Abstract Alzheimer's disease is a neurodegenerative disorder characterized by a decline in cognitive function. The β-amyloid (Aβ) hypothesis suggests that Aβ peptides can spontaneously aggregate into β-fragment-containing oligomers and protofibrils, and this activation of the amyloid pathway alters Ca^2+ signaling in neurons, leading to neurotoxicity and thus apoptosis of neuronal cells. In our study, a blood-brain barrier crossing flavonol glycoside hyperoside was identified with anti-Aβ aggregation, BACE inhibitory, and neuroprotective effect in cellular or APP/PSEN1 double transgenic Alzheimer's disease mice model. While our pharmacokinetic data confirmed that intranasal administration of hyperoside resulted in a higher bio-availability in mice brain, further in vivo studies revealed that it improved motor deficit, spatial memory and learning ability of APP/PSEN1 mice with reducing level of Aβ plaques and GFAP in the cortex and hippocampus. Bioinformatics, computational docking and in vitro assay results suggested that hyperoside bind to Aβ and interacted with ryanodine receptors, then regulated cellular apoptosis via endoplasmic reticulum-mitochondrial calcium (Ca^2+) signaling pathway. Consistently, it was confirmed that hyperoside increased Bcl2, decreased Bax and cyto-c protein levels, and ameliorated neuronal cell death in both in vitro and in vivo model. By regulating Aβ-induced cell death via regulation on Ca^2+ signaling cascade and mitochondrial membrane potential, our study suggested that hyperoside may work as a potential therapeutic agent or preventive remedy for Alzheimer's disease. Keywords: Alzheimer's disease, Aβ aggregates, Hyperoside, Neurotoxicity, Endoplasmic reticulum-mitochondrial-Ca^2+, Ryanodine receptors, Calcium signal, Mitochondrial membrane potential Graphical abstract [55]Image 1 [56]Open in a new tab Highlights * • HYP inhibits Aβ aggregation-induced neurotoxicity via direct binding to Aβ. * • HYP targets the endoplasmic reticulum-mitochondrial Ca^2+ signal cascade through RyR2. * • Nasal administration increases bioavailability of HYP in APP/PS1 mice brain. * • HYP attenuates Aβ plaque deposition in the cortex and hippocampus of APP/PS1 mice. * • HYP improves learning and cognitive abilities in APP/PS1 mice. Abbreviations AD Alzheimer's disease Aβ amyloid β-protein APP amyloid-beta precursor protein HYP Hyperoside APP/PS1 APP/PSEN1 WT Wild-Type BACE Beta-secretase cyto-c cytochrome c Fig Figure ATP adenosine triphosphate ER endoplasmic reticulum ThT Thioflavin T ThS Thioflavin S BLI Biolayer Interferometry MRM multiple reaction monitoring mRNA Messenger RNA MCU mitochondrial calcium uniporter Q-PCR Quantitative Polymerase Chain Reaction MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide PI Propidium Iodide DMSO Dimethyl sulfoxide IS internal standard ISO Isorhamnetin BBB blood-brain barrier CNS central nervous system NMDAR N-methyl-d-aspartate receptor MEM Memantine Hydrochloride GFAP glial fibrillary acidic protein CA1 cornuammonis region 1 CA3 cornuammonis region 3 DG dentate gyrus region OMM outer mitochondrial membrane VDAC1 voltage-dependent anion channel 1 HFIP hexafluoroisopropanol SSA super streptavidin JC-1 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolyl cobalt iodide CCCP Carbonyl cyanide m-chlorophenyl hydrazine EP tube Eppendorf Micro Test Tubes 1. Introduction Alzheimer's disease (AD) is the most common form of dementia, and the number of people with dementia is expected to reach 152 million by 2050 [[57]1]. While most patients with AD have amnestic problems, a significant proportion of young-onset cases have atypical phenotypes including predominant visual, language, executive, behavioral or motor dysfunction [[58]2,[59]3]. Currently, scientists have proposed multiple hypothesis for the pathogenesis of AD, for example, the amyloid β (Aβ) hypothesis has suggested amyloid biomarkers as the earliest evidence of detectable neuropathological changes in AD patients [[60][4], [61][5], [62][6]]. With the continuous research and development on AD biomarker and its application in diagnostic setting, Jack and colleagues [[63]7] proposed the “ATN” framework, which classified biomarkers into A (amyloid), T (phosphorylated tau) and N (neurodegeneration and neuronal injury) in 2018. In this research framework, the diagnosis of Alzheimer's disease is defined by the presence of phosphorylated tau and amyloid β. In fact, the earliest cellular pathogenesis of Alzheimer's disease included both the existence and accumulation of Aβ and p-tau [[64]8]. Therefore, although the role of Aβ on AD remains controversial, diagnosis (by PET scans and plasma assays) on the level of Aβ and phosphorylated tau remains one of the most classical way for clinical and research work on AD [[65]1]. Since Aβ may not be the only culprit