Morpholine-based chalcones as dual-acting monoamine oxidase-B and acetylcholinesterase inhibitors: synthesis and biochemical investigations

Abstract

Nine compounds (MO1–MO9) containing the morpholine moiety were assessed for their inhibitory activities against monoamine oxidases (MAOs) and acetylcholinesterase (AChE). Most of the compounds potently inhibited MAO-B; MO1 most potently inhibited with an IC₅₀ value of 0.030 µM, followed by MO7 (0.25 µM). MO5 most potently inhibited AChE (IC₅₀ = 6.1 µM), followed by MO9 (IC₅₀ = 12.01 µM) and MO7 most potently inhibited MAO-A (IC₅₀ = 7.1 µM). MO1 was a reversible mixed-type inhibitor of MAO-B (K_i = 0.018 µM); MO5 reversibly competitively inhibited AChE (K_i = 2.52 µM); and MO9 reversibly noncompetitively inhibited AChE (K_i = 7.04 µM). MO1, MO5 and MO9 crossed the blood–brain barrier, and were non-toxic to normal VERO cells. These results show that MO1 is a selective inhibitor of MAO-B and that MO5 is a dual-acting inhibitor of AChE and MAO-B, and that both should be considered candidates for the treatment of Alzheimer’s disease.

1. Introduction

Due to the extreme complexities of brain systems and their diverse dysfunctions, research focus is being directed towards the design of multi-target directed ligands (MTDLs)¹. In fact, it has been suggested that functional equilibrium of brain after a neurologic disorder is unlikely to be achieved by focussing on a single molecular target². The major challenge presented by the development of MTDLs is to preserve balance between the effects drugs have by acting at their individual molecular targets³. On the other hand, some ligands can target two or more specific entities within numerous biological networks. The likelihood of successful MTDL design can be enhanced by considering the design of molecular scaffolds via suitable molecular hybridisation and by understanding the pathophysiologies of multifaceted diseases. The design process can be accelerated by selecting pharmacophores based on pre-clinical studies, emphasising structure–activity relationships (SARs) and by performing in silico-based virtual screening⁴.

Monoamine oxidases (MAOs) play prominent roles in the inactivations of various biogenic amines in central and peripheral tissues. The isoenzymes of MAO-A and MAO-B are considered major therapeutic targets in various neuropsychiatric illnesses and neurodegenerative disorders like Alzheimer’s disease (AD) and Parkinson’s disease (PD)⁵. During the degradations of various biogenic neurotransmitters catalysed by MAO-A and MAO-B, hydrogen peroxide and reactive oxygen species (ROS) are produced as major by-products, which might cause oxidative damage in brain tissues. MAO inhibitors are considered potential neuroprotective agents and up-regulating agents for neurotransmitter amines⁶. Numerous studies have documented that selective and reversible/irreversible MAO-B inhibitors are likely to play pivotal roles in AD-related therapeutic strategies. Acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) inhibitors also play significant roles in the maintenance of cholinergic functions and are used to provide symptomatic relief in AD. Over the past two decades, many researchers have tried to develop multi-acting MAO-B and AChE/BChE inhibitors^7–10.

Morpholine is a tetrahydro-1,4-oxazine with a saturated heterocyclic ring and provides a promising developmental starting point due to its biological profile with metabolic stability. The presence of a heteroatom like oxygen or nitrogen facilitates hydrogen bonding, and thus, interactions with enzymes, and the presence of electron-deficient atoms may also impart hydrophobic interactions with morpholine. From the synthetic perspective, various molecular scaffolds have been added to morpholine by replacing its secondary nitrogen¹¹. Moclobemide (1) and reboxetine (2) (both antidepressants) provide examples of FDA-approved drugs containing the morpholine moiety (Figure 1). These drugs reversibly inhibit MAO-A and selectively inhibit norepinephrine reuptake in the central nervous system (CNS), and thus, block the human a4b2 nicotinic acetylcholine receptor. In addition, more than 20 drugs containing the morpholine moiety have been FDA approved; they include mycophenolate mofetil (an immunosuppressant), linezolid and finafloxacin (antibiotics), geftinib (an antineoplastic and epidermal growth factor inhibitor), rivaroxaban (an anticoagulant and factor Xa inhibitor), and eteplirsen, which is used to treat Duchenne muscular dystrophy¹². Considering the importance of morpholine nucleus, it is worthwhile to design morpholine derived compounds of medicinal chemistry interest.

