Bioactivity-guided synthesis of tropine derivatives as new agonists for melatonin receptors

Abstract

Twenty-three tropine derivatives as new melatonin receptor (MT₁ and MT₂) agonists were synthesized and evaluated on HEK293 cells in vitro. Derivatives 1f, 1i, 1j, 1m–1s and 1t exhibited increased agonisting activities on MT₁ and MT₂ receptors compared to the substrate tropine. Particularly, compound 1r showed significant agonistic activities on MT₁ and MT₂ receptors with EC₅₀ values of 0.20 and 0.24 mM, respectively. The preliminary structure–activity relationships (SARs) of tropine derivatives were summarized for further investigation on melatonin receptor agonists.

---

Melatonin (MT), 5-methoxy-N-acetyltryptamine, released from the pineal gland at night, can regulate physiology and behavior by targeting MT₁ and MT₂ receptors.¹ MT₁ and MT₂ receptors belonging to the G-protein-coupled receptor (GPCR) superfamily are widely distributed in brain and peripheral nervous systems of mammals,^2–7 which are fascinating targets for drug discovery due to their multiply regulatory functions.^8–10 MT₁ receptor, generally expressed in the central nervous system and many peripheral tissues,¹¹ modulates neuronal firing. MT₂ receptor is observed in brain and lung, closely related to circadian rhythms of neuronal firing in the suprachiasmatic nucleus and dopamine release in retina.¹² In recent years, different types of MT₁ and MT₂ receptors agonistic ligands were reported, some of which had been used for psychiatric purposes.^13,14 For example, TAK-375 and VEC-162 were marketed for the treatment of insomnia,¹⁵ and circadian rhythm sleep disorders.¹⁶ However, most of the MT₁ and MT₂ receptors agonists are synthetic compounds possessing high similarity with melatonin,¹⁷ and exploring novel types of agonists is still needed.

In the 19th and early 20th centuries, tropane alkaloids attracted particular interest due to their potent and extensive biological activities,¹⁸ such as regulating secretion of the monoamine neurotransmitter,^19–22 influencing expression of the glycine receptor,²³ and modulating activation of the acetylcholine receptor.^24–26 Tropine as a natural tropane alkaloid mainly distributed in Solanaceae plants, showed pleiotropic physiological effects in humans and animals,²⁷ however, its agonistic effects on MT receptors have not been revealed.

Within the G-protein coupled receptor family of proteins, the MT₁ and MT₂ receptors can couple to multiple and distinct signal transduction cascades whose activation lead to unique cellular response.²⁸ MT₁ (or MT₂)-expressing cell line was made in the HEK 293-G_α15 host cells, which supported high levels of recombinant MT₁ (or MT₂) expression on the cell surface and contained high levels of the promiscuous G protein G_α15 to enhance coupling of the receptor to the calcium signaling pathway. Based on the Ca²⁺ influx, the receptor channel pore could be controlled by agonists using fluorimetric techniques, monitored by Ca²⁺-sensitive dyes.²⁹ Test data for detecting changes in intracellular calcium was the transient calcium flux observed after activation of G-coupled protein receptors.³⁰ The details of Fluo-8 calcium assay were listed.³¹ Our previous bio-assay evaluation suggested that tropine exhibited moderate agonistic activities on MT₁ and MT₂ receptors with the values of 90.50% and 77.48% at 1 mM. Thus, a series of tropine derivatives were synthesized for developing new MT receptors agonists. The preliminary structure–activity relationships (SARs) were also discussed based on the bioassay on HEK293 cells in vitro.

With an objective to obtain target compounds which have agonistic activities on HEK293 cells in vitro, the synthetic strategies for target compounds are flexible. Tropine was reacted with various organic acids by Steglich esterification in the mixed solution of a catalytic amount of 4-dimethylaminopyridine (DMAP), dehydrating agent N,N-dicyclohexylcarbodiimide (DCC) and anhydrous CH₂Cl₂ to generate compounds 1a–f and 1i–l (Scheme 1).