in the pathology of AD, other emerging putative causes of AD ranging from inflammation to metabolic dysfunction, and also tau pathology have also been suggested as the possible pathogenic mechanisms for AD [[66]9]. Tau, the microtubule-associated protein becomes hyperphosphorylated and aggregates into neurofibrillary tangles in AD patients' brains [[67]10]. Insoluble tau aggregates are highly associated with the cognitive and clinical symptoms of AD [[68]11]. Therefore, current pharmacological research and treatments are still poised at the advanced stages of clinical trials by targeting on anti-amyloid β, anti-tau, and anti-inflammatory strategies [[69]12]. In current clinical diagnosis strategy, the presence of amyloid β (regardless of the presence of phosphorylated tau or neurodegeneration) which is cleaved from the larger precursor amyloid precursor protein (APP), remains one of the key markers for the diagnosis of AD pathological change [[70]13]. Cleavage of APP by β-secretase (BACE) produces sAPPβ and C99. γ-secretase is also proposed to play a role in protein hydrolysis of Notch [[71]14], which cleaves into the pathogenic Aβ (1–42) [[72]15]. Aβ (1–42) spontaneously aggregates into β-sheet-rich oligomers and fibrils, which the Aβ oligomers are reported to be transient intermediates in the formation of protofibrils [[73]15] and did not exist as stable entities [[74]16]. With the uncertain identity and pathogenic mechanisms of Aβ aggregates in AD [[75]17], recent studies showed a robust correlation between the soluble Aβ oligomer levels and severity of cognitive impairment [[76][18], [77][19], [78][20]]. There is also evidence that the accelerated Aβ fibrillation process greatly enhanced the toxicity of Aβ in vitro [[79]21,[80]22]. Furthermore, Aβ neurotoxicity has also been associated with intraneuronal Ca^2+ dyshomeostasis [[81]23]. The calcium hypothesis [[82]24,[83]25] of AD suggested that the activation of amyloid pathway remodels neuronal Ca^2+ signaling, since Aβ aggregates can bind strongly to neuronal cell membranes [[84]26,[85]27]. Following an increase in cytoplasmic Ca^2+, excess Ca^2+ uptake into mitochondria leads to overload of Ca^2+, inhibition of ATP synthesis, opening of the mitochondrial permeability transition pore (mPTP), release of cytochrome c and apoptosis [[86]28]. Reduction of intracellular calcium significantly lowered the cytotoxicity level. In addition, it is also shown that a normal Ca^2+ balance is responsible for the mechanisms of learning and memory [[87]29]. As shown by quantitative imaging [[88]30], 20% of neuronal Ca^2+ was elevated (calcium overload) in APP/PSEN1 (APP/PS1) mice with cortical plaques, compared with less than 5% in wild-type mice. The ryanodine receptor (RyR) 2 protein is a major component of Ca^2+ channels located in the sarcoplasmic reticulum, allowing calcium release from the sarcoplasmic reticulum into the cytoplasmic matrix [[89]31]. RyR2 channels are implicated in many cellular functions, particularly mitochondrial metabolism, and disruption of RyR2 regulation in the endoplasmic reticulum (ER) mediated the key signal transduction cascade responses associated with AD [[90]24]. Hyperoside (HYP), as the main component of Crataegus pinnatifida Bunge, is a folk medicine and multifunctional medicine food homology agent reported for its antidepressive, antitumor and antiviral effect [[91]32], however, its neuroprotective effect remains unexplained. In this study, the protective role and mechanisms of HYP in reducing neuronal cell death through anti-Aβ and targeting ER-mitochondrial Ca^2+ signaling cascade response were firstly reported. In addition, behavioral evaluation on APP/PS1 double transgenic AD mice model showed that HYP improved cognitive and learning functions, reduced Aβ plaques, and attenuated apoptosis in the cortex and hippocampus of mice via modulation of Ca^2+ signaling cascade. In summary, this study has provided scientific evidence for development of HYP as a potential natural therapeutic agent for AD. 2. Materials and methods 2.1. Rengeats All chemicals and reagents were purchased from Sigma unless otherwise stated. The following reagents from other suppliers were used: Hyperoside (PuFei De, Chengdu, CHINA), Aβ (1–42) (DGpeptites, Hangzhou, CHINA), BAPTA/AM (Santa Cruz, CA, USA), Thapsigargin (Sigma, MO, USA), Dantrolene sodium salt (Sigma, MO, USA), Thioflavin T (Sigma, MO, USA), Thioflavin S (Sigma, MO, USA), RIPA (CST, MA, USA), BACE1 Kit (Bioscience, CA, USA), FLIPR® Calcium 6 Evaluation Kit (Molecular Device, CA, USA). Apoptosis was detected by Annexin V staining kit (BD Biosciences, CA, USA). FlexiTube RyR2 siRNA target sequence: 5′-CTCGTCGTATTTCTCAGACAA-3′, Lipofectamine™ RNAiMAX Transfection Reagent (Thermofisher, CA, USA), Enhanced mitochondrial membrane potential assay kit