Search Results (104)

Neurodegenerative diseases are a serious problem for modern society, and their treatment remains an important issue discussed by the scientific community. One of the promising potential directions for modulating neurodegenerative processes is the use of acylhydrazones, a class of compounds that combine different bioactive fragments linked by an acylhydrazone moiety. So far, the biological properties of these compounds have been proven. They show antibacterial, antiviral, antifungal, antiparasitic, anticancer, anti-inflammatory and antioxidant activity. Many research papers focus on designing acylhydrazones that will find use in the treatment of neurodegenerative diseases by inhibiting the enzymatic activity of acetylcholinesterase (AChE), butyrylcholinesterase (BuChE), β-secretase 1 (BACE1) and monoamine oxidase (MAO), as well as inhibiting β-amyloid aggregation, exhibiting metal chelation and antioxidant properties. Recent studies have described the acylhydrazone-based dual (multi-target) inhibitors, which have demonstrated encouraging outcomes during in vitro evaluations. This review covers recent articles published in the years 2020–2025 and offers a comprehensive overview of the biological properties of the acylhydrazones and their multifunctional derivatives on neurodegenerative processes and/or neuroprotection, while emphasizing their universal nature, structural versatility and role as leading structures in the search for new drugs. Full article

Objectives: Using high-performance liquid chromatography (HPLC), we developed and validated an in vitro assay for the quantitative determination of beta-site amyloid precursor protein cleaving enzyme 1 (BACE1) activity, supplementing limited current methodologies to assess the efficacy of BACE1 inhibitor compounds. A hexa-histidine tagged peptide substrate of BACE1 was used as the analyte for the determination of in vitro BACE1 activity; it was validated according to ICH guidelines. Methods: The HPLC analysis was performed on the Agilent 1290 Series Infinity II UHPLC System equipped with a Phenomenex Kinetex EVO C18 (100 × 3 mm) 5 µm column. The method was developed using a gradient programme comprising 10% aqueous acetonitrile (0.02 M TFA) to 30% aqueous acetonitrile (0.02 M TFA) for 5 min at a flow rate of 0.6 mL/min. Results: The method showed linearity over the range of 14.92 to 72 µM with ${\mathrm{r}}^2=0.9997$. The accuracy of the method in terms of mean recovery ranged between 96.62 and 98.38%. The %RSD for intra- and inter-day precision was less than 5%. Two commercial inhibitors, AZD3839 and OM99-2, were used to evaluate the performance of the method at their respective IC₅₀, resulting in inhibition of 53.46 and 50.74%, respectively. The described method addresses the void for a practical and cheap alternative to quantitatively determine the activity of BACE1 compared to current commercially available detection assays. Conclusions: We have successfully developed an HPLC method to measure the inhibitory function of two commercial inhibitors of BACE1, indicating the suitability of the method for the identification and characterisation of novel BACE1 inhibitors. Full article

Traffic-related air pollution (TRAP) is known to contribute to oxidative stress in the central nervous system (CNS) and has been linked to increased risk of Alzheimer’s disease (AD). Alterations in the renin–angiotensin system (RAS), specifically increased angiotensin II (Ang II) signaling via the angiotensin II type 1 (AT₁) receptor, are implicated in increased oxidative stress in the CNS via activation of NADPH oxidase (NOX). As exposure to TRAP may further elevate AD risk, we investigated whether exposure to inhaled mixed gasoline and diesel vehicle emissions (MVE) promotes RAS-dependent expression of factors that contribute to AD pathophysiology in an apolipoprotein E-deficient (ApoE^−/−) mouse model. Male ApoE^−/− mice (6–8 weeks old) on a high-fat diet were treated with either an ACE inhibitor (captopril, 4 mg/kg/day) or water and exposed to filtered air (FA) or MVE (200 µg PM/m³) for 30 days. MVE exposure elevated plasma Ang II, inflammation, and oxidative stress in the hippocampus, associated with increased levels of Aph-1 homolog B (APH1B), a gamma-secretase subunit, and beta-secretase 1 (BACE1), involved in Aβ production. Each of these endpoints was normalized with ACEi treatment. These findings indicate that TRAP exposure in ApoE^−/− mice drives a RAS- and NOX-dependent oxidative and inflammatory response and shifts Aβ processing towards an amyloidogenic profile before overt Aβ deposition, suggesting a potential therapeutic approach for air pollution-induced AD risk. Full article

