SAR and molecular mechanism studies of monoamine oxidase inhibition by selected chalcone analogs

Abstract The present study describes the synthesis of a series of 22 chalcone analogs. These compounds were evaluated as potential human MAO-A and MAO-B inhibitors. The compounds showed varied selectivity against the two isoforms. The IC50 values were found to be in the micromolar to submicromolar range. The Ki values of compound 16 were determined to be 0.047 and 0.020 μM for the inhibition of MAO-A and MAO-B, respectively. Dialysis of enzyme-inhibitor mixtures indicated a reversible competitive mode of inhibition. Most of the synthesized chalcone analogs showed a better selectivity toward MAO-B. However, introducing of 2,4,6-trimethoxy substituents on ring B shifted the selectivity toward MAO-A. In addition, we investigated the molecular mechanism of MAO-B inhibition by selected chalcone analogs. Our results revealed that these selected chalcone analogs increased dopamine levels in the rat hepatoma (H4IIE) cells and decreased the relative mRNA expression of the MAO-B enzyme.


Introduction
Neuronal signaling is mediated by several neurotransmitters, including monoamine neurotransmitters such as norepinephrine, dopamine, and serotonin. The levels of monoamines in the brain are controlled by a balance between their enzyme-catalyzed synthesis and breakdown, while both reuptake carriers and vesicular transporters regulate the concentrations of these neurotransmitters in the synapse. Alterations of the monoamines homeostasis are associated with many neurological conditions such as depression, autism, drug addiction, and neurodegenerative diseases [Alzheimer's disease (AD) and Parkinson's disease (PD)] 1 . Monoamine oxidases (MAOs, EC 1.4.3.4) are among a family of flavin adenine dinucleotide (FAD)-dependent enzymes that play a key role in the breakdown of endogenous and exogenous amines 2 . These enzymes are located on the mitochondrial outer membrane, mostly in the brain, and are also present in the liver, gut, intestine, skin, placenta, lymphocytes, and platelets 3 .
MAOs are responsible for catalyzing the oxidative deamination of various biogenic amine neurotransmitters and a variety of xenobiotic amines, and modulate their concentrations in the brain and peripheral tissues 4 . The MAO enzymes exist as two isoforms, MAO-A and MAO-B, with a sequence similarity of 73% but with different inhibitor selectivity, substrate specificity, and tissue distribution 5 . The MAO-A isoform predominantly deaminates serotonin, norepinephrine, and epinephrine, whereas MAO-B has substrate specificity for benzylamine and b-phenylethylamine. Tyramine and dopamine are common substrates for both isoforms 6 . The amine in the deprotonated form binds to the active site of the enzyme and is oxidized to the corresponding imine with the reduction of the FAD cofactor to its hydroquinone form.
The reduced FAD cofactor reacts with oxygen (O 2 ) to regenerate FAD to its oxidized form. During this process, hydrogen peroxide is produced, which may generate highly reactive hydroxyl radicals that may lead to the development of neurodegenerative processes and neuronal death 7,8 . As mentioned, the imbalance in the brain neurotransmitters concentration is linked with various neurodegenerative diseases and psychiatric disorders, including depression, PD, AD, and Huntington's disease 9,10 . Several studies have confirmed that MAO-B is overexpressed in the brains of AD patients and is attributed to the loss of cognitive functions 11 . Other reports revealed positive correlation between the low level of neurotransmitters in patients with severe depression and the increased concentrations of MAO-A.
MAO inhibitors block the enzyme catalytic activities and halt the oxidative deamination of these neurotransmitters. Inhibition of MAOs can lead to increase in the concentrations of neurotransmitters stored in the nerve terminals (e.g. serotonin and dopamine). Thus, MAO inhibitors can be developed as therapeutic agents, particularly for disease states where MAO enzyme is overexpressed. As aforementioned, MAO catalyzes metabolism of monoamine neurotransmitters, which may generate byproducts potentially injurious species (e.g. hydrogen peroxide and aldehydes), and lead to neuronal damage. MAO inhibitors can reduce the production of the reactive oxygen species (ROS) by halting the MAO-catalyzed oxidation process and hence prevent their neurotoxic effect. This is in agreement with a prior study demonstrating that MAO inhibitors are potential neuroprotective agents 12 .
The X-ray crystal structures of human MAO-A and MAO-B revealed that human MAO-A has a single substrate cavity with a volume of $500Å 3 , while human MAO-B has a bipartite cavity structure with an entrance cavity ($300Å 3 ) and a substrate cavity ($400Å 3 ). These two cavities are separated by the side chains of Ile199 and Tyr326 13 . Mutagenesis experiments with the human MAO-B mutant, Ile199Phe, indicated that the bulky Phe side chain reduces conformational flexibility of this residue in MAO-B, and as a result impedes the binding of larger cavity-spanning MAO-B selective inhibitors to both the active and entrance cavities 14 . The smaller side chain of the Ile199 residue, in turn, may rotate from the active site cavity to permit the binding of cavity-spanning inhibitors. Thus, Ile199 serves as gating residue and a structural determinant for substrate and inhibitor recognition by MAO-B 15 . A single Ile199Ala mutation to MAO-B permanently opens the "gate." Tyr326, another key residue that determines substrate and inhibitor specificity, also exhibits conformational changes on the inhibitor binding and restricts the binding of certain inhibitors (e.g. harmine) to human MAO-B. In MAO-A, the corresponding residue is Ile335, which is much smaller and this allows for the binding of inhibitors such as harmine. The double mutations Ile199Ala-Tyr326Ala in MAO-B exhibit inhibitor binding properties similar to those of MAO-A rather than to MAO-B.
This analysis indicates the importance of the bipartite cavity structure of MAO-B to substrate and inhibitor recognition, and in distinguishing its specificities from those of MAO-A. It has been reported that Ile199 and Tyr326 can act as gating residues in human MAO-B. With long inhibitors, the side chain of Ile 199 rotates into an open conformation leading to the merging of the two cavities to form one large cavity of about 700Å 3 to accommodate the long inhibitor. This information would be helpful in designing selective MAO-A and MAO-B inhibitors ( Figure 1).
Most of the marketed selective MAO-B inhibitors are irreversible and can cause potential adverse side effects due to long-lasting enzyme inhibition. However, safinamide was recently approved as the first reversible selective MAO-B inhibitor for the treatment of PD. There is an increase interest in chalcone as a new promising scaffold for the development of novel reversible MAO-B inhibitors. In our previous work, we explored the effect of introducing a benzodioxol ring in the chalcone molecule in combination with different substituents 13 . Computer-aided docking studies confirmed the role of the carbonyl group in forming a hydrogen bond with the SH group of Cys172. Based on our previous findings, we designed and prepared a series of chalcone analogs that showed different MAO inhibitory activities and selectivities. In addition, the molecular mechanism of MAO inhibition by chalcone analogs remains unclear with respect to their effects of protein expression. The main question to be answered is whether chalcone analogs inhibit the enzyme expression in addition to the inhibition of MAO activity.

