Alzheimer’s disease (AD) is a progressive neurodegenerative disorder with characteristic features of memory impairment, cognitive dysfunction, behavior disturbances, and deficits in activity of daily living [1
]. The increasing prevalence of AD has resulted in more interest in identifying risk factors associated with the development of AD [2
]. AD poses a huge economic burden worldwide, because there is no cure for this disease, which becomes progressively worse and eventually leads to death [3
]. Two major hypotheses have been proposed regarding the molecular mechanism of AD pathogenesis: the cholinergic hypothesis and the amyloid cascade hypothesis [4
]. However, these two major hypotheses cannot explain all pathological pathways of AD. Numerous recent studies have reported correlations between AD, inflammation, and oxidative stress [5
]. AD has been reported to be highly associated with cellular oxidative stress, including augmentation of protein oxidation, protein nitration, and accumulation of amyloid-β peptide (Aβ) [6
]. In vivo formation of ONOO−
has been implicated in Aβ formation and accumulation, with high levels of Aβ also augmenting ONOO−
generation in brains of AD patients [8
]. Simultaneous studies on cholinesterase (ChEs), Aβ accumulation inhibitory effects, and antioxidant effects are considered integral to the development of promising anti-AD agents. Several of the BACE1 (AZD3293, E2609, MK-8931, RG7129 10 and LY2886721) and cholinesterase (donepezil, tacrine, rivastigmine, and galantamine) inhibitors currently in use have adverse side-effects, including gastrointestinal disturbances, such as nausea, vomiting, and diarrhea; they also have bioavailability issues [11
For these reasons, there is growing scientific interest in identifying natural sources of AChE, BChE, and BACE1 inhibitors with safer profiles.
6-Formyl umbelliferone (2
), an uncommon coumarin derivative in nature, has been isolated from Angelica decursiva
, a perennial herb distributed on hillsides, grasslands, and sparse forests within China, Japan, and Korea. This plant has long been used in traditional Korean medicine as an antitussive, analgesic, antipyretic, and cough remedy. In traditional Chinese medicine, A
is used as to treat thick phlegm, asthma, and upper respiratory tract infections [14
]. We previously reported that coumarins from A
have promising anti-AD, anti-diabetic, antioxidant, and anti-inflammatory activity [15
]. Coumarin and its derivatives have been reported to show a wide range of pharmacological activity, such as anticoagulant, estrogenic, dermal photosensitizing, vasodilator, molluscicidal, anthelmintic, sedative, hypnotic, analgesic, hypothermic, antimicrobial, anti-inflammatory, antifungal, and antiulcer activity [22
]. Caffieri et al. [23
] reported the apoptogenic effects of 6-formyl umbelliferone on mitochondria. Other than their study, the biological properties of 2
have not been further investigated.
In the present study, we evaluated the anti-AD potential of coumarins by assessing their ability to inhibit AChE, BChE, BACE1, and ONOO−-mediated tyrosine nitration. Because there is currently no detailed information regarding 3D molecular interactions between these coumarins and BACE1 and AChE, we performed molecular docking analyses and detailed enzyme kinetic analysis to investigate the potential of coumarins as potent anti-AD drug candidates.