with JC-1 (Beyotime, Shanghai, CHINA), Calcein AM Cell Viability Assay Kit (Beyotime, Shanghai, CHINA), Purified anti-β-Amyloid 1–16 Antibody (clone 6E10) (Biolegend, CA, USA), anti-amyloidogenic protein oligomer A11 (Invitrogen, CA, USA), anti-Amyloid Fibril antibody [mOC87] (Abcam, MA, USA), APP (E4H1U) Rabbit mAb against total APP protein (CST, MA, USA), BACE1 polyclonal antibody (Proteintech, IL, USA), purified anti-sAPPβ antibody (Biolegend, CA, USA) were used. Bax antibody (CST, MA, USA), Bcl-2 (D17C4) rabbit mAb (CST, MA, USA), cytochrome c antibody (A-8) (Santa Cruz, CA, USA), GFAP (D1F4Q) XP® Rabbit (CST, MA, USA), anti‐β‐actin mouse monoclonal IgG1 (Santa Cruz, USA), Agilent Zorbax Eclipse Plus C-18 column (Agilent, CA, USA) were also adopted. 2.2. Cell culture Unless otherwise stated, all cells were obtained from the American Type Culture Collection (Thermofisher, CA, USA). All media were supplemented with 10% FBS and the antibiotics penicillin (50 U/mL) and streptomycin (50 μg/mL, Invitrogen, CA, USA). All cell cultures were incubated at 37 °C in a 5% humidified CO^2 incubator. 2.3. Computational docking Structure of HYP from Pubchem and Aβ (1–42) co-crystal structure from RCSB PDB were downloaded from the databases. In ligand preparation, Schrodinger Ligprep was used to prepare high quality ligand for further molecular docking. Protein was prepared using Schrodinger Protein Preparation Wizard for pre-processing, optimization, water removal and minimization procedures. Sitemap tools were used to evaluate potential protein binding sites. Receptor grid was generated using results of SiteMap in Receptor grid generation in the Glide application (Glide, version 9.1, Schrödinger) of Maestro (Maestro, version12.8, Schrödinger). The receptor grid for Aβ (1–42) was generated by specifying the binding (active) site residues, which was identified by SiteMap tool. Once the receptor grid is generated, the ligands are docked to the protein Aβ (1–42) using Glide version 9.1 (Grid based LIgand Docking with Energetics) docking protocol. Energy was calculated using the Calculate Energy module in MacroModel application (BatchMin V13.2). 2.4. Aβ peptide preparation 1 mg of Aβ (1–42) peptide was dissolved in 400 μL of hexafluoroisopropanol (HFIP; Sigma) and sonicated for 5 min. The Aβ peptide solution was aliquoted into 1.5 mL tubes (100 μL/tube) and nitrogen blown to dryness to produce a peptide film to yield Aβ monomer and stored at −80 °C. Aβ was re-dissolved in 10 μL of dimethyl sulfoxide (DMSO) (Sigma, USA) and an appropriate amount of PBS (pH = 7.4) to the desired final concentration before use. The Aβ peptides were incubated at 37 °C for 5 days to promote the aggregation to form Aβ aggregates. 2.5. Characterization of enriched Aβ aggregates The intensity-weighted average hydrodynamic diameter of Aβ aggregates (30 μM) in PBS at 25 °C was measured with the Zetasizer Nano ZSP (Malvern Instruments) at a backscattering angle of 173° versus the polydispersity index PDI (a parameter used to describe the width of the size distribution) and the diameter number distribution curve [[92]33]. 2.6. Thioflavin T (ThT) fluorescence assay HYP was incubated with Aβ (30 μM) for 7 days at 37 °C at a final volume of 100 μL. ThT fluorescence measurement was measured for every 24 h. ThT (20 μM) dissolved in PBS (pH = 7.4) was added in a black 96-well plate with 10 μL of aggregated Aβ (with or without HYP) and incubated for 1 h. Fluorescence measurement was then performed using a microplate reader (SpectraMax Paradigm, Molecular Devices, CA, USA) with an excitation wavelength of 450 nm and an emission wavelength of 490 nm. Background fluorescence was measured in control samples containing PBS and 0.02 % DMSO. 2.7. Biolayer interferometry (BLI) analysis 200 μL of solution containing 200 μg of the Aβ peptide was incubated at 37 °C for 5 days. EZ-Link NHS-LC-LC-Biotin (Thermo Scientific, USA) was dissolved in DMSO to a concentration of 10 mM. Aβ aggregates was biotinylated in a 1:0.5 M ratio of biotin reagent and incubated for 30 min at room temperature before being added to a 96-well plate (Greiner Bio-One, PN:655,209). Biotinylation was ascertained by loading the mixture onto super streptavidin (SSA) capacity tips (ForteìBIO, CA, USA) and detected by the FortéBIO Octet Red instrument. Additionally, SSA biosensors were pre-wetted with PBS for the recording of baselines. Successful biotinylated Aβ aggregates solution was collected and immobilized onto SSA tips overnight at 4 °C. HYP dissolved in DMSO was diluted to an appropriate concentration with PBS to a final 200 μL/well volume. Control wells were added with an equal amount of DMSO. All experiments consisted of repeated cycles of four significant steps: wash (300 s), baseline (120 s), association (120 s), and dissociation (120 s). The association, dissociation plot, and kinetic constants were analyzed with ForteìBIO data analysis software. 