General methods
All solvents and fine chemicals were purchased from Sigma-Aldrich (unless specified otherwise) and were used without further purification. For the MAO inhibition studies, recombinant human MAO-A and human MAO-B (5 mg protein/mL) from Sigma-Aldrich were used. Analytical thin layer chromatography was carried out on precoated silica gel 60 F254 aluminum plates purchased from Merck and visualized by UV. Melting points were measured on a Stuart SMP40 automatic melting point apparatus and are uncorrected. For all of the prepared compounds, infrared (IR) spectra were recorded with a Perkin Elmer Spotlight 400 Fourier transform infrared (FTIR) imaging system. GC-MS was performed on a Thermo instrument (ISQELTL). Proton ( 1 H) and carbon ( 13 C) NMR spectra were recorded on JEOL 600 MHz using CDCl 3 as a solvent at the Central Lab Unit, Qatar University and a Bruker Avance III spectrometer (700 MHz) at the King Abdullah Institute for Research & Consulting Studies, King Saud University, Riyadh, Saudi Arabia. The chemical shifts were reported as parts per million (d) relative to the solvent peak and coupling constants are given in hertz (Hz). The original spectra are provided in the Supporting information. Purity of compounds were confirmed with C, H and S analysis performed on Thermo Scientific FLASH 2000 CHNS/O analyzer.