Scheme 1 Reagents and conditions: (a) DMAP, DCC, CH₂Cl₂, r.t., 60–70% for compounds 1a–x. (b) PCC, CH₂Cl₂, r.t., 95% for compound 2. (c) CH₃CN, reflux, 50–65% for compounds 1m.

The results of preliminary bioassay on MT₁ receptor showed that introduction of a benzoyl group and a cinnamoyloxy group improved agonistic activity compared with that of tropine (Fig. 1), and thus, derivatives 1g–h and 1n–x were gained for further study. Derivative 2 was synthesized starting from tropine with the oxidizing agent pyridinium chlorochromate (PCC) by oxidation reaction, which was no activity on MT₁ receptor. Derivative 1c and piperazine were dissolved in CH₃CN with reflux condition to give derivative 1m, and agonistic activity of derivative 1m on MT₁ receptor increased. The structure of target compounds were identified by ¹H-NMR, ¹³C-NMR, MS and HRMS. Physicochemical properties (including clog p – calculated logarithm of partition coefficient between n-octanol and water, TPSA-polar surface area, log S – a unit stripped logarithm of the solubility measured in mol liter⁻¹) and toxicity profiles (including mutagenic effect, tumorigenic effect, irritating effect and reproductive effect) of these compounds were calculated and predicted using OSIRIS Property Explorer software at URL http://www.organic-chemistry.org/prog/peo/.^32,33 The calculated data were shown in Table 1.

Fig. 1 Agonistic activities of derivatives 2, 1a–f and 1i–m on MT₁ receptor. The agonistic activities of derivatives (1.00 mM) and agomelatine (3.33 μM, the control group) were performed in triplicate, and the results were presented as X̄ ± SD (n = 3). The agonistic activity of the control group was 100%.

Table 1 Calculated physicochemical properties and predicted toxicity of compounds 2 and 1a–x

Compd	clog p^a	TPSA^b (Å^2)	log S^c	Toxicity risks^d
				M/T/I/R
a clog p, calculated logarithm of partition coefficient between n-octanol and water.b TPSA, polar surface area.c S, solubility.d Toxicity risks: M, mutagenic effect; T, tumorigenic effect; I, irritating effect; R, reproductive effect; L, low; M, medium; H, high.
2	0.65	20.31	−1.26	L/L/L/L
1a	0.99	29.54	−1.62	L/L/L/L
1b	1.89	29.54	−2.16	L/L/L/L
1c	2.12	29.54	−2.56	M/H/L/H
1d	2.57	29.54	−2.92	L/L/H/L
1e	6.44	29.54	−4.86	L/L/L/L
1f	2.43	29.54	−2.79	L/L/L/L
1g	3.04	29.54	−3.52	L/L/L/L
1h	3.16	29.54	−3.62	L/L/L/L
1i	2.25	29.54	−2.05	L/L/H/L
1j	2.76	29.54	−3.16	L/L/H/L
1k	2.46	38.77	−2.84	L/L/H/L
1l	1.42	67.87	−2.49	L/L/L/L
1m	2.32	32.78	−2.24	L/L/L/L
1n	3.10	29.54	−3.50	L/L/L/L
10	3.10	29.54	−3.50	L/L/L/L
1p	2.86	29.54	−3.47	L/L/L/M
1q	2.86	29.54	−3.47	L/L/L/L
1r	2.86	29.54	−3.47	L/L/L/L
1s	3.37	29.54	−3.89	L/L/L/L
1t	2.69	38.77	−3.17	L/L/H/M
1w	3.97	29.54	−4.63	L/L/L/L
1x	2.62	48.00	−3.19	L/L/L/L

---

As shown in Table 2, tropine displayed moderate agonistic activities on MT₁ and MT₂ receptors with the values of 90.50% and 77.48% at 1 mM. When the hydroxyl group was changed to be a carbonyl group, derivative 2 obviously reduced activities on MT₁ and MT₂ receptors with the values of 18.42% and 26.73% at 1 mM respectively, suggesting C=O group was unfavorable activity for maintaining activities (Fig. 2). In an effort to gain more information of the SARs of tropine derivatives, the esterified products were further obtained.

