Fluorine substituted methoxyphenylalkyl amides as potent melatonin receptor agonists

A series of fluorine substituted methoxyphenylalkyl amides were prepared with different orientations of the fluorine and methoxy groups with respect to the alkylamide side chain and with alkyl sides of differing lengths (n = 1–3).

Abstract

A series of fluorine substituted methoxyphenylalkyl amides were prepared with different orientations of the fluorine and methoxy groups with respect to the alkylamide side chain and with alkyl sides of differing lengths (n = 1–3). β-Dimethyl and α-methyl derivatives were also synthesised. The compounds were tested as melatonin agonists and antagonists using the pigment aggregation of Xenopus melanophores as the biological assay. A number of these compounds were potent melatonin agonists, the potency depending on the length of the alkyl chain, the orientation of the methoxy and fluorine substituents, the amide chain length and, for the ethyl side-chain analogues, the presence of β-substituents.

Introduction

Melatonin (1, Fig. 1) is the hormone secreted by the human pineal gland and appears to be involved in sleep onset1 and other functions associated with the body clock, the suprachiasmatic nucleus (SCN).2 It is ubiquitous throughout both the animal and plant kingdoms and must have a long evolutionary history as a hormone. Melatonin has a major role in the regulation of circadian rhythms in non-mammalian vertebrates and forms part of their control in mammals.3–5 Administration of melatonin to humans and other mammals has a sleep-promoting action.6 Sleep problems become more common in the elderly7,8 in whom there is also a loss in the production of melatonin. A prolonged-release melatonin formulation has been approved for short-term treatment of primary insomnia in elderly patients,9 and an analogue, ramelteon, has recently been approved as a treatment for insomnia characterised by difficulty with sleep onset.10 Ramelteon is the first prescription medication for insomnia with a novel therapeutic mechanism of action to be licensed for 35 years, and the only hypnotic indicated for long-term treatment of insomnia as it does not have hangover, addiction or withdrawal effects.11 Tasimelteon has been recently licensed for the treatment of non-24 hour sleep–wake disorder.12 Another melatonin analogue, agomelatine, was earlier introduced as an antidepressant and also appears to have few side effects.13 Melatonin treatment has therapeutic value in some blind subjects, restoring their disturbed circadian rhythm.14 It also has potential in the treatment of seasonal affective disorder (SAD) which afflicts some people during short winter days,15 and it has been used to reset the body clock of people suffering from jet lag.16,17 Melatonin has been implicated in a range of other conditions, including Parkinson's disease,18 Alzheimer's and other neurological conditions,19–21 and in certain cancers.22,23 There is an ongoing controversy with regard to the action of melatonin as an antioxidant and, if it is, how it functions.24–26

Fig. 1. Melatonin (1).

The melatonin receptor belongs to the family of G protein-coupled receptors (GPCR). MT₁ and MT₂ receptors share high sequence homology, while GPR50 is an orphan receptor and is considered the mammalian ortholog of Mel1c, a melatonin receptor found exclusively in fish, Xenopus species, and chickens.27 An atypical melatonin-binding site has been discovered to be the enzyme quinine reductase 2 (QR2) and it has been proposed that inhibition of this enzyme may be responsible for the anti-oxidant actions of the hormone.28 A nuclear melatonin receptor, member of the RZR/ROR receptor superfamily,29 has also been reported. Moreover, it has recently been shown that melatonin can be synthesised by mammalian skin, where it may be important in regulating hair growth and pigmentation physiology.30

We have been interested in the interaction of melatonin with its receptors and have prepared a large number of analogues in an attempt to map the receptor requirements.31 We prepared a series of substituted phenylalkylamides and showed that these simple derivatives could have high binding affinities for the melatonin receptor.32 Following this work, other groups have recently reported similar compounds with high affinities for the melatonin receptor,33–36 and have shown that enantiomers have different affinities.33,37 We now describe the preparation of a series of fluorine containing phenylalkylamides, which show high agonist activity in the Xenopus assay and further define the steric requirements of the melatonin receptor. It has been reported that fluorinated tryptamine has increased selectivity and functional activity.38 Fluorine substitution modifies the biophysical and chemical features such as lipophilicity and acidity, as well as the reactivity and conformation of the resulting derivatives.39