A series of novel granatane–triazole hybrid molecules was designed, synthesized, and evaluated as dual acetylcholinesterase (AChE) and β-secretase 1 (BACE1) inhibitors. The compounds were obtained through a convergent synthetic route involving azide formation, triazole construction via dipolar cycloaddition, and final coupling with a granatane scaffold to give a pseudopelletierine (3-granatanone) analogue. In vitro assays demonstrated that all target compounds inhibited both AChE and BACE1. Molecular docking and molecular dynamics simulations revealed stable interactions with key catalytic residues, suggesting distinct binding modes compared to reference ligands. QSAR-based pharmacokinetic predictions indicated favorable blood–brain barrier permeability and compliance with key drug-likeness filters. These findings identify granatane–triazole hybrids as promising multi-target directed ligand (MTDL) candidates with potential for further optimization in the search for new anti-Alzheimer therapeutics. Full article

Background: β-site amyloid precursor protein cleaving enzyme 1 (BACE1) inhibitors demonstrated amyloid-lowering efficacy but failed in phase II/III clinical trials due to adverse effects and limited disease-modifying outcomes. This study employed an integrated network pharmacology and molecular docking approach to quantitatively elucidate the multitarget mechanisms of 4 (phase II/III) discontinued BACE1 inhibitors (Verubecestat, Lanabecestat, Elenbecestat, and Umibecestat) and the preclinical compound AM-6494 in Alzheimer’s disease (AD). Methods: Drug-associated targets were intersected with AD-related genes to construct a protein–protein interaction (PPI) network, followed by topological analysis to identify hub proteins. Gene Ontology (GO) and KEGG pathway enrichment analyses were performed using statistically significant thresholds (p < 0.05, FDR-adjusted). Molecular docking was conducted using AutoDock Vina to quantify binding affinities and interaction modes between the selected compounds and the identified hub proteins. Results: Network analysis identified 10 hub proteins (CASP3, STAT3, BCL2, AKT1, MTOR, BCL2L1, HSP90AA1, HSP90AB1, TNF, and MDM2). GO enrichment highlighted key biological processes, including the negative regulation of autophagy, regulation of apoptotic signalling, protein folding, and inflammatory responses. KEGG pathway analysis revealed significant enrichment in the PI3K–AKT–MTOR signalling, apoptosis, and TNF signalling pathways. Molecular docking demonstrated strong multitarget binding, with binding affinities ranging from approximately −6.6 to −11.4 kcal/mol across the hub proteins. Umibecestat exhibited the strongest binding toward AKT1 (−11.4 kcal/mol), HSP90AB1 (−9.5 kcal/mol), STAT3 (−8.9 kcal/mol), HSP90AA1 (−8.5 kcal/mol), and MTOR (−8.3 kcal/mol), while Lanabecestat showed high affinity for AKT1 (−10.6 kcal/mol), HSP90AA1 (−9.9 kcal/mol), BCL2L1 (−9.2 kcal/mol), and CASP3 (−8.5 kcal/mol), respectively. These interactions were stabilized by conserved hydrogen bonding, hydrophobic contacts, and π–alkyl interactions within key regulatory domains of the target proteins, supporting their multitarget engagement beyond BACE1 inhibition. Conclusions: This study demonstrates that clinically failed BACE1 inhibitors engage multiple non-structural regulatory proteins that are central to AD pathogenesis, particularly those governing autophagy, apoptosis, proteostasis, and neuroinflammation. The identified ligand–hub protein complexes provide a mechanistic rationale for repurposing and optimization strategies targeting network-level dysregulation in Alzheimer’s disease, warranting further in silico refinement and experimental validation. Full article