Medicinal plants have long provided a reliable source of new drugs to combat diseases. There have also been new trends in the preparation and marketing of drugs based on medicinal plants. Their scientific and commercial significance appears to be gathering momentum in health-relevant areas. Plant-derived products are carefully standardized, and medicinal plants offer several options to modify the progress and symptoms of AD [25
AD presents as the progressive, inexorable loss of cognitive function associated with the presence of senile plaques in the hippocampal area of the brain. The major hurdle in understanding AD is lack of knowledge regarding the etiology and pathogenesis of selective neuron death. The two most common hypotheses used to explain the pathology of AD are the “cholinergic hypothesis” and “amyloid hypothesis.” The cholinergic hypothesis suggests that AD is caused by a deficiency at the brain level of the cerebral neurotransmitter acetylcholine (ACh), which is hydrolyzed by AChE [27
]. Similarly, BChE activity increases by 40–90% during progression of AD [29
], and BChE inhibition is therefore considered a potentially important aspect of AD treatment. In addition to the AChE and BChE hypotheses, accumulation of amyloid-β peptide (Aβ) in the brain is widely considered to be critically involved in the pathogenesis of AD [30
]. Aβ plaques emerge roughly 15 years before the symptoms of AD appear [31
]. Once AD develops, the cognitive decline caused by neuronal damage cannot be reversed, even if the Aβ level in the brain is lowered by immunotherapy [32
]. Thus, prevention of Aβ accumulation is considered an important part of AD prevention. Aβ is excised from amyloid-β precursor protein (APP) through sequential cleavage by aspartic protease β-secretase 1 (BACE1) [28
]. Because BACE1 initiates Aβ processing, inhibition of BACE1 activity may be an effective way to prevent Aβ accumulation [34
In recent years, considerable data has accumulated that indicates that brains with AD are under increased oxidative stress. This stress may have a role in the pathogenesis of neurodegeneration and death due to AD. Oxidative stress refers to undue oxidation of biomolecules, and often leads to cellular damage. Histopathological and experimental evidence support the significant impact of oxidation on the pathogenesis of AD [35
Many researchers over the past few decades have focused their efforts on designing ChEs and BACE1 inhibitors. However, efforts to discover naturally occurring ChEs and BACE1 inhibitors have been relatively limited. Several plant-derived ChEs and BACE1 inhibitors, including coumarins, anthraquinones, triterpenoids, lignin, flavonoids, and alkaloids [19
] have been reported. Most of the previously discovered natural BACE1 and ChEs inhibitors have a low molecular weight, exhibit structural similarities, such as aromatic rings, and may have some adverse side effects. Nonetheless, research into effective agents from natural sources is still in its infancy, and there is a crucial need for better ChEs and BACE1 inhibitors from natural resources.
Coumarins are naturally occurring compounds present in a large number of plants. Coumarin and its derivatives are widespread in nature. Coumarin is a benzopyrone; benzopyrones are compounds composed of benzene rings connected to a pyrone moiety. Dietary exposure to benzopyrones is quite significant, as these compounds are found in fruits, vegetables, seeds, nuts, and higher plants. It is estimated that the average Western diet contains ~1 g/day of mixed benzopyrones [42
]. Coumarin-containing higher plant extracts are also widely popular in Chinese medicine. Natural coumarins have recently been reported to show promising anti-diabetic, anti-AD, anti-inflammatory, and antioxidant activity [15
]. Coumarins are also considered a promising group of bioactive compounds, because they exhibit a wide range of biological activity, including antibacterial [45
], antioxidant [46
], anti-inflammatory, and anticoagulant [47
] activity. In this study, 6-formyl umbelliferone (2
), a benzopyrone-type coumarin, was isolated for the first time from A. decursiva
, and showed activity against electric eel AChE, horse serum BChE, and human recombinant BACE1. We assumed that the inhibitory activity of 2
was mainly due to the formyl moiety at the C-6 position, therefore, we further synthesized 8-formyl umbelliferone (3
) to determine structure–activity relationships (SARs). To establish SARs between the three coumarins and the target enzymes, the inhibitory effects of the three coumarins against electric eel AChE, horse serum BChE, and human recombinant BACE1 were investigated. 2
inhibited AChE and BChE with an IC50
range of 16.70–19.13 µM for AChE and 27.90–87.67 µM for BChE, whereas umbelliferone (1
) showed weak inhibitory activity. In addition, 2
strongly inhibited BACE1 with an IC50
of 1.31 µM. Among the molecules investigated, both 2
, which have a free aldehyde group at the 6 or 8 position, significantly inhibited ChEs and BACE1. A comparison of the three coumarins indicated that the presence of an aldehyde group markedly increased inhibitory activity (Table 1
has a free hydroxyl group at the 7 position, but no aldehyde group. The inhibitory activity of ChEs and BACE1 therefore appears to be largely dependent on the presence of the aldehyde group at the 6 or 8 position.
Further enzyme kinetic studies in the presence of varying substrate and inhibitor concentrations revealed that 2
are noncompetitive BACE1 inhibitors. We next analyzed the molecular structure of AChE-inhibitors (Figure 3
) and BACE1-inhibitors (Figure 4
) to determine specific functional groups involved in these interactions.
Computational molecular docking analysis can provide insight into the mechanism underlying active site binding interactions. 2 could interact with both CAS (His440 and Ser200) and PAS (Tyr334) of AChE, whereas 3 bound only to CAS (His440). In the BACE1 docking simulation, we observed multiple hydrogen bonds in the docked BACE1–2 complex. However, 3 did not form any hydrogen bonds with BACE1, even though it docked in a similar region to 2. These docking data showed that hydrogen bonds between the coumarins and major active residues of the target enzyme play a crucial role in enzyme inhibition. Moreover, we also confirmed the inhibitory activity of 2 against formation of ONOO− mediated tyrosine nitration by Western blotting.
