Chromeno[3,4-b]xanthones as First-in-Class AChE and Aβ Aggregation Dual-Inhibitors

Alzheimer’s disease (AD) is a complex multifactorial disorder, mainly characterized by the progressive loss of memory and cognitive, motor, and functional capacity. The absence of effective therapies available for AD alongside the consecutive failures in the central nervous system (CNS) drug development has been motivating the search for new disease-modifying therapeutic strategies for this disease. To address this issue, the multitarget directed ligands (MTDLs) are emerging as a therapeutic alternative to target the multiple AD-related factors. Following this concept, herein we describe the design, synthesis, and biological evaluation of a family of chromeno[3,4-b]xanthones as well as their (E)-2-[2-(propargyloxy)styryl]chromone precursors, as first-in-class acetylcholinesterase (AChE) and β-amyloid (Aβ) aggregation dual-inhibitors. Compounds 4b and 10 emerged as well-balanced dual-target inhibitors, with IC50 values of 3.9 and 2.9 μM for AChE and inhibitory percentages of 70 and 66% for Aβ aggregation, respectively. The molecular docking showed that most of the compounds bound to AChE through hydrogen bonds with residues of the catalytic triad and π-stacking interactions between the main scaffold and the aromatic residues present in the binding pocket. The interesting well-balanced activities of these compounds makes them interesting templates for the development of new multitarget compounds for AD.


Introduction
Alzheimer's disease (AD) is the most common form of dementia and contributes to 60-70% of all cases, affecting especially the elderly population, with ages above 65 [1]. This neurodegenerative disorder is mainly characterized by the neurodegeneration of several areas of the brain, particularly in the cerebral cortex and hippocampus, which leads to the progressive and unremitting loss of memory and cognitive, motor, and functional capacity [2,3]. Presently, the clinical diagnosis of AD is mainly based on mental and cognitive evaluation and its pathological confirmation is only achieved through the postmortem histopathological examination of brain tissue [4].
Despite many efforts from the scientific community, the exact cause of AD remains unclear; however, it has been stated that several hypothesis are associated with the pathological development of the disease, including acetylcholinesterase (AChE) and amyloid-β (Aβ) [5,6]. AChE is the enzyme responsible for the hydrolysis and inactivation of the neurotransmitter acetylcholine (ACh) and subsequent regulation of its concentration in the synaptic cleft [7]. The ACh-mediated neurotransmission is crucial for the nervous system function since its blockade and/or loss is correlated with the progressive deterioration of cognitive, autonomic, and neuromuscular functions, as in AD. Over the years, several studies have been reporting that lower levels of ACh are straightly correlated with AChE activity and, thus, AChE is considered one of the main targets for AD therapy. In 1990s cholinesterase inhibitors were introduced to the market as the first therapeutics against AD. These drugs increase acetylcholine levels in the brain by inhibiting AChE; however, they offer only time-limited symptomatic relief for the patients. Nevertheless, cholinesterases are still considered important biological targets in the design of multifunctional ligands. Usually, the inhibitory activity against cholinesterases is combined with the activity towards disease-modifying targets such as Aβ or tau aggregation [2,8,9].
On the other hand, the Aβ-peptides are generated by the proteolysis of its amyloid precursor protein (APP) [10][11][12]. Once in the cell surface, this protein can be cleaved at its Nterminus by β-secretase, leading to the production of the soluble amyloid precursor protein β-fragment (APPsβ) and the membrane-bound C-terminus, C99 [10,13,14]. This remaining fragment is then cleaved by γ-secretase, which generates Aβ-peptides [13,15,16]. These peptides are characterized by a long, usually ribbon-like morphology and share a common "cross-β" structure, at a molecular level, which consists of laminated β-sheets [17,18]. The accumulation of Aβ-peptides leads to the production of insoluble fibrils, which ultimately aggregate into the characteristic Aβ plaques of AD [15].
The current clinical therapy for AD is mainly divided in acetylcholinesterase inhibitors (AChEIs), represented by Donepezil, Galantamine, Rivastigmine, and Tacrine, which has already been discontinued due to its hepatoxicity, and the antagonist of the N-methyl-D-aspartic acid (NMDA) receptor, Memantine, all approved by the US Food and Drug Administration (FDA) and European Medicines Agency (EMA) [1,5]. This approach only provides symptomatic relief at short-term and do not decreases the disease progression, leading to the damage of the neurons and ultimately to the patient's death [5].
The relationship between the histopathological hallmarks of AD, i.e., Aβ plaques, neurofibrillary tangles (NFTs) and AD-associated cognitive impairment has been strengthening its multifactorial nature and may help to explain the consecutive failures in the disease-modifying drug discovery [19]. To address this issue, the multitarget directed ligands (MTDLs) have been emerging as an alternative approach to focus on the multiple AD-related factors [20,21]. This strategy is based on the simultaneous molecular recognition of multiple targets associated with the same pathology by a single molecule, which combines different pharmacophores of different bioactive molecules [22,23].
Natural products often play important roles in drug discovery and development processes. Xanthones are important examples of that, presenting a wide range of biological activities, particularly as AChE inhibitors ( Figure 1) [24,25]. Furthermore, several xanthone derivatives have been reported for their dual/multitarget activity as potential agents to treat AD [26]. Alongside xanthones, chromenes are also important pharmacophores, reported to have AChE inhibition properties ( Figure 1) [27]. An interesting example is the natural macluraxanthone (Figure 1), which contains a chromene nucleus naturally overlapped to a xanthone scaffold, presenting good AChE inhibition properties [28].
Inspired by the above examples and following the MTDLs concept, in this work, we describe the design, synthesis, and biological evaluation of a series of chromeno [3,4b]xanthones and their structurally flexible precursors (E)-2-[2-(propargyloxy)styryl]chromones ( Figure 2), as first-in-class dual-inhibitors of both AChE and Aβ aggregation.
The target compounds-chromeno [3,4-b]xanthones-feature xanthone and chromene pharmacophoric units, fused to each other in one new molecular entity, expected to have enhanced anticholinesterase potency ( Figure 2). Their (E)-2-styrylchromone precursors were also evaluated against both targets, to establish structure-activity relationships (SAR) and to get insights whether structurally flexible or rigid chromone-derived compounds are better templates for the development of AChE and Aβ aggregation dual-inhibitors. Beyond that, the hybrid structure of chromeno [3,4-b]xanthones represent highly aromatic/hydrophobic flat extended frameworks, which is a common feature of the existing Aβ aggregation inhibitors, together with hydrogen bonding spots such as methoxy or hydroxy groups (Figure 2), and therefore they can also be effective against Aβ fibrillation [29].  The target compounds-chromeno [3,4-b]xanthones-feature xanthone and chromene pharmacophoric units, fused to each other in one new molecular entity, expected to have enhanced anticholinesterase potency ( Figure 2). Their (E)-2-styrylchromone precursors were also evaluated against both targets, to establish structure-activity relationships (SAR) and to get insights whether structurally flexible or rigid chromonederived compounds are better templates for the development of AChE and Aβ aggregation dual-inhibitors. Beyond that, the hybrid structure of chromeno [3,4-b]xanthones represent highly aromatic/hydrophobic flat extended frameworks, which is a common feature of the existing Aβ aggregation inhibitors, together with hydrogen bonding spots such as methoxy or hydroxy groups (Figure 2), and therefore they can also be effective against Aβ fibrillation. [29]