2.8. Cell membrane permeability assay HT22 cells were co-treated with Aβ aggregates and HYP (20–80 μM) for 24 h and washed with PBS for 3 times. The cell membrane was disrupted with 200 μL methanol for the collection and centrifugation of cellular content at 20,000 rpm for 10 min. The supernatant was then collected for LC-MS/MS analysis. 2.9. BACE-1 inhibition assay The assay was performed in a 96-well flat-bottom white plate using β-Secretase Activity Fluorometric Assay Kit (Biovision, CA, USA). To begin, HT22 cells were treated with HYP (20–80 μM) for 24 h and then collected by centrifugation with ice-cold extraction buffer added for homogenization. After centrifugation, 50 μL supernatant (cell lysate) was transferred to each well in the 96-well plate. 2 μL of active β-secretase (protein concentration: 4 μg/μL) was added to the 50 μL of extraction buffer as the positive control. For negative control, 2 μL of the β-secretase inhibitor was added to the 50 μL of sample well. 2 μL of each of the tested samples were added into each sample well for inhibitory activity evaluation. Following compounds addition, 50 μL of 2X reaction buffer was added with a gentle mix and incubation for 20 min at 37 °C before adding 2 μL of the β-secretase substrate. The plate was then covered and incubated in the dark at 37 °C for 1 h. Samples were then measured in a fluorescent 96-well plate reader (Ex/Em = 345/500 nm). Background readings produced from the substrate without adding secretase were subtracted from all samples. 2.10. Protein extraction and western blotting After HYP treatment, the cells were lysed with RIPA. Protein concentrations were determined by Bio-Rad protein assay (Bio-Rad Laboratories, CA, USA). After electrophoretic separation, the gels were blotted and stained with primary antibodies. Binding of antibodies was visualized with peroxidase-coupled secondary antibodies using ECL Western Blotting detection reagents (Invitrogen, Scotland, UK). Band intensities were quantified using ImageJ software. Data were obtained from three independent experiments. 2.11. Dot blot assay 4 μL of prepared cell or animal sample protein solutions were dotted for adsorption onto methanol-pre-activated PVDF membranes and then blocked with 5% non-fat dry milk in Tris-buffered saline and Tween 20 for 1 h. The samples were incubated overnight at 4 °C with anti-amyloid fibril primary antibody [mOC87] (Abcam, MA, USA), anti-amyloid oligomer A11 (Invitrogen, CA, USA) or purified anti-β-amyloid (1–16) antibody (clone 6E10) (Biolegend, CA, USA) (1:1000), followed by HRP-conjugated secondary antibody. Protein bands were detected using Super Signal Sensitive ECL Western Blotting Detection Reagent (Beyotime, Beijing, China) and observed using a gel imaging device (Amersham Imager 800, GE, Tokyo, Japan). 2.12. Cytotoxicity assays Cytotoxicity was assessed using the MTT (5.0 mg/mL) assay. 4 × 10^3 HT22 cells per well were seeded in 96-well plates. After overnight incubation, cells were treated with Aβ aggregates (30 μM) for 48 h with or without the presence of HYP (20–80 μM). Cells were treated with DMSO as a control. Subsequently, MTT (10 μL) was added to each well for 4 h, followed by the addition of 100 μL of lysis buffer (10% SDS in 0.01 mol/L hydrochloric acid) and overnight incubation. The absorbance was measured at 570 nm the next day. The formula was calculated as a percentage of cell viability: Cell viability (%) = A (treatded) / A (control) × 100%. Data were obtained from three independent experiments in triplicate. 2.13. LIVE/DEAD cells analysis According to the manufacturer's instructions, cell death was detected by Calcein AM Cell Viability Assay Kit (Beyotime, Shanghai, CHINA). 5 × 10^4 HT22 cells per well were seeded in 24-well plates, and HT22 was treated with HYP (20–80 μM) or PBS for 24 h. After overnight incubation, treated cells were stained with Calcein/Propidium Iodide (PI) dye for 20 min. Cells images were then visualized and captured by using Olympus IX71 fluorescence microscope with FITC and TRITC filters consecutively. Green and red fluorescence images were merged and analyzed with cellSens Standard 1.8.1 software. The percentage of cell death was quantified by dividing the number of dead cells (red fluorescence) by the total number of cells. 2.14. Annexin V detection by flow cytometry analysis Apoptosis was detected with Annexin V staining kit (BD Biosciences, CA, USA). Briefly, HT22 cells were treated with HYP (20–80 μM) and Aβ aggregates (30 μM) for 24 h were stained with FITC-Annexin V and PI for flow cytometric detection according to the manufacturer's instructions. The number of apoptotic cells was quantified by flow cytometry (BD FACSAria III, CA, USA). Data acquisition and analysis were performed with CellQuest (BD Biosciences) from three independent experiments. 