Biological Evaluation and Docking Analysis of Potent BACE1 Inhibitors from Boesenbergia rotunda

Alzheimer’s disease (AD) is an irreversible neurodegenerative disorder characterized by progressive impairment of cognitive functions. Beta-site amyloid precursor protein cleaving enzyme1 (BACE1) is essential for the formation of β-amyloid peptide (Aβ), a major constituent of amyloid plaques that represent a neuropathological hallmark of this disorder. To find alternative therapies for AD sourced from natural products, the present study focused on three flavonoids from Boesenbergia rotunda, namely, cardamonin, pinocembrin, and pinostrobin. Biological evaluation showed that cardamonin presented the strongest BACE1 inhibition, with an The half maximal inhibitory concentration (IC50) value of 4.35 ± 0.38 µM, followed by pinocembrin and pinostrobin with 27.01 ± 2.12 and 28.44 ± 1.96 µM, respectively. Kinetic studies indicated that the inhibitory constants (Ki) for cardamonin, pinocembrin, and pinostrobin against BACE1 were 5.1, 29.3, and 30.9 µM, respectively. Molecular docking studies showed that the tested compounds did not bind to the BACE1 active site, consistent with the biological results, illustrating non-competitive inhibitory activity for all three compounds. In addition, the lowest binding energy of the most proposed complexes of cardamonin, pinocembrin, and pinostrobin with BACE1 were −9.5, −7.9, and −7.6 kcal/mol, respectively. Overall, we provide the first evidence that these flavonoids from B. rotunda may be considered as promising AD preventative agents through inhibition of Aβ formation.


Introduction
Alzheimer's disease (AD) is an irreversible neurodegenerative disorder that results in progressive behavioral and cognitive impairment. The pathological hallmarks of AD are extracellular beta-amyloid (Aβ) plaque deposits and intracellular neurofibrillary tangles (NFTs) of the microtubule-binding protein tau [1]. Numerous studies strongly support a major role for Aβ aggregates in the initiation phase of AD pathogenesis, although the precise mechanisms of Aβ-induced neurotoxicity remain unclear. Aβ is derived from sequential proteolytic cleavage of the amyloid precursor protein (APP) by two aspartic proteases, βand γ-secretase. β-Secretase (BACE1) initiates amyloidogenic APP processing, generating sAPPβ and CTF99, and resulting in the processing of CTF99 in the transmembrane region by the γ-secretase complex and then produces the Aβ 40 and Aβ 42 peptides [2][3][4]. Increasing evidence suggests that Aβ generated by BACE1 triggers neuroinflammatory responses by inducing the expression of inflammatory cytokines and mediators via nuclear factor-κB (NF-κB) activation [5]. In addition, reactive oxygen species (ROS) have been revealed to cause the formation of Aβ fibrils, which in turn accelerates oxidative stress, inflammatory responses,

In Vitro Enzyme Inhibitory Assay for Biological Evaluation
Enzyme assays for BACE1, TACE, chymotrypsin, trypsin, and elastase were carried out according to previously described methods [27]. Briefly, reaction mixtures containing human recombinant BACE1 (1.0 U/mL), a specific substrate (Rh-EVNLDAEFK-Quencher in 50 mM ammonium bicarbonate), and samples dissolved in an assay buffer (50 mM sodium acetate, pH 4.5) were incubated for 60 min at 25 • C in well plates. The fluorescence intensity produced by substrate hydrolysis was observed on a microplate reader with excitation and emission wavelengths of 545 and 590 nm, respectively. The inhibition ratio was obtained using the following equation: where C was the fluorescence of control (enzyme, assay buffer, and substrate) after 60 min of incubation, C 0 was the fluorescence of control at time 0, S was the fluorescence of tested samples (enzyme, sample solution, and substrate) after 60 min of incubation, and S 0 was the fluorescence of the tested samples at time 0.
A human recombinant TACE (0.1 ppm in 25 mM Tris buffer), the substrate (APP peptide YEVHHQKLV using EDANS/DABCYL), and samples were dissolved in an assay buffer, which were then combined and incubated for 60 min in the dark at 25 • C. The increase in fluorescence intensity produced by substrate hydrolysis was measured by using a microplate reader with excitation and emission wavelengths of 320 and 405 nm, respectively.
Absorbance assays for trypsin, chymotrypsin, and elastase were estimated using N-benzoyl-L-Arg-pNA, N-benzoyl-L-Tyr-pNA, and N-succinyl-Ala-Ala-Ala-pNA as substrates, respectively. Enzyme, Tris-HCl buffer (0.05 M, in 0.02 M CaCl 2 , pH 8.2), and tested samples were incubated for 10 min at 25 • C then added to the substrate for 30 min at 37 • C. The absorbance was recorded at 410 nm. The inhibition ratio was obtained using the following equation: where A was the absorbance of control (enzyme, assay buffer, and substrate) after 60 min of incubation, and B was the absorbance of the tested sample (assay buffer and sample solution) after 60 min of incubation.