General synthetic procedure
The proposed Chalcone analogs were synthesized using the Claisen-Schmidt condensation reaction. To a stirred solution of the appropriate acetophenone (1 mmol) and the substituted benzaldehyde (1 mmol) in ethanol at 0 C, a sodium hydroxide solution (0.12 g; 3 mmol in minimum amount of water) was added dropwise. The reaction mixture was stirred at room temperature overnight. The obtained precipitate was filtered, washed with cold water and recrystallized from ethanol to produce chalcones 1-22.

Enzymatic screening
Screening for the inhibition of MAO All prepared compounds were screened against both human MAO-A and MAO-B to assess their inhibitory activities. This was done spectrophotometrically using MMTP as a nonselective substrate. MMTP is oxidized by MAO to yield the corresponding MMDP þ species, which absorbs maximally at 420 nm. DMSO was used as a co-solvent at 3.3% (v/v) final concentration, which does not affect MAO activity. DMSO was included since the test compounds exhibit limited aqueous solubility.
Each assay was performed in a quartz cuvette (150 mL final volume). A mixture composed of 5 mL enzyme (25 mg of MAO-A or MAO-B), 95 mL buffer (0.1 M potassium phosphate buffer), and 5 mL test compound (10 mM final concentration) was incubated at 37 C for 5 min. The reactions were initiated by adding 45 mL MMTP at a final concentration of 0.15 mM. The increase in absorbance was monitored at 37 C and recorded at 420 nm. The enzyme initial velocity (v 0 ) was determined by using the extinction coefficient of 25000 M À1 cm À1 for the dihydropyridinium species. The inhibitory activity of the prepared compounds was determined as the ratio between the obtained v 0 value for each chalcone analog and that of the control experiment containing no inhibitor (100% activity). Control experiments were prepared by following the same procedure except for the addition of 5 mL DMSO instead of the test compounds. Compounds that showed more than 60% enzyme inhibition were subjected to IC 50 determination.

Determination of IC 50 values for MAO inhibition
IC 50 values for the inhibition of MAO were determined by using the recombinant human MAO-A and MAO-B enzymes (Sigma-Aldrich). Enzyme reactions were performed at a final volume of 200 mL in white 96-well microplates (Eppendorf), with the reaction solvent consisting of potassium phosphate buffer (100 mM, pH 7.4, adjusted to isotonicity with KCl). The reactions contained the nonselective MAO substrate, kynuramine (50 lM), and the test inhibitors at 0.003-100 mM. The test compounds were initially dissolved in DMSO (100%), and after addition to the reactions yielded a concentration of 4% DMSO. Control reactions were included, which contained substrate and 4% DMSO but no test inhibitor. The MAO reactions were started by adding MAO-A (0.0075 mg protein/mL) or MAO-B (0.015 mg protein/mL), and were subsequently incubated for a period of 20 min (37 C) in a convection oven. A volume of 80 mL 2 N sodium hydroxide was added to the reactions, and the fluorescence intensity of 4hydroxyquinoline (the end-product of the MAO-catalyzed metabolism of kynuramine) in the reactions was measured at k ex 310 and k em 400 nm. To quantify 4-hydroxyquinoline, a calibration curve (linear) ranging from 0.047 to 1.56 lM, was employed. From these data, the rates of 4-hydroxyquinoline formation were determined, sigmoidal plots of activity versus inhibitor concentration [Log (I)] were prepared, and the IC 50 values were recorded (Prism 5, GraphPad). These measurements were carried out in triplicate and the IC 50 values are presented as the mean ± standard deviation (SD) 18 .