2.15. Fluorescent probe method (Fluo-3) to detect calcium ion concentration 2 × 10^5 HT22 cells were incubated in 35 mm confocal dishes for 24 h at 37 °C in a CO^2 incubator. A 5 mM Fluo-3 AM stock solution was diluted with Hanks-Balanced Salt Solution (HBSS) to a 5 μM working solution and then added to the cells for 30 min at 37 °C. HT22 cells were washed three times with HEPES-buffered saline and then incubated in an imaging chamber at 37 °C for an additional 10 min. After the addition of 30 μM of Aβ aggregates in HBSS buffer, changes in cellular Ca^2+ levels were monitored by epifluorescence microscopy (Applied Precision DeltaVision Elite, Applied Precision, Inc., United States) using the real-time mode for tracking Fluo-3 changes for 5 min. The data inspection program provided by DeltaVision software was used to measure the intensity of Fluo-3 fluorescence. 2.16. Measurement of cytoplasmic calcium dynamics Intracellular Ca^2+ dynamics were determined using the FLIPR Calcium 6 Assay Kit (Molecular Devices) according to the manufacturer's instructions. Briefly, HT22 was housed at 4 × 10^4 cells per well in a black-walled and clear-bottom 96 multi-well plate (Costar, MA, USA) and treated with Calcium 6 reagent for 1 h. A mixture of HYP and Aβ aggregates or thapsigargin (TG) was added to the wells and data were immediately collected at room temperature using SpectraMax Paradigm multi-mode microplate reader (Molecular Devices, CA, USA) using 5-s read intervals in five independent experiments. 2.17. siRNA transfection HT22 were transfected with siRNAs using Lipofectamine™ RNAiMAX Transfection Reagent (Thermofisher, CA, USA) according to the manufacturer's protocol. To maximize the knockdown efficiency for the RyR2, the siRNA of different RyR2 isoforms were used for transfection. RyR2 siRNA target sequence 5′-CTCGTCGTATTTCTCAGACAA-3′ was purchased from Qiagen (CA, USA). 2.18. Real‐time quantitative PCR RNA was extracted from the HT22 cell using FavorPrep™ Blood/Cultured Cell Total RNA Purification Mini Kit (Favorgen Biotech Corp.). RNA concentration was determined using the NanoDrop 2000c Spectrophotometer (Thermo Scientific). 1 μg of total RNA was used to reverse transcribe to its corresponding cDNA by using the Transcriptor Universal cDNA Master mix (Roche, USA). Quantitative PCR was then performed with the addition of PowerUp™ SYBR® Green Master Mix (Applied Biosystems) using the ViiA™ 7 Real-Time PCR System (Applied Biosystems). Specific primers (Tech Dragon Ltd., Hong Kong) were designed by employing ThermoFisher Scientific's online OligoPerfect™ Designer software and then verified with NCBI's Primer‐BLAST software to confirm specific recognition of the target genes. Gene expression levels were normalized to actin (control) and analyzed using the ΔΔCT method. Three independent experiments with three replicates per group were analyzed for each primer. Primer sequences are specified as in [93]Table 1: Table 1. Primer sequences. Gene Name Forward Reverse RyR2 5′-AGAAGGAGAGGCCAGAGGAG-3′ 5′-GGACAGGGTTGGTCATGAGG-3′ APP 5′-CCTCCGTGTGATCTACGAGC-3′ 5′-GAACCTGGTCGAGTGGTCAG-3′ MCU 5′-ACGACAACTGCAAGAGGAGG-3′ 5′-CAGGGTCTTCACGTCGTTCA-3′ [94]Open in a new tab 2.19. Measurement of mitochondrial membrane potential 5 × 10^4 HT22 cells per well in 24-well plates were co-treated with Aβ aggregates and HYP (20–80 μM) or PBS for 24 h 10 mM of Carbonyl cyanide m-chlorophenyl hydrazine (CCCP) was used as a positive control. The mitochondrial membrane potential of HT22 stained with 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolyl cobalt iodide (JC-1) was observed by laser scanning confocal microscopy. 2.20. ATP content measurement HT22 cells were co-treated with Aβ aggregates and HYP (20–80 μM) for 24 h and washed with PBS. Each digested sample was re-suspended in 10 mL of ice-cold PBS, followed by centrifugation for 5 min (300 g, 4 °C). Cell pellets were then treated with 85% cold methanol (2 × 10^6 cells / 100 μL) containing 2 μM ATP-^13C10,^15N5, then vortexed for 1 min and placed on ice for 10 min. After centrifugation (13,000 g) at 4 °C for 15 min, the supernatant was transferred into another tube. The derivatization reaction was initiated by adding 75 μL of derivatization reagent MTBSTFA into 200 μL of cell lysis with 85% methanol and completed over 5 min with consistent vortex. Derivatization samples were then centrifuged (13,000 g) at 4 °C for 10 min, and 25 μL of the supernatant was injected into LC-MS/MS system for analysis. 