Synthesis of Target Compounds
The synthetic strategy for chromeno [3,4-b]xanthones 4a-i encompasses two reaction steps (Scheme 1). First, the base-promoted aldol reaction between 2-methylchromones 1ad and aldehydes 2a-c, gave the intermediate (E)-2-styrylchromones 3a-i in 40-91% yield   The target compounds-chromeno [3,4-b]xanthones-feature xanthone and chromene pharmacophoric units, fused to each other in one new molecular entity, expected to have enhanced anticholinesterase potency ( Figure 2). Their (E)-2-styrylchromone precursors were also evaluated against both targets, to establish structure-activity relationships (SAR) and to get insights whether structurally flexible or rigid chromonederived compounds are better templates for the development of AChE and Aβ aggregation dual-inhibitors. Beyond that, the hybrid structure of chromeno [3,4-b]xanthones represent highly aromatic/hydrophobic flat extended frameworks, which is a common feature of the existing Aβ aggregation inhibitors, together with hydrogen bonding spots such as methoxy or hydroxy groups (Figure 2), and therefore they can also be effective against Aβ fibrillation. [29]
The (E)-2-styrylchromone nitrogen-containing analogues 6 and 8, were prepared in good yields (60% and 82%, respectively), following the same reaction conditions as described for derivatives 3a-i, using the 2-methylchromone 1a and aldehydes 5 and 7, respectively (Scheme 1). The 2-methylchromones 1a-d as well as the aldehydes 2a-c, 5 and 7 are commercially unavailable, and therefore they were prepared following reported methodologies (see SI for full details, Schemes S1-S3).

Biological Evaluation
All the synthesized chromeno [3,4-b]xanthones 4a-i, as well as their precursors (E)-2propargyloxy-2-styrylchromones 3a-i were evaluated in vitro for their potential as dualtarget agents, against the selected AD targets-AChE and Aβ aggregation. Both nitrogencontaining analogues 6 and 8, as well as (E)-2-styrylchromone 10 were also evaluated to explore the possibility to prepare more efficient structural analogues of chromeno [3,4b]xanthones 4a-i. Molecular modelling studies were conducted to explain the enzymeinhibitors interaction and to support the SAR analysis discussion.

AChE Inhibition
The synthesized compounds were evaluated for their inhibitory activity against AChE (Electrophorus electricus-eeAChE), using an adaptation of Ellman's method [30]. The eeAChE is an excellent model system as this enzyme and the human enzyme display a remarkably extensive sequence identity (>85%). Initially, we performed a screening assay for the synthesized compounds to determine their enzyme inhibition at 20 µM as a threshold maximum concentration. Based on this preliminary study, the compounds with inhibitory potency higher than 50% were established, for which the IC50 values were determined. The obtained results from the in vitro assays are represented in Table 1. Donepezil was used as reference compound. inhibitors interaction and to support the SAR analysis discussion.

AChE Inhibition
The synthesized compounds were evaluated for their inhibitory activity against AChE (Electrophorus electricus-eeAChE), using an adaptation of Ellman's method. [30] The eeAChE is an excellent model system as this enzyme and the human enzyme display a remarkably extensive sequence identity (>85%). Initially, we performed a screening assay for the synthesized compounds to determine their enzyme inhibition at 20 μΜ as a threshold maximum concentration. Based on this preliminary study, the compounds with inhibitory potency higher than 50% were established, for which the IC50 values were determined. The obtained results from the in vitro assays are represented in Table 1. Donepezil was used as reference compound. b]xanthones 4a-i. Molecular modelling studies were conducted to explain the enzymeinhibitors interaction and to support the SAR analysis discussion.

AChE Inhibition
The synthesized compounds were evaluated for their inhibitory activity against AChE (Electrophorus electricus-eeAChE), using an adaptation of Ellman's method. [30] The eeAChE is an excellent model system as this enzyme and the human enzyme display a remarkably extensive sequence identity (>85%). Initially, we performed a screening assay for the synthesized compounds to determine their enzyme inhibition at 20 μΜ as a threshold maximum concentration. Based on this preliminary study, the compounds with inhibitory potency higher than 50% were established, for which the IC50 values were determined. The obtained results from the in vitro assays are represented in Table 1. Donepezil was used as reference compound. OMe inhibitors interaction and to support the SAR analysis discussion.

