Isorhynchophylline ameliorates cognitive impairment via modulating amyloid pathology, tau hyperphosphorylation and neuroinflammation: Studies in a transgenic mouse model of Alzheimer’s disease

Brain, Behavior, and Immunity 82 (2019) 264–278

Contents lists available at ScienceDirect

Brain, Behavior, and Immunity journal homepage: www.elsevier.com/locate/ybrbi

Isorhynchophylline ameliorates cognitive impairment via modulating amyloid pathology, tau hyperphosphorylation and neuroinflammation: Studies in a transgenic mouse model of Alzheimer’s disease

T

Hui-Qin Lia, Siu-Po Ipa,b, Qiu-Ju Yuana,b, Guo-Qing Zhengc, Karl K.W. Tsimd, Tina T.X. Dongd, ⁎ ⁎ Ge Line, Yifan Hanf, Yue Liug, Yan-Fang Xiana,b, , Zhi-Xiu Lina,b,h, a

School of Chinese Medicine, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong Special Administrative Region Brain Research Centre, School of Chinese Medicine, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong Special Administrative Region Department of Neurology, The Second Affiliated Hospital and Yuying Children’s Hospital of Wenzhou Medical University, Zhejiang Province, PR China d Division of Life Science, The Hong Kong University of Science and Technology, Hong Kong Special Administrative Region e School of Biomedical Sciences, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong Special Administrative Region f Department of Applied Biology & Chemical Technology, The Hong Kong Polytechnic University, Hong Kong Special Administrative Region g Cardiovascular Disease Centre, Xiyuan Hospital of China Academy of Chinese Medical Sciences, Beijing, PR China h Hong Kong Institute of Integrative Medicine, The Chinese University of Hong Kong, Hong Kong Special Administrative Region b c

A R T I C LE I N FO

A B S T R A C T

Keywords: Isorhynchophylline Alzheimer’s disease Aβ deposition Tau hyperphosphorylation Neuroinflammation TgCRND8 mice

Isorhynchophylline (IRN) has been demonstrated to have distinct anti-Alzheimer’s disease (AD) activity in several animal models of AD. In this study, we aimed at evaluating the preventive effect of IRN on the cognitive deficits and amyloid pathology in TgCRND8 mice. Male TgCRND8 mice were administered with IRN (20 or 40 mg/kg) by oral gavage daily for 4 months, followed by assessing the spatial learning and memory functions with the Radial Arm Maze (RAM) test. Brain tissues were determined immunohistochemically or biochemically for changes in amyloid pathology, tau hyperphosphorylation and neuroinflammation. Our results revealed that IRN (40 mg/kg) significantly ameliorated cognitive deficits in TgCRND8 mice. In addition, IRN (40 mg/kg) markedly reduced the levels of Aβ40, Aβ42 and tumor necrosis factor (TNF-α), interleukin 6 (IL-6) and IL-1β, and modulated the amyloid precursor protein (APP) processing and phosphorylation by altering the protein expressions of β-site APP cleaving enzyme-1 (BACE-1), phosphorylated APP (Thr668), presenilin-1 (PS-1) and anterior pharynx-defective-1 (APH-1), as well as insulin degrading enzyme (IDE), a major Aβ-degrading enzyme. IRN was also found to inhibit the phosphorylation of tau at the sites of Thr205 and Ser396. Immunofluorescence showed that IRN reduced the Aβ deposition, and suppressed the activation of microglia (Iba-1) and astrocytes (GFAP) in the cerebral cortex and hippocampus of TgCRND8 mice. Furthermore, IRN was able to attenuate the ratios of p-c-Jun/c-Jun and p-JNK/JNK in the brains of TgCRND8 mice. IRN also showed marked inhibitory effect on JNK signaling pathway in the Aβ-treated rat primary hippocampus neurons. We conclude that IRN improves cognitive impairment in TgCRND8 transgenic mice via reducing Aβ generation and deposition, tau hyperphosphorylation and neuroinflammation through inhibiting the activation of JNK signaling pathway, and has good potential for further development into pharmacological treatment for AD.