Table 2 Agonistic activities of tropine derivatives (general molecular formula 1, Fig. 3) on melatonin receptors^a

Comp.	R	Agonistic activities (%)
		MT1^b	MT2^c
a Agomelatine was tested at the concentration of 3.33 μM and other compounds were tested at the concentration of 1.00 mM.b The agonistic activities expressed as X̄ ± SD (n = 3) were obtained by comparing to the highest agonistic activity that was achieved for the agomelatine (3.33 μM) and was set as 100%.c
Ago.	—	100.00 ± 3.54	100.00 ± 2.00
2	—	18.42 ± 5.72	26.73 ± 1.84
Tropine	H	90.50 ± 3.33	77.48 ± 5.67
1a	Acetyl	69.51 ± 3.43	98.45 ± 4.35
1b	Butyryl	21.63 ± 3.70	149.97 ± 33.86
1c	4-Chloro butyryl	−3.06 ± 0.45	106.34 ± 8.71
1d	2-Methylvaleryl	73.52 ± 4.62	141.05 ± 13.91
1e	Myristoyl	4.20 ± 1.26	88.60 ± 14.28
1f	Benzoyl	194.22 ± 3.15	219.61 ± 7.28
1g	4-Chlorlbenzoyl	100.44 ± 3.50	127.05 ± 15.81
1h	4-Bromoxybenzoyl	47.65 ± 1.74	88.69 ± 2.03
1i	Senecioyl	163.13 ± 5.70	173.01 ± 50.85
1j	Cinnamoyl	197.18 ± 14.55	215.67 ± 14.01
1k	3-Phenoxy-propionyl	−2.55 ± 0.24	193.87 ± 7.35
1l	Boc-glycyl	96.37 ± 2.97	95.64 ± 1.91
1m	4-(N-Piperidyl)butyryl	173.02 ± 2.35	217.84 ± 12.70

---

Fig. 2 Structure–activity relationships of tropane derivatives.

Fig. 3 General molecular formula 1 and 2.

Most of esterified derivatives showed attractively agonistic activities on MT₁ and MT₂ receptors (Table 2). The acetylated derivative 1a displayed similar agonistic activities compared with tropine. Derivative 1b, with a butyryl group at C-3 position, exhibited increased agonistic activity on MT₂ receptor. Derivative 1c with a chlorinated group at the C-4 position of the butyryl group showed similar agonistic activity with derivative 1b. Derivative 1d with a 2-methyl-valeryl group owned an approximately 2-fold agonistic activity on MT₂ receptor and similar agonistic activity on MT₁ receptor compared to tropine. However, the myristoyl derivative 1e resulted in a significantly decreased activities on MT₁ and MT₂ receptors. The above analyses suggested that bearing –OCO– ester group at C-3 position and a side chain (less than 5 carbons) are crutial for agonistic activities (Fig. 2). Followingly, the effect of some bulky aromatic groups at C-3 position was investigated. As shown in Table 2, introduction of a benzoyl group improved agonistic activities on both receptors. Derivative 1f endowed significantly agonistic activities on MT₁ (194.22% at 1.00 mM) and MT₂ (219.61%, at 1.00 mM) receptors. These results encouraged us to further explore derivatives 1g and 1h by introducing diverse substitutions on a phenyl group. Interestingly, the unsubstituted benzoyl derivative 1f, showed higher agonistic activities than derivatives 1g and 1h with para-substituted at the phenyl ring. Agonistic activities of derivative 1i with a conjugate double bond was two times higher than that of the tropine on both receptors. Cinnamoyloxy substituted derivatives 1j with a conjugate double bond exhibited excellent agonistic activities on both receptors (Fig. 2), which encouraged us to further explore derivatives 1o–1x. Derivative 1k with a phenoxylpropionyl group showed a 3-fold agonistic activity on MT₂ receptor but no activity on MT₁ receptor. Derivative 1i with a N-(tert-butoxycarbonyl)-glycinyl group had similar agonistic activities compared with that of tropine. Derivative 1m with a 4-butyryl piperidyl group at C-4 position, displayed an approximately 2-fold agonistic activity on MT₁ receptor and 3-fold on MT₂ receptor compared with tropine.

