Therapeutic Potential of Quercetin in the Treatment of Alzheimer’s Disease: In Silico, In Vitro and In Vivo Approach

Abstract

Background: The pathophysiology of Alzheimer’s disease (AD) is strongly linked to damage to the cholinergic systems of the central nervous system (CNS), mainly due to the formation of β-amyloid peptide plaques, which trigger intense inflammatory responses and are currently the main cause of the symptoms of the disease. Among the therapeutic strategies under investigation, classes of natural products with immunomodulatory properties, action on the CNS, and potent antioxidant activity, which contribute to neuroprotection, stand out. Methods: We aimed to evaluate the flavonoid quercetin using in silico, in vitro, and in vivo methods for the treatment of AD. Initially, the compounds were selected, and molecular dynamics simulations were performed. The in vitro assays included tests of antioxidant activity (DPPH), enzymatic inhibition of acetylcholinesterase (AChE), and prediction of oral toxicity. The in vivo studies investigated the effects on scopolamine-induced learning deficits and conducted histopathological analysis of the brain. Results: Quercetin showed structural stability in the complex with (AChE), with no significant alterations in the Root Mean Square Deviation (RMSD), SASA and radius of gyration (Rg) parameters. Through the same method it was possible to predict stability between the quercetin and inducible nitric oxide synthase (iNOS) complex, a possible mechanism for quercetin immunomodulation in the CNS. In the AChE inhibition test, the IC₅₀ obtained for quercetin was 59.15 μg mL⁻¹, while in the antioxidant test with DPPH, the concentration of 33.1 µM exhibited 50% of the scavenging of reactive oxygen species. This corroborates the perspective of quercetin having neuroprotective activity. This activity was also corroborated in vivo, in a zebrafish model, in which quercetin reduced the cognitive deficit induced by scopolamine. Histopathological analysis revealed its ability to prevent atrophy, caused by scopolamine, in the nervous tissue of animals, reinforcing the potential of quercetin as a neuroprotective agent. Conclusions: The results of the tests carried out with quercetin suggest that this molecule has antioxidant, AChE inhibitory, and neuroprotective activities, making it a good candidate for use in future clinical trials to ensure its efficacy and safety.

1. Introduction

Alzheimer’s disease (AD) is a neurodegenerative disease that mainly affects the elderly population and is characterized by a progressive decline in cognitive function [1]. AD is one of the most common causes of dementia, accounting for approximately 60% to 80% of cases [2].

AD is clinically characterized by initial memory deficits and cognitive decline that ultimately affect other functional abilities, including speech, behavior, spatial orientation, sleep, and the motor system, associated with neuropsychological manifestations [3]. Pathologically, it is marked by extracellular deposition of β-amyloid (Aβ) and intracellular accumulation of hyperphosphorylated tau, forming senile plaques and neurofibrillary tangles (NFTs), respectively [4]. However, the main pathological features of AD include not only β-amyloid plaques and neurofibrillary tangles, but also neuroinflammation, oxidative stress, and cerebrovascular abnormalities, among others [5].

Several pieces of evidence suggest that oxidative stress plays a fundamental role in the pathogenesis of AD. Oxidative stress occurs early in the course of AD, which would corroborate its role in its pathogenesis [6]. The brain is the most sensitive organ in the body to the damaging effects of free radicals, and this vulnerability increases with age. The association between AD and oxidative stress is widely investigated as a potential therapeutic target [7].

Another mechanism involved in the development of AD is inflammation. A set of many cells, including astrocytes, microglia cells, and proteins such as cytokines and chemokines, cause neuroinflammation and damage the environment around neurons. This leads to their damage and thus contributes to the development of oxidative stress or leads to cellular apoptosis, which causes the onset of symptoms indicative of AD [8].

In neuronal signaling, AChE is also an important target, as it is responsible for the degradation of ACh, which in turn blocks the transmission of the postsynaptic signal. Cholinergic neurotransmission plays a key role in neuronal plasticity and cell survival in the central nervous system [9].

Regarding treatment, the medications currently available are only symptomatic, and aim only to slow the progression of the disease [10]. The symptomatic treatment focuses on three therapies: cholinesterase inhibitors, N-methyl D-aspartate (NMDA) receptor antagonists, and combination therapy. The main cause of AD is neurotransmitter depletion, hence the use of cholinesterase inhibitors designed to increase the amount of ACh. This is achieved by the administration of cholinergic inhibitors (rivastigmine, donepezil, tacrine, galantamine). These inhibitors limit the reduction in ACh concentration in the brain [11].

In recent years, however, therapeutic advances have increasingly focused on disease-modifying strategies, particularly through the development of antibodies aimed at removing beta-amyloid plaques or modulating other pathological targets. Among them, lecanemab (Leqembi) received traditional FDA approval on 6 July 2023, becoming the first beta-amyloid-targeting antibody to obtain such approval [12]. More recently, donanemab (Kisunla) was approved by the FDA on 2 July 2024, further reinforcing the therapeutic potential of this class [13]. In contrast, aducanumab (Aduhelm) was discontinued by Biogen in January 2024, following a strategic re-evaluation by the company [14]. These events highlight the translational relevance of disease-modifying therapies and underscore the need for new experimental approaches.

Recent studies have shown that polyphenols play an important role and are known to protect against cancer, and neurodegenerative and cardiovascular diseases. They act as potent antioxidants and play a defensive role against oxidative stress [15]. Polyphenols are a group of phytonutrients found in fruits and vegetables that have shown promise as facilitators of cognitive enhancement or neuroprotection. In particular, the polyphenolic subclass of flavonoids has been extensively studied for its ability to influence cognition and delay cognitive aging due to its known bioactivity [16].

Quercetin is a flavonoid with important therapeutic and pharmacological properties. The neuroprotective properties of quercetin have been investigated extensively. The molecular structure of the compound is shown in Figure 1, highlighting its flavonoid core. The diagram emphasizes the hydroxyl groups essential for acetylcholinesterase AChE inhibition, the catechol moiety responsible for antioxidant activity, and the regions of electron delocalization that contribute to its chemical reactivity. Key transformations upon the activation of molecular targets, along with brief pharmacokinetic and pharmacodynamic characteristics, are also presented. It reduces cellular toxicity caused by oxidative stress in neurons at low micromolar doses. Also, it inhibits neuroinflammation by inhibiting pro-inflammatory cytokines, including NF-kB and iNOS, while promoting neuronal regeneration [17]. Furthermore, quercetin has been reported to utilize multiple mechanistic targets for neuroprotection in AD [18].

This study sought to better understand the neuroprotective effects of quercetin through in silico, in vitro, and in vivo tests. The results obtained could pave the way for new therapies with greater efficacy and fewer side effects. To illustrate the pathophysiological mechanisms involved in Alzheimer’s disease and the possible neuroprotective pathways of quercetin, including its molecular structure and pharmacological properties, Figure 2 was prepared.