Figure 1.

Morpholine-containing, CNS active, FDA approved drugs. 1, moclobemide; 2, reboxetine.

The chalcones framework is considered an excellent starting point for the design for MAO-B and AChE enzyme inhibitors^13–15. The presence of an α-β unsaturated ketone and three rotatable bonds in chalcones can provide different bonding orientations¹⁶. The literature amply demonstrates that most chalcones are potent, reversible and selective MAO-B inhibitors¹⁷. In addition, the presence and orientation of various electron-donating and withdrawing groups on the phenyl/heteroaryl A and B rings of chalcones can impart electrophilic character to the Michael acceptor present. The introductions of electron-donating/lipophilic halogens onto the phenyl B ring of chalcones have resulted in the syntheses of highly selective MAO-B inhibitors^18–27. Many studies have reported that the presence of various alkylamino groups on the A ring provides AChE inhibitory activity^28–31. The FDA-approved drugs like flurbiprofen and rivastigmine linked with chalcone moiety were developed as selective ChE/MAO-B inhibitors for the prophylactic agents for AD³². The design strategy of the current study is depicted in Figure 2 and the synthetic route is shown in Scheme 1. A recent study showed that the presence of a pendant morpholine ring on the chalcone A ring favoured hMAO-B inhibitory activity^33,34. The difference in MAO-B inhibition profiles became more obvious when chalcone B ring was substituted using a fluoro or trifluoromethyl group³³.

Figure 2.

MTDL design of MAO-B and AChE inhibitors.

Scheme 1.

The synthetic route used to produce the target compounds (MO1–MO9).

2. Results and discussion

2.1. Chemistry

Morpholine-containing α,β-unsaturated ketones were prepared by condensation between various aromatic/para substituted aldehydes and 4-morpholine acetophenone in the presence of an alcoholic basic medium³³. The H¹ NMR showed that the ring protons N-CH₂ and O-CH₂ of morpholine ring resonated at 3.32–3.31 and 3.75–3.88 ppm as triplets, respectively. The large coupling constant (J = 15 Hz) value of Hα and Hβ protons confirmed the trans configuration of the morpholine chalcones. The deshielded values between 186.88–186.07 in C¹³-NMR clearly evidenced the formation of sp² carbonyl functional group of α,β-unsaturated ketones. The identities of the compounds were determined by ¹H-NMR, ¹³C-NMR and Mass spectrometry (Supplementary materials).

2.2. Analysis of enzyme inhibitory activities

2.2.1. Inhibitory activities

In this study, nine morpholine-based compounds (MO1–MO9) were analysed with respect to their abilities to inhibit MAO-A, MAO-B and AChE. Most of the compounds inhibited MAO-B by ∼ 50% at 1.0 µM, except MO3 and MO5 (Table 1). Compound MO1 most potently inhibited MAO-B (IC₅₀ = 0.030 µM), followed by MO7, MO8, MO4, MO9, MO6, MO2, MO3 and MO5 (IC₅₀ = 0.25, 0.32, 0.33, 0.36, 0.64, 0.70, 1.01 and 1.31 µM, respectively). In addition, MO5 most potently inhibited AChE (IC₅₀ = 6.1 µM), followed by MO9 and MO8 (IC₅₀ = 12.01 and 12.07 µM, respectively). MO7 most potently inhibited MAO-A (IC₅₀ = 7.1 µM), followed by MO6 (IC₅₀ = 8.7 µM). MO5 most potently inhibited BChE (IC₅₀ = 18.09 µM), with weaker inhibitory activity compared to AChE, followed by MO7 (IC₅₀ = 24.83 µM) (Table 1).

Table 1.