Abbreviations: Aβ, beta-amyloid; AD, Alzheimer’s disease; ADAM, A disintegrin and metalloproteinase; APH-1, anterior pharynx-defective-1; APP, amyloid precursor protein; BACE-1, β-site APP cleaving enzyme 1; BBB, blood-brain barrier; CMC-Na, sodium carboxymethyl cellulose; ELISA, enzyme-linked immunosorbent assay; GFAP, glial fibrillary acidic protein; HFIP, hexafluoroisopropanol; IDE, insulin degrading enzyme; IL-6, interleukin 6; IL-1β, interleukin-1beta; IRN, isorhynchophylline; JNK, C-Jun N-terminal kinase; MTT, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2-H-tetrazolium bromide; NEP, Neprilysin; PCR, polymerase chain reaction; PHFs, paired-helical filaments; PS-1, presenilin-1; RAM, radial Arm Maze; SD, sprague-Dawley; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; Tg, TgCRND8; TNF-α, tumor necrosis factor; URCU, Uncariae Ramulus Cum Uncis; WT, wild-type ⁎ Corresponding authors at: School of Chinese Medicine, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong Special Administrative Region. E-mail addresses: [email protected] (H.-Q. Li), [email protected] (S.-P. Ip), [email protected] (Q.-J. Yuan), [email protected] (G.-Q. Zheng), [email protected] (K.K.W. Tsim), [email protected] (T.T.X. Dong), [email protected] (G. Lin), [email protected] (Y. Han), [email protected] (Y. Liu), [email protected] (Y.-F. Xian), [email protected] (Z.-X. Lin). https://doi.org/10.1016/j.bbi.2019.08.194 Received 14 March 2019; Received in revised form 14 August 2019; Accepted 27 August 2019 Available online 30 August 2019 0889-1591/ © 2019 Elsevier Inc. All rights reserved.

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

1. Introduction

and spatial working memory and reference memory deficits are present at 3 months of age (Chishti et al., 2001). Apart from Aβ plaque deposition and cognitive impairments, neuroinflammation and tau hyperphosphorylation were also found in the brains of 5.5-month-old TgCRND8 mice (Chishti et al., 2001; Durairajan et al., 2012). All these features render TgCRND8 mice an accurate and valuable tool for examining new therapeutics for early stage of AD, as well as for evaluating the underlying mechnisms including Aβ pathology and Aβ plaqueassociated pathogenesis (Woodhouse et al., 2009). In this study, we aimed to investigate the preventive and therapeutic effects of IRN on AD using the TgCRND8 mice as an experimental tool, and to unravel the molecular mechanisms underlying the actions of IRN on Aβ production and deposition, tau hyperphosphorylation, and Aβ plaque-associated neuroinflammation.

Alzheimer’s disease (AD), the most common neurodegenerative disease with inexorable deterioration of memory, is pathologically characterized by intercellular β-amyloid (Aβ) plaques and intracellular neurofibrillary tangles (Blennow et al., 2006). The Aβ is generated by sequential cleavages of amyloid precursor protein (APP) by β- and γsecretase. The dominant components of neurofibrillary tangles are the paired-helical filaments (PHFs) and straight filaments, both of which are composed predominantly of insoluble polymers of abnormally hyperphosphorylated microtubule-associated protein τ (tau) (Lee et al., 2001). Treatment strategies for AD based on the amyloid and tau hypotheses mainly target β- and/or γ-secretase, tau kinase inhibition, and Aβ immunotherapy; however, a number of clinical trials conducted recently to test therapeutic agents on these targets failed because of either serious side effects or lack of therapeutic efficacy (Hampel et al., 2009; Egan et al., 2018; Honig et al., 2018). Given multiple pathologies are associated with AD, development of novel therapeutics that target these multiple pathologies may be desirable and advantageous in combating AD progression. Chinese herbal medicines are known to possess multiple components and exert therapeutic effects through multiple pathways for disease prevention and treatment, rendering it an attractive alternation for AD treatment (Howes and Perry, 2011). Isorhynchophylline (IRN) (Chemical structure of IRN is shown in Fig. 1) is a c-22 oxindole alkaloid isolated from a commonly prescribed Chinese herb named Uncariae Ramulus Cum Uncis (URCU) (Gou-Teng in Chinese) (Chinese Pharmacopoeia Commission, 2015). In Chinese medicine practice, URCU is widely prescribed in many formulae such as Choto San and Yi-Gan San for the treatment of several types of dementia (Terasawa et al., 1997; Iwasaki et al., 2005). Alkaloids are the predominant pharmacologically active components of URCU, and IRN is one of the major alkaloidal constituents (Kang et al., 2004; Yuan et al., 2009). The total alkaloid content in URCU is about 0.2%, among which IRN accounting for about 15% (Shi et al., 2003). IRN can easily pass through the blood–brain barrier (BBB) (Imamura et al., 2011; Zhang et al., 2017, 2019). Moreover, previous studies revealed that IRN possessed potential neuroprotective effects against Aβ25-35-, D-gal- and AlCl3-induced cognitive impairments in a number of experimental AD models, and the underlying molecular mechanisms involved inhibition of acetylcholinesterase activity, neuronal apoptosis, tau protein hyperphosphorylation, oxidative stress and neuro-inflammation (Kang et al., 2002; Yuan et al., 2009; Xian et al., 2012a, 2013, 2014a,b; Li et al., 2018). These findings suggest that IRN is a promising multifunctional phytochemical with neuroprotective properties warranting further investigation. However, the in vivo efficacy of IRN on Aβ production and clearance in AD brains is still unclear. TgCRND8 transgenic mouse strain is a well-characterized line of APP transgenic mouse of AD, and expresses a transgene incorporating both the Indiana mutation (V717F) and the Swedish mutation (K670N/ M671L) in the human APP695 gene on a hybrid C3H/He-C57BL/6 background (Chishti et al., 2001). This model shows progressive, agerelated cognitive impairments that parallels with increases in Aβ deposition (Hyde et al., 2005). In TgCRND8 mice, Aβ plaque deposition