2. Materials and Methods

The compound Quercetin was selected for this study due to its satisfactory results in the predictions of molecular properties, pharmacophoric pattern, molecular docking, and pharmacokinetic and toxicological properties. Thus, it is indicated as a promising prototype for the design of drug candidates for the treatment of AD [19]. Furthermore, its effects as a neuroprotective substance are already well reported in the literature. The Figure 3 below presents a schematic of the methodology used in this study to investigate quercetin as a neuroprotective compound. First, molecular dynamics is employed to simulate and analyze the molecular interactions of the compound of interest. Next, in vitro DPPH and AChE assays are performed to evaluate, respectively, its antioxidant potential and acetylcholinesterase inhibitory activity. Finally, the passive avoidance test is conducted in an animal model to investigate the behavioral effects on memory and learning.

2.1. Molecular Dynamics Procedures

For the molecular dynamics simulations, the GROMACS 2022.3 software was used, applying the Charmm36 force field with periodic boundary conditions (PBC) [20], using the PDB2gmx module of the software itself. The topology of ligands and cofactors were performed by the CGenFF server [21]. The complexes manipulated were AChE (PDB ID: 4EY6) and iNOS (PDB ID: 1NSI), interacting with the previously screened compound: Quercetin, obtained from molecular docking studies [19]. In addition to the complexes, proteins in their APO forms were evaluated

Both systems were solvated with sufficient amounts of water in TIP3P format, sufficient to fill a cubic system with dimensions of 10 nm. The charges were neutralized using 10 sodium atoms for AChE and 2 for iNOS. NVT and NPT simulations were used to balance the system, with a pressure of 1 bar (Berendsen barostat and Parrinello-Rahman pcoupl) and a temperature of 300 K.

The following information was analyzed: solvent access surface area (SASA), radius of gyration (Rg), root mean square deviation (RMSD) and root mean square fluctuation (RMSF). The graphs illustrating the results of these analyses were generated using Excel version 2025.

2.2. In Vitro Tests

2.2.1. DPPH Antioxidant Assay

The in vitro antioxidant assay for quercetin was carried out in collaboration with the Quality Control and Bromatology Laboratory of the Federal University of Amapá. The experiment was coordinated by Professor Doctor Fabio Rodrigues De Oliveira.

The assay was based on the DPPH neutralization method, which is based on the electron donation of antioxidants to neutralize the DPPH radical. The reaction is accompanied by a change in the color of DPPH measured at 517 nm, and the discoloration acts as an indicator of antioxidant activity [22].

The free radical scavenging activity was determined using the 2,2-diphenyl-1-picrylhydrazyl (DPPH) method, as described by Sousa et al. [23], with modifications. DPPH (2,2-diphenyl-1-picrylhydrazyl, ≥95% purity, Sigma-Aldrich, St. Louis, MO, USA) was used to assess free radical scavenging activity. Trolox (≥97% purity, Sigma-Aldrich, St. Louis, MO, USA) was employed as a positive control. Methanol of spectrophotometric grade (Synth, Diadema, SP, Brazil) was used to prepare the solutions.

A methanolic DPPH solution (40 μg mL⁻¹) was prepared and kept protected from light until analysis. Samples were initially dissolved in methanol to obtain stock solutions (1 mg mL⁻¹), which were then diluted to final concentrations of 10, 25, 50, 100, 125, and 250 μg mL⁻¹. Trolox was prepared at the same concentrations and under the same experimental conditions as a positive control.

For each determination, 2.7 mL of the DPPH solution was added to 0.3 mL of the sample or positive control. The mixture was homogenized and kept in the dark at rest for 30 min. The blank consisted of 2.7 mL of methanol and 0.3 mL of each tested concentration solution. Absorbance readings were performed using a spectrophotometer (Biospectro SP-22 was sourced from Biospectro, located in Curitiba, Paraná, Brazil) at 517 nm. Calibration was verified using the Trolox curve, ensuring the linearity of the response across the tested concentration range. IC₅₀ values (with 95% CIs) were obtained by fitting a four-parameter logistic model.

The antioxidant capacity of the substance was expressed by the free radical scavenging potential, using the following formula:

$$Scavenging \left(\%\right)= \left({\displaystyle \frac{Abs sample-Abs blank}{Abs standard-Abs blank}}\right)\times 100$$

% scavenging: (Abs sample − Abs blank) × 100/(Abs standard − Abs blank)

2.2.2. Inhibitory Enzymatic Evaluation in AChE

The enzyme inhibitory evaluation in AChE was carried out in collaboration with the Phytopharmaceutical Nanobiotechnology Laboratory of the Federal University of Amapá. The experiment was coordinated by Professor Doctor Rodrigo Alves Soares Cruz.

In order to verify the quantitative inhibitory activity of the samples, a spectrophotometric assay was performed based on Ellman’s colorimetric method (1961) as described by Rhee et al. [24], with adaptations. The colorimetric test occurs according to the enzymatic reaction of hydrolysis of the enzyme substrate acetylthiocholine iodide (ATCI) catalyzed by the enzyme AChE. The hydrolysis product, thiocholine, reacts with the colorimetric agent, 5,5′-dithiobis (2-nitrobenzoic) (DTNB), promoting the formation of a yellow-colored substance, thionitrobenzoic acid (TNB).

The reagents used in this study were obtained from recognized commercial suppliers. AChE enzyme from electric eel (Electrophorus electricus, Sigma-Aldrich) with analytical-grade purity (≥95%) was used. Bovine serum albumin (BSA) fraction V, used as a stabilizer, was supplied by Merck (Darmstadt, Germany), with a minimum purity of 96%. DTNB (5,5′-dithiobis(2-nitrobenzoate)) and acetylthiocholine iodide (ATCI) were obtained from Sigma-Aldrich with a purity of ≥98%. Triton X-100 detergent, of reagent grade, was purchased from Merck. Methanol and acetone, used for solubilizing physostigmine and quercetin samples, were from J.T. Baker (Radnor, PA, USA), and were HPLC grade, ensuring high purity.

The monobasic sodium phosphate-buffer solution was prepared using analytical grade reagents, and the pH was adjusted with analytical grade NaOH, both from Sigma-Aldrich. Purity specifications and other characteristics of the reagents were verified using the respective technical data sheets and commercial catalogs provided by the manufacturers.

Accordingly, a 25 mL solution was prepared with AChE enzyme (4.7 mg), containing 25 mg of bovine serum albumin fraction V. Monosodium phosphate monohydrate solution was made with 13.79 g dissolved in 950 mL of distilled water. The pH was adjusted to 7.5 with a NaOH solution. This buffer solution was used to dissolve DTNB (134.75 mg) and Triton-X (1.7 g), obtaining two 500 mL solutions.

Acetylcholine iodide solution was prepared at a concentration of 17 mM for use at the time of the experiment, without storage.