Derivative 1j with a cinnamoyloxy group at C-3 position exhibited excellent agonistic activities on MT₁ (197.18%, at 1.00 mM) and MT₂ (215.67%, at 1.00 mM) receptors, suggesting that the cinnamoyl group is preferable for gaining high agonistic activity. Followingly, derivatives with diverse substitutions on the cinnamoyl group were synthesized in the following investigation. Derivatives 1n–x, with different substituents on the cinnamoyl ring, displayed agonistic activities from 56.65 to 190.04% on MT₁, and from 30.53% to 255.23% on MT₂, respectively (Table 3). Agonistic activities of methyl substituted analogues at ortho- and para-positions of phenyl ring (1n, 1o) resulted in a significantly decreased activity on MT₁ receptor and maintained almost no change on MT₂ receptor compared with derivative 1j. Fluorinated analogous (1p, 1q, 1r) with ortho-, meta-, and para-substituted patterns at the phenyl ring, displayed agonistic potency on MT₂ receptor with the values of 177.57%, 196.72% and 255.23%, respectively. Agonistic activity of fluorinated analogue at para-position of phenyl ring exhibited higher activity than derivative 1j on MT₂ receptor. Chlorinated derivative (1s) at meta-position of phenyl ring displayed slightly lower activity than derivative 1j on both receptors. Derivative 1t with a methoxy group at the C-3 position of the phenyl ring showed similar agonistic activities on both receptors compared with derivative 1j. The dichloro substituted analogue (1w) and the dimethoxy substituted analogue (1x) resulted in remarkably reduced activity on both receptors compared with derivative 1j.

Table 3 Agonistic activities of cinnamoyloxytropane derivatives (general molecular formula 2, Fig. 3) on melatonin receptors^a

Comp.	R1	R2	R3	Agonistic activities (%)
				MT1^b	MT2^c
a Agomelatine was tested at the concentration of 3.33 μM and other compounds were tested at the concentration of 1.00 mM.b The agonistic activities expressed as X̄ ± SD (n = 3) were obtained by comparing to the highest agonistic activity that was achieved for the agomelatine (3.33 μM) and was set as 100%.c
Ago.	—	—	—	100.00 ± 3.54	100.00 ± 2.00
1n	Me	H	H	158.60 ± 9.63	200.66 ± 4.50
1o	H	H	Me	145.56 ± 2.76	223.44 ± 10.94
1p	F	H	H	142.46 ± 6.25	177.57 ± 12.79
1q	H	F	H	160.17 ± 9.02	196.72 ± 3.75
1r	H	H	F	190.04 ± 9.13	255.23 ± 10.01
1s	H	Cl	H	125.35 ± 2.64	183.36 ± 2.35
1t	H	H	OMe	146.35 ± 10.32	193.24 ± 8.02
1w	Cl	H	Cl	56.65 ± 8.75	36.56 ± 10.02
1x	H	OMe	OMe	63.00 ± 6.26	30.53 ± 2.81

---

The dose–response curves for the most potent derivative 1r was investigated to provide EC₅₀ values of 0.20 and 0.24 mM on MT₁ and MT₂ receptors, respectively (Fig. 4). Dose–response of calcium activity was performed in triplicate and monitored with FlexStation plate reader. EC₅₀ values for the derivative 1r were determined from the dose–response curves obtained with eight concentrations from the range of 7.81 to 1000 μM for MT₁ and MT₂ receptors, and calculated by the software of Graphpad Prism 5.0.

Fig. 4 The dose-dependent effects of derivative 1r on MT₁ and MT₂ receptors.