2. Methods 2.1. Drugs and chemical reagents IRN (purity ≥ 98%) was obtained from Chengdu Mansite Pharmaceutical Co. Ltd. (Chengdu, Sichuan, China). Its identity was confirmed by comparing its H1 NMR spectra with that published in the literature (Haginiwa et al., 1973). Donepezil hydrochloride (referred to simply as donepezil thereafter) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Aβ42 peptide was obtained from GL Biochem Ltd. (Shanghai, China). The JNK inhibitor SP600125 was purchased from Abcam (Cambridge, MA, USA). Penicillin-streptomycin-neomycin (PSN) antibiotic mixture (Cat. 15640055), GlutaMAX™ Supplement (Cat. 35050061), MEM with HEPES (glutamin(−)) (Cat. 12360038), B27(TM) plus supplement (50X) (Cat. A3582801) and Neurobasal (TM) plus medium (Cat. A3582901) were obtained from Thermo Fisher Scientific Co. (Fair Lawn, NJ, USA). 2.2. Animals The genetic background of TgCRND8 mice is (C57BL/6J) × (C3H/ HeJ × C57BL/6J), so the male TgCRND8 mice and female non-transgenic mice on the hybrid C3H/He-C57BL/6 background were used to breed a colony of experimental animals. Non-transgenic littermates that do not express human APP transgene were identified as wild-type mice and used as negative controls for experiments. The mice were bred in the Run Run Shaw Science Building, The Chinese University of Hong Kong, and routinely maintained on a 12 h light/dark cycle under controlled humidity (50 ± 10%) and temperature (22 ± 2 °C), with access to food and water ad libitum. The experimental procedures were approved by the Animal Experimentation Ethics Committee of The Chinese University of Hong Kong (Ref. No. 15/018/MIS-5-C). 2.3. Polymerase chain reaction (PCR) for genotyping All mice were subjected to genotyping for the APP transgene. DNA was isolated from the mouse tail. The APP transgene cassette was detected by a transgene-specific PCR reaction using the following primers: F - TGTCCAAGATGCAGCAGAACGGCTAC, R - GGCCGCGGAGAAATG AAGAAACGCCA. Briefly, a visible amount of tail was incubated in the non-SDS tissue digesting buffer (500 mM KCl, 100 mM Tris-HCl, 0.1 mg/ml gelatin, 0.45% NP-40 (IgepalTM CA-630) and 0.45% Tween 20) containing proteinase K (Cat. V900887, Sigma) at 55 °C overnight. After heated at 98 °C for 10 min to inactivate proteinase K, the supernatant was collected and then underwent PCR reaction with a TaKaRa Taq™ package (Cat. R001A, TaKaRa) and the primers. The reactions were run at 95 °C for 5 min, followed by 45 cycles at 95 °C for 30 s, 54 °C for 40 s, 72 °C for 80 s and 72 °C for 10 min. PCR products were separated by 1% agarose gel, and visualized under UV light for 30 s. Those mice with APP transgene were identified as transgenic mice, while those without APP transgene as wild type ones.

Fig. 1. Chemical structure of isorhynchophylline (IRN), a tetracyclic oxindole alkaloid. 265

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

Fig. 2. Experimental design and schedule for evaluating the neuroprotective effects of IRN on TgCRND8 transgenic mice.

entries to complete the test, (2) the number of reference memory errors (RMEs), meaning the entries into a non-baited arm during the test, and (3) the number of working memory errors (WMEs), meaning entries into the baited arms that had already been visited during the same test.