Taken together, the in vitro results and in silico molecular docking data indicate that 2 has significant potential to inhibit, and may prevent AD by targeting Aβ formation through inhibition of BACE1 and AChE, as well as ONOO−-mediated tyrosine nitration.
4. Material and Methods
4.1. General Experimental Procedures
1H- and 13C-NMR spectra were measured using a JEOL JUM ECP-400 spectrometer (Jeol, Tokyo, Japan) at 400 MHz for 1H-NMR and 100 MHz for 13C-NMR, with the compounds dissolved in deuterated dimethyl sulfoxide (DMSO-d6) and chloroform (CDCl3). Column chromatography was carried out using 70–230 mesh silica gel (Merck, Darmstadt, Germany). Thin-layer chromatography (TLC) was performed on pre-coated Merck Kiesel gel 60 F254 plates (0.25 mm), and 25% H2SO4 was used as a spray reagent. All solvents for column chromatography were of reagent grade and were acquired from commercial sources.
4.2. Chemicals and Reagents
Electric eel acetylcholinesterase (AChE, EC184.108.40.206), horse serum butyrylcholinesterase (BChE, EC 220.127.116.11), acetylthiocholine iodide (ACh), butyrylthioline chloride (BCh), 5,5′-dithobis [2-nitrobenzoic acid] (DTNB), ethylenediaminetetraacetic acid (EDTA), berberine, and quercetin were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). The BACE1 FREF assay kit (β-secretase) was purchased from Pan Vera Co. (Madison, WI, USA). ONOO− was purchased from Molecular Probes Cayman (Ann Arbor, MI, USA). All other chemicals and solvents used were purchased from E. Merck, Fluka, or Sigma- Aldrich, unless otherwise stated.
4.3. Isolation of Coumarins from A. Decursiva
Powder of whole A. decursiva
plants was refluxed with methanol (MeOH) for 3 h (3 × 10 L). The total filtrate was then concentrated until dry in vacuo at 40 °C to render the MeOH extract. This extract was suspended in distilled water and then successively partitioned with dichloromethane (CH2
), ethyl acetate (EtOAc), and n
-BuOH) to yield CH2
, EtOAc, n
-BuOH, and H2
O fractions, respectively, according to Zhao et al. [15
]. Initially, the EtOAc fraction was also subjected to silica gel column chromatography using CH2
–MeOH (10:1→0:1, gradient) to obtain 20 subfractions (F-1 to F-20). Repeated chromatography of subfraction-6 over a silica gel column using CH2
–MeOH (20:1→0:1, gradient) yielded 6-formyl umbelliferone (2
) (30 mg) and umbelliferone (1
) (3.9 g), respectively. The structure of the compounds was confirmed by 1
H- and 13
C-NMR spectroscopy and comparison with published data [18
]. Structures are shown in Figure 1
6-Formyl umbelliferone (2). Colorless needles; 1H-NMR (400 MHz, DMSO-d6): δH 11.7 (1H, brs, -OH), 10.24 (1H, s,-CHO), 8.03 (1H, s, H-5), 8.04 (1H, d, J = 9.6 Hz, H-4), 6.84 (1H, s, H-8), 6.32 (1H, d, J = 9.6 Hz, H-3); 13C-NMR (100 MHz, DMSO-d6): δC 189.2 (-CHO), 163.4 (C-7), 159.5 (C-2), 158.7 (C-9), 144.4 (C-4), 130.0 (C-5), 120.4 (C-6), 113.1 (C-3), 111.96 (C-10), 103.4 (C-8).
4.4. Synthesis of 8-Formyl Umbelliferone
8-Formyl umbelliferone (3
) was prepared from umbelliferone (2.0 g, 9.8 mmol) and urotropine (2.06 g, 14.69 mmol) in an ice-bath, as described by Qiao et al. [48
], and purified by recrystallization and Si gel column chromatography (n
-hexane/EtOAc = 10:1, v
) to yield 3
as a light yellow powder (0.30 g, yield 15%).
8-Formyl umbelliferone (3). Light yellow powder, 1H-NMR (400 MHz, CDCl3): δ 12.22 (1H, brs, -OH), 10.61 (1H, s, -CHO), 7.66 (1H, d, J = 9.6 Hz, H-4), 7.60 (1H, d, J = 8.8 Hz, H-5), 6.89 (1H, d, J = 8.8 Hz, H-6), 6.33 (1H, d, J = 9.6 Hz, H-3), 13C-NMR (100 MHz, CDCl3): 192.94 (-CHO), 165.51 (C-7), 159.12 (C-2), 156.74 (C-9), 143.36 (C-4), 135.98 (C-5), 114.70 (C-10), 113.42 (C-8), 110.87 (C-3), 108.68 (C-6).