Reversibility of MAO-B inhibition
The reversibility of inhibition of MAO by compound 16 was investigated by dialysis. Dialysis cassettes (Slide-A-Lyzer V R ; Thermo Scientific) with a volume of 0.5-3 mL and a molecular weight cutoff of 10,000 were used for the dialysis. Mixtures of MAO-A or MAO-B (0.03 mg protein/mL) and the test inhibitor (concentration ¼ 4 Â IC 50 ), were prepared in potassium phosphate buffer (100 mM, pH 7.4) with 5% sucrose added. The volume of these mixtures was 0.8 mL, and contained 4% DMSO. The mixtures were pre-incubated for 15 min at 37 C and then dialyzed at 4 C in buffer (80 mL), using the above buffer as dialysis buffer. After 3 and 7 h of dialysis, the dialysis buffer was changed with fresh buffer.  20 were used at concentrations equal to fourfold their IC 50 values. After 24 h of dialysis, samples of the dialysis mixtures (250 lL) were diluted twofold by adding kynuramine (250 lL) to yield a concentration of 2 Â IC 50 for the inhibitor and 50 mM for kynuramine. After incubation of these enzyme reactions for 20 min at 37 C, the remaining MAO activities were subsequently recorded by fluorescence spectrophotometry as described above for the IC 50 value determination. For comparison, incubations of the MAOs and the test inhibitor were prepared as above but were not dialyzed and kept at 4 C over the same time period. These experiments were carried out in triplicate, and the remaining enzyme catalytic rates are reported as the mean (± SD).

Lineweaver-Burk plots and K i value calculations
Sets of six Lineweaver-Burk plots each were prepared to determine if the inhibition of MAO-A and MAO-B by compound 16 is competitive. For the first plot, MAO activity was recorded without inhibitor present; while for the other five plots, MAO activity was recorded at inhibitor concentration of 1 / 4 Â IC 50 , 1 / 2 Â IC 50 , 3 / 4 Â IC 50 , 1 Â IC 50 , and 1 1 /4 Â IC 50 . The substrate, kynuramine, was used at concentrations ranging from 15 to 250 lM, whereas the final enzyme concentration was 0.015 mg protein/mL. All incubations were performed at a volume of 500 lL and the MAO activities were determined by fluorescence spectrophotometry as described for the IC 50 value determination. K i values were estimated by graphing the slopes of the Lineweaver-Burk plots against inhibitor concentration. The intercept of the x-axis is equal to -K i .

Molecular docking studies
To gain a better understanding of inhibitor-enzyme interactions at the molecular level, and to explain the structural requirements for MAO inhibition activity of the synthesized compounds, molecular docking was carried out using Discovery Studio's CDOCKER 21 . The module uses a CHARMm-based molecular dynamics algorithm to generate random ligand conformations. The conformations are then docked into the binding site using random rigid-body rotations and simulated annealing, followed by a final minimization step to refine the ligand poses. The X-ray crystal structure of human MAO-B in complex with safinamide (code: 2V5Z) 22 and human MAO-A in complex with harmine (code: 2Z5X) 23 were downloaded from the PDB. Both structures were prepared by removing the co-crystalized ligand and water molecules, fixing the missing residues and adding hydrogen atoms and partial charges. As in previous docking protocols reported by our group, we performed molecular docking in the presence of preserved water molecules in proximity of the FAD, which seem important for mediating interactions with the inhibitors 11 . Furthermore, to determine the binding site, a sphere with a radius of 13Å was centered on the co-crystalized safinamide in human MAO-B or harmine in human MAO-A. Default CDOCKER parameters were used and each docking run returned ten poses, which were clustered and visually inspected and the top-ranked pose in the largest cluster was chosen. To assess the ability of CDOCKER and to predict the native binding pose, co-crystalized safinamide and harmine were re-docked in the binding site of human MAO-B (2V5Z) and MAO-A (2Z5X), respectively, starting from a random conformation. The docked poses were compared with the orientation of co-crystalized ligand. In addition, crossdocking was performed: various co-crystalized MAO-B ligands were docked into the binding pocket of human MAO-B (2V5Z) and RMSD (Root Mean Square Deviation) values against their original crystalized poses were measured after superimposing the different crystal structures. All synthesized compounds were drawn and prepared by adding hydrogen atoms and partial charges using the Discovery Studio software. All single bonds of the compounds were considered rotatable during the docking runs where residues of the binding site were considered rigid. Visualization of key binding pocket residues and docked poses were performed using UCSF Chimera 24 .