Inhibitions of MAO-A, MAO-B, and AChE by MO1–MO9^a

Compounds	Residual activities (%)				IC50 (µM)				SIb
	MAO-A (10 µM)	MAO-B (1.0 µM)	AChE (10 µM)	BChE (10 µM)	MAO-A	MAO-B	AChE	BChE
MO1	94.2 ± 0.62	38.1 ± 9.50	53.4 ± 0.99	62.85 ± 2.97	>40	0.030 ± 0.062	16.1 ± 2.24	>40	>1333.3
MO2	93.8 ± 1.34	54.5 ± 5.28	72.3 ± 1.34	63.55 ± 3.30	>40	0.70 ± 0.23	30.2 ± 3.24	>40	> 57.1
MO3	91.0 ± 3.17	79.9 ± 5.28	79.3 ± 2.82	56.78 ± 2.97	>40	1.01 ± 0.08	>40	>40	>39.6
MO4	70.2 ± 1.24	49.0 ± 1.45	66.0 ± 8.79	80.30 ± 6.43	25.8 ± 1.58	0.33 ± 0.03	28.42 ± 0.02	>40	78.2
MO5	88.1 ± 3.43	58.6 ± 3.39	34.5 ± 9.51	60.86 ± 1.07	>40	1.31 ± 0.26	6.1 ± 0.0048	18.09 ± 0.38	>30.5
MO6	12.3 ± 1.58	64.1 ± 1.37	76.9 ± 3.54	70.96 ± 0.36	8.7 ± 1.32	0.64 ± 0.04	>40	>40	13.6
MO7	−16.5 ± 4.53	26.1 ± 5.44	65.3 ± 1.55	70.21 ± 4.76	7.1 ± 0.41	0.25 ± 0.05	20.48 ± 1.10	24.83 ± 0.34	28.4
MO8	83.7 ± 4.02	34.6 ± 5.44	50.6 ± 3.61	74.61 ± 0.73	>40	0.32 ± 0.16	12.07 ± 1.18	>40	>125.0
MO9	79.8 ± 1.54	51.3 ± 2.42	50.4 ± 2.42	70.21 ± 5.50	>40	0.36 ± 0.16	12.01 ± 2.13	>40	>111.1
Toloxatone					1.08 ± 0.025	—
Lazabemide					—	0.063 ± 0.015
Clorgyline					0.007 ± 0.00070	—
Pargyline					—	0.028 ± 0.0043
Tacrine							0.27 ± 0.019	0.060 ± 0.0022

^a Values above are the means ± SEs of duplicate or triplicate experiments.

^b SI = IC₅₀ of MAO-A/IC₅₀ of MAO-B.

AChE values ​were determined after pre-incubation compounds and enzymes for 15 min.

Morpholine-containing compounds in this study shared a 1–(4-morpholinophenyl) prop-2-en-1-one structure. Introduction of various electron-donating and withdrawing groups onto the para position of the phenyl B ring of the basic chalcone scaffold afforded different derivatives. Surprisingly, unsubstituted MO1 inhibited MAO-B more potently than the other eight derivatives. All nine derivatives exhibited MAO-B selectivity. In particular, MO1 (IC₅₀ = 0.030 µM) was ∼2 times more potent than the reversible MAO-B inhibitor lazabemide (IC₅₀ = 0.063 µM), and had a high selectivity index (SI = >1333.3), which meant that it was the best inhibitor in the series. Regarding MAO-B inhibition, MO1 was slightly less potent than the reference irreversible MAO-B inhibitor pargyline (IC₅₀ = 0.028 µM). The presence of dimethylamino, chloro and bromo substituents at the para position of the chalcone B ring conferred moderate AChE inhibition. Recently, it was reported that shifting of morpholine ring to the phenyl B ring of chalcones showed moderate AChE inhibition³⁵. Results of SAR analysis of MAO-B/AChE inhibitions by the nine compounds are provided in Figure 3.

Figure 3.

SAR analysis of MAO-B and AChE inhibitors.

2.2.2. Kinetics of MAO-B inhibition

Lineweaver–Burk plots and secondary plots of MAO-B inhibition by MO1 showed that MO1 is a mixed inhibitor of MAO-B (K_i = 0.018 ± 0.002) (Figure 4(A,B)). These results suggested that MO1 binds to an allosteric site other than the substrate-binding site of MAO-B.

Figure 4.

Lineweaver–Burk plots of MAO-B inhibition by MO1 (A), and respective secondary plots (B) of slopes vs. inhibitor concentrations.