AChE Inhibition
The synthesized compounds were evaluated for their inhibitory activity against AChE (Electrophorus electricus-eeAChE), using an adaptation of Ellman's method. [30] The eeAChE is an excellent model system as this enzyme and the human enzyme display a remarkably extensive sequence identity (>85%). Initially, we performed a screening assay for the synthesized compounds to determine their enzyme inhibition at 20 μΜ as a threshold maximum concentration. Based on this preliminary study, the compounds with inhibitory potency higher than 50% were established, for which the IC50 values were determined. The obtained results from the in vitro assays are represented in Table 1. Donepezil was used as reference compound. Among the 21 tested compounds, eight displayed interesting anticholinesterase profile, with IC50 values ranging from 2.1 to 9.5 μΜ (Table 1), just two orders of magnitude higher than Donepezil. Given that the inhibition constant of Donepezil for AChE is in the low-nanomolar range, the IC50 of some of the tested compounds is remarkable, in partic- Among the 21 tested compounds, eight displayed interesting anticholinesterase profile, with IC50 values ranging from 2.1 to 9.5 μΜ (Table 1), just two orders of magnitude higher than Donepezil. Given that the inhibition constant of Donepezil for AChE is in the low-nanomolar range, the IC50 of some of the tested compounds is remarkable, in partic- Donepezil 0.0156 ± 0.0016 [a] Values are expressed as the means ± the standard error of the mean (SEM) of at least three experiments (n = 3), each performed in triplicate.

of 19
Among the 21 tested compounds, eight displayed interesting anticholinesterase profile, with IC 50 values ranging from 2.1 to 9.5 µM (Table 1), just two orders of magnitude higher than Donepezil. Given that the inhibition constant of Donepezil for AChE is in the lownanomolar range, the IC 50 of some of the tested compounds is remarkable, in particular because they have not been optimized yet.
The obtained results showed that the most active compounds were the (E)-2styrylchromones 3d and 10, and the chromeno[3,4-b]xanthones 4a, 4b, 4e, and 4f (highlighted in blue, Table 1). The first SAR analysis to the (E)-2-styrylchromones 3a-i revealed that the overall introduction of methoxy groups throughout the main core decreased their anticholinesterase activity, when compared to the unsubstituted derivative 3a (Table 1, Figure 3), except for derivative 3d, in which the presence of ortho-dimethoxy group (donepezil fragment) in A-ring increased its anticholinesterase activity (Table 1, Figure 3). The (E)-2-styrylchromone 10 also displayed interesting inhibitory activity against AChE, which seems to indicate that the addition of the D-ring may be advantageous for AChE activity in comparison with the parent compounds 3a-i (Table 1, Figure 3). The replacement of the terminal triple bond (compound 3a) for a nitrile group in compound 6 decreased the AChE activity, suggesting the triple bond as an essential pharmacophoric fragment. Finally, the replacement of the O-propargyl group for the N,N-dipropargyl moiety in compound 8 did not show meaningful difference in AChE activity, when compared to the parent compound 3a (Table 1, Figure 3). Conversely, the rigid structure of chromeno [3,4-b]xanthones 4a-i showed better overall results, when compared to the respective (E)-2-styrylchromone precursors 3a-i ( Table  1). The unsubstituted chromeno [3,4-b]xanthone 4a exhibited the lowest IC50 of the series, being the introduction of methoxy groups disadvantageous for AChE activity, mainly for derivatives 4c, 4d, 4g and 4h (Table 1, Figure 3). The introduction of Br at C-4′ and C-3 of derivatives 3i and 4i, respectively, decreased their inhibitory potential against AChE, when compared to their respective unsubstituted analogues 3a and 4a (Table 1, Figure 3).
Generically, the presence of methoxy groups in chromeno [3,4-b]xanthones 4a-h seems to be more important for AChE activity than in their precursors 3a-h. On the other hand, chromeno [3,4-b]xanthones 4a-i can be understood as a structural rigidification of their precursors (E)-2′-propargyloxy-2-styrylchromones 3a-i. In this way, the obtained inhibition potencies clearly show that the rigid structures of chromeno [3,4-b]xanthones 4ai are more effective AChE inhibitors in detriment to the flexible structures of their precursors (Table 1, Figure 3).

Molecular Docking
To understand the structural basis of the AChE inhibitory activity determined for (E)-2′-styrylchromones 3a-i, 6, and 10 as well as for the chromeno[3,4-b]xanthones 4a-i, the binding mode to AChE was studied, using the GOLD software [31].
First, the accuracy of the protocol used in the molecular docking was validated. To do so, we re-docked the huprine ligand co-crystallized in the human acetylcholinesterase Conversely, the rigid structure of chromeno [3,4-b]xanthones 4a-i showed better overall results, when compared to the respective (E)-2-styrylchromone precursors 3a-i ( Table 1). The unsubstituted chromeno [3,4-b]xanthone 4a exhibited the lowest IC 50 of the series, being the introduction of methoxy groups disadvantageous for AChE activity, mainly for derivatives 4c, 4d, 4g and 4h (Table 1, Figure 3). The introduction of Br at C-4 and C-3 of derivatives 3i and 4i, respectively, decreased their inhibitory potential against AChE, when compared to their respective unsubstituted analogues 3a and 4a (Table 1, Figure 3).
Generically, the presence of methoxy groups in chromeno [3,4-b]xanthones 4a-h seems to be more important for AChE activity than in their precursors 3a-h. On the other hand, chromeno [3,4-b]xanthones 4a-i can be understood as a structural rigidification of their precursors (E)-2 -propargyloxy-2-styrylchromones 3a-i. In this way, the obtained inhibition potencies clearly show that the rigid structures of chromeno [3,4-b]xanthones 4a-i are more effective AChE inhibitors in detriment to the flexible structures of their precursors (Table 1, Figure 3).