Molecular mechanism investigation
H4IIE cell culture and dopamine assay Rat hepatoma (H4IIE) cells were purchased from American Type Culture Collection (ATCC). H4IIE cells are from rat liver, and contain both MAO-A and MAO-B in approximately equal levels. Cells were cultured in Eagle's minimum essential medium (EMEM) supplemented with 10% fetal bovine serum (FBS) and 1% streptomycin/penicillin at 37 C in a 5% CO 2 humidified incubator. Cells were plated at a cell density of 3.0 Â 10 5 cells in each well of six well plates. Upon 80% confluence, the cells were washed with phosphate buffered saline (PBS) and the media were changed to HBSS supplemented with 10 mM HEPES and 5 mM D-glucose, pH 7.4. These media also contained calcium chloride (1.26 mM) and magnesium chloride (0.5 mM). Cells were then treated with tolcapone (1 0 À6 M), a catechol-O-methyltransferase (COMT) inhibitor to protect dopamine from enzymatic degradation, 20 min prior to starting the experiment. The cells were treated with the negative vehicle control (1% DMSO) or 70 mM of the MAO-B inhibitors (compounds 12, 14, 16, and 17) with the DMSO concentration also being 1%. The cells were subsequently treated with L-DOPA (50 mM). To measure dopamine release and MAO-B gene expression, both the cells and media were collected at 0, 30, 60, and 120 min after treatment. Dopamine level was measured in the supernatant aspirated from the cells by using high performance liquid chromatography (HPLC), while the cells were lysed with trizol for RNA extraction and subsequent gene expression studies.
Quantitative real-time polymerase chain reaction Total RNA was isolated using the Trizol agent (Invitrogen). One lg of RNA from each sample was reverse transcribed to cDNA using the M-MLV reverse transcriptase kit, according to the manufacturer's instruction (Promega). PowerUp SYBR green master mix (Life technologies) was used for quantitative real-time polymerase chain reaction (qRT-PCR) for analysis of gene expression. The following primers were used (IDTs) MAO-B forward, ACCAGTGTGGA ATCCAATCACTTA; MAO-B reverse, TTGTCATGTAGTCCCACTCTTCAG; b-actin forward, GCGAGTACAACCTTCTTGCAGCTC and b-actin reverse, AGCGCAGCGATATCGTCATCCAT. qRT-PCR was performed in a 384 well plates with the Quant Studio-5 instrument (Applied Biosystems). Samples were carried out in triplicate in a 10 mL reaction volume using the following cycling conditions: UDG activation for one cycle at 50 C for 2 min, followed by a single cycle at 95 C for 2 min, then 40 cycles of denaturation at 95 C for 15 s, and annealing at 60 C for 1 min. b-actin was used as a loading control and gene expression was calculated using the 2 -DDCT method as described previously 25 .
High performance liquid chromatography with electrochemical detection Rat hepatoma cells (H4IIE) were treated with 1% DMSO (vehiclecontrol), compounds 12, 14, 16, or 17 for different time points at a concentration of 70 mM of each compound along with 50uM L-DOPA.
The concentrations of dopamine were detected using a CoulArray electrochemical array detector system as described previously 24 . Briefly, supernatants obtained from cell culture were filtered using 0.22 lm filter. The filtrates were subsequently injected onto a C18 column (3.2 Â 150 mm, 3 lm particle size, Thermo Scientific). The mobile phase consisted of a mixture of 54.3 mM sodium phosphate, 0.215 mM octyl sodium sulfate, 0.32 mM citric acid and 11% methanol. Dopamine concentrations were measured with the CoulArray Electrochemical Array Detector. The chromatograms were analyzed using the CoulArray software. A standard calibration curve was determined using different concentrations of an external standard dopamine (SigmaAldrich).

Statistical analysis
One-way ANOVA followed by Dunnett's test (compared to the vehicle-control) was used to analyze the data obtained for the dopamine concentrations for compounds 12, 14, 16, and 17targeting MAO-B at 0, 30, 60, and 120 min. One-way ANOVA followed by Newman-Keuls multiple comparison tests was used to analyze time-dependent effects of compounds 14 and 17 on the relative mRNA expression (triplicate) of MAO-B enzyme.