Physostigmine and quercetin samples were solubilized in methanol or acetone solvent. A total of 25 μL of these solutions were added to 475 μL of 0.1 M phosphate buffer, pH 7.5. We used 50 μL of the enzyme solution (final activity around 30 mU); 250 μL of the 0.68 mM DTNB solution (final concentration 0.2 mM); and 50 μL of the 17 mM ATCI solution in water (final concentration 1 mM). The reagents were mixed in this order, and the readings were taken on a spectrophotometer at 412 nm with a 1 cm cuvette. Water was used as a blank and methanol or acetone as a control. Readings in the absence of the enzyme were taken to check for the occurrence of parallel reactions. The initial reaction rate was measured over the first 5 min of the reaction. The enzyme activity measurement was calculated according to the following equation:

$$\mathrm{A}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{i}\mathrm{t}\mathrm{y} (\mathrm{U}/\mathrm{m}\mathrm{L})={\displaystyle \frac{\Delta \mathrm{A}\mathrm{b}\mathrm{s}/\mathrm{m}\mathrm{i}\mathrm{n}\times \mathrm{V}}{\mathsf{\epsilon } \mathrm{b}/1000}}$$

where ΔAbs/min: Variation in absorbance per minute. ε: molar absorptivity (14,150 M⁻¹ cm⁻¹) of thionitrobenzoic acid at 412 nm, formed from the reaction of thiocholine and DTNB. b: optical path of the cuvette.

The calculation of the percentage of inhibition of the samples or standard in relation to the control was performed as follows:

$$\% ofInhibition=100-\left(A2÷A1\right)\times 100$$

where A1: Control activity (mU); A2: Sample activity (mU). IC₅₀ values were calculated in molar concentration and are presented with 95% confidence intervals.

2.3. Prediction of Oral Toxicity

The prediction of the median lethal dose (LD₅₀), using the SMILES (Simplified Molecular Input Line Entry System) format, was performed on the ProTox 3.0 web server (version 3.0, https://comptox.charite.de/protox3/index.php?site=home (accessed on 23 July 2025). Results were expressed according to the GHS (Globally Harmonized System of Classification), which includes six categories: the first five indicate increasing degrees of toxicity, and the sixth refers to non-toxic compounds [25,26,27].

ProTox 3.0 predicts acute oral toxicity based on 2D structural similarity comparisons and the recognition of toxic fragments from approximately 38,000 compounds, providing an estimated accuracy by comparing the test set with the submitted input. In addition to LD₅₀, the platform allows the prediction of multiple toxicological endpoints, including organ-specific toxicity, molecular initiating events (MIEs), adverse outcome pathways (AOPs), and toxicity targets, all derived from QSAR models validated with independent external sets, demonstrating robust performance (accuracy~75–90%) [28].

Limitations: Despite their utility, QSAR models depend on the quality and coverage of the training data, may have limited generalizability to compounds outside the training set, and cannot fully capture the complexity of biological mechanisms underlying toxicity [29,30,31].

2.4. In Vivo Tests

This study was conducted in accordance with the ARRIVE 2.0 guidelines for research involving animal models, ensuring methodological rigor and adherence to ethical principles in experimentation [32]. The in vivo evaluation of quercetin followed the methodology described by Kim et al. [33]. The experimental protocol was approved by the Ethics Committee on Animal Use (CEUA) of the Federal University of Amapá under number 07/2021. It was carried out at the institution’s Zebrafish Platform. For transparency and verification purposes, the checklist confirming compliance with the ARRIVE 2.0 guidelines is available as Supplementary Material for this article.

2.4.1. Substances Used

The substances used in the in vivo assays, including scopolamine, physostigmine, and quercetin, were obtained from Sigma-Aldrich (St. Louis, MO, USA). Scopolamine and physostigmine had a purity of ≥98% (HPLC grade) and quercetin had a purity of ≥95% (HPLC grade), according to the technical data sheets provided by the manufacturer. Catalog and batch numbers were not specified. The treatment was performed by immersion. The compounds were diluted in the following concentrations: scopolamine 200 μM (60.67 µg/mL), physostigmine 20 μM (5.51 µg/mL) and quercetin 20 μM (6.05 µg/mL), in distilled water, and then diluted to the final concentrations in a container with water taken from the aquarium. The zebrafish were placed individually in a 250 mL beaker filled with 100 mL of water containing the drug for administration.

2.4.2. Animals

Adult zebrafish were obtained from a specialized supplier (Power Fish, located in Rio de Janeiro), and subsequently acclimated in the laboratory for a period of 2 weeks prior to the experiments. They were kept at a temperature of 28.0 ± 1.0 °C with a 14 h light, 10 h dark cycle. In the aquariums where the fish were kept, the tap water passed through a multi-stage filtration system, equipped with a sediment filter, a post-carbon filter, and a fluorescent UV light sterilizing filter. The water in the aquariums was aerated and kept at pH 6.5–7.5. Total ammonia was below 0.05 mg/L, and nitrite was below 0.02 mg/L. They were fed three times a day with flake feed suitable for fish.

2.4.3. Experimental Aquarium

A glass aquarium (18.5 cm long × 7.0 cm wide × 10.0 cm high) was divided into two compartments by a circular door that was 3 cm in diameter. One side of the aquarium was black (except for the top). The other side was white, with a transparent circular door that was 3 cm in diameter to allow the entry of a flashing light [33].

2.4.4. Passive Avoidance Response Test

One zebrafish was placed in the dark compartment of the aquarium and allowed to acclimate for 3 min. After acclimation, a flashing light was turned on and the sliding divider was removed. The time taken for the zebrafish to cross into the white compartment was measured from the moment the partition was removed until the fish fully passed through, by using a stopwatch.

As an aversive stimulus, a small stone (~0.5 g) was gently dropped in front of the zebrafish 3 s after it crossed the divider. The stone was released from a height of approximately 5 cm using a custom-made apparatus to ensure consistent trajectory and minimal variability. The fish was then carefully returned to the dark compartment using a small net. This crossing procedure was repeated at 3 min intervals twice more for acclimatization. Each crossing was considered a test, and a training session consisted of three tests. The maximum allowed crossing time was 300 s; fish exceeding this limit on the first test were excluded from further trials. For each test session, a single crossing was performed 2 h after the end of a training session.

The pebble-drop method has been previously described as a reliable aversive stimulus capable of eliciting robust avoidance learning in zebrafish [33]. In the present study, we provide full methodological details of the stimulus (apparatus dimensions, stone mass, release height, timing, and duration) to ensure reproducibility and compliance with ARRIVE 2.0 guidelines.

2.4.5. Experimental Procedure

To evaluate the passive avoidance response test immersion treatment, the compound was diluted in the following concentration: scopolamine 200 μM, physostigmine 20 μM and quercetin 20 μM. The negative control was immersed only in distilled water. A 250 mL beaker with 100 mL of water containing the substances was used. Each group was exposed to its respective compound for 60 min. Then, each fish was individually subjected to the passive avoidance response test.