The present study provides a comprehensive comparative evaluation of three Greek fig cultivars through integrated instrumental, computational, and chemometric approaches. Fresh fig peel and flesh samples were analyzed to determine total soluble solids, total phenolic and flavonoid content, as well as antioxidant and antiradical activities, complemented by attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy for structural profiling. Significant varietal and tissue-dependent differences were observed, with fig peel exhibiting higher levels of phenolic compounds and antioxidant activity compared to flesh. ATR-FTIR spectral patterns revealed the presence of characteristic functional groups associated with carbohydrates, phenolic compounds, carboxylic acids, and volatile compounds, reflecting the influence of variety, pollination requirements, and geographical origin. In parallel, to explore potential neuroprotective relevance, 30 phytochemicals reported in figs were subjected to molecular docking against human β-secretase 1 (hBACE1), a key enzyme in Alzheimer’s disease (AD) pathogenesis. Phenolic acids and flavonoids displayed favorable binding affinities and interaction profiles with the catalytic Asp32–Asp228 dyad and with the flap domain. A machine learning model (XGBoost) trained on known BACE1 inhibitors further classified all examined fig metabolites as active candidates. Collectively, these findings highlight Greek figs as chemically rich fruits with potential biological properties, supporting future targeted studies on their bioactive potential. Full article

Background/Objective: Alzheimer’s disease (AD) is characterized by the accumulation of amyloid-β plaques, derived from the amyloid precursor protein through sequential cleavage by β-secretase 1 (BACE-1) and γ-secretase. BACE-1 is therefore a key drug target for designing of selective inhibitors to avoid off-target effects associated with BACE-2 inhibition. The objective of this study was to design novel BACE-1 inhibitors using a structure-based drug design approach. Methods: A focused compound library was designed based on the SAR of N-(4-fluorophenyl)formamide derivatives. In silico ADME predictions were performed to assess pharmacokinetic suitability. Compounds showing favorable ADME profiles were subjected to molecular docking against the BACE-1 enzyme. The top-scoring hit, compound 9.7 (−5.48 (kcal/mol), was further evaluated using a 200 ns MD simulation to assess the stability of its binding interactions with BACE-1. Results: Designed compounds indicated acceptable physicochemical and ADME characteristics. Molecular docking identified compound 9.7 as exhibiting favorable binding interactions with binding pocket residues of BACE-1. The 200 ns MD simulation further confirmed the stability of the docked complex. MD simulations confirmed that 9.7 forms stable interactions with the catalytic residue ASP32 and key hydrophobic residues TRP115 and PHE108 of BACE-1. These important interactions are absent in the reference compound verubecestat. Conclusions: The multi-step computational analysis suggests that compound 9.7 is a promising and selective BACE-1 inhibitor. Its favorable ADME profile, favorable docking interactions, and stable MD simulation behavior highlight its potential as a hit compound for further optimization in the development of anti-Alzheimer’s agents. Full article

β-site amyloid precursor protein cleaving enzyme 1 (BACE1) is a central therapeutic target in Alzheimer’s disease, as it catalyzes the rate-limiting step in amyloid-β production. Verubecestat (VER), a clinical BACE1 inhibitor, failed in late-stage trials due to limited efficacy and safety concerns. This study employed an integrative computational approach to design VER derivatives with improved binding affinity, stability, and pharmacokinetic profiles. Structural similarity analysis, Molecular docking, frontier molecular orbital (FMO) analysis, pharmacophore modeling, 200 ns molecular dynamics (MD) simulations, MM/PBSA free energy calculations, and per-residue decomposition were performed. In silico ADMET profiling assessed drug-likeness, absorption, and safety parameters. Docking and pharmacophore analyses identified derivatives with stronger complementarity in the BACE1 catalytic pocket. MD simulations revealed that VERMOD-33 and VERMOD-57 maintained low root mean square deviations (RMSDs) and stable binding orientations and induced characteristic flexibility in the flap and catalytic loops surrounding the catalytic dyad (Asp93 and Asp289), consistent with inhibitory activity. MM/PBSA confirmed the superior binding free energies of VERMOD-33 (−51.12 kcal/mol) and VERMOD-57 (−43.85 kcal/mol), both outperforming native VER (−35.33 kcal/mol). Per-residue decomposition highlighted Asp93, Asp289, and adjacent flap residues as major energetic contributors. ADMET predictions indicated improved oral absorption, BBB penetration, and no mutagenicity or toxicity alerts. Rationally designed VER derivatives, particularly VERMOD-33 and VERMOD-57, displayed enhanced binding energetics, stable inhibitory dynamics, and favorable pharmacokinetic properties compared with native VER. These findings provide a computational framework for rescuing VER and support further synthesis and experimental validation of next-generation BACE1 inhibitors for Alzheimer’s disease. Full article