Chemistry
The 22 chalcone analogs ( Figure 2) were designed and prepared in moderate to good yields by Claisen-Schmidt condensation reaction. The synthesis was achieved by the reaction of equimolar amounts of the substituted acetophenones and appropriate benzaldehydes in ethanol and in presence of three equivalent of NaOH (in the minimum amount of water) as shown in Scheme 1.
The crude products were collected by filtration and crystallization from ethanol. The structures of the prepared compounds were elucidated by various spectroscopic techniques (e.g. FT-IR, GC-MS, and NMR). The chemical shifts of all protons and carbons of compound 4 were fully determined and assigned by 2D-NMR (COSY, HSQC, and HMBC; Figure 3). The trans configuration of the olefinic protons was confirmed by the large coupling constant (J ¼ 16 Hz).

Structure-activity relationships of the prepared analogs
The synthesized compounds were initially screened at 10 lM final concentration against recombinant human MAO-A and MAO-B by using 1-methyl-4-(1-methylpyrrol-2-yl)-1,2,3,6-tetrahydropyridine (MMTP) as a nonselective substrate for both MAO-A and B. It is important to note that the K m values for human MAO-A and MAO-B were found to be 88 lM and 101 lM, respectively, and this is in accordance with previously reported values 26,27 . The formation of the dihydropyridinium species (MMDP þ ) as a MAO-catalyzed oxidation product of MMTP was monitored at 420 nm. The results showed that most of the chalcones, indeed, have a marked inhibitory effect on the MAO enzymes at this concentration. Based on this, the IC 50 values were determined using kynuramine as a nonselective substrate. The rate of oxidation of kynuramine by MAO was monitored by measuring the fluorescence of the oxidation product, 4-hydroxyquinoline (k ex ¼ 310 nm; k em ¼ 400 nm). Kynuramine and the chalcone analogs did not interfere with the measurement, as these drugs did not show any fluorescence.
The IC 50 values were determined for all the synthesized compounds against recombinant human MAO-A and B to determine their activities and selectivity ( Table 1

Molecular docking studies
In general, CDOCKER was able to predict the native binding pose of different MAO-B ligands with a high degree of accuracy. Table 2 shows the calculated RMSD values between the docked and the co-crystalized ligands in absence and presence of water molecules in the binding pocket. Harmine was redocked in the active site of human MAO-A with RMSD of 0.848Å.
It is observed that retaining the conserved water molecules leads to lower RMSD values for the co-crystalized ligands that were re-docked, indicating a higher capability of predicting the native binding mode. This could be attributed to the reduced binding site size and hence the space available for possible docked poses, however, the persistence of these water molecules in all MAO-B crystal structures indicates that their displacement is unfavorable, and this case can be generalized to our synthesized compounds 8,9 . In the human MAO-B binding pocket, residues Tyr398 and Tyr435, together with FAD, form an aromatic cage, while Ile199 plays a key role as a gating residue between the entrance and substrate-binding cavities by occupying either a closed or an open conformation based on the length of the inhibitors. The docking study showed that the aromatic amino acids (Tyr398 and Tyr435) of the binding site accommodate the electron-deficient aromatic rings of chalcones. In all poses, the carbonyl oxygen anchors the inhibitors to the bottom of the binding   site by forming a hydrogen bond with Cys172 as shown in Figure 7.