The division of the groups (n = 15/group) is shown here. Group I—negative control, treated with distilled water. Group II—positive control, treated with muscarinic receptor antagonist (scopolamine 200 µM + distilled water), in which the fish were immersed for 1 h in scopolamine and then 1 h in distilled water. Group III—group treated with AChE inhibitor (physostigmine 20 µM). Group IV—experimental group treated with scopolamine + physostigmine (scopolamine 200 µM + physostigmine 20 µM), in which the zebrafish were immersed for 1 h in scopolamine and then 1 h in physostigmine. Group V—experimental group treated with Quercetin (quercetin 20 µM). Group VI—Experimental group treated with Scopolamine + Quercetin) (scopolamine 200 µM + quercetin 20 µM), in which the fish were immersed for 1 h in scopolamine and then 1 h in quercetin. One hundred and twenty minutes after the treatments, the animals were subjected to training. The doses used in this study were chosen as described by Kim et al. [33].

2.4.6. Ethical Considerations and Experimental Design

The allocation of animals to experimental groups was performed by simple randomization. Data collection and the outcome assessment were conducted by a researcher who was blinded to the group assignments. The sample size (n = 15/group) was determined based on previous studies using similar protocols [33], ensuring a sufficient number of animals to detect biologically relevant differences, in accordance with the ethical and rational use of animal models. Humane endpoints were predefined, including signs of severe stress (e.g., prolonged immobility, loss of balance, refusal to feed, or abnormal swimming behavior), in which case the animals would be immediately removed from the experiment. No unexpected mortality was observed.

2.5. Histopathological Analysis

For histopathological analysis of the brain, the animals were fixed in Bouin’s solution for 24 h, decalcified in EDTA solution (ethylenediamine tetraacetic acid, Sigma Co., São Paulo, Brazil) for 24 h. The samples were dehydrated in a series of alcohols (70, 80, 90 and 100%). Subsequently, they were diaphonized by impregnation with xylene and embedded in paraffin. The samples were sectioned at 5 µm using a microtome (Brand Rotary Microtome Cut 6062, Slee Medical, Nieder-Olm, Germany), and histopathological analysis was performed after the tissue sections were stained with hematoxylin and eosin, as described by Melo [34]. The images were analyzed using an Olympus BX41-Micronal Microscope (Olympus, Tokyo, Japan) and photographed using an MDCE-5C USB 2.0 (digital) camera.

2.6. Evaluation of Histopathological Alterations

The Histopathologic Alteration Index (HAI) was calculated from the levels of tissue alterations observed in the brain. Alterations can be classified into levels I, II and III, and the HAI value indicates whether the organ is healthy (0 to 10), has mild to moderate alterations (11 to 20), has moderate to severe alterations (21 and 50), or contains irreversible alterations (>100) [35]. Level I alterations are edema and vascular dilation; level II alterations are neuronal atrophy, gemistocytic astrocytes, central chromatolysis, neuronal atrophy, adipose granule cells, hyperemia and cerebral aneurysm; level III alterations are necrosis [34]. Thus, the indexes were calculated according to the following equation:

$$I={\displaystyle \frac{{\displaystyle {\sum }_{i-1}^{na}ai}+10{\displaystyle {\sum }_{i-1}^{nb}bi+{10}^2{\displaystyle {\sum }_{i-1}^{nc}ci}}}{N}}$$

where

• a: first stage alterations
• b: second stage alterations
• c: third stage alterations
• na: number of alterations considered to be first stage
• nb: number of alterations considered to be second stage
• nc: number of alterations considered to be third stage
• N: number of fish analyzed per treatment

2.7. Statistical Analysis

For the in vivo experiments, the results were expressed as mean ± Standard Error of the Mean (SEM) of each experimental group. Comparisons between group means were performed using ANOVA, followed by Tukey’s post hoc test. Additionally, a paired t-test was applied to analyze time differences within the same groups. The significance level was set at 5% (p < 0.05), and all analyses were conducted using GraphPad Prism^® (version 10).

For the histopathological analysis, results were expressed as mean ± Standard Deviation (SD) of each experimental group. Comparisons between groups were performed using ANOVA, followed by Tukey’s test, with a significance level of 5% (p < 0.05). Analyses were conducted using GraphPad Prism^® (version 5.03).

3. Results

3.1. Analysis of Molecular Dynamics Simulations

With the final molecular docking conformations [19] for the complexes, it was possible to perform molecular dynamics, analyzing the conformational stability between the quercetin ligand and the targets AChE and iNOS, and investigate their interactions. The conformation of the proteins studied were obtained from the PDB, passing through simulations of relaxation of the bindings in the GROMACS system itself.

The results of RMSD, RMSF, SASA and Radius of Gyration for AChE APO and complexed with the studied ligands are shown in Figure 4.

The analysis of the RMSD graph of the complexes and APO showed that these systems did not present instability in 100 ns. The variations did not exceed 0.1 nm of difference between the initial poses. APO (black) remains the same and shows no tendency for change, as does the quercetin complex (yellow). The greatest changes were for the physostigmine complex (blue), which in 20 ns changed by approximately 0.05 nm.

Research by Gao et al. [36] demonstrates a variation of approximately 0.1 nm for the AChE complex and its clinical inhibitor rivastigmine in 200 ns, using GROMACS and CHARMM force field. The variation occurred between 0 and 50 ns and then the complex remained stable. Another simulation between donepezil and AChE was performed with the Desmond software version 5.6 with the OPLS2005 force field, and the result showed a variation of 0.2 nm between the initial and final conformation in 100 ns, within an acceptable range for AChE ligands [37]. These results resemble the variations in physostigmine (blue).

Figure 4B of RMSF shows that the largest variations, which were above 0.3 nm, were for the regions of residues 80–90; 250–270; and 380–390. In the first region (80 to 90), all ligands demonstrated deformation. In this region, there is the residue Trp86—responsible for π-π interactions with the aromatic portions of such ligands. This interaction was predicted in molecular docking.

Quercetin (yellow) appears to alter this residue less. This may be because the hydroxyl at C3 of quercetin interacts with other residues around the site, leaving it in a conformation less favorable to form a parallelism with the Trp86 ring. This residue is extremely important because it interacts with the cationic portion of acetylcholine at the time of its cleavage, and the region in which it is located is part of a highly flexible portion [38].

Physostigmine is the only ligand to cause a relevant deformation in the region from 250 to 270. This region can undergo large changes, probably due to interactions of the ligands with residues Trp286 and Ser293.