In Silico Docking Studies
Molecular docking analysis of the interaction between BACE1 and our tested compounds was performed using the AutoDock Vina program to display the binding conformations of the compounds in the enzyme [28]. The amino acid sequence of BACE1 and P-glycoprotein (P-gp) (PDB, 2WJO and 3G60) was acquired from the Protein Data Bank (http://www.rcsb.org). The grid dimension was set to 30 Å × 30 Å × 30 Å with a cluster radius of 1 Å. The Cα coordinates for every used backbone-binding residue of the protein receptor were used for the center of docking space. Default options were selected for other docking simulations. The atomic coordinates of the ligands were drawn and visualized using Marvin sketch (5.11.4, 2012, ChemAxon, Cambridge, MA, USA). The calculated geometries were ranked in terms of the lowest energy, and the top-ranked poses were selected. The hydrogen bond calculation was evaluated using Chimera 10.0.2.

Statistics
All results were presented as the mean ± standard deviation (SD) of three independent experiments. Significant differences were determined by Duncan s multiple range tests using Statistical Analysis System (SAS) version 9.3 (Cary, NC, USA).

Statistics
All results were presented as the mean ± standard deviation (SD) of three independent experiments. Significant differences were determined by Duncan′s multiple range tests using Statistical Analysis System (SAS) version 9.3 (Cary, NC, USA).
To determine their specificity for BACE1, the compounds were tested on α-secretase (TACE), involved in non-amyloidogenic processes, as well as on other proteases. None of the three compounds significantly inhibited the activity of TACE or serine proteases, including trypsin, chymotrypsin, and elastase, even at a concentration of 100 µM ( Table 2).   To determine their specificity for BACE1, the compounds were tested on α-secretase (TACE), involved in non-amyloidogenic processes, as well as on other proteases. None of the three compounds significantly inhibited the activity of TACE or serine proteases, including trypsin, chymotrypsin, and elastase, even at a concentration of 100 µM (Table 2). Resveratrol 3 14.59 ± 0.79 1 The IC 50 values (µM) were calculated from a log dose inhibition curve and expressed as the mean ± SD. 2 K i has the dimension of a concentration with the unit of µmol/L (µM). 3 Resveratrol was used as a positive reference control in the BACE1.