2.2.3. Kinetics of AChE inhibitions

Lineweaver–Burk plots and secondary plots showed that MO5 and MO9 competitively and non-competitively, respectively, inhibited AChE (Figure 5(A,C)) with K_i values of 2.52 ± 0.17 and 7.04 ± 0.32 µM, respectively (Figure 5(B,D)), which suggest that MO5 is a potent, selective and competitive inhibitor of AChE that binds to the active site of AChE, and that MO9 binds to a site other than the active site of AChE and changes the 3D structure of the enzyme, thus inhibiting enzyme activity.

Figure 5.

Lineweaver–Burk plots of AChE inhibition by MO5 (A) and MO9 (C), and respective secondary plots (B and D) of slopes vs. inhibitor concentrations.

2.2.4. Analysis of the reversibilities of MAO-B and AChE inhibitions

Inhibition assays were carried out after preincubating MAO-B or AChE with the inhibitors for 15 min. The reversibilities of MAO-B inhibition by MO1 and of AChE inhibitions by MO5 and MO9 were investigated by dialysis³⁴. Dialysis recovered the inhibition of MAO-B by MO1 from 21.6% (A_U) to 77.1% (A_D), which was similar to that shown by lazabemide (from 38.8 to 90.0%), a reversible MAO-B inhibitor (Figure 6(A)). On the other hand, inhibition of MAO-B by pargyline (an irreversible inhibitor) was recovered from 10.7 to 16.5%. In addition, inhibitions of AChE by MO5 and MO9 were recovered by dialysis from 35.8% (A_U) to 79.2% (A_D) and from 33.8 to 80.7%, respectively, which were similar to that observed for tacrine (from 32.8 to 91.7%), a reversible AChE inhibitor (Figure 6(B)). These results indicate that MO1 is a reversible inhibitor of MAO-B, and that MO5 and MO9 are reversible inhibitors of AChE.

Figure 6.

Dialysis recoveries of MAO-B inhibition by MO1 (A) and AChE inhibition by MO5 and MO9 (B). The concentrations of the inhibitors used were approximately 2 × IC₅₀ values: MO1, 0.06 µM; MO5, 12.0 µM; MO9, 24.0 µM; lazabemide, 0.12 µM; pargyline, 0.06 µM; and tacrine, 0.54 µM (tacrine was used as a reference reversible inhibitor). For recovery experiments, preincubated enzyme mixtures were dialysed as described in the text.

2.3. Blood–brain barrier (BBB) permeation

The synthesised derivatives were screened for their abilities to cross the BBB using the parallel artificial membrane permeability assay (PAMPA) because this ability is a critical developmental requirement for any drug targeting neurodegenerative disorders or depression. The PAMPA used was validated by comparing the experimentally determined permeabilities of eight commercial drugs with reported values (Table 2). According to reported BBB permeation limits, compounds were classified as follows:³⁶ CNS+ (high): Pe (10⁻⁶cm s⁻¹) > 4.00, CNS- (low): Pe (10⁻⁶cm s⁻¹) <2.00 and CNS± (uncertain): Pe (10⁻⁶cm s⁻¹) from 4.00 to 2.00. Our results suggest that all nine synthesised derivatives could cross the BBB, and that MO1, MO5 and MO9 would be the most effective.

Table 2.

Experimental membrane permeabilities of the synthetic and reference inhibitors.

Compounds	Bibliography Pe (×10−6 cm s−1)	Experimental Pe (×10−6 cm s−1)	Prediction
Testosterone	17.0	16.86 ± 0.67	CNS+
Verapamil	16.0	15.69 ± 0.44	CNS+
β-estradiol	12.0	11.88 ± 0.53	CNS+
Progesterone	9.3	9.10 ± 0.11	CNS+
Piroxicam	2.5	2.35 ± 0.30	CNS+/-
Hydrocortisone	1.8	1.71 ± 0.05	CNS-
Lomefloxacin	1.1	1.26 ± 0.01	CNS-
Dopamine	0.2	0.21 ± 0.01	CNS-
MO1	—	16.34 ± 0.16	CNS+
MO5	—	14.44 ± 0.81	CNS+
MO9	—	14.61 ± 0.08	CNS+

2.4. Cytotoxic studies

The biocompatibilities of MO1, MO5 and MO9 were investigated using an MTT (3–(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay and normal VERO cells³⁷,³⁸. Results showed that MO1, MO5 and MO9 were non-toxic to normal VERO cells with IC₅₀ values of 195.14, 185.44 and 188.34 µg/mL, respectively, indicating that the therapeutic potentials of these compounds would be suitable.