The SASA and Rg graphs demonstrate the stability of the complex and APO in the simulated system. Therefore, both results showed high constancy in the variation in the graphs. This can be interpreted as indicating that the structures remained compacted and the protein did not break in the established system.

iNOS was evaluated with its APO form and complexed with quercetin and an ITU inhibitor (2-ethyl-2-thiopseudourea). The initial poses were the results of molecular docking performed by de Souza et al. [19]. The results of RMSD (Figure 5A), RMSF (Figure 5B), SASA (Figure 5C), Rg (Figure 5D), and RMSD of the ligands only (Figure 5E) are shown in Figure 5.

The analysis of the first RMSD figure of the protein with the complexes shows that there was stability for the APO protein (black) and the complexes with quercetin (yellow) and ITU inhibitor (blue).

A simulation using GROMACS and the CHARMM force field27 was published [39]. The quercetin 3,7-dirhamnoside molecule was docked in the active site of iNOS and the stability was evaluated by molecular dynamics for 100 ns. This quercetin analog showed stability calculated by RMSD, similar to the quercetin result of this thesis. Both had a fluctuation up to 40 ns and stabilized completely between 50 and 100 ns, with an average equilibrium around 0.250 nm. It is noteworthy that this research also evaluated the in vitro capacity of the quercetin analog to regulate iNOS. The result demonstrated that this molecule has the capacity to reduce the production of nitric oxide (NO) in the macrophage line RAW-264.7.

In general, quercetin did not show a significant difference in residue changes, as it presented a profile very similar to that of the APO protein (black). Boumezber and Yelekçi [40] report similar results to those described for Figure 5B. This indicates a lack of stronger interactions between the ligands and the active site of the molecule or even a disturbance in the integrity of the ligand structure.

However, the Rg and SASA results prove that the system with complex and APO remained intact during the 100 ns analyzed.

Quercetin (yellow) demonstrated good stability at the iNOS site, according to the established method. The ITU inhibitor showed regular stability at the beginning, but after 80 ns it presented a large fluctuation and then continued with little change.

Better optimization and standardization in research with the iNOS enzyme are important, since its mechanism of nitric oxide production from L-arginine involves at least two cofactors—heme domain and tetrahydrobiopterin [41]. Likewise, few selective clinical inhibitors for NOS have been evaluated by molecular dynamics and published in the scientific literature. Therefore, more robust validations of in silico models for this enzyme are needed.

The molecules showed greater stability in the active site of the AChE enzyme, with fluctuations not exceeding 0.3 nm and not breaking their intramolecular bindings. It was similar to what occurred in molecular docking, as the molecules obtained higher score values for AChE in relation to iNOS. Although these compounds exhibited a greater number of interactions with iNOS, this was not noted in molecular dynamics according to the RMSF graph, evidencing greater emphasis on the interaction with AChE.

The results obtained through molecular dynamics calculations give more consistency to the results of subsequent studies that investigated the interaction of the studied molecules with AChE enzymes. This is due to the greater consistency observed in the results of theoretical studies in relation to the molecules in question. Thus, in vitro and in vivo experiments were carried out in order to evaluate the degree of interaction between these structures.

3.2. Results of In Vitro Tests

3.2.1. DPPH Antioxidant Activity Results

Quercetin (Figure 6) exhibited significant antioxidant activity in the DPPH assay: radical scavenging exceeded 50% at the lowest tested concentration (10 µM, equivalent to 3.02 µg/mL) and reached over 90% inhibition at concentrations of 50 µM (15.11 µg/mL). The IC₅₀ was determined to be 18.7 µM (5.65 µg/mL), with a 95% confidence interval of 17.2–20.3 µM. The difference compared to the control was statistically significant, with a large effect size. In contrast, physostigmine showed an IC₅₀ above the tested range, with a maximum mean inhibition of 46.89 ± 2.64%, remaining below 70% even at the highest tested concentration (Figure 7).

Figure 7 shows that quercetin presented a positive result in relation to its tested antioxidant capacity. This antioxidant potential of polyphenols has already been widely evaluated in plant extracts and in foods rich in this phytochemical class. Also, theoretical physical–chemical studies demonstrate that flavonoids have characteristics that are favorable to antioxidant activity, due to their electronic density and high electron delocalization [42,43,44].

Samra et al. [45] described the chemical characteristics through which flavonoids have a propensity for high antioxidant activity. The three main structural patterns for this effect are as follows: (I) ring B being a catechol; (II) an unsaturation between carbon two and three of ring C—so that there can be a delocalization between the electrons of ring B and ring C; and (III) hydroxyls in ring A and C are of great importance because they act as radical-scavenging groups and stabilize the structure when there is delocalization of electrons. When the carbonyl group of ring C is in a vicinal position with a hydroxyl, it can exert greater stability on the molecule at the time of proton transfer by such hydroxyl [45].

Tian et al. [46] found similar results to our study in the DPPH assay for quercetin. The authors demonstrated satisfactory DPPH radical scavenging activity. Quercetin is more effective than physostigmine in its DPPH radical scavenging activity. It exhibits better values than the antioxidant standard BHT (butylhydroxytoluene); the flavonoid kaempferol, which lacks the catechol structure in the B ring; and apigenin, another flavonoid without the catechol ring and the hydroxyl in position three of the C ring [45].

This result and the sources found in the literature define quercetin as a potential compound for the control of oxidative stress. Also, the scaffold of this polyphenol can serve as a basis for future structural modifications that result in improvements, potentially in the metabolic aspect. This is because this class is easily susceptible to enzymatic degradation when administered orally [42].

Despite its already established effects in AD therapy, the antioxidant activity of physostigmine is not widely investigated in the literature. However, the results expressed in this research suggest that there is a low capacity of this molecule to sequester the DPPH- radical or transfer a proton to neutralize reactive species. Some compounds containing the carbamate group lack the direct ability to disrupt the balance of oxidative stress. However, carbamates can bind to the antioxidant molecule-reduced glutathione (GSH), transforming it into oxidized glutathione (GSSG), or even interact with enzymes that regulate the glutathione pathway, demonstrating a tendency to induce oxidative stress [47,48]. Therefore, there is a need for more robust investigations between physostigmine and the glutathione antioxidant pathway.

3.2.2. Inhibitory Activity on AChE

The results generated for AChE inhibition by quercetin are shown in Figure 8, as well as the result expressed by the standard enzyme inhibitor, physostigmine. The results are expressed in IC₅₀ (μg mL⁻¹) and the graphs reflect the % of activity by the concentration of the compounds evaluated.

In the acetylcholinesterase inhibition assay, physostigmine exhibited an IC₅₀ of 0.56 µg/mL (1.77 µM) and was used as a reference compound. Quercetin showed an IC₅₀ of 59.15 µg/mL (195.6 µM), with a 95% confidence interval. The inhibition by quercetin was statistically significant compared to the control. This suggests that quercetin can inhibit the catalysis of acetylcholine at a relevant concentration.