BACE1 Kinetics of Cardamonin, Pinocembrin, and Pinostrobin
The type of inhibition and kinetic analysis was determined using a Michaelis-Menten curve, Dixon plot, and Lineweaver-Burk plot ( Figure 2). The three lines of the Dixon plots intersected on the x-axis of the same plot (Figure 2b,e,h). In addition, the Lineweaver-Burk plots showed decreased V max and unchanged K m at increasing inhibitor concentrations (Table 1). This pattern indicated non-competitive inhibitory activity against BACE1, thereby interacting with locations other than the active binding site for these compounds. For noncompetitive inhibition of enzymes, the K i value of inhibitor is essentially the same numerical value as the IC 50 [29]. As shown in Table 1, the K i values for cardamonin, pinocembrin, and pinostrobin were 5.1, 29.3, and 30.9 µM, respectively, which were similar to their IC 50 values. Because lower K i values represent tighter binding with the enzyme, cardamonin was determined to be the most potent BACE1 inhibitor of the three compounds tested.
Vmax and unchanged Km at increasing inhibitor concentrations (Table 1). This pattern indicated noncompetitive inhibitory activity against BACE1, thereby interacting with locations other than the active binding site for these compounds. For noncompetitive inhibition of enzymes, the Ki value of inhibitor is essentially the same numerical value as the IC50 [29]. As shown in Table 1, the Ki values for cardamonin, pinocembrin, and pinostrobin were 5.1, 29.3, and 30.9 µM, respectively, which were similar to their IC50 values. Because lower Ki values represent tighter binding with the enzyme, cardamonin was determined to be the most potent BACE1 inhibitor of the three compounds tested.

Molecular Docking Study of Cardamonin, Pinocembrin, and Pinostrobin with BACE1 and P-Glycoprotein (P-gp)
Based on the biological evaluation results, we performed molecular docking simulations of enzyme-inhibitor conformations and lowest binding energies. As indicated in Table 3, the docking results for cardamonin, pinocembrin, and pinostrobin showed negative binding energies of −9.5, −7.9, and −7.6 kcal/mol, respectively, suggesting that all our three compounds are high-affinity BACE1 inhibitors that can bind tightly to the enzyme. In addition, the pinocembrin-BACE1 complex was stabilized by the formation of a hydrogen bond between residue LYS75 and the hydroxyl group of pinocembrin at a distance of 3.088 Å (Figure 3d-f and Table 3). Pinostrobin is bound at the PHE108 residue of BACE1, connected by one hydrogen bond with a bonding distance of 1.990 Å (Figure 3g-i and Table 3). In cardamonin-BACE1 complex, Gln12, Leu30, Asp32, Val69, Tyr71, Lys75, Trp76, Phe108, Ile110, Trp115, and Gly230 of BACE1 participated in hydrophobic interactions with cardamonin without any hydrogen bonding interactions (Figure 3a-c and Table 3).

