity and peripheral selectivity than TRK-820.

The μ-opioid receptor (MOR) is available for analgesia. However, MOR agonists cause drug dependence and respiratory depression. Therefore, in the clinical setting, analgesics that avoid these adverse effects are strongly desired. The κ-opioid receptor (KOR) agonist is particularly noteable because it has an analgesic effect but causes minimal physical dependence and respiratory depression.¹⁾ Recent studies have shown that KOR agonists are effective for peripheral pain, for example, visceral pain.^2,3) However, KOR agonists in the central nervous system sometimes cause unpleasant sensations and sedation.⁴⁾ Therefore, we expected that a highly peripheral selective KOR agonist could serve as a novel analgesic.

Several pharmaceutical companies have examined peripheral KOR agonists; for example, asimadoline and CR-845 are currently in clinical trials for treatment of pruritus by oral administration and for treatment of pain. Peripheral selectivity of asimadoline was examined, and its brain/peripheral (B/P) ratio was found to be 0.73.⁵⁾ CR-845 is a tetra peptide that is presumed cannot easily penetrate the blood–brain barrier (BBB) (Fig. 1).

Fig. 1. Structure of Asimadoline and CR-845

We previously developed the KOR agonist TRK-820 (nalfurafine hydrochloride) as an antipruritic agent for uremic pruritus⁶⁾ (Fig. 2). Its B/P ratio is 0.41.⁷⁾

Fig. 2. Structure and KOR/MOR Agonist Activity of TRK-820

For the purpose of creating oral analgesics with highly peripheral selective KOR agonist, we focused on TRK-820 as a key compound. We considered controlling BBB permeability by increasing the polarity of TRK-820. The introduction location of polar substituents is important in terms of retaining KOR agonist activity and selectivity. First, we investigated the placement of a hydroxy group into the available space around TRK-820 using docking simulation between TRK-820 and the crystal structure of KOR (pdbid:4DJH).⁸⁾ According to the docking model of TRK-820 with KOR, there is enough space around the 10α-position and the 17-cyclopropylmethyl position, so we attempted to introduce a hydroxy group at these locations (Fig. 3). In this letter, we report the synthesis of the 17-hydroxy-cyclopropylmethyl derivative, as well as the pharmacological activities and B/P ratio of 10α-hydroxy and 17-hydroxy-cyclopropylmethyl derivatives.

Fig. 3. Docking Model of Crystal Structure of KOR with 10α-Hydroxy TRK-820 (Left) or 17-Hydroxy-cyclopropylmethyl TRK-820 (Right)

Introduction of hydroxy group at the 10α and 17-cyclopropylmethyl position is permissible (black dotted line).

The 17-hydroxy-cyclopropylmethyl derivative was synthesized as shown in Chart 1. The phenolic hydroxyl group of naloxone was protected with a benzyl group. The allyl group of 2 was removed using Wilkinson’s catalyst. Amidation of 3 and 1-hydroxycyclopropane-1-carboxylic acid using HATU gave 4. Reductive amination by treatment with N-methylbenzylamine afforded 5. Reduction of the amide moiety and subsequent removal of the benzyl protecting group gave 7. Finally, 8 was synthesized by N-acylation with (E)-3-(furan-3-yl)acryloyl chloride.

Chart 1. Synthesis Route to the 17-Hydroxy-cyclopropylmethyl Derivative

Synthesis of the 10α-hydroxy derivatives have been reported previously.^9,10) ortho- and meta-Phenol derivatives were synthesized as shown in Chart 2. Amine precursor 9 was reacted with 2- or 3-hydroxycinnamic acid to give 11 and 12, respectively.¹⁰⁾ Compound 10 is available for assessing the effect of adding the 10α-hydroxy group to TRK-820. Additionally, a phenol group was introduced at the 6-acyl position instead of furan. These phenol derivatives (11–13) were expected to be more peripherally selective than 10.

Chart 2. Synthesis of Phenol Derivatives

KOR agonist activity and selectivity against MOR are shown in Table 1.¹¹⁾ Though 10, 11, 12 and 13 showed slightly weak KOR agonist activity, as with TRK-820, 8 exhibited strong KOR agonist activity equal to that of TRK-820. Because there was enough space around the 10α-position and 17-cyclopropylmethyl position, the results for 8 and 10 support the probability of the docking model. Additionally, there was space around the furan, so the KOR agonist activity of the phenol derivatives (11–13) was not decreased drastically. However, these compounds had weak MOR agonist activity; in particular, 8 and 13 showed no agonist activity up to 30000 nM. We presume that the 17-hydroxy group of 8 or the p-phenol group of 13 form hydrogen bonds with MOR thus preventing modification of the receptor to the agonist structure.¹⁰⁾ We confirmed that introducing a hydroxy group to the 10α- or 17-cyclopropylmethyl position of TRK-820 maintains KOR agonist activity and improves KOR selectivity.