Xie et al. [43] evaluated the AChE inhibitory activity of 20 flavonoids, aiming to evaluate a structure–activity relationship. Quercetin was one of the flavonoids with the lowest IC₅₀. The other compounds with low IC₅₀ also presented a hydroxylation pattern in ring A. In contrast, the authors did not find a correlation indicating a greater anticholinesterase activity for flavonoids with a higher degree of hydroxylation in ring B, with hydroxyls in ring A being more relevant. Comparing quercetin with the IC₅₀ result of kaempferol with a single hydroxyl in ring B, the inhibitory activity in AChE increased by 6.3 times for quercetin that has a catechol as ring B. This shows a high activity, but not high enough when compared with baicalein that does not have any hydroxyl in ring B; however, ring A is a pyrogallol [43].

Research by Liao et al. [44] reported that the unsaturation of the binding between carbon C2 and C3 in quercetin may be important for the inhibitory activity in AChE. This is because, when comparing this molecule with its analog taxifolin, the IC₅₀ of quercetin is much lower than that expressed by its analog without C2–C3 unsaturation. The authors also report results of a kinetic test of quercetin inhibition on the AChE enzyme. The results demonstrated a mixed competitive type of inhibition, without an inhibition that completely inhibited the enzyme [44].

Therefore, when comparing the results of other researchers and those of this research, it is clear that quercetin is a promising molecule for drug development, as structural modifications can be made to its structure to improve its inhibitory activity on the enzyme in question. The structure–activity relationship analyzed provided information that helps in the modeling of new compounds based on quercetin. They include analogs with more hydroxyl groups in ring A and maintenance of unsaturation between C2 and C3, which increases the activity in AChE and eliminates the existence of diastereoisomers. Also, as demonstrated in (Figure 8), the presence of hydroxyl in the C4 carbon position can have a positive influence on the inhibition activity of AChE, considering its IC₅₀.

3.3. In Silico Evaluation of Lethal Dose

The results for quercetin are shown in Figure 9, as well as the results of average similarity with the server database and accuracy of the results.

Quercetin showed significant accuracy results, indicating that the molecule is part of the server’s molecular library. The acute toxicity classification of quercetin was class III, referring to the name: toxic by ingestion (50 < LD₅₀ ≤ 300). In the data report, quercetin was one of the flavonoids with the highest toxicity rate, despite having a wide range of LD₅₀. In the study by Lucida et al. [49], the dose of 1600 mg/kg was not lethal to the rats evaluated during the 24 h observation period. However, it caused sound insensitivity and changes in heart rate. It is noteworthy that the authors tested quercetin as a dispersed solid mixed with polyvinylpyrrolidone (1:9).

Another study revealed an LD₅₀ in rats of 161 mg/kg [50]. This research reported the LD₅₀ of quercetin, gallic acid, and curcumin in zebrafish embryos. The results led to the conclusion that quercetin was the most tolerable among these natural products tested. Microscopic analysis revealed that quercetin was responsible, at its highest dose, for pericardial edema and irregular spinal curvature in zebrafish embryos, but the authors suggest further evaluation.

The main toxicological evaluations are related to extracts [51], further demonstrating the need for LD₅₀ definitions for the compound in its isolated form.

The toxicological properties of the compounds were assessed using the ProTox 3.0 platform, which performs computational predictions based on the quantitative structure–activity relationship (QSAR). Although these predictions provide a preliminary overview of the toxicological profile, we highlight the inherent limitations of the in silico approach, including the need for future experimental validations to confirm the results.

Our study conducted in vivo tests to evaluate the neuroprotective effects of quercetin, using adult zebrafish as an experimental model. While these assays provide important information about the compound’s biological activity, further studies are needed to assess pharmacological safety in clinical models. Therefore, future clinical investigations are essential to confirm the tolerability and toxicological profile of quercetin, ensuring its safe use in the development of therapies for neurodegenerative diseases.

To better illustrate the structure–activity relationship (SAR), only quercetin was evaluated in this study. However, a comparative scheme including quercetin, taxifolin, and kaempferol, presented in Figure 10, allows for a visualization of subtle structural differences that may influence biological activities, serving as a reference for interpreting their antioxidant and enzyme inhibitory effects.

3.4. Evaluation of In Vivo Tests

This study investigated the effects of quercetin on learning deficits induced by scopolamine. Zebrafish (Danio rerio) were used as an animal model, since they share orthologous genes with those of humans, including important regions for the expression of proteins relevant to AD, such as: β-secretase decoding genes BACE-1 and BACE-2; microtubule-associated tau protein; and the APOE gene. More detailed genomic evaluations revealed that zebrafish is an excellent model for evaluating BACE inhibitors, as well as selectivity between BACE-1 and BACE-2 [47].

The passive avoidance experimental procedure resulted in the following mean crossing times (seconds) in the training session for the negative control: 85.13 ± 12.9, 113.06 ± 14.9, and 115.4 ± 19.06 (mean ± SEM/n = 15). Meanwhile, in the test session the result was 170.86 ± 10.1 (mean ± SEM/n = 15), performed 2 h after the training session (Figure 11a). Trainings one (p < 0.001), two (p = 0.0046) and three (p = 0.0031) exhibited different crossing times in relation to the test crossing time (paired t test).

The increased crossing time in the test session, compared to the training session, suggests that the zebrafish learned and retained the avoidance response. These results indicate the formation of aversive memory, since the animal avoided crossing the hole, remembering the previous aversive stimulus (shock with the stone).

The positive control group, treated with scopolamine for 1 h and distilled water for 1 h before the training session, demonstrated the following mean ± SEM of crossing time (seconds) in training one, two and three: 176.4 ± 6.4, 174.6 ± 6.3 and 183.6 ± 4.39, respectively. In the test session, the result was 57.7 ± 3.34 (Figure 11b). The comparison between the training and the negative group test shows a difference (paired t test, p < 0.001, n = 10), indicating that treatment with scopolamine alone impaired the acquisition of the avoidance response after two hours.

This alkaloid is a competitive and non-selective antagonist of muscarinic receptors and is widely used as an inducer of neurotoxicity in models of cognitive impairment in preclinical studies. Scopolamine administration in healthy individuals was investigated and they showed learning deficits, impaired passive avoidance response, and impaired object discrimination, as well as attention. These impairments are caused by the influence of scopolamine on the cholinergic system, making it difficult to encode new memories [48]. This is similar to the finding in the test carried out with the zebrafish model.

The group treated only with physostigmine 1 h before the training session presented significantly lower crossing times during the training session (113.4 ± 3.14 s, 117.3 ± 3.6 s and 108.6 ± 2.7 s) compared to the test session, 2 h later, which increased the crossing time (163 ± 4.94 s, paired t test, p < 0.001, n = 15) (Figure 11c). This result is similar to the study by Cho et al. [52], in which the administration of physostigmine was shown to attenuate anxiety-like effects in a zebrafish model. Also, in the research by Kim et al. [33], in zebrafish physostigmine, it was able to reduce the learning deficit effects caused by scopolamine.