Molecular mechanism investigations
Effects of compounds 12, 14, 16, and 17 on dopamine concentrations (ng/mL) at 0, 30, 60, and 120 min We assessed the concentrations of dopamine in rat hepatoma (H4IIE) cells treated with 1% DMSO (vehicle-control), and compounds 12, 14, 16, and 17 at 0, 30, 60, and 120 min. One way ANOVA followed by Dunnett's test did not show any significant changes in dopamine concentrations between the compounds and vehicle-control groups at 0 min (Figure 8(A)). However, Oneway ANOVA showed significant differences in the dopamine concentrations among the groups at 30 min [F(4, 10) ¼ 10.26863, p < .01, (Figure 8(B))]. One-way ANOVA followed by Dunnett's test revealed that treatment with compounds 14 and 17 increased the concentrations of dopamine compared to the vehicle control group at 30 min. We further investigated the effects of the selected compounds on the dopamine concentrations at 60 min. One-way ANOVA revealed significant changes in the dopamine concentrations among the groups at 60 min [F(4, 10) ¼ 4.477, p < .05, (Figure 8(C))]. One-way ANOVA followed by Dunnett's test showed that compounds 14 and 17 caused elevation in the dopamine concentration when compared to the vehicle-control group. Moreover, one-way ANOVA revealed a significant difference at 120 min among groups [F(4, 10) ¼ 4.544, p < .05, (Figure 8(D))]. One way ANOVA followed by Dunnett's test indicates that both compounds 14 and 17 were able to increase the dopamine levels at 120 min. These data indicate that compounds 14 and 17 have rapid onset on the elevation of dopamine levels using H4IIE cells.
Other compounds showed a positive trend (but not significant) on  dopamine levels, and these compounds may be effective to increase the dopamine concentrations when exposed for a longer duration.   The PDB codes of the crystal structures from which ligands were extracted are indicated. Docking was performed with the protein structure 2V5Z. compared to the negative control. Surprisingly, one-way ANOVA followed by Dunnett's test did not show any significant changes in the MAO-B mRNA expression between the compounds and negative control at 60 min ( Figure 9(C)). However, one-way ANOVA showed significant differences in MAO-B mRNA expression among groups at 120 min [F(2, 21) ¼ 15.62921, p < .0001, ( Figure  9(D))]. One-way ANOVA followed by Dunnett's test showed that only compound 14 decreased the relative mRNA expression compared to the negative control at 120 min. The data confirm that compound 17 has a short duration on reducing MAO-B mRNA expression since this compound only reduced mRNA expression of the enzyme at 30 min as compared to negative control. This may suggest a compensatory mechanism for the reduction of mRNA expression of MAO-B enzyme that has been occurred after 30 min from the compound exposure. Interestingly, compound 14 was able to reduce MAO-B mRNA expression in H4IIE cells at 30 and 120 min time points but not 60 min. We suggest here that higher concentrations might be required for the compound to exhibit stable reduction effects on the MAO-B mRNA expression in H4IIE cells.  (Figure 10(B))]. Oneway ANOVA followed by Newman-Keuls analysis showed that compound 14 decreased mRNA expression of MAO-B only at 120 min compared to the 0 min time point. These findings further support the previous data obtained in Figure 9, which indicate that compound 14 reduced the mRNA expression of MAO-B enzyme at 120 min. However, our work showed that the effects of compound 17 on reducing MAO-B mRNA expression started at 30 min. This indicates that the onset of inhibitory effects of compound 14 is slower than compound 17.

Conclusions
Chalcones are the natural precursors for the biosynthesis of flavonoids. These drugs have interesting chemical structures that include two aromatic rings linked with an enone linker (C 6 -C 3 -C 6 ).   In this study, 22 new chalcone analogs were prepared via Claisen-Schmidt condensation. Their inhibitory activities were evaluated against human MAO-A and MAO-B. The inhibition data showed differing activities and selectivities for MAO-A and MAO-B. Compound 13 showed a high degree of selectivity and activity against MAO-B with an IC 50 value of 34 nM and selectivity index >1338. Interestingly, the introduction of 2,4,6-trimethoxy substituents on the B aromatic ring shifted the selectivity toward MAO-A as exemplified by compounds 18 and 19. These compounds were potent and selective MAO-A inhibitors with IC 50 values of 0.56 lM (selectivity index <0.005) and 0.444 lM (selectivity index <0.004), respectively. As shown with compound 16, the chalcones are reversible MAO inhibitors as inhibitory activity is reversed by dialysis. Four compounds (12, 14, 16, and 17) were selected to assess their effects on cellular dopamine concentrations using H4IIE cells, with compounds 14 and 17 showing a significant increase in intracellular dopamine concentrations. Furthermore, these two compounds were found to inhibit the relative mRNA expression of MAO-B.