2.21. Pharmacokinetics analysis After tail vein injection or nasal administration, blood was collected at 5, 15 and 30 min, 1, 2, 4, 8, 12, 24, 48, 72, 120 and 168 h, for immediate centrifugation at 14,000 rpm for 5 min. The serum was harvested and stored at −20 °C until further processing. Brain tissue was collected and was thoroughly rinsed in saline to eliminate blood and blotted dry with filter paper. Each tissue sample was homogenized in saline (1:5, w/v) and stored at −20 °C until analysis. 2.22. Determination of serum and brain tissue drug concentrations 50 μL of serum or 100 μL of brain tissue was put in a 1.5 mL EP tube, followed by the addition of 20 μL of internal standard solution (10 μM ISO). The protein was precipitated with 1 mL of ethyl acetate, vortexed for 5 min and centrifuged at 20,000 g (4 °C and 10 min). The supernatant solution was nitrogen blown to dry and re-suspended with 100 μL of 50% acetonitrile for centrifugation again. The supernatant was taken for analysis by using LC-MS/MS. 2.23. LC-MS/MS measurement of HYP HYP content was quantified using UHPLC-MS/MS system, which includes Agilent 1290 Infinity UHPLC, and Agilent 6460 Triple Quadrupole, equipped with an electrospray ionization interface used to generate positive ions for the determination of HYP. The compounds were analyzed by using the Agilent Zorbax Eclipse Plus C-18 column with a particle size of 1.8 μM (flow rate: 0.35 mL/min). The mobile phase was set as follows: mobile phase A (0.1% formic acid in water) and mobile phase B (0.1% formic acid in ACN): 0–3 min, 10–30% B; 3–4 min, 70% B; 4–5 min, 30–70% B; 5–6 min, 70-10% B. The column and auto-sampler temperature were maintained at 30 °C and 4 °C, respectively. Data were analyzed by using Agilent MassHunter Workstation software B.01.03. The gas temperature was set at 300 °C with a flow rate of 11 L/min. Gases were set at 30 psi for the nebulizer, capillary, at 4000 V. For the HYP, the fragment was 170. The collision energy was set at 30; The mass transitions were as follows based on multiple reaction monitoring: m/z 464.38 → 299.9. For the Isorhamnetin (ISO), the fragment was 150. The collision energy was set at 30; The mass transitions were as follows based on multiple reaction monitoring: m/z 315.4 → 300.1, The measurements of HYP were done using the standard and linear least-squares regression curve. 2.24. Animal experiment All animal care and experimental procedures were performed in accordance with the “Institutional Animal Care and User Committee Guidelines” of the Macau University of Science and Technology. APP/PSEN1 (APP/PS1) mice (N000175) were purchased from the Nanjing Institute of Biomedical Sciences, Nanjing University, China, weighing 25 ± 5 g. The animals were housed in a temperature-controlled room with a 12 h light/dark cycle and given ad libitum food and water. Male mice were randomly divided into the following 6 experimental groups. (a) wild-type group (n = 6), (b) APP/PS1 group (n = 6), (c) HYP (20 mg/kg group) (n = 7), (d) HYP (40 mg/kg) group (n = 8), (e) HYP (80 mg/kg) group (n = 8) and (f) memantine (MEM) (10 mg/kg) positive control group (n = 8). HYP was dissolved in the solvent containing 10% DMSO, 40% PEG300 and 5% Tween-80 in PBS, and administrated via nasal injection for every 3 days with a volume of 0.5 μL/g of body weight for 8 consecutive weeks. 2.25. Y-maze tests Y-maze tests assess cognitive changes, short-term spatial working memory (by spontaneous alternation), and exploratory activity (by total number of arm choices) of mice. The Y-maze apparatus (Yuyan, Shanghai, China) is a three-arm horizontal maze (40 cm long, 10 cm wide with 12 cm high walls) in which the arms are symmetrically disposed at 120° angles from each other. To begin, mice were placed at the end of one arm and allowed to move freely through the maze during a 9-min session. The number of total arm choices and sequence of arm choices were recorded. The percentage of alternation can be calculated as the ratio of total number of alternations / the number of arms entered x 100%. Before each trial, the interior of the maze was sprayed with 70% ethanol solution to erase any scent cues [[95]34]. 2.26. Fear-conditioning tests The fear conditioning test measures the ability of mice to learn and remember associations between aversive experiences and the environment. On the first day of the fear conditioning test (Yuyan, Shanghai, China), mice were placed in a chamber for 5 min to acclimatize and explore. On the second day, the training (fear conditioning phase) was performed by placing the mice into a standard operating chamber with sound attenuation for 3 min. A 30-s tone (3 kHz, 85 dB) was then delivered, followed by a 2-s electric shock (0.5 mA). The training was repeated twice over a 4-min period. 24 h later, the mice were placed in the same conditioning chamber for a context retention test which the conditions were consistent with the training except that the 2-s electric shock was not allowed. Autonomic activity was performed for 90 s after the tone stimulus, for a total of 5 min. Responses were recorded with a video camera and scored for resting time, which was defined as the absence of any movement other than breathing [[96]35,[97]36]. 