In the next group, the animals were treated for 1 h with scopolamine and then placed in another beaker with physostigmine for another hour. Crossing times were 107.2 ± 3.77 s, 105.93 ± 4.98 s and 101.33 ± 3.81 s in the training session and 154 ± 5.59 s in the test session (Figure 11d). The results demonstrate that the crossing time in the test session was significantly greater (paired t test, p < 0.001, n = 15). This suggests that the scopolamine-induced learning deficits were reversed by physostigmine treatment.

Thus, the ability to retain the avoidance response appears to have fully recovered under these conditions, since the crossing time in the test session was greater than the three times in the training session.

In the group treated with quercetin 1 h before the training session, the crossing times were 76.67 ± 2.21 s, 75.80 ± 1.64 s and 76.20 ± 1.81 s, while in the test session they were significantly increased to 206.33 ± 3.88 s (Figure 11a). This result indicates that quercetin enhanced memory retention, as shown by the longer crossing time in the test compared to the training sessions (paired t test, p < 0.001, n = 15).

Similarly, in the group treated with scopolamine for 1 h and subsequently exposed to quercetin for another 1 h, the crossing times during training were 77.53 ± 2.38 s, 73.53 ± 1.66 s and 76.87 ± 2.30 s. In the test session, the crossing time significantly increased to 158 ± 4.16 s (Figure 11b, paired t test, p < 0.001, n = 15). These findings suggest that, even under scopolamine-induced memory impairment, quercetin was able to improve avoidance memory. Although the effect was less pronounced than in the group treated with quercetin alone, these animals were still able to remember the aversive stimulus and avoid the associated area. Therefore, a longer crossing time in the test session is indicative of stronger aversive memory.

Statistical analysis revealed that crossing times were significantly increased in the quercetin and physostigmine groups compared to the control (Figure 12, *** p < 0.001).

The study by Richetti et al. [53] reached similar results regarding quercetin as a modulator of cognition deficit. However, the method of modeling AD in zebrafish was different, using scopolamine as an inducing agent and quercetin was administered intraperitoneally. Quercetin demonstrated an activity of preventing the effect of scopolamine, increasing the avoidance time when compared to the control group. This is a model of evaluation of learning and memory in rats. Pretreatment with quercetin decreased forgetfulness-related behavior, and histological tests demonstrated restoration of hippocampal cell proliferation that was impaired due to chronic dexamethasone administration [54]. It is worth noting that neuroprotection investigations in zebrafish were carried out and quercetin exhibited immunostimulating properties in the central nervous system and increased the resistance of the animals. Also, inflammation markers decreased with the administration of quercetin [55].

In summary, scopolamine, a muscarinic receptor antagonist, induced cognitive deficits in zebrafish, demonstrated by the reduced crossing time in the passive avoidance task that indicates memory impairment. When treated with physostigmine, the animals showed recovery in performance, suggesting a reversal of cognitive deficits. Quercetin, administered before the task, significantly increased avoidance time in both scopolamine-treated and control zebrafish. These results indicate that quercetin may have neuroprotective and cognition-modulating effects. They are similar to other studies in Alzheimer’s models, suggesting its therapeutic potential for the treatment of neurodegenerative conditions.

3.5. Histopathological Findings

The zebrafish’s nervous system extends throughout the body as an interconnecting system of integration centers and communication pathways: neurons and their axonal and dendritic processes. The largest concentrations of nervous tissue are in the brain and its posterior extension, the spinal cord, which together comprise the central nervous system (CNS). The peripheral nervous system (PNS) comprises the nerves that exit the CNS and their nerve endings or special sense organs [56].

According to Baatrup [57], fish depend on an intact nervous system and sense organs to perform behaviors relevant to their survival, especially searching for food, recognizing predators, communication, and orientation. However, the nervous system is extremely vulnerable to xenobiotics, due to its high lipid content and high mitochondrial activity. Lesions in this tissue may drastically alter the behavior and, consequently, the survival of the fish [34].

Studies evaluating the histopathology of the zebrafish brain caused by oral treatment or the immersion of substances of natural or synthetic origin are still scarce. The identification of histopathological changes in this work was possible thanks to a study carried out by Melo and Carvalho [34] that identified and classified the changes into three levels of severity.

In the histopathological study of the brain of group C (control treated with maintenance water), no changes were recorded. In the CE group (treated with a concentration of 0.88 mg/mL of scopolamine), the highest HAI was observed, equal to 23.4 ± 3.71 (Figure 13). This characterizes this organ with moderate to severe alterations, capable of compromising the normal functioning of this organ. The alterations observed were vascular dilation (I), hyperemia (II), neuronal atrophy (II), and cerebellar aneurysm (II) (Figure 14). According to Roberts and Ellis [56], histopathological changes related to neurons are rarely observed in the nervous system of teleosts. In this study, neuronal changes, such as atrophy, were only observed in animals treated with scopolamine.

Groups F (physostigmine), EF (scopolamine + physostigmine), Q (quercetin) and EQ (scopolamine + quercetin) presented low HAI (1.4 ± 0.54, 1.8 ± 0.44, 1.2 ± 0.44, 1.8 ± 0.44, 0.8 ± 0.44, 1.8 ± 0.44, 1 ± 0.70, 1.8 ± 0.55, respectively). This characterizes the organ as normal, since according to the index, the observed alterations are not capable of altering the normal functioning of the organ. Only level I alterations were observed, such as edema and vascular dilation.

Roberts and Ellis [56] emphasize that neuronal dysfunctions in teleosts are related to histopathological changes in the meninges and blood vessels. In this study, blood vessels located in the nervous tissue underwent dilation, and in the group of animals exposed only to scopolamine, cerebellar aneurysm formation occurred.

4. Discussion

The recent literature suggests that quercetin exerts multifactorial neuroprotective effects, making it a promising candidate for the prevention and attenuation of cognitive decline associated with neurodegenerative diseases. Evidence from in vitro and in vivo studies points to antioxidant action, modulation of inflammatory pathways, mitochondrial protection, and inhibition of pathological protein aggregation, among others. Recent reviews summarize these mechanisms and highlight that, although the preclinical body of evidence is robust, clinical evidence remains preliminary [58,59].

Meta-analyses and reviews of preclinical studies show consistent cognitive improvement effects in different AD models (e.g., Aβ-induced, scopolamine-induced, transgenic models). They show reductions in oxidative stress, decreased Aβ deposition, and improved performance in behavioral tests. Individual studies using scopolamine models report memory recovery and associated biochemical parameter improvements. However, heterogeneity in doses, administration routes, and formulations complicates direct extrapolations to humans [60,61].

Clinical studies in humans are limited, with small sample sizes and populations of elderly individuals in preclinical stages. Examples include trials with onion powder rich in quercetin at an approximate dose of 50 mg of quercetin per day, which showed signs of cognitive preservation or improvement in subgroups after continuous intake. Nevertheless, these trials were generally not conducted in patients with an established diagnosis of dementia, and present methodological limitations. Therefore, from a translational perspective, clinical results are promising but still preliminary [62,63].