Results and discussion

A number of fluorine substituted phenylalkylamides were prepared using the appropriately substituted benzaldehyde, as shown in Scheme 1. Treatment of the aldehyde with hydroxylamine followed by reduction of the oxime gave the amine, which was acylated with the appropriate anhydride to give compounds 2a–c, 3a–c. Reaction of the aldehyde with p-toluenesulphonylmethyl isocyanide (TosMIC) gave phenyl methyl cyanide which on treatment with the appropriate anhydride together with RANEY® nickel and hydrogen gave compounds 4a–f and 5a–c. A Wittig reaction of the aldehyde with (cyanomethyl)triphenylphosphonium chloride followed by treatment with the appropriate anhydride with RANEY® nickel and hydrogen gave compounds 8a–f, 9a–c. The β,β-substituted derivatives 6a–f and 7a–c were prepared by a previously described method,31 as shown in Scheme 1. The appropriately substituted phenylacetonitrile was treated with NaH followed by methyl iodide and the resulting dimethylcyanide on reaction with the appropriate anhydride with RANEY® nickel and hydrogen gave the desired dimethyl derivatives. The α-methyl derivative 10 was synthesised from 4-fluoro-3-methoxybenzaldehyde by treatment with nitroethane followed by reduction of the nitro group with LAH and treatment of the resulting amine with butyric anhydride. The compounds are shown in Fig. 2.

Scheme 1. Reagents and conditions: (a) H₂NOH·HCl, CH₃CO₂Na/EtOH–H₂O, 70 °C, 1.5 h; (b) H₂/10% Pd–C/EtOH–EtOAc, 4 atm, 6 h; (c) appropriate anhydride, Et₃N, DCM, 0 °C, after addition rt, 1.5 h; (d) TosMIC/DME, tert-ButOK, –30 °C, after addition of appropriate aldehyde, –60 °C, 1.5 h; (e) appropriate acid anhydride/dry THF, H₂/RANEY®-Ni, 4 atm, 50 °C, 9–14 h; (f) Ph₃P⁺CH₂CN,Cl^–, DBU, toluene, reflux, 1 h; (g) MeI, NaH, DMF.

Fig. 2. Fluorine substituted methoxyphenylalkyl amides.

The biological activity of the analogues was assessed in a well-established model of melatonin action, the pigment aggregation response of Xenopus laevis melanophores.40 In these cells several thousands of black pigment granules are normally distributed throughout the cell and the addition of melatonin induces their rapid movement to the center of the cell. This response is termed pigment aggregation and can be quantified by measuring the change in light absorbance of the cells as the pigment concentrates near the cell centre. In the present study, a clonal melanophore cell line, generously provided by Dr. Michael Lerner (Department of Dermatology, University of Texas), was used.

The results are shown in Table 1. The majority of the compounds examined were full agonists in the melanophore model, with a few showing considerable partial agonist activity with some antagonism at the highest concentration tested (10^–4 M). As expected from our previous observations,32 the benzyl derivatives 2a–c and 3a–c are poor agonists, the side chain being too short to adequately bind into the receptor pocket. Increasing the side chain length increases the agonist effect, the ethylamine derivatives being better agonists and now the relative position of the methoxy and fluorine substituents has an effect. The introduction of methyl groups at the beta position of the ethylacetamide side chain to give the methylpropylacetamide derivatives 6 and 7 provides a surprising increase in melatonin activity since the side chain length is not optimal. This substantial increase in agonist activity appears to arise from a number of factors. Thus, the 2-fluoro-5-methoxy derivatives 6a–c have very similar agonist activities to the analogues 4a–c with no methyl substituents, whereas the 4-fluoro-3-methoxy compounds 6d–f show a substantial increase in agonist activity compared to the non-methylated derivatives 4d–f, with the N-propanoyl analogue 6f having an agonist activity greater than melatonin. This increase appears to result from a combination of increasing the population of the optimal binding conformation with an increase in the length of the N-acyl group. Thus, 6d (R = Me) and 6e (R = Me) are both ca. 150 fold more active than the non-methylated analogues 4d and 4e, while 6f (R = Pr) is ca. 2900 fold more active than 4f. The beta-dimethyl-3-fluoro-4-methoxy derivatives 7a–c are ca. 30 times more active than the non-methylated compounds 5a–c, but there is no increase in activity with increasing length of the N-acyl group, suggesting that this group is not accommodated in its normal binding site. The 3-methoxy-4-fluoro derivatives (4d–e) are better agonists than the corresponding 5-methoxy-2-fluoro derivatives (4a–c), while the 3-fluoro-4-methoxy derivatives (5a–c), in which the methoxy group is para to the side chain are still poorer, though these latter compounds are still agonists. Increasing the side chain to three carbon atoms leads to a further improvement as a melatonin agonist, as expected from previous studies. Again, the 4-fluoro-5-methoxy compounds (8d–f) are better agonists than the 2-fluoro-5-methoxy analogues (8a–c), with the 3-fluoro-4-methoxy compounds (9a–c) considerably poorer agonists. It is also notable that the position of the fluorine atom at 2 has little or no effect on the agonist activity of the non-methyl substituted analogues 4a–f and 8a–f, whereas it has a major effect on the methyl substituted analogues 6a–f. Compound 10, with an alpha methyl group, has a similar agonist activity to the corresponding non-methylated derivative 4f.