Molecular Docking
To understand the structural basis of the AChE inhibitory activity determined for (E)-2 -styrylchromones 3a-i, 6, and 10 as well as for the chromeno [3,4-b]xanthones 4a-i, the binding mode to AChE was studied, using the GOLD software [31].
First, the accuracy of the protocol used in the molecular docking was validated. To do so, we re-docked the huprine ligand co-crystallized in the human acetylcholinesterase (hAChE) and the results indicated that the docking algorithm perfectly reproduced the x-ray structure of the hAChE:huprine complex. Subsequently we docked the huprine ligand into the eeAChE structure and observed that the binding modes are comparable in the two organisms. The two enzymes are very similar, sharing as much as >85% sequence identity, and display fully overlapping binding pockets. Thus, it is not surprising that the docking algorithm would behave similarly in both cases. We concluded that it was suitable to proceed with the analysis of the chosen compounds (see SI, Figures S1 and S2).
To determine the substituents importance for the binding, the docking results were visually inspected by dividing the compound series in agreement with their main scaffold. First, it was studied the pattern of interactions observed for (E)-2-styrylchromones 3a-i, 6, and 10, and a few examples are illustrated in Figure 4A. Here, it was observed that either Ser203 or His443 formed a hydrogen bond with the ether moiety in most of the compounds, except for compounds 3b, 3e, and 3f. Both Ser203 and His443 are key residues of the catalytic triad and are implicated on the hydrolysis of the carboxyl ester during the enzymatic mechanism. Moreover, the ether substituent in the (E)-2 -propargyloxy-2styrylchromones 3 was stabilized by the Tyr124 side chain and by a π-stacking interaction either with Trp286 or with Tyr341. The compounds that showed the best inhibition were 3d and 10, both forming a perfect stacking with the Tyr341 side chain. Compound 3b was unable to perform this interaction and stacked with Trp286 instead. Compound 3e tried to bind in a similar position to 3d but the results implied that the methoxy group at 3e was destabilized by the vicinity with Glu202. This may justify why this resulted in a weaker inhibition of the enzyme ( Figure 4A).
Finally, the docking results obtained for chromeno [3,4-b]xanthones 4a-i reinforced, once again, the role of Ser203, and also Phe294, which formed hydrogen bonding interactions with the heterocyclic rings, in all the studied compounds, except for compound 4e, which formed an additional hydrogen bond with the adjacent residue His443. The chromeno [3,4-b]xanthones 4a-i binding was also stabilized by several π-stacking interactions between the main scaffold and aromatic residues present in the binding pocket Trp86, Trp286, Phe338, and Tyr341. Moreover, the results did not show any specific interactions between the enzyme and the methoxy groups. However, there was a preferential trend to accommodate the methoxy group towards the region occupied by Trp286. Only when two methoxy substituents are present, the second group occupies the anionic subsite of the enzyme, surrounded by Glu202, Trp86 and Tyr337 ( Figure 4B). The enzyme can accommodate one or two methoxy groups in the binding pocket but only if the substituents do not occupy orto-positions; otherwise, it leads to steric clashes. The presence of the bromide group in compound 4i also destabilized the binding as it led to electrostatic repulsions. The experimental binding affinity, alongside the molecular docking results, suggested that there is a similar probability for the compounds to bind in two equivalent positions, related by a 180 • flip around the ligand center. The small differences in the binding affinity are probably due to the cost of the methoxy accommodation, resulting in small re-adjustments of the residues of the binding pocket. This explains why compound 4a, which perfectly formed three π-stacking interactions and two hydrogen bonds, was the best compound exploited among all. was destabilized by the vicinity with Glu202. This may justify why this resulted weaker inhibition of the enzyme ( Figure 4A).  4a, 4b, 4e, a 4f. The hydrogen bonds are showed as black dash lines and the π-stacking interactions in green dash lines.
Finally, the docking results obtained for chromeno [3,4-b]xanthones 4a-i reinfor once again, the role of Ser203, and also Phe294, which formed hydrogen bonding inte tions with the heterocyclic rings, in all the studied compounds, except for compound which formed an additional hydrogen bond with the adjacent residue His443.  4a, 4b, 4e, and 4f. The hydrogen bonds are showed as black dash lines and the π-stacking interactions in green dash lines.