Alzheimer’s disease (AD) is the most common cause of dementia, characterized by progressive cognitive decline and neuropathological hallmarks, including amyloid-β (Aβ) plaques, neurofibrillary tangles (NFTs), and neurodegeneration. Since the amyloid cascade hypothesis was proposed, Aβ has remained a central therapeutic target, with interventions aiming to reduce Aβ production, aggregation, or downstream toxicity. This review first outlines the historical development of the Aβ hypothesis and the two major APP processing pathways (α-cleavage and β-cleavage), highlighting the role of biomarkers in early diagnosis, patient stratification, and regulatory approval. We then summarize the development and clinical outcomes of anti-Aβ small-molecule drugs, including β-secretase inhibitors, γ-secretase modulators, Aβ aggregation inhibitors, receptor/synapse modulators, and metabolic or antioxidant modalities. We further review the progression of biologic therapies, with a particular focus on monoclonal antibodies, vaccines, and emerging gene-silencing strategies, such as small interfering RNA (siRNA) and antisense oligonucleotides. Finally, we discuss future perspectives, including next-generation biologics, multi-target approaches, optimized delivery platforms, and early-prevention strategies. Collectively, these efforts underscore both the challenges and opportunities in translating anti-Aβ therapies into meaningful clinical benefits for patients with AD. Full article

The emergence of human coronaviruses has led to three epidemics or pandemics in the last two decades, collectively causing millions of deaths and thus highlighting a long-term need to identify new antiviral drug targets and develop antiviral therapeutics. In this study, a compound library was screened to uncover novel potential inhibitors of coronavirus replication. Three lead compounds, designated #16-14, #16-23, and #16-24, which were Ixazomib and its analogs, were identified based on their potent antiviral activity and minimal cytotoxicity. These compounds were found to inhibit the immunoproteasome subunit LMP7, a target whose subcellular localization and expression are altered in Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2)-infected Huh7 cells. Yeast two-hybrid assays and co-immunoprecipitation further revealed that LMP7 interacts with the viral proteins Nsp13 and Nsp16. In addition, Nsp13 and Nsp16 disrupted the expression of LMP7 in response to pathogen attacks. Functional studies showed that LMP7 knockout in BEAS-2B-ACE2 cells resulted in enhanced replication of attenuated SARS-CoV-2, highlighting the role of this subunit in restricting viral replication. Taken together, these findings position LMP7 as a novel therapeutic target and highlight Ixazomib and its analogs as potential antiviral agents against current and future coronavirus threats. Full article