Table 1. In Vitro KOR and MOR Agonist Activities and Selectivity

Next, we determined the brain–plasma concentration ratio (K_p,brain) of 8, 10 and 13 in mice; these compounds were selected with reference to the KOR selectivity results. The K_p,brain values were calculated from the brain and plasma concentrations of these compounds measured 15 min after intravenous administration to male ICR mice (n=3). The K_p,brain results are shown in Table 2. These compounds were more peripherally selective than TRK-820; in particular, 10 and 13 showed K_p,brain <0.1. We calculated Log P with Advanced Chemistry Development (ACD)/Percepta and found these compounds have higher polarity compared with TRK-820 (Table 2). Additionally, increasing the number of hydrogen bond donors made it more difficult for them to penetrate the BBB and thereby making them more peripherally selective.¹²⁾

Table 2. Brain–Plasma Concentration Ratio (K_p,brain) of Compounds at 15 min after Intravenous Administration to Male ICR Mice and Calculated Log P Values

No.	Kp,brain	Calculated Log P
TRK-820	0.41	2.4
8	0.11	1.2
10	0.065	1.7
13	0.063	1.8

Finally, we evaluated the analgesic effect on peripheral pain using an acetic acid writhing model. Compounds 8 and 10, which showed high KOR agonist activity and selectivity, were intravenously administered to mice, and 0.6% acetic acid (10 mL/kg) was immediately administered intraperitoneally.¹³⁾ Writhing behavior was recorded for 10 min beginning 10 min after acid injection. Results are shown in Fig. 4. ED₅₀ was 0.12 mg/kg for 8 and 0.31 mg/kg for 10. Central nervous system effects were not observed at the effective dose.

Fig. 4. Anti-allodynic Effects of 8 (Left) and 10 (Right) in Acetic Acid-Induced Writhing in Mice

* p<0.025 versus vehicle-treated group (Williams’ test).

In conclusion, we have demonstrated the biological properties of hydroxy TRK-820 derivatives. Compounds 8 and 10 showed especially good efficacy in an in vivo model of pain as well as better KOR selectivity and peripheral selectivity than TRK-820, which has been approved for clinical use. These compounds are therefore presumed to have lower risk of MOR and central nervous system adverse effects and better potential for use in medications. Further pharmacological investigations of these compounds are in progress and will be reported in due course.

Conflict of Interest

The authors declare no conflict of interest.

References and Notes

• 1) Millan M. J., Trends Pharmacol. Sci., 11, 70–76 (1990).
• 2) Rivière P. J. M., Br. J. Pharmacol., 141, 1331–1334 (2004).
• 3) Hughes F. M. Jr., Shaner B. E., Brower J. O., Woods R. J., Dix T. A., Open Med. Chem. J., 7, 16–22 (2013).
• 4) Pfeiffer A., Brantl V., Herz A., Emrich H. M., Science, 233, 774–776 (1986).
• 5) Jonker J. W., Wagenaar E., van Deemter L., Gottschlich R., Bender H. M., Dasenbrock J., Schinkel A. H., Br. J. Pharmacol., 127, 43–50 (1999).
• 6) Kumada H., Miyakawa H., Muramatsu T., Ando N., Oh T., Takamori K., Nakamoto H., Hepatol. Res., 47, 972–982 (2016).
• 7) Inventors; Toray Industries, Inc, assignee. WO patent 2016152953, 2016.
• 8) Wu H., Wacker D., Mileni M., Katritch V., Han G. W., Vardy E., Liu W., Thompson A. A., Huang X.-P., Carroll F. I., Mascarella S. W., Westkaemper R. B., Mosier P. D., Roth B. L., Cherezov V., Stevens R. C., Nature (London), 485, 327–332 (2012).
• 9) Horikiri H., Hirano N., Tanaka Y., Oishi J., Hatakeyama H., Kawamura K., Nagase H., Chem. Pharm. Bull., 52, 664–669 (2004).
• 10) Suzuki S., Sugawara Y., Tsuji R., Tanimura R., Kaneko C., Yuzawa N., Yagi M., Kawai K., Bioorg. Med. Chem. Lett., 27, 3920–3924 (2017).
• 11) We examined the inhibitory effects on forskolin-induced cAMP accumulation in Chinese hamster ovary (CHO)-K1 and CHO-dhfr(−) cells expressing the human KOR and MOR, respectively, as indicators of in vitro KOR and MOR agonist activities of compounds. Cellular concentration of cAMP was measured using the LANCE cAMP 384 assay kit (PerkinElmer, Inc.) according to the manufacturer’s instructions.
• 12) Zoran R., J. Med. Chem., 60, 5943–5954 (2017)
• 13) The experiments were conducted according to the Guidelines for Animal Experiments, Research & Development Division of Toray Industries, Inc.