Table 1. Agonist and antagonist activity in the Xenopus melanophore assay of the fluorine substituted phenylalkyl amides.

Compound	Agonist potency pEC50 ± SEM a	Agonist efficacy IA ± SEM a	Antagonist activity (% inhibition)
Melatonin	9.91 ± 0.03	100
Luzindole41	—	—	100%
2a	5.71 ± 0.02	100 ± 1	NA b
2b	5.62 ± 0.15	69 ± 1	23%
2c	5.87 ± 0.04	43 ± 1	44%
2d	5.82 ± 0.05	77 ± 1	NA
2e	5.98 ± 0.01	51 ± 1	24%
2f	6.18 ± 0.06	57 ± 1	37%
3a	5.68 ± 0.02	83 ± 1	NA
3b	5.54 ± 0.01	86 ± 1	NA
3c	5.60 ± 0.08	75 ± 1	NA
4a	5.61 ± 0.24	100 ± 1	NA
4b	6.10 ± 0.28	100 ± 1	NA
4c	6.60 ± 0.15	100 ± 1	NA
4d	6.56 ± 0.16	100 ± 1	NA
4e	7.07 ± 0.12	100 ± 1	NA
4f	7.18 ± 0.03	100 ± 1	NA
5a	5.62 ± 0.14	80 ± 1	NA
5b	5.87 ± 0.08	80 ± 1	NA
5c	5.92 ± 0.11	52 ± 1	48%
6a	6.74 ± 0.12	100 ± 1	NA
6b	6.57 ± 0.10	100 ± 1	NA
6c	7.01 ± 0.20	100 ± 1	NA
6d	8.74 ± 0.01	100 ± 1	NA
6e	9.27 ± 0.02	100 ± 1	NA
6f	10.64 ± 0.02	100 ± 1	NA
7a	7.09 ± 0.04	100 ± 1	NA
7b	7.03 ± 0.01	100 ± 1	NA
7c	7.10 ± 0.01	100 ± 1	NA
8a	7.69 ± 0.07	100 ± 1	NA
8b	8.43 ± 0.02	100 ± 1	NA
8c	8.02 ± 0.04	100 ± 1	NA
8d	8.41 ± 0.01	92 ± 1	NA
8e	8.96 ± 0.02	89 ± 1	NA
8f	9.05 ± 0.02	89 ± 1	NA
9a	5.74 ± 0.02	79 ± 1	NA
9b	6.06 ± 0.01	78 ± 1	8%
9c	6.38 ± 0.03	73 ± 1	19%
10	6.98 ± 0.04	89 ± 1	NA

^aConcentration–response curves were analyzed by non-linear regression. The agonist potency (pEC₅₀) is the mean ± SEM of triplicate determinations of log EC₅₀, while the agonist efficacy (IA ± SEM) was expressed as a percentage of the maximal pigment aggregation response observed with melatonin (= 100%). Data are mean of at least 3 independent experiments.

^bNA indicates no antagonist activity up to 10^–4 M, or the percentage inhibition of aggregation induced by melatonin (10^–9 M). Data from triplicate assays.

Conclusions

The most potent sets of compounds are 6d–f and 8d–f with the fluorine atom in position 4. The compounds 6d–f, with the dimethyl group on the side chain, presumably have a higher probability to fit into the melatonin receptor than compounds 8d–f with a linear side chain, although the length of the side chain in the latter compounds is probably more optimal. It will be of interest to examine the lifetime affinity of 6d–f compared to melatonin to determine whether the fluorine atom slows down the metabolism of the compounds.