2.5. Total ROSs assay

ROS are considered to be the radicals primarily responsible for the neuronal death observed in various neurodegenerative disorders³⁹. The lead compounds MO1, MO5 and MO9 were subjected to ROS assays using Hela cells. ROS levels significantly increased in H₂O₂-treated Hela cells as determined by fluorescent image analyses (Figure 7(B)). MO1 and MO5 (Figure 7(C,D)) at 40 µg/mL resulted in significantly lower ROS levels than MO9 (Figure 7(E)). Interestingly, the colony formation induced by H₂O₂ was markedly reduced by the lead MAO-B inhibitor MO1, which suggested that the molecule can be highly recommended for the arrest of free radicals in the neurodegenerative disorders.

Figure 7.

Inhibition of H₂O₂-induced ROS generation in human cervical cancer cells. (A) Treatment-naïve control cells; (B) Cells treated with 100 μg/mL of H₂O₂; (C) Cells treated with 40 μg/mL of MO1; (D) Cells treated with 40 μg/mL of MO5; (E) Cells- treated with 40 μg/mL of MO9.

2.6. Computational studies

Initially, MO1 and MO5 were used as bases for profiling the pharmacological spectra of putative protein targets using the free web platform MuSSel^40,41. Interestingly, MAO-B and AChE were predicted to be 13th and 14th targets for MO1 and 7th and 15th targets for MO5, respectively. Details are provided in Supporting Information.

Docking studies were performed on MO1 and MO5 using the X-ray resolved structures of MAO-A, MAO-B and AChE in the Protein Data Bank (PDB) as entries 2Z5X, 2VZ5 and 4EY7^42–45. Docking protocols were performed as we described in a previous study¹³. Molecular docking simulations were performed to investigate interactions between MO1 and MO5 and MAO-A, MAO-B and AChE. The docking scores resulting from in silico simulations are reported in Table 3. Results showed that the carbonyl and the 4-morpholin-phenyl groups of the chalcone scaffolds of MO1 and MO5 hydrogen-bonded with F295 and formed π − π interactions with the side chain of W286 at the peripheral anionic subsite (PAS) of AChE (Figure 8(A)). Also, the dimethylamino group of MO5 established a weak hydrophobic interaction with W86 at the catalytic anionic subsite (CAS). MO1 and MO5 formed π − π interactions with the selective MAO-B residue Y326 (Figure 8(B)). In addition, the styryl and dimethylaminophenyl groups of MO1 and MO5, respectively, established π–π interactions with Y398 and were sandwiched within an aromatic cage formed by Y328, Y435 and FAD. MO1 engaged in a π − π interaction with Y407 and was entangled in a hydrophobic cage consisting of Y407, Y444 and FAD of MAO-A (Figure 8(C)), whereas MO5 established π–π interactions with a selective MAO-A residue F208 using its 4-morpholine-phenyl group. In order to point out the selectivity of MO1 and MO5, molecular docking analyses were performed on BChE crystal structure. Docking score values of MO1 and MO5 for BChE were worse than those for AChE (Table 3). Interestingly, MO1 and MO5 can predominantly establish hydrophobic interactions in the BChE binding pocket; in particular MO1 engaged a π–π interaction with F329, but both compounds cannot make crucial interactions with the catalytic W82 residue (Figure 8(D)).

Table 3.

Docking score values for bindings between MO1 and MO5 and MAO-B or AChE.

Compounds	Docking score (kcal/mol)
	AChE	MAO-B	BChE
MO1	−10.31	−8.67	−7.42
MO5	−10.12	−9.34	−6.66

Figure 8.

Top scored poses of MO1 and MO5 for binding with AChE (A), MAO-B (B) and MAO-A (C) are shown on left and right sides, respectively. MO1, MO5 and key enzyme residues are displayed as green, yellow and grey sticks, respectively. Green and red arrows indicate π–π interactions and hydrogen bonds, respectively.