Conclusions

In conclusion, twenty-three derivatives of tropine were synthesized and evaluated on MT₁ and MT₂ receptors, of which, derivatives 1f, 1j and 1r exhibited promising agonistic activities. The preliminary SARs suggested that the hydroxyl group at C-3 position was crucial for maintaining agonistic activities, and esterification on the hydroxyl group will generate different activities. This investigation will provide valuable insights for further developing tropine derivatives as potential melatonin receptors agonists.

Author contributions

Xiu-juan Yin and Ji-jun Chen contributed to the design of experiments and the manuscript preparation. Xiu-juan Yin, Chang-an Geng, Xiao-yan Huang, Yun-bao Ma and Ji-jun Chen performed the experimental studies and analyzed the data. Xiao-yan Huang and Yun-bao Ma contributed to the design and test of biological activity. Xing-long Chen was responsible for data testing of HRMS. Chang-li Sun, Tong-hua Yang, Jun Zhou and Xue-mei Zhang contributed in critical reading and discussion on the manuscript. All the authors approved the final version.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 81573322), the Program of Yunling Scholar, the Hundred Talents Program of the Chinese Academy of Sciences (CAS) and the Youth Innovation Promotion Association, CAS.

Notes and references

1. D. Sugden, K. Davidson, K. A. Hough and M. T. Teh, Pigm. Cell Res., 2004, 17, 454.
2. A. Ghanemi, Saudi Pharm. J., 2015, 23, 115.
3. J. M. Karasinska, S. R. George and B. F. O'Dowd, Brain Res. Rev., 2003, 41, 125.
4. B. L. Roth, D. L. Willins and W. K. Kroeze, Drug Alcohol Depend., 1998, 51, 73.
5. J. A. Salon, D. T. Lodowski and K. Palczewski, Pharmacol. Rev., 2011, 63, 901.
6. R. Jockers, P. Maurice, J. A. Boutin and P. Delagrange, Br. J. Pharmacol., 2008, 154, 1182.
7. V. Audinot, F. Mailliet, C. Lahaye-Brasseur, A. Bonnaud, A. Le Gall, C. Amossé, S. Dromaint, M. Rodriguez, N. Nagel, J. P. Galizzi, B. Malpaux, G. Guillaumet, D. Lesieur, F. Lefoulon, P. Renard, P. Delagrange and J. A. Boutin, Naunyn-Schmiedeberg's Arch. Pharmacol., 2003, 367, 553.
8. M. L. Dubocovich, P. Delagrange, D. N. Krause, D. Sugden, D. P. Cardinali and J. Olcese, Pharmacol. Rev., 2010, 62, 343.
9. M. A. Shehata, H. B. Christensen, V. Isberg, D. S. Pedersen, A. Bender, H. Bräuner-Osborne and D. E. Gloriam, RSC Adv., 2015, 5, 48551.
10. S. M. Reppert, C. Godson, C. D. Mahle, D. R. Weaver, S. A. Slaugenhaupt and J. F. Gusella, Proc. Natl. Acad. Sci. U. S. A., 1995, 92, 8734.
11. K. H. Chan and Y. H. A. Wong, Int. J. Mol. Sci., 2013, 14, 18385.
12. M. L. Dubocovich and M. Markowska, Endocrine, 2005, 27, 101.
13. D. P. Zlotos, Arch. Pharm., 2005, 338, 229.
14. L. Naji, A. Carrillo-Vico, J. M. Guerrero and J. R. Calvo, Life Sci., 2004, 74, 2227.
15. K. Kato, K. Hirai, K. Nishiyama, O. Uchikawa, K. Fukatsu, S. Ohkawa, Y. Kawamata, S. Hinuma and M. Miyamoto, Neuropharmacology, 2005, 48, 301.
16. S. M. W. Rajaratnam, M. H. Polymeropoulos, D. M. Fisher, T. Roth, C. Scott, G. Birznieks and E. B. Klerman, Lancet, 2009, 373, 482.
17. D. P. Zlotos, Curr. Med. Chem., 2012, 19, 3532.