Background and Objectives: Over the past few years, there has been a significant shift in focus from developing better diagnostic tools to detecting Alzheimer’s disease (AD) earlier and initiating treatment interventions. This review will explore four main objectives: (a) the role of biomarkers in enhancing the diagnostic accuracy of AD, highlighting the major strides that have been made in recent years; (b) the role of neuropsychological testing in identifying biomarkers of AD, including the relationship between cognitive performance and neuroimaging biomarkers; (c) the amyloid hypothesis and possible molecular mechanisms of AD; and (d) the innovative AD therapeutics and the challenges and limitations of AD research. Materials and Methods: We have searched PubMed and Scopus databases for peer-reviewed research articles published in English (preclinical and clinical studies as well as relevant reviews and meta-analyses) investigating the molecular mechanisms, biomarkers, and treatments of AD. Results: Genome-wide association studies (GWASs) discovered 37 loci associated with AD risk. Core 1 biomarkers (α-amyloid Aβ42, phosphorylated tau, and amyloid PET) detect early AD phases, identifying both symptomatic and asymptomatic individuals, while core 2 biomarkers inform the short-term progression risk in individuals without symptoms. The recurrent failures of Aβ-targeted clinical studies undermine the amyloid cascade hypothesis and the objectives of AD medication development. The molecular mechanisms of AD include the accumulation of amyloid plaques and tau protein, vascular dysfunction, neuroinflammation, oxidative stress, and lipid metabolism dysregulation. Significant advancements in drug delivery technologies, such as focused Low-Ultrasound Stem, T cells, exosomes, nanoparticles, transferin, nicotinic and acetylcholine receptors, and glutathione transporters, are aimed at overcoming the BBB to enhance treatment efficacy for AD. Aducanumab and Lecanemab are IgG1 monoclonal antibodies that retard the progression of AD. BACE inhibitors have been explored as a therapeutic strategy for AD. Gene therapies targeting APOE using the CRISPR/Cas9 genome-editing system are another therapeutic avenue. Conclusions: Classic neurodegenerative biomarkers have emerged as powerful tools for enhancing the diagnostic accuracy of AD. Despite the supporting evidence, the amyloid hypothesis has several unresolved issues. Novel monoclonal antibodies may halt the AD course. Advances in delivery systems across the BBB are promising for the efficacy of AD treatments. Full article

β-secretase 1 (BACE1), known for its role in amyloid-β production associated with Alzheimer’s disease (AD), has also been suggested to be elevated in patients with Type 2 diabetes mellitus (T2DM). Notably, BACE1 could cleave the insulin receptor (InsR), leading to reduced InsR levels, which may impair insulin signaling and contribute to insulin resistance. Presently, we observed decreased InsR levels and impaired glucose disposal in the livers of mice with systemic overexpression of BACE1 (HUBC mice). This suggests that elevated BACE1 could contribute to insulin resistance by shedding membrane InsR. Additionally, mice fed a high-fat diet (HFD), a well-established model of T2DM, displayed increased BACE1 levels and decreased InsR. To further investigate whether inhibiting BACE1 could enhance insulin sensitivity and alleviate symptoms of diabetes, we treated HFD mice with the BACE1 inhibitor Elenbecestat. Remarkably, the administration of Elenbecestat restored InsR levels and improved their downstream signaling pathways, leading to increased insulin sensitivity and enhanced glucose tolerance. In summary, our findings suggest that inhibiting BACE1 can restore InsR expression and improve insulin-signaling sensitivity, ultimately resulting in enhanced diabetic phenotypes. Full article