Aβ Aggregation Inhibition
The formation of amyloid aggregates is crucial for the path mechanism of AD. Therefore, we also decided to check the inhibitory activity of the synthesized compounds against the Aβ aggregation. To do so, we conducted a thioflavin T (ThT)-based fluorimetric assay to evaluate their ability to inhibit β-amyloid aggregation in vitro, using 1,1,1,3,3,3,-hexafluoro-2-propanol (HFIP)-treated Aβ1-42 peptides. This assay is based on the interaction between benzothiazole dye (ThT) and cross β-sheet quaternary structure of the amyloid protein. The intensity of fluorescence is weak when the dye is free and increases when the dye binds to β-sheet structures [32]. Cumulative evidence has been shown that aromatic/hydrophobic areas and hydrogen bonding components (such as hydroxy or methoxy groups) along flat extended structures in small-molecule inhibitors [18,29] as in chromeno [3,4-b]xanthones 4a-i, are important structural features for Aβ aggregation inhibitors. However, in line with objectives of the present work and encouraged by the obtained results so far, we selected the best compounds from the AChE inhibition assay and performed an in vitro inhibition assay at 20 µM, to evaluate the compounds inhibition potencies against Aβ aggregation and their subsequent dual-target profile. The ThT fluorescence data was recorded over 24 h and the obtained results were then plotted against time, as illustrated in Figure 5. Phenol red was used as the reference inhibitor for in vitro Aβ aggregation. Almost all the tested compounds were found to have good antiaggregating activity when compared to control [ThT, Aβ and dimethylsulfoxide (DMSO) as control vehicle] as well as to the reference compound, except for chromeno [3,4-b]xanthone 4a, and (E)-2-styrylchromones 3a, 3b, and 3d, as depicted in Figure 5A. In a more detailed analysis, we highlighted the compounds with the best results in Figure 5B. The amyloid fibrillation comprises two main stages: (i) the nucleation phase, in which the monomers undergo a conformational change and associate to form oligomers, and (ii) the elongation phase, in which the oligomers rapidly grow by further addition of more monomers and form larger fibrils, until saturation (blue box, Figure 5B). [33] As shown in Figure 5B, the Aβ1-42 peptides alone were completely aggregated after 20-24 h and consequently the ThT fluorescence intensity reached its highest value, while the treatment with the assay compounds resulted in a delay of the nucleation phase of the Aβ formation and, subsequently, in a decrease of the fluorescence intensity. Moreover, almost all compounds showed an even bigger delay in the nucleation phase (except for compound 8), when compared to the reference compound, Phenol Red.
The analysis of the averaged fluorescence intensities at the plateau showed that compounds 3f, 8, 10, 4b, 4e, and 4f inhibited the Aβ aggregation in a range of 18-70%, even though the rest of them turned out to be weak inhibitors ( Table 2). The best results were obtained for chromeno[3,4-b]xanthones 4b, 4e, 4f, and (E)-2-styrylchromone 10, particularly the first one with an inhibition percentage of 70% (highlighted in blue, Table 2). Table 2. Inhibitory activity towards Aβ aggregation of (E) -2-styrylchromones 3a, 3b, 3d, 3f, 6, 8 and 10, and chromeno[3,4-  In a more detailed analysis, we highlighted the compounds with the best results in Figure 5B. The amyloid fibrillation comprises two main stages: (i) the nucleation phase, in which the monomers undergo a conformational change and associate to form oligomers, and (ii) the elongation phase, in which the oligomers rapidly grow by further addition of more monomers and form larger fibrils, until saturation (blue box, Figure 5B) [33]. As shown in Figure 5B, the Aβ 1-42 peptides alone were completely aggregated after 20-24 h and consequently the ThT fluorescence intensity reached its highest value, while the treatment with the assay compounds resulted in a delay of the nucleation phase of the Aβ formation and, subsequently, in a decrease of the fluorescence intensity. Moreover, almost all compounds showed an even bigger delay in the nucleation phase (except for compound 8), when compared to the reference compound, Phenol Red.
The analysis of the averaged fluorescence intensities at the plateau showed that compounds 3f, 8, 10, 4b, 4e, and 4f inhibited the Aβ aggregation in a range of 18-70%, even though the rest of them turned out to be weak inhibitors ( Table 2). The best results were obtained for chromeno [3,4-b]xanthones 4b, 4e, 4f, and (E)-2-styrylchromone 10, particularly the first one with an inhibition percentage of 70% (highlighted in blue, Table 2). Table 2. Inhibitory activity towards Aβ aggregation of (E) -2-styrylchromones 3a, 3b, 3d, 3f, 6, 8 and 10, and chromeno[3,4b]xanthones 4a, 4b, 4e, 4f, and  shown in Figure 5B, the Aβ1-42 peptides alone were completely aggregated after 20-24 h and consequently the ThT fluorescence intensity reached its highest value, while the treatment with the assay compounds resulted in a delay of the nucleation phase of the Aβ formation and, subsequently, in a decrease of the fluorescence intensity. Moreover, almost all compounds showed an even bigger delay in the nucleation phase (except for compound 8), when compared to the reference compound, Phenol Red. The analysis of the averaged fluorescence intensities at the plateau showed that compounds 3f, 8, 10, 4b, 4e, and 4f inhibited the Aβ aggregation in a range of 18-70%, even though the rest of them turned out to be weak inhibitors ( Table 2). The best results were obtained for chromeno [3,4-b]xanthones 4b, 4e, 4f, and (E)-2-styrylchromone 10, particularly the first one with an inhibition percentage of 70% (highlighted in blue, Table 2). A brief SAR analysis indicates that among the (E)-2-styrylchromone-type compounds, only derivatives 3f, 8 and 10 were capable of inhibit the Aβ aggregation, within a range of 18-66% (Table 2, Figure 6). The replacement of O-propargyl for the N,N-dipropargyl moiety seems to improve the capacity to inhibit the Aβ aggregation, since it slightly enhanced the antiaggregating activity of compound 8, when compared to the parent compound 3a (highlighted in orange, Figure 6). Additionally, the presence of methoxy groups at R 2 and R 4 in (E)-2-styrylchromone 3f and the presence of D-ring in 10 increased the antiaggregating activity when compared to (E)-2-styrylchromone 3a, with especial relevance for the last case (highlighted in yellow and green, respectively, Figure 6). Figure 6. SAR profile towards Aβ aggregation of (E) -2′-styrylchromones 3a, 3b, 3d, 3f, 8 and 10 and chromeno[3,4-b]xanthones 4a, 4b, 4e, 4f, and 4h. A brief SAR analysis indicates that among the (E)-2-styrylchromone-type compounds, only derivatives 3f, 8 and 10 were capable of inhibit the Aβ aggregation, within a range of 18-66% (Table 2, Figure 6). The replacement of O-propargyl for the N,N-dipropargyl moiety seems to improve the capacity to inhibit the Aβ aggregation, since it slightly enhanced the antiaggregating activity of compound 8, when compared to the parent compound 3a (highlighted in orange, Figure 6). Additionally, the presence of methoxy groups at R 2 and R 4 in (E)-2-styrylchromone 3f and the presence of D-ring in 10 increased the antiaggregating activity when compared to (E)-2-styrylchromone 3a, with especial relevance for the last case (highlighted in yellow and green, respectively, Figure 6). Figure 6. SAR profile towards Aβ aggregation of (E) -2′-styrylchromones 3a, 3b, 3d, 3f, 8 and 10 and chromeno[3,4-b]xanthones 4a, 4b, 4e, 4f, and 4h A brief SAR analysis indicates that among the (E)-2-styrylchromone-type compounds, only derivatives 3f, 8 and 10 were capable of inhibit the Aβ aggregation, within a range of 18-66% (Table 2, Figure 6). The replacement of O-propargyl for the N,N-dipropargyl moiety seems to improve the capacity to inhibit the Aβ aggregation, since it slightly enhanced the antiaggregating activity of compound 8, when compared to the parent compound 3a (highlighted in orange, Figure 6). Additionally, the presence of methoxy groups at R 2 and R 4 in (E)-2-styrylchromone 3f and the presence of D-ring in 10 increased the antiaggregating activity when compared to (E)-2-styrylchromone 3a, with especial relevance for the last case (highlighted in yellow and green, respectively, Figure 6). Figure 6. SAR profile towards Aβ aggregation of (E) -2′-styrylchromones 3a, 3b, 3d, 3f, 8 and 10 and chromeno[3,4-b]xanthones 4a, 4b, 4e, 4f, and 4h. On the other hand, almost all the tested chromeno[3,4-b]xanthones 4a-i were found to be good Aβ aggregation inhibitors, except for derivative 4a, probably due to its highly A brief SAR analysis indicates that among the (E)-2-styrylchromone-type compounds, only derivatives 3f, 8 and 10 were capable of inhibit the Aβ aggregation, within a range of 18-66% (Table 2, Figure 6). The replacement of O-propargyl for the N,N-dipropargyl moiety seems to improve the capacity to inhibit the Aβ aggregation, since it slightly enhanced the antiaggregating activity of compound 8, when compared to the parent compound 3a (highlighted in orange, Figure 6). Additionally, the presence of methoxy groups at R 2 and R 4 in (E)-2-styrylchromone 3f and the presence of D-ring in 10 increased the antiaggregating activity when compared to (E)-2-styrylchromone 3a, with especial relevance for the last case (highlighted in yellow and green, respectively, Figure 6).
On the other hand, almost all the tested chromeno [3,4-b]xanthones 4a-i were found to be good Aβ aggregation inhibitors, except for derivative 4a, probably due to its highly hydrophobic nature. Although the aromatic interactions seem to play a crucial role in Aβ aggregation inhibition, the absence of hydrogen bonding groups may limit the compounds interaction with mature fibrils. Among the chromeno [3,4-b]xanthone series, derivative 4b exhibited the best inhibition against the Aβ aggregation, probably due to its flat aromatic moiety, as well as the presence of methoxy groups, as in both 4e and 4f derivatives (highlighted in red, yellow, and blue, respectively, Figure 6). a range of 18-66% (Table 2, Figure 6). The replacement of O-propargyl for the N,N-dipropargyl moiety seems to improve the capacity to inhibit the Aβ aggregation, since it slightly enhanced the antiaggregating activity of compound 8, when compared to the parent compound 3a (highlighted in orange, Figure 6). Additionally, the presence of methoxy groups at R 2 and R 4 in (E)-2-styrylchromone 3f and the presence of D-ring in 10 increased the antiaggregating activity when compared to (E)-2-styrylchromone 3a, with especial relevance for the last case (highlighted in yellow and green, respectively, Figure 6). Figure 6. SAR profile towards Aβ aggregation of (E) -2′-styrylchromones 3a, 3b, 3d, 3f, 8 and 10 and chromeno[3,4-b]xanthones 4a, 4b, 4e, 4f, and 4h. On the other hand, almost all the tested chromeno [3,4-b]xanthones 4a-i were found to be good Aβ aggregation inhibitors, except for derivative 4a, probably due to its highly hydrophobic nature. Although the aromatic interactions seem to play a crucial role in Aβ Figure 6. SAR profile towards Aβ aggregation of (E) -2 -styrylchromones 3a, 3b, 3d, 3f, 8 and 10 and chromeno[3,4b]xanthones 4a, 4b, 4e, 4f, and 4h. Overall, these results seem to be in line with several previous investigations, which stated that small molecules containing single or multiple aromatic rings with hydroxy or methoxy groups, as the chromeno[3,4-b]xanthones 4 are capable of inhibit the Aβ aggregation [18,29]. As in the AChE inhibition, the structural rigidification of (E)-2-styrylchromones 3 into chromeno[3,4-b]xanthones 4 was pivotal for the enhancement of compounds' antiaggregating activity ( Figure 6).