4.5. In Vitro BACE1 Enzyme Assay
A BACE1 fluorescence resonance energy transfer (FRET) assay kit (β-secretase, human recombinant) was purchased from PanVera Co. (Madison, WI, USA). The assay was carried out according to the provided manual with slight modifications, as described in a previous study [19
]. Quercetin was used as a positive control. The tested concentration range was 0.4–200 µM for 1
and 2–50 µM for quercetin.
4.6. In Vitro ChE Enzyme Assay
The inhibitory activity of the isolated coumarins toward ChE was measured using the spectrophotometric method developed by Ellman et al. [49
]. ACh and BCh were used as substrates to assay the inhibition of AChE and BChE, respectively. Each reaction mixture consisted of 140 µL sodium phosphate buffer (pH 8.0), 20 µL of test sample solution, and 20 µL of either AChE or BChE solution, which were then combined and incubated for 15 min at room temperature. 1
and the positive control (berberine) were dissolved in 10% DMSO. The tested concentration range was 4–250 µM for 1
and 0.32–20 µM for berberine. Reactions were initiated upon addition of 10 µL of DTNB and 10 µL of either ACh or BCh. Enzymatic hydrolysis, mediated by AChE or BChE, was monitored according to the formation of yellow 5-thio-2-nitrobenzoate anions at 412 nm for 15 min, which were generated by the reaction of DTNB with thiocholine released from ACh or BCh. All reactions were performed in 96-well plates in triplicate, and recorded using a microplate spectrophotometer (Molecular Devices, Sunnyvale, CA, USA).
4.7. Kinetics of BACE1 Inhibition
Complementary kinetic methods were employed: Lineweaver–Burk and Dixon plots [50
]. The mechanism of BACE1 inhibition was evaluated by monitoring the effects of different concentrations of substrate (750, 375, and 250 nM). The test concentrations of 2
in BACE1 inhibition kinetic experiments were 0, 0.4, 2.0, and 10 µM, while those for 3
were 0, 25, 50, and 100 µM. Inhibition constants (Ki
) were determined by interpreting the Dixon plot, where the value on the x
-axis represents Ki
4.8. Molecular Docking Simulation to Investigate AChE and BACE1 Inhibition Using AutoDock 4.2
Molecular docking was carried out with AutoDock 4.2 program using the default parameters, with slight modification [53
]. The target proteins, AChE and BACE1, were obtained from the RCSB Protein Data Bank (PDB, http://www.rcsb.org/
), with the respective accession codes 1acj and 2wjo, respectively. The co-crystallized ligands, THA and QUD, were used to generate the grid box for a catalytic inhibition mode. The reported allosteric inhibitors, donepezil and PMF, were also used to compare the interaction residues [24
], and their 3D structures were downloaded from PubChem Compound (NCBI), with compound CIDs of 3152 and 79,730, respectively. The 3D structures of 1
were drawn with Chemdraw Ultra 12.0 (CambridgeSoft, Cambridge, MA, USA), and their pKa
values were computed at crystallographic pH (pH = 7.5) using the MarvinSketch (v17.1.30, ChemAxon, Budapest, Hungary). Docking simulation of 1
with AChE and BACE1 were performed individually using AutoDock Tools (ADT), and all torsions were allowed to rotate. The grid box size was 40 × 40 × 40 with a default spacing of 0.375 Å, and the x, y, z, center was 6.312, 71.239, 69.019 for AChE, and 18.167, 36.716, 40.55 for BACE1. The docking parameters were used as defaults of the ADT and Lamarckian genetic algorithm method was employed. The results were analyzed using PyMOL 1.7.4 and Ligplot+.
4.9. Inhibition of ONOO−-Mediated Protein Tyrosine Nitration
We evaluated ONOO−
-mediated protein tyrosine nitration using the method of Ali et al. [20
], with slight modifications. The tested concentration range of 2
was 12.5–100 µM.
4.10. Statistical Analysis
All results are expressed as means ± SEMs of triplicate samples. Results were analyzed using one-way ANOVA and Student’s t-test where appropriate (Systat Inc., Evanston, IL, USA). Values of p < 0.05 were considered statistically significant.