Overall, computational analyses provided a sound explanation of the experimental data. Docking scores and the binding modes obtained demonstrated that MO1 and MO5 can meaningfully interact with MAO-B as indicated by IC₅₀ experimental values. Interestingly, even if MO1 had adopted a similar binding mode in MAO-A and MAO-B binding pockets, the two docking score values would have been quite different, due to a lack of π–π interaction with the MAO-A selective residue F208 and lesser hydrophobic interactions. On the other hand, MO5 assumed a switched pose within the MAO-A binding pocket, with its morpholine ring facing FAD, presumably because the dimethylamino substituent on the phenyl ring did not allow adequate access to the binding site. As regards AChE, molecular docking awarded a better score for MO5 than MO1, which also agreed with experimental data, despite the resemblance between their binding poses.

3. Conclusion

We describe the synthesis of morpholine-containing α,β-unsaturated ketones, and the results of an investigation of their MAOs and AChE inhibition profiles. Most of the nine compounds synthesised exhibited potent MAO-B inhibition with moderate AChE inhibition. Interestingly, MO1 (the lead compound) inhibited MAO-B inhibition in the low nanomolar range and was more potent than lazabemide (the reference compound). The low cytotoxicity and the ability of MO1 to transit the BBB support our drug design strategy. In addition, the ROS scavenging efficacy of MO1 suggests improved neuroprotective effects.

4. Experimental

4.1. Enzyme assays

MAO-A activity was measured continuously for 20 min at 316 nm using 0.06 mM kynuramine as substrate, as described previously^46,47, whereas MAO-B activity was measured for 30 min at 250 nm using 0.3 mM benzylamine as substrate. MAO activity assays were performed using recombinant human MAO-A or MAO-B. A slightly modified version of the method developed by Ellman⁴⁸ was used to measure AChE and BChE activities, using 0.5 mM acetylthiocholine iodide (ATCI) and S-butyrylthiocholine iodide (BTCI), respectively, as substrates for 10 min at 412 nm⁴⁹. Colour development was performed using 0.5 mM 5,5′-dithiobis (2-nitrobenzoic acid) (DTNB), which reacts with thiocholine (a product of ATCI by AChE or of BTCI by BChE) to produce 5-thio 2-nitrobenzoic acid. Preincubation was treated for 15 min before adding ATCI and DTNB.

4.2. Analysis of enzyme inhibitions and kinetics

MAO-A, MAO-B, AChE and BChE activities were measured after exposure to inhibitors at a concentration of 10 µM. Inhibitions of MAO-B at 10 µM of these compounds tested were too excessive, and thus, a concentration of 1.0 µM was used. IC₅₀ values were determined by measuring the residual enzyme activities. Toloxatone, lazabemide and tacrine, were used as reference reversible inhibitors of MAO-A, MAO-B and AChE, respectively, and clorgyline and pargyline as reference irreversible inhibitors of MAO-A and MAO-B, respectively. K_i values and inhibitor types were determined by kinetic testing, as previously described⁵⁰. Kinetic tests were conducted at 5 different substrate concentrations, and the inhibitor concentrations used were 0, ∼1/2 × IC₅₀, IC₅₀, and 2 × IC₅₀ values. Lineweaver–Burk plots and their secondary plots were used to determine K_i values and inhibitor types.

4.3. Analysis of inhibitor reversibilities

Inhibition types (reversible or irreversible) were determined by dialysis, as previously described^51,52, but by using 0.06 mM kynuramine and 0.3 mM benzylamine. For the experiment of AChE, 0.5 mM ATCI was used. Dialysis experiments were performed by preincubating enzymes and inhibitors (or reference inhibitors) at ∼2 × IC₅₀ in 0.1 M sodium phosphate buffer (pH 7.2) for 30 min, and then dialysing solutions with stirring for 6 h with a buffer change at 3 h. Residual activities were calculated using the measured activities of undialyzed (A_U) and dialysed (A_D) solutions and the activities of untreated controls (i.e. without inhibitor).

Funding Statement

This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Republic of Korea government (Grant No. NRF-2019R1A2C1088967 to H. Kim).

Disclosure statement

No potential conflict of interest was reported by the authors.