Dual-Target Activity Analysis
The main objective of this work was to design, synthesize and evaluate the biological potential of a small library of chromeno [3,4-b]xanthones as first-in-class dual-target agents for AD. The obtained results from the AChE and Aβ aggregation inhibition assays showed that many of the synthesized compounds emerged as single-target agents. Nevertheless, a few other compounds stood out as lead structures for the development of more effective dual-target agents against AD, as shown in Table 3. The obtained results revealed IC 50 values ranging from 2.9 to 8.7 µM, and inhibitory ranges of 18-70% (at 20 µM) against the AChE and Aβ aggregation, respectively (Table 3). Most compounds were found to have enough binding affinity towards both targets to be considered to be first-in-class leads, being (E)-2-styrylchromone 8 the weakest and chromeno [3,4-b]xanthone 4b and (E)-2-styrylchromone 10 the strongest compounds.
Conversely, the best results were obtained for compounds 4b and 10, which exhibited moderate to good inhibitory activity against AChE and Aβ, with IC 50 values of 2.9-3.9 µM and inhibitory percentages of 66-70%, respectively ( Table 3). The presence of the D-ring in compound 10 seems to be crucial for both anticholinesterase and antiaggregating activity enhancement, as well as the presence of hydrophilic groups (such as methoxy groups) throughout the flat aromatic moiety of compound 4b ( Table 3).
As an overall SAR analysis, the structural rigidification was proven to be of paramount importance also for dual-target activity, since the best dual-target compounds were chromeno[3,4-b]xanthones 4. As discussed above, the structural modification promotes conformational restraint, which may block unwanted conformations, enhancing the compound's activity towards AChE and Aβ. Additionally, this type of compounds containing single or multiple aromatic rings and hydrophilic groups, such as methoxy groups, have been underlined for their good antiaggregating properties. Finally, the introduction of rigid and hydrophobic groups, such as the extra phenyl ring can promote the interaction with both targets, especially with the hydrophobic region in the active site of AChE. Table 3. Dual-target activity towards eeAChE and Aβ aggregation of (E)-2-styrylchromones 8 and 10, and chromeno [3,4-b] xanthones 4b, 4e, and 4f. dual-target agents against AD, as shown in Table 3. The obtained results revealed IC50 values ranging from 2.9 to 8.7 μM, and inhibitory ranges of 18-70% (at 20 μM) against the AChE and Aβ aggregation, respectively (Table 3). Most compounds were found to have enough binding affinity towards both targets to be considered to be first-in-class leads, being (E)-2-styrylchromone 8 the weakest and chromeno [3,4-b]xanthone 4b and (E)-2styrylchromone 10 the strongest compounds. Table 3. Dual-target activity towards eeAChE and Aβ aggregation of (E)-2-styrylchromones 8 and 10, and chromeno [3,4b] xanthones 4b, 4e, and 4f. values ranging from 2.9 to 8.7 μM, and inhibitory ranges of 18-70% (at 20 μM) against the AChE and Aβ aggregation, respectively (Table 3). Most compounds were found to have enough binding affinity towards both targets to be considered to be first-in-class leads, being (E)-2-styrylchromone 8 the weakest and chromeno [3,4-b]xanthone 4b and (E)-2styrylchromone 10 the strongest compounds. AChE and Aβ aggregation, respectively (Table 3). Most compounds were found to have enough binding affinity towards both targets to be considered to be first-in-class leads, being (E)-2-styrylchromone 8 the weakest and chromeno [3,4-b]xanthone 4b and (E)-2styrylchromone 10 the strongest compounds. Table 3. Dual-target activity towards eeAChE and Aβ aggregation of (E)-2-styrylchromones 8 and 10, and chromeno [3,4b] xanthones 4b, 4e, and 4f.