Molecular Docking Study of Cardamonin, Pinocembrin, and Pinostrobin with BACE1 and P-Glycoprotein (P-gp)
Based on the biological evaluation results, we performed molecular docking simulations of enzyme-inhibitor conformations and lowest binding energies. As indicated in Table 3, the docking results for cardamonin, pinocembrin, and pinostrobin showed negative binding energies of −9.5, −7.9, and −7.6 kcal/mol, respectively, suggesting that all our three compounds are high-affinity BACE1 inhibitors that can bind tightly to the enzyme. In addition, the pinocembrin-BACE1 complex was stabilized by the formation of a hydrogen bond between residue LYS75 and the hydroxyl group of pinocembrin at a distance of 3.088 Å (Figure 3d-f and Table 3). Pinostrobin is bound at the PHE108 residue of BACE1, connected by one hydrogen bond with a bonding distance of 1.990 Å (Figure 3g-i and Table 3 ). In cardamonin-BACE1 complex, Gln12, Leu30, Asp32, Val69, Tyr71, Lys75, Trp76, Phe108, Ile110, Trp115, and Gly230 of BACE1 participated in hydrophobic interactions with cardamonin without any hydrogen bonding interactions (Figure 3a-c and Table 3).  The multidrug transporter P-glycoprotein (P-gp) is an important component of the BBB, acting as an ATP-driven efflux pump, controlling the movement of structurally diverse molecules across the BBB [30]. To predict the possibility of passing the BBB, the docking simulation between P-gp and our tested compounds was performed and the results are presented in Table 4 and Figure 4. Four hydrogen bonds were found between cardamonin and the P-gp residues, with the lowest binding energy of −9.78 kcal/mol. In addition, Tyr303, Gln721, and Ser725 participated in hydrogen-interactions (bonding distances of 3.17, 2.82, 3.25, and 2.96 Å, respectively). The pinocembrin-P-gp complex (the lowest binding energy: −10.64 kcal/mol) was stabilized by a hydrogen bond at Ser725 and Ser975 (bonding distance: 2.90 and 3.06 Å, respectively). Like pinocembrin, pinostrobin formed two hydrogen bonds with Ser725 and Ser975 (bonding distance: 2.89 and 3.03 Å, respectively). best docking poses between BACE1 and cardamonin (a), pinocembrin (d), and pinostrobin (g). Magnified view of the binding site for cardamonin (b), pinocembrin (e), and pinostrobin (h). The hydrophobic interaction of cardamonin (c). Hydrogen interaction diagram of pinocembrin (f), and pinostrobin (i). The blue dotted lines show hydrogen bonds between ligands and BACE1.
The multidrug transporter P-glycoprotein (P-gp) is an important component of the BBB, acting as an ATP-driven efflux pump, controlling the movement of structurally diverse molecules across the BBB [30]. To predict the possibility of passing the BBB, the docking simulation between P-gp and our tested compounds was performed and the results are presented in Table 4 and Figure 4. Four hydrogen bonds were found between cardamonin and the P-gp residues, with the lowest binding energy of −9.78 kcal/mol. In addition, Tyr303, Gln721, and Ser725 participated in hydrogeninteractions (bonding distances of 3.17, 2.82, 3.25, and 2.96 Å, respectively). The pinocembrin-P-gp complex (the lowest binding energy: −10.64 kcal/mol) was stabilized by a hydrogen bond at Ser725 and Ser975 (bonding distance: 2.90 and 3.06 Å, respectively). Like pinocembrin, pinostrobin formed two hydrogen bonds with Ser725 and Ser975 (bonding distance: 2.89 and 3.03 Å, respectively). Table 4. Lowest energies and the number of hydrogen and hydrophobic bond interactions of cardamonin, pinocembrin, and pinostrobin with P-glycoprotein (P-gp).

Discussion
In the present study, we investigated the anti-AD potential of three flavonoids from B. rotunda, namely cardamonin, pinocembrin, and pinostrobin, by evaluating their capacity to inhibit BACE1. These compounds exhibited powerful inhibitory activities on BACE1 with IC50 values in the range of 4.4-28.4 µM. Several other natural BACE1 inhibitors including genistein, biochanin A, daidzein, formononetin, and glycitein had IC50 values ranging from at a range of 28->100 µM [27,31]. The results showed that cardamonin, containing an α, β-unsaturated ketone with two aromatic rings, was the most potent BACE1 inhibitor of the three flavonoids. Similar results for the correlation between structure and enzyme inhibition properties were shown in a previous study [32]. Using an antiviral assay, the authors showed that cardamonin exhibited the strongest inhibition of HIV-1 protease activity, with an IC50 value of 31.0 µg/mL, compared to pinocembrin (>100 µg/mL) and pinostrobin (>100 µg/mL). Further enzyme kinetic evaluations using various substrates and inhibitor concentrations indicated that all three compounds are noncompetitive BACE1 inhibitors. The docking results indicated that hydrogen bond formation played a key role in the interaction of