18. J. W. Medley and M. Movassaghi, Chem. Commun., 2013, 49, 10775.
19. L. Melzig, A. Gavryushin and P. Knochel, Org. Lett., 2007, 26, 5529.
20. K. C. Schmitt, J. Zhen, P. Kharkar, M. Mishra, N. Chen, A. K. Dutta and M. E. A. Reith, J. Neurochem., 2008, 107, 928.
21. R. H. Kline, S. Izenwasser, J. L. Katz, D. B. Joseph, W. D. Bowen and A. H. Newman, J. Med. Chem., 1997, 40, 851.
22. J. L. Katz, G. E. Agoston, K. L. Alling, R. H. Kline, M. J. Forster, W. L. Woolverton, T. A. Kopajtic and A. H. Newman, Psychopharmacology, 2001, 154, 362.
23. G. Maksay, P. Nemes, Z. Vincze and T. Bíró, Bioorg. Med. Chem., 2008, 16, 2086.
24. R. L. Papke, H. C. Schiff, B. A. Jack and N. A. Horenstein, Neurosci. Lett., 2005, 378, 140.
25. K. Sun and Z. D. Guo, Int. J. Pharm. Res., 2008, 35, 87.
26. D. I. Lainé, H. B. Xie, N. Buffet, J. J. Foley, P. Buckley, E. F. Webb, K. L. Widdowson, M. R. Palovich and K. E. Belmonte, Bioorg. Med. Chem. Lett., 2007, 17, 6066.
27. W. J. Griffin and G. D. Lin, Phytochemistry, 2000, 53, 623.
28. P. A. Witt-Enderby, J. Bennettb, M. J. Jarzynkaa, S. Firestinea and M. A. Melan, Life Sci., 2003, 72, 2183.
29. J. Walstab, S. Combrink, M. Brüss, M. Göthert, B. Niesler and H. Bönisch, Anal. Biochem., 2007, 368, 185.
30. P. Gribbon, C. Chambers, K. Palo, J. Kupper, J. Mueller and A. Sewing, J. Biomol. Screening, 2006, 11, 511.
31. In vitro agonist activity: the Fluo-8 calcium assay could provide a fast, simple and reliable fluorescence-based assay for detecting changes in intracellular calcium. HEK293 cell lines stably expressing the human melatonin MT₁ or MT₂ receptor was grown in Dubecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), and cultured with 95% O₂/5% CO₂ at 37 °C. The cells were seeded in a Matrigel coated 96-well black plate with a plating volume of 100 μL per well at a density of 4 × 10⁴ per well, and incubated in CO₂ incubator (Thermo Forma 3310, US) for overnight. Then the cells were dyed by HDB Wash Free Calcium Assay Kit, and placed in CO₂ incubator for 1 h. Tested compounds and positive drug were dissolved in 10 μL dimethyl sulfoxide (DMSO) and 990 μL HBSS Buffer respectively, and extracted a plating volume of 100 μL per well in a Matrigel coated 96-well plate. Two 96-well plates were put into Flexstation 3 Benchtop Multi-Mode Microplate Reader (Molecular Devices, Sunnyvale, California, USA). The absorption values were readed by Flexstation 3 Benchtop Multi-Mode Microplate Reader at room temperature with wavelength (excitation: 485 nm; emission: 525 nm; emission cut-off: 515 nm). The agonistic activities expressed as X̄ ± SD (n = 3) were obtained by comparing to the highest agonistic activity that was achieved for melatonin at the highest concentration and was set as 100%. The results were calculated by the software of Graphpad Prism 5.0.
32. S. Y. Wang, L. J. Wang, B. Jiang, N. Wu, X. Q. Li, J. Luo, B. C. Wang, R. S. Zhang, Q. Xu and D. Y. Shi, RSC Adv., 2015, 5, 91795.
33. OSIRIS Property Explore, http://www.organic-chemistry.Org/prog/peo/, accessed 23-11-2015.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra06748d

---