2.27. Rotarod test The rotarod test was used to monitor motor coordination. The test consisted of placing the animal on a rotating rod with a rotational speed of 25 rpm (Yuyan, Shanghai, China) until the animal fell to the ground and the movement time was recorded. The test was performed three times with an interval of 300 s, and the mean of the test results were calculated. 2.28. Hematoxylin and eosin (HE) staining Pathological assessment on the cornuammonis region 1 (CA1), cornuammonis region 3 (CA3) and dentate gyrus region (DG) regions of the mouse hippocampus was done by HE staining. Brain sections were dehydrated with 70%, 80% and 90% alcohol respectively. Sections were then subjected to staining with hematoxylin (50 °C) for 30 s, incubation with 1% hydrochloric acid alcohol for 10–20 s, washing with 0.5% ammonia hydroxide for 10 s, staining with eosin for 3–5 s, and finally dehydration in 70%, 80%, and 90% alcohol respectively. After transparentized within dimethyl benzene, the sections were sealed with neutral gum for microscopic observation. Images were captured by light microscopy (Leica, WZ, GER). 2.29. Nissl staining The CA1, CA3 and DG regions of the mouse hippocampus was assessed by Nissl staining for Nissl body detection. Brain sections were dehydrated as the same as in HE staining, followed by 1% tar violet staining for 1 h, and with distilled water washing followed by 70% alcohol separation for 1 min. Tissues were then dehydrated in 70%, 80% and 90% of alcohol. After transillumination in dimethylbenzene, sections were sealed with neutral adhesive for microscopic observation. Images were captured by a light microscope (Leica, WZ, GER). 2.30. Thioflavin S (THS) staining Paraffin sections of the brain were washed three times with PBS after dewaxing and then stained with THS dissolved in 50% alcohol for 7 min, followed by 50% alcohol solution for washing. Sections were then air-dried and sealed with neutral gel for microscopic observation. Images were captured by API DeltaVision Live‐cell Imaging System (Applied Precision Inc., DC, USA). 2.31. Immunofluorescent staining of brain tissues Brain tissues from all treatment groups were fixed and embedded in paraffin for microtome sectioning and immunofluorescence staining. After deparaffinization, the tissue sections were subjected to antigen retrieval (EnVision™ FLEX Target Retrieval Solution, High pH), soaked in PBS, dehydrated, deparaffinized and rehydrated according to the standard protocols. Next, the sections were incubated in 1% Triton X-100 for 40 min, rinsed with PBS, and blocked with 5% bovine serum albumin (BSA) for 40–60 min. The sections were incubated with anti-GFAP rabbit polyclonal antibodies (1:200, CST, CA, USA) at 4 °C overnight. After rinsed with PBS, the sections were incubated with fluorescein secondary antibodies (1:200, CST, CA, USA) for 1 h, then counterstained with DAPI for 10 min. The coverslips were then mounted with FluorSave™ mounting media (Calbiochem, San Diego, CA, USA) for fluorescence imaging. The expression of GFAP was captured by API DeltaVision Live‐cell Imaging System (Applied Precision Inc., GE, DC, USA). 2.32. Microarray data analysis An Agilent Microarray Scanner (Agilent Technologies) was used in the present study. Data were obtained using Feature Extraction software 10.7 (Agilent Technologies). Raw data were normalized by Quantile algorithm, Gene Spring Software 11.0 (Agilent Technologies). The mRNAs were considered to be differentially expressed when the fold change was >2 (p < 0.05). A volcano plot was used to visualize differentially expressed genes and was subsequently processed for hierarchical clustering analysis using Gene Spring Software 11.0 (Agilent Technologies). Finally, Pearson correlation coefficients between differentially expressed mRNAs were calculated, and co-expression networks of mRNAs were constructed. 2.33. Gene ontology (GO) and kyoto encyclopedia of genes and genomes (KEGG) pathway analysis GO terms were used to annotate and classify gene function. The differentially expressed genes were put into the Database for Annotation, Visualization and Integrated Discovery (DAVID; [98]http://david.abcc.ncifcrf.gov/) v6.8, which utilizes GO to identify the molecular function represented in the gene profile. Furthermore, KEGG was used to analyze the potential functions of these genes in metabolic pathways. p < 0.05 was recommended as a cut-off value. 