Discussion
In the present study, we investigated the anti-AD potential of three flavonoids from B. rotunda, namely cardamonin, pinocembrin, and pinostrobin, by evaluating their capacity to inhibit BACE1. These compounds exhibited powerful inhibitory activities on BACE1 with IC 50 values in the range of 4.4-28.4 µM. Several other natural BACE1 inhibitors including genistein, biochanin A, daidzein, formononetin, and glycitein had IC 50 values ranging from at a range of 28->100 µM [27,31]. The results showed that cardamonin, containing an α, β-unsaturated ketone with two aromatic rings, was the most potent BACE1 inhibitor of the three flavonoids. Similar results for the correlation between structure and enzyme inhibition properties were shown in a previous study [32]. Using an antiviral assay, the authors showed that cardamonin exhibited the strongest inhibition of HIV-1 protease activity, with an IC 50 value of 31.0 µg/mL, compared to pinocembrin (>100 µg/mL) and pinostrobin (>100 µg/mL). Further enzyme kinetic evaluations using various substrates and inhibitor concentrations indicated that all three compounds are noncompetitive BACE1 inhibitors. The docking results indicated that hydrogen bond formation played a key role in the interaction of pinocembrin and pinostrobin with BACE1. However, cardamonin did not form any hydrogen bonds with BACE1 even though it presented the strongest BACE1 inhibition, suggesting that hydrophobic bonds were involved in the interaction between this compound and BACE1. The cardamonin result is in agreement with the report by Castro et al. (2015), which showed that cardamonin inhibited SmATPDase 1 through hydrophobic interactions [33].
Together, the above results indicate that all three compounds are beneficial BACE1 inhibitors, with potential as candidates for use in the prevention and/or treatment of AD. However, few BACE1 inhibitors have been marketed to treat AD due to the difficulty in developing therapeutic agents that can efficiently cross the BBB and reach appropriate concentrations in the cerebral parenchyma, and, moreover, can do so without side effects [34]. Recently, QSAR study approaches have been used to generate predictive models for BBB permeability. To cross the BBB, an active compound must possess the following physicochemical properties: a molecular weight (MW) of less than 450, polar surface area (PSA) of less than 70 Å 2 , and a clog P value, used as an assessment of lipophilicity, between 0 and 5 [35]. Cardamonin has a MW of 270.28, a PSA value calculated as 66.76, and a clog P of 4.54, obtained by determining the partition coefficient between n-octanol and water [19]. Furthermore, a previous study showed that some synthetic indolyl chalcones could cross the BBB, as determined by a parallel artificial membrane permeability assay [36]. Therefore, cardamonin is predicted to cross the BBB and exert its activity in the brain. An in vitro study by Yang et al. demonstrated that pinocembrin could cross the BBB by passive transport partly through P-glycoprotein [37].
Abnormal phosphorylated tau protein promotes the loss of microtubule-stabilizing ability and may contribute to neuronal degeneration as well as NFT formation [38]. To date, however, it is not known whether our three flavonoids can inhibit tau phosphorylation. However, several studies have revealed aspects in regard to the effect of flavonoids on the tau protein. Myricetin and epicatechin-5-gallate have been shown to suppress heparin-induced tau production and administration of EGCG leads to regulated tau profiles in AD transgenic mouse model, with inhibition of sarkosyl-soluble phosphorylated tau isoforms [39,40]. The authors plan to examine the attenuation effect of the three flavonoids on tau pathology in future studies.
Chronic oral administration of cardamonin (10 mg/kg) for 30 weeks did not elicit any apparent toxicity [41]. Additionally, Charoensin et al. reported that pinostrobin and pinocembrin also did not induce any toxicity or mutagenic effects following oral administration of 100 mg/kg for 7 days in Wister rats, suggesting that all three compounds are safe for consumption [42].
Overall, our tested compounds appear to be safe and potent natural BACE1 inhibitors that hold promise for use in the prevention and/or treatment of AD.

Conclusions
To the best of our knowledge, the present study is the first to indicate that cardamonin, pinocembrin, and pinostrobin are potent and selective BACE1 inhibitors. In a biological study, cardamonin was six-fold more effective than the two flavanones, pinocembrin and pinostrobin, at inhibiting BACE1. In addition, the inhibition kinetics and docking studies confirmed the allosteric nature of BACE1 inhibition by the tested compounds. Although further in vivo studies of these compounds are required to confirm our present results, our compounds may be good candidates for development as preventative agents against AD through inhibition of Aβ formation.