The present Perspective analyzes the remarkable evolution of the Amyloid Cascade Hypothesis 2.0 (ACH2.0) theory of Alzheimer’s disease (AD) since its inception a few years ago, as reflected in the diminishing role of amyloid-beta (Aβ) in the disease. In the initial iteration of the ACH2.0, Aβ-protein-precursor (AβPP)-derived intraneuronal Aβ (iAβ), accumulated to neuronal integrated stress response (ISR)-eliciting levels, triggers AD. The neuronal ISR, in turn, activates the AβPP-independent production of its C99 fragment that is processed into iAβ, which drives the disease. The second iteration of the ACH2.0 stemmed from the realization that AD is, in fact, a disease of the sustained neuronal ISR. It introduced two categories of AD—conventional and unconventional—differing mainly in the manner of their causation. The former is caused by the neuronal ISR triggered by AβPP-derived iAβ, whereas in the latter, the neuronal ISR is elicited by stressors distinct from AβPP-derived iAβ and arising from brain trauma, viral and bacterial infections, and various types of inflammation. Moreover, conventional AD always contains an unconventional component, and in both forms, the disease is driven by iAβ generated independently of AβPP. In its third, the current, iteration, the ACH2.0 posits that proteolytic production of Aβ is suppressed in AD-affected neurons and that the disease is driven by C99 generated independently of AβPP. Suppression of Aβ production in AD seems an oxymoron: Aβ is equated with AD, and the later is inconceivable without the former in an ingrained Amyloid Cascade Hypothesis (ACH)-based notion. But suppression of Aβ production in AD-affected neurons is where the logic leads, and to follow it we only need to overcome the inertia of the preexisting assumptions. Moreover, not only is the generation of Aβ suppressed, so is the production of all components of the AβPP proteolytic pathway. This assertion is not a quantum leap (unless overcoming the inertia counts as such): the global cellular protein synthesis is severely suppressed under the neuronal ISR conditions, and there is no reason for constituents of the AβPP proteolytic pathway to be exempted, and they, apparently, are not, as indicated by the empirical data. In contrast, tau protein translation persists in AD-affected neurons under ISR conditions because the human tau mRNA contains an internal ribosomal entry site in its 5′UTR. In current mouse models, iAβ derived from AβPP expressed exogenously from human transgenes elicits the neuronal ISR and thus suppresses its own production. Its levels cannot principally reach AD pathology-causing levels regardless of the number of transgenes or the types of FAD mutations that they (or additional transgenes) carry. Since the AβPP-independent C99 production pathway is inoperative in mice, the current transgenic models have no potential for developing the full spectrum of AD pathology. What they display are only effects of the AβPP-derived iAβ-elicited neuronal ISR. The paper describes strategies to construct adequate transgenic AD models. It also details the utilization of human neuronal cells as the only adequate model system currently available for conventional and unconventional AD. The final alteration of the ACH2.0, introduced in the present Perspective, is that AβPP, which supports neuronal functionality and viability, is, after all, potentially produced in AD-affected neurons, albeit not conventionally but in an ISR-driven and -compatible process. Thus, the present narrative begins with the “omnipotent” Aβ capable of both triggering and driving the disease and ends up with this peptide largely dislodged from its pedestal and retaining its central role in triggering the disease in only one, although prevalent (conventional), category of AD (and driving it in none). Among interesting inferences of the present Perspective is the determination that “sporadic AD” is not sporadic at all (“non-familial” would be a much better designation). The term has fatalistic connotations, implying that the disease can strike at random. This is patently not the case: The conventional disease affects a distinct subpopulation, and the basis for unconventional AD is well understood. Another conclusion is that, unless prevented, the occurrence of conventional AD is inevitable given a sufficiently long lifespan. This Perspective also defines therapeutic directions not to be taken as well as auspicious ways forward. The former category includes ACH-based drugs (those interfering with the proteolytic production of Aβ and/or depleting extracellular Aβ). They are legitimate (albeit inefficient) preventive agents for conventional AD. There is, however, a proverbial snowball’s chance in hell of them being effective in symptomatic AD, lecanemab, donanemab, and any other “…mab” or “…stat” notwithstanding. They comprise Aβ-specific antibodies, inhibitors of beta- and gamma-secretase, and modulators of the latter. In the latter category, among ways to go are the following: (1) Depletion of iAβ, which, if sufficiently “deep”, opens up a tantalizing possibility of once-in-a-lifetime preventive transient treatment for conventional AD and aging-associated cognitive decline, AACD. (2) Composite therapy comprising the degradation of C99/iAβ and concurrent inhibition of the neuronal ISR. A single transient treatment could be sufficient to arrest the progression of conventional AD and prevent its recurrence for life. Multiple recurrent treatments would achieve the same outcome in unconventional AD. Alternatively, the sustained reduction/removal of unconventional neuronal ISR-eliciting stressors through the elimination of their source would convert unconventional AD into conventional one, preventable/treatable by a single transient administration of the composite C99/iAβ depletion/ISR suppression therapy. Efficient and suitable ISR inhibitors are available, and it is explicitly clear where to look for C99/iAβ-specific targeted degradation agents—activators of BACE1 and, especially, BACE2. Directly acting C99/iAβ-specific degradation agents such as proteolysis-targeting chimeras (PROTACs) and molecular-glue degraders (MGDs) are also viable options. (3) A circumscribed shift (either upstream or downstream) of the position of transcription start site (TSS) of the human AβPP gene, or, alternatively, a gene editing-mediated excision or replacement of a small, defined segment of its portion encoding 5′-untranslated region of AβPP mRNA; targeting AβPP RNA with anti-antisense oligonucleotides is another possibility. If properly executed, these RNA-based strategies would not interfere with the protein-coding potential of AβPP mRNA, and each would be capable of both preventing and stopping the AβPP-independent generation of C99 and thus of either preventing AD or arresting the progression of the disease in its conventional and unconventional forms. The paper is interspersed with “validation” sections: every conceptually significant notion is either validated by the existing data or an experimental procedure validating it is proposed. Full article