2.34. GSEA analysis of differentially expressed genes The GSEA analysis was done using GSEA software version 2.2.2.0 (49, 70), which used predefined gene sets from the Molecular Signatures Database (MSigDB v5.0) (70). A gene set is a group of genes that shares pathways, functions, chromosomal localization, or other features. For the present study, the C collection sets for GSEA analysis (i.e., C1–C7 collection in MsigDB) and list of ranked genes based on a score calculated as −log10 of P value multiplied by sign of fold-change were used. The minimum and maximum criteria for selection of gene sets from the collection were 10 and 500 genes, respectively. 3. Results 3.1. Verification of the binding propensity of HYP to Aβ Quercetin-3-O-β-d-galactopyranoside (Hyperoside, HYP), belongs to the class of flavonol glycosides and is derived from Crataegus pinnatifida Bunge [[99]37] (a hawthorn in the Rosaceae family), was shown in [100]Fig. 1A. Aβ is a well-known biomarker correlated to AD, therefore, the effect of HYP in targeting Aβ was investigated by molecular docking. As shown in [101]Fig. 1B, a binding energy of −8.9 kcal/mol between Aβ fibers and HYP was predicted by BatchMin V13.2. HYP with Aβ molecular surface interaction and hydrogen bonding information were shown in supplementary 1A and 1B. The ligand interaction diagram indicated that the hydroxyl group of HYP bind to Aβ by chain E-F (supplementary 1C). The formation of Aβ aggregate and its particle size was verified by DLS ([102]Fig. 1C). While the average particle size of the Aβ monomer is 365.62 nm, the particle size of Aβ aggregate is 1877 nm ([103]Fig. 1D). The detection on the formation of Aβ aggregate was then performed by using the ThT fluorescence assay. ThT is a benzothiazole-like small molecule that specifically binds to amyloid fibrils [[104]38]. As shown in [105]Fig. 1E, HYP reduced Aβ fibrillation and aggregation. The direct binding affinity of HYP to Aβ aggregate was confirmed by the BLI assay. By applying an increasing concentration of HYP (6.25–200 μM), a dose-dependent direct association of HYP to the biotinylated Aβ aggregate was shown by the association / dissociation binding curve of HYP towards Aβ aggregate ([106]Fig. 1F). Together with [107]Fig. 1G presenting the steady-state analysis of the binding curves and the binding affinity (KD), association rate constant (Kon) and dissociation rate constant (Kids) of HYP to Aβ, the results showed that the KD value of HYP binding to Aβ aggregates was 14.5 μM. To further confirm the Aβ binding propensity of HYP, increasing concentration of HYP was incubated with 30 μM of Aβ aggregate. By using HYP alone (without Aβ aggregate incubation) as the control, all incubation mixtures were analyzed by using LC/MS/MS under the same chromatographic conditions according to our previous reported detection method [[108]39]. The quantitative reduction (%) in the peak area, representing the Aβ binding propensity of the tested compound [[109]39] is used for statistical analysis. [110]Fig. 1H showed the MRM of HYP alone (S1), HYP (20–80 μM) with the incubation of Aβ aggregates (S2), and the merged image (S3), with the peak area of all the identified components calculated and analyzed. As the concentration of HYP increased, [111]Fig. 1I showed the percentage (%) of decrease in the peak area of HYP in MRM increased, indicating the dose-dependent binding affinity of HYP to Aβ aggregate. These data confirmed the Aβ binding propensity of HYP by both spectroscopic and chemical approaches. Fig. 1. [112]Fig. 1 [113]Open in a new tab The binding propensity of HYP to Aβ (A) Chemical structure of HYP. (B) Computational docking of HYP with Aβ (2MXU). HYP and Aβ (2MXU) were represented as green sticks and red ribbons, respectively. (C) ZetaView measurements of the mean hydrodynamic diameter of Aβ monomer and Aβ aggregates. (D) Particle size analysis of Aβ monomer and Aβ aggregates. (E). Measurement of Aβ (1–42) fibrillation. Fluorescence intensity (A.U.) of fibrillated Aβ (1–42) from the indicated treatment groups were monitored from day 1–7 by ThT assay. (F) Kinetic binding sensorgrams of increasing concentrations of HYP from 6.25 to 200 μM. The Y axis (nm) represented the optical thickness changes of the Aβ sensor layer during the interaction of HYP and Aβ. (G) Steady-state analysis of the binding curves and measured parameters of the binding interaction: binding rate constant (Kon), dissociation rate constant (Kids), and affinity constant (KD). (H) MRM chromatograms of HYP-treated HT22 cells. S1: HYP (20, 40 or 80 μM) alone groups. S2: 30 μM aggregated Aβ co-incubated with 20, 40 or 80 μM of HYP groups; S3: Overlaid chromatograms of S1 and S2. (I) The quantitative reduction (%) of HYP peak area in MRM chromatograms of HYP (20–80 μM) with or without the co-incubation of 30 μM aggregated Aβ. All data were mean ± S.D. compared with the HYP alone groups, n = 3. *P < 0.05, **P ≤ 0.01, ***P ≤ 0.001, one-way ANOVA analysis. (For interpretation of the references to colour in this