General Remarks
Melting points were measured with a Büchi B-540 apparatus. NMR spectra were recorded with a Bruker Avance 300 (300.13 MHz for 1 H and 75.47 MHz for 13 C) spectrometer. Chemical shifts (δ) are reported in ppm and coupling constants (J) in Hz; the internal standard was TMS. Unequivocal 13 C assignments were made with the aid of 2D gHSQC and gHMBC (delays for one-bond-long range J C/H couplings were optimized for 145 and 7 Hz, respectively) experiments. Positive-ion ESI mass spectra were acquired with a QTOF 2 instrument [dilution of 1 µL] of the sample in chloroform solution (ca. 10-5 mL) in 200 µL of 0.1% trifluoroacetic acid/methanol solution. Nitrogen was used as the nebuliser gas and argon as the collision gas. The needle voltage was set at 3000 V, with the ion source at 80 • C and the desolvation temperature at 150 • C. The cone voltage was 35 V. Other lowand high-resolution mass spectra (EI, 70 eV) were measured with VG Autospec Q and M spectrometers. Preparative thin-layer chromatography was performed with Merck silica gel (60 DGF254). All chemicals and solvents used were obtained from commercial sources and used as received or dried using standard procedures. MW-assisted reactions were carried out in a CEM Discover SP apparatus.

Synthesis of (E)-2-Styrylchromones 3a-i, 6, and 10
To a stirred solution of sodium (110 mg, 5 mmol) in ethanol (5 mL) were added the appropriate 2-methylchromone 1a-d (1.25 mmol) and the respective aldehydes 2a-c, 5 or 7 (1.56 mmol). The resulting mixture was stirred at room temperature for 3-6 h (completion of the reaction monitored by TLC). After that period, the reaction mixture was poured into cold water (10 mL) and the pH adjusted to 4 with 10% aqueous HCl. The crude precipitate was collected by filtration, taken in dichloromethane, and purified by silica gel preparative thin-layer chromatography (TLC), using dichloromethane as eluent.