The development of BACE-1 (β-site amyloid precursor protein cleaving enzyme 1) inhibitors is a crucial focus in exploring early treatments for Alzheimer’s disease (AD). Recently, graph neural networks (GNNs) have demonstrated significant advantages in predicting molecular activity. However, their reliance on graph structures alone often neglects explicit sequence-level semantic information. To address this limitation, we proposed a Graph and multi-level Sequence Fusion Learning (GSFL) model for predicting the molecular activity of BACE-1 inhibitors. Firstly, molecular graph structures generated from SMILES strings were encoded using GNNs with an atomic-level characteristic attention mechanism. Next, substrings at functional group, ion level, and atomic level substrings were extracted from SMILES strings and encoded using a BiLSTM–Transformer framework equipped with a hierarchical attention mechanism. Finally, these features were fused to predict the activity of BACE-1 inhibitors. A dataset of 1548 compounds with BACE-1 activity measurements was curated from the ChEMBL database. In the classification experiment, the model achieved an accuracy of 0.941 on the training set and 0.877 on the test set. For the test set, it delivered a sensitivity of 0.852, a specificity of 0.894, a MCC of 0.744, an F1-score of 0.872, a PRC of 0.869, and an AUC of 0.915. Compared to traditional computer-aided drug design methods and other machine learning algorithms, the proposed model can effectively improve the accuracy of the molecular activity prediction of BACE-1 inhibitors and has a potential application value. Full article

Alzheimer’s disease (AD), a leading neurodegenerative disorder, is closely associated with the accumulation of amyloid-beta (Aβ) peptides in the brain. The enzyme β-secretase (BACE1), pivotal in Aβ production, represents a promising therapeutic target for AD. While bioactive peptides derived from food protein hydrolysates have neuroprotective properties, their inhibitory effects on BACE1 remain largely unexplored. In this study, we evaluated the inhibitory potential of protein hydrolysates from gliadin, whey, and casein proteins prepared using bromelain, papain, and thermolysin. Through in vitro and cellular assays, bromelain-hydrolyzed gliadin (G-Bro) emerged as the most potent BACE1 inhibitor, with an IC₅₀ of 0.408 mg/mL. G-Bro significantly reduced BACE1 expression and amyloid precursor protein (APP) processing in N2a/PS/APP cell cultures, suggesting its potential to attenuate Aβ aggregation. The unique peptide profile of G-Bro likely contributes to its inhibitory effect, with proline residues disrupting β-sheets, lysine residues introducing positive charges that hinder aggregation, hydrophobic residues stabilizing binding interactions, and glutamine residues enhancing solubility and stability. These findings highlight gliadin hydrolysates, particularly G-Bro, as potential natural BACE1 inhibitors with applications in dietary interventions for AD prevention. However, further studies are warranted to elucidate specific peptide interactions and their bioactivity in neural pathways to better understand their therapeutic potential. Full article