This study observed that quercetin exhibited a significant theoretical affinity for the catalytic site of AChE, with ligand–target complex stability comparable to that of physostigmine, as indicated by RMSD, SASA, Rg, and RMSF analyses. This is consistent with previous studies reporting quercetin’s ability to interact with critical AChE residues, resulting in the inhibition of acetylcholine hydrolysis and the potential enhancement of cholinergic transmission in AD models. Some earlier studies also describe that this interaction is associated with hydroxyl groups in the B ring, which favor hydrogen bonding with the active site, supporting the SAR rationale discussed in the literature [44,64].

Molecular dynamics simulations with iNOS demonstrated complex stability with quercetin, although RMSF analysis revealed that interaction with AChE promoted changes in a greater number of relevant residues compared to iNOS. This result suggests that, although quercetin may have modulatory potential on iNOS, its more pronounced action may be related to cholinesterase inhibition. Recent animal model studies indicate that quercetin’s modulation of iNOS can reduce neuroinflammation, but its direct impact on cognitive parameters seems less consistent than the cholinergic pathway [65,66,67].

The antioxidant assay with DPPH confirmed the strong free radical scavenging potential of quercetin, promoting more than 50% inhibition. This value is consistent with the literature associating its antioxidant activity with radical stabilization and transition metal chelation. Considering that oxidative stress is a major contributor to AD pathogenesis, this effect constitutes an additional pillar of its neuroprotective action.

Regarding in vitro inhibition of AChE, the observed IC₅₀ value (59.15 µg/mL) was higher than that of physostigmine. We emphasize that these values may be partly related to the previously documented antioxidant and anti-inflammatory properties of quercetin. Studies indicate that the compound exerts neuroprotective action through multiple mechanisms. They include direct neutralization of reactive oxygen species, activation of the Nrf2-ARE pathway, negative modulation of the pro-inflammatory NF-κB pathway, and regulation of cytokines such as IL-1β, IL-6, IL-10, and TNF-α [68,69]. Thus, we adopted a cautious interpretation that considers multiple mechanisms, especially the antioxidant and anti-inflammatory effects of quercetin, as important contributors to neural protection and memory improvement observed in the zebrafish model, in addition to its AChE inhibitory action.

The in vivo model with Danio rerio (zebrafish) was crucial in demonstrating functional neuroprotective activity, as quercetin prevented scopolamine-induced learning deficits and reduced neuronal atrophy in histological analyses. These results reinforce findings from other animal AD models, in which quercetin attenuated cognitive deficits, reduced β-amyloid deposition, and modulated inflammatory and oxidative biomarkers. Furthermore, the use of positive control drugs, such as physostigmine, validated the results obtained. Its mechanisms of action are widely described in the literature, such as in the study by Cho et al. [52], in which physostigmine administration attenuated anxiety-like effects in a zebrafish model. Similarly, in the study by Kim et al. [33], physostigmine reduced scopolamine-induced learning deficits in zebrafish. These studies provide support for comparing the effects of quercetin across the different biological systems analyzed.

Taken together, the findings of this work indicate that quercetin acts through multiple mechanisms cholinesterase inhibition, inflammatory modulation, antioxidant action, and neuronal structural preservation. This reinforces its relevance as a prototype molecule for the development of new candidates against AD. However, inherent study limitations, such as the low bioavailability of quercetin, its extensive metabolism, and the methodological particularities employed, still represent barriers to its direct clinical application. In addition, the experimental model used, although effective for initial evaluation, has limitations regarding direct extrapolation to humans. Thus, complementary studies are necessary to validate the observed effects.

Recent strategies have sought to improve the oral bioavailability of quercetin by combining nanostructured systems with metabolism inhibitors. Li et al. [66] showed that quercetin nanosuspensions with soybean lecithin and piperine significantly increased bioavailability in rats compared to the conventional form. This effect was associated with greater intestinal absorption and reduced metabolism, indicating potential for expanding its clinical application.

Future perspectives include investigating structural optimization of quercetin to improve its stability and penetration into the central nervous system, as well as developing innovative pharmaceutical formulations that enhance its bioavailability. Controlled clinical trials with representative samples and improved administration strategies are essential to confirm its efficacy and safety in AD patients.

Additionally, research exploring the effects of quercetin in combination with other therapies and its role in different stages of AD may broaden the understanding of its therapeutic potential and contribute to significant advances in the field.

5. Conclusions

Considering that AD is the most prevalent form of dementia, with high treatment costs for both the public health system and the patients’ families, it is essential to identify and develop promising molecules for its prevention and treatment. Thus, the results of this study demonstrate that quercetin has significant theoretical affinity for the catalytic site of AChE, with stability of the ligand–target complex comparable to that of physostigmine. This is evidenced by the analyses of RMSD, SASA, Rg and RMSF. Molecular dynamics simulations with the iNOS enzyme also indicated the stability of the complex with quercetin. However, the RMSF results revealed that the interaction with AChE promoted changes in a greater number of relevant residues, while the interaction with iNOS caused less expressive changes. The DPPH antioxidant assay corroborated the antioxidant potential already described in the literature for quercetin. At a concentration of 33.10 µM, it presented a free radical scavenging percentage greater than 50%. Regarding in vitro inhibition of AChE, quercetin presented an IC₅₀ value equal to 59.15 µg/mL. Although this value is higher than that observed for physostigmine, the data obtained provide a basis for future structural optimizations of quercetin, preserving its scaffold. Furthermore, the in vivo zebrafish model was essential to demonstrate the neuroprotective activity of quercetin. This is because its administration prevented scopolamine-induced learning deficits, as well as neuronal atrophy observed in histological analyses of animals exposed only to scopolamine. Thus, the findings of this work highlight quercetin as a molecule with relevant neuroprotective effects, especially due to its ability to attenuate oxidative damage. They consolidate its potential as a promising starting structure for the development of new candidate hits. In real clinical scenarios, such properties may contribute to slowing cognitive decline and reducing oxidative stress: factors involved in the pathophysiology of neurodegenerative diseases such as Alzheimer’s disease. This study provided preliminary data and results suggesting that quercetin has potential as a neuroprotective agent, given its ability to inhibit AChE activity and act as an antioxidant. However, clinical application requires further investigations in in vivo models and controlled clinical trials to confirm efficacy, safety, and appropriate therapeutic dosages.

To synthesize the main neuroprotective mechanisms of quercetin, an integrative schematic was developed (Figure 15). This diagram compiles evidence obtained from in silico, in vitro, and in vivo approaches, highlighting its potential effects on AchE inhibition, oxidative stress reduction, inflammatory modulation, and mitochondrial protection. In addition, future directions are outlined, including pharmacological validation, formulation development, and molecular optimization.