AChE Inhibition
The inhibitory activity of the synthesized compounds towards AChE was evaluated spectrophotometrically by a 96-well microplate modified Ellman's method. The enzyme (AChE) from Electrophorus electricus (eeAChE) stock solution was prepared as 7.4 U/mL in phosphate buffer, at pH 7. The stock solutions for the target compounds (0.1 M) were prepared in DMSO. The assay solution consisted of 0.0005 M 5,5 -dithiobis(2-nitrobenzoic acid) (DTNB), 0.013 U/mL eeAChE, and 0.0025 M acetylthiocholine iodide ATC (substrate for AChE) in 0.1 M phosphate buffer, at pH 7. The reaction mixture was prepared with 100 µL of the enzyme (eeAChE) and 50 µL of the test compound (or DMSO/water, not exceeding 0.2% of DMSO in the wells; i.e., blank samples). After 5 min of the pre-incubation period at 37 • C, 50 µL of acetylthiocholine iodide ATC and 50 µL of 5,5 -dithiobis(2nitrobenzoic acid) (DTNB) were added to each well, thus initiating the enzymatic reaction.
The absorbance values were measured at 415 nm, for a total of 7.5 min, at every 2.5 min, using a microplate reader (Synergy multi-mode reader; BioTek). All compounds were tested at a screening concentration of 20 µM. The enzyme inhibition was calculated using the 100 (S/B) × 100 formula, where S and B are the enzyme activities with and without the test compound, respectively. The IC 50 values were determined for the compounds with an inhibitory activity higher than 50% at 20 µM, using seven different concentrations of 12.5, 10, 7.5, 5, 2.5, 1.0 and 0.5 µM. The absorbance was measured at seven different concentrations of inhibitor, converted to % inhibition of the enzyme and plotted against the applied inhibitor concentration, using a nonlinear regression. Donepezil was used as the reference compound. All the reactions were performed in triplicate.

Molecular Docking
Molecular docking calculations, using the GOLD software were carried out to determine the molecular interactions between the compounds and acetylcholinesterase [31]. The pdb codes 1C2O and 4BDT were used as structure inputs for the eeAChE and hAChE, respectively [35,36]. The missing atoms and hydrogen atoms were added using the tleap software from the Amber18 package and the protonation states carefully determined using Propka3.1 program [37,38]. The results indicated that all residues have a standard pKa and all the histidines were protonated at epsilon nitrogen. The compounds were designed using the Marvin Suite software. To ensure enough space for the docking search we defined a cavity with a radius of 15 Å centered on the position occupied by the huprine ligand. For each compound, 20 runs were requested using the ChemScore scoring function, allowing full flexibility of the ligand and the protein active site.

Aβ Aggregation Inhibition
The inhibition of Aβ1-42 aggregation was determined through a ThT-based fluorimetric assay. Initially, the Aβ-peptides (1 mg) were pre-treated with 1,1,1,3,3,3,-hexafluoro-2propanol (HFIP), kept at room temperature overnight, aliquoted, and dried under vacuum. The Aβ-peptides stock solutions (111 µM) were then prepared in ammonium hydroxide 0.1% (NH 4 OH) prior to the fluorescence assay. The stock solutions of all tested compounds (400 µM) were prepared in DMSO, while the stock solution of ThT (400 µM) was prepared in distilled water (1 mL). The Aβ aggregation was monitored using a standard 96-well plates, in the Synergy multi-mode reader (BioTek) and setting the excitation and emission wavelengths at 440 and 485 nm, respectively. Samples were prepared by diluting the stock solutions in PBS pH 7.4 to a final Aβ peptides concentration of 10 µM, 20 µM of ThT and 20 µM of the test compound (maximum final DMSO content: 5% v/v), with a final volume of 100 µL. The assays were performed at room temperature. The fluorescence data was recorded over 24 h with 5 s shaking (360 cpm) prior to each reading and then plotted against the time, using a nonlinear regression. Phenol red was used as the reference compound. Each condition was performed in triplicate, and values were averaged. The estimation of the inhibitory potency (%) was carried out by comparing fluorescence values at the plateau.

Conclusions
In this investigation, we have focused our efforts on the design, synthesis, and biological evaluation of a family of chromeno [3,4-b]xanthones and their respective precursors, (E)-2-styrylchromones, as first-in-class dual-target inhibitors of AChE and Aβ aggregation. We found that almost half of the 21 tested compounds exhibited inhibitory activity against the selected biological targets in the micromolar range, and some of them even showed dual activity towards AChE and Aβ aggregation. Their binding modes towards AChE were determined through molecular docking and the results showed that most of the compounds bound to the enzyme through hydrogen bonds and π-stacking interactions between the main scaffold and Tyr341 and Trp286 aromatic sidechains present in the binding pocket. The best results were obtained for compounds 3d and 4a, which effectively bound to AChE, mostly through the formation of a perfect stacking with its Tyr341 side chain in the first case and through the extensive stacking with three aromatic side chains, Phe295, Tyr337 and Trp86, in the second case. The obtained results for each biological test individually showed that chromeno [3,4-b]xanthones 4a, 4b, 4e, 4f, (E)-2 -propargyloxy-2-styrylchromone 3d and (E)-2-styrylchromone 10 exhibited good inhibition against AChE with IC 50 values 2.1-6.9 µM, while chromeno [3,4-b]xanthones 4b, 4e, 4f and (E)-2-styrylchromone 10 were moderate inhibitors of Aβ aggregation, within a percentage range of 52-70%, at 20 µM. Further investigation to establish the mode of action of such compounds is ongoing in our laboratory, as well as a complete SAR profile with a larger number of analogues. Nonetheless, we have identified a new family of compounds as a template that needs to be developed to find lead molecules. Thereafter, the identification of a lead compound that is currently underwent in our lab, will be furthered by a critical evaluation of its blood-brain barrier (BBB) penetrance and its toxicological effects using specific approaches. Among the tested compounds, chromeno [3,4-b]xanthone 4b and (E)-2-styrylchromone 10 should be outlined as the most promising ones, due to their well-balanced activity towards both selected targets. These two derivatives will be used as templates for the development of more potent multitarget compounds for AD-related targets.