Experimental section

General procedure for the preparation of arylacetonitriles from aldehydes

A solution of p-toluenesulfonylmethyl isocyanide (TosMIC) (2.04 g, 10.4 mmol) in dimethoxyethane (DME) (11 mL) was added dropwise to a suspension of potassium tert-butoxide (2.35 g, 21.2 mmol) in DME (11 mL) at –30 °C. After the addition, the mixture was cooled to –60 °C and a solution of the appropriate aldehyde (9.74 mmol) in DME (22 mL) was cautiously added. The reaction was stirred at this temperature for 1.5 h and methanol (33 mL) was added. The mixture was allowed to thaw and then refluxed for 0.5 h. The solvent was removed in vacuo and H₂O (33 mL) and then acetic acid (2 mL) was added. The resulting suspension was extracted with dichloromethane (3 × 50 mL), and the combined organic extracts were washed with saturated aqueous NaHCO₃, brine and then dried (Na₂SO₄). The solvent was evaporated under reduced pressure and the residue was purified by column chromatography (cyclohexane/ethyl acetate) to give the desired compound as an off-yellow solid.

General procedure for the preparation of arylacrylonitriles from aldehydes

1,8-Diazabicyclo[5.4.0]undec-7-ene (DBU) (746 mg, 4.91 mmol) and cyanomethyltriphenylphosphine chloride (1.54 mg, 4.55 mmol) were added to a stirred solution of the appropriate aldehyde (3.25 mmol) in toluene (6.5 mL) under argon. The resulting suspension was stirred under reflux for 1 h, cooled to room temperature and the solvent was removed in vacuo. The residue was dissolved in CH₂Cl₂ (30 mL), washed with saturated aqueous NH₄Cl, dried (Na₂SO₄) and concentrated under reduced pressure to give a brownish oil, which was purified by flash column chromatography (cyclohexane/AcOEt). The desired arylacrylonitrile (yellowish light oil) was obtained as a 1 : 1 cis/trans mixture.

General procedure for reductive acylation

A solution of phenylacetonitrile or phenylacrylonitrile (1.03 mmol) and the appropriate acid anhydride (10.3 mmol in the case of phenylacetonitriles or 16.9 mmol in the case of phenylacrylonitriles) in anhydrous THF (20 mL) was stirred under hydrogen (4 atm) at 50 °C in the presence of RANEY®-Ni for 9–14 h. The reaction mixture was then diluted with AcOEt (15 mL), filtered through Celite and the filtrate concentrated in vacuo to give the crude product, which was purified by flash column chromatography (cyclohexane/AcOEt) followed by trituration with AcOEt.

General procedure for the preparation of 1° amines from aldehydes

Sodium acetate (23.3 mmol) was added to a mixture of the appropriate aldehyde (7.73 mmol) and hydroxylamine hydrochloride (15.5 mmol) in ethanol (15 mL) and water (15 mL). The resulting suspension was heated at 70 °C for 1.5 h and the solvents were then removed under vacuum. The residue was treated with a small quantity of water and extracted with ether (3 × 25 mL). The organic layers were dried (Na₂SO₄) and the solvent was removed in vacuo to give the desired oxime in crystalline form as a 9 : 1 anti/syn-mixture. A solution of the appropriate oxime (6.98 mmol) in anhydrous methanol (20 mL) and AcOEt (20 mL) was hydrogenated in the presence of 10% Pd/C (728 mg) at ambient temperature for 6 h under a pressure of 4 atm. The reaction mixture was then diluted with AcOEt (15 mL) and filtered through Celite and the filtrate was concentrated in vacuo to give the desired primary amine, which was then used without further purification.

General procedure for the preparation of amides

A solution of corresponding amine (2.32 mmol) in dry dichloromethane (10 mL) was treated with triethylamine (0.5 mL) at 0 °C. The appropriate anhydride (3.48 mmol) was then added dropwise at the same temperature, and the resulting reaction mixture was left stirring at room temperature for 1.5 h. The solution was then poured into a separatory funnel, CH₂Cl₂ was added, and the yellow organic layer was washed with H₂O (20 mL). After drying over Na₂SO₄, the solvent was evaporated under reduced pressure to give the crude product, which was chromatographed (flash column) to give the desired amide.

Conflicts of interest

There are no conflicts to declare.