2.4. Experimental design and drugs treatment in TgCRND8 mice At the age of two months, the male mice were randomly assigned to five groups: (a) wild type (WT); (b) TgCRND8 (Tg) + vehicle; (c) and (d) Tg + IRN (20 mg/kg and 40 mg/kg); and (e) Tg + donepezil (5 mg/ kg). The dosages of IRN were selected based on our previous studies (Xian et al., 2014a, b; Li et al., 2018). Donepezil was chosen as a positive control as published previously (Xian et al., 2014a, b; Li et al., 2018). IRN was suspended in 0.5% sodium carboxymethyl cellulose (CMC-Na) and donepezil was dissolved in normal saline. Each mouse with IRN treatment received IRN at a dose of 20 or 40 mg/kg body weight once daily by oral gavage for 4 months. The mice with donepezil treatment received donepezil at a dose of 5 mg/kg, whereas the mice in the WT group and Tg + vehicle group received an equal volume of 0.5% CMC-Na. After the treatment, the spatial learning and memory functions were assessed using the Radial Arm Maze (RAM) test. Fig. 2 shows the experimental design and schedule.

2.6. Preparation of brain samples Twenty-four hours after the RAM test, the brain samples were collected. A portion of the mice (n = 6/group) were decapitated under deep anesthesia and the brain tissues removed quickly. After washed with ice-cold normal saline, the brains were bisected in the midsagittal plane. One hemisphere was used for the enzyme-linked immunosorbent assay (ELISA) kits analysis, while the opposite hemisphere used for western blotting analysis. These samples were immediately stored at −80 °C until used. On the other hand, for immunofluorescence analysis, the other portion of mice (n = 5/group) were transcardially perfused with 0.9% saline followed by buffered formalin (4 g NaH2PO4, 6.5 g Na2HPO4, 100 mL formaldehyde, 900 mL H2O) under deep anesthesia. Then, the brain tissues were collected and fixed in the buffered formalin for 24 h, dehydrated with alcohol, and embedded in paraffin. These paraffin samples were stored at room temperature.

2.5. RAM test The spatial learning and memory functions of mice were assessed using a spatial task involving a RAM. The modular radial arm maze with a video tracking software of SuperMaze V2.0 was purchased from Xinruan Information Technology Co. Ltd (Xinruan, Shanghai, China). The RAM consists of a central platform (22 cm in diameter) and eight radial arms (35 cm long, 5 cm wide and 10 cm high), numbered from 1 to 8. The RAM test was conducted as described previously (Li et al., 2018). During the behavioral assessment, to stimulate hunger, the animals were kept on restricted diet with only water being available ad libitum. Their body weights were maintained at 85–90% of free-feeding level. The RAM test lasted for a total of 8 days: 2 days for habituation trials, 5 days for training trials, 1 day for tasks test. At the beginning of each habituation trial, all arms were baited with several food pellets of 10 mg each, and 3 or 4 mice were simultaneously placed in the central platform. After two days of habituation trial, mice were trained for 5 consecutive days, with 1 trial per day. At the beginning of each training trial, only one mouse was placed in the central platform, and only 4 constant arms were baited with one 10 mg food pellet, which was placed in the nontransparent food cup to prevent visual detection. The mice were trained to run to the end of the baited arms and consume all the food pellets. At the eighth day, the mice were subjected to working and reference memory tasks test. In the tasks test, the same four arms were baited with one 10 mg food pellet as that in each training trial. An arm entry was counted when all four limbs of the mice were within an arm. A trial or a test was ended after a maximum of 10 min had elapsed or when all of food pellets had been consumed. During the tasks test, the following data were recorded using a video-tracking system of SuperMaze V2.0 (Xinruan, Shanghai, China): (1) the number of total

2.7. Preparation of Aβ42 oligomer Soluble Aβ42 oligomer was prepared as described previously (Chen et al., 2017). Briefly, synthetic Aβ42 peptide was dissolved in hexafluoroisopropanol (HFIP, Sigma, St. Louis, MO, USA) to become Aβ42 monomers. Aβ monomers were spin-vacuumed just before the experiments, dissolved in HFIP solution (final concentration: 10% (v/v) HFIP) and incubated for 20 min at room temperature. The solution was centrifuged at 14,000 rpm for 15 min at 4 °C and the supernatant was collected. Aβ solution was obtained after thorough evaporation of HFIP. The solution was kept under constant stirring for 48 h at room temperature, and then quantified by Bradford method with protein assay dye reagent (Bio-Rad, USA). 2.8. Primary rat hippocampal neurons and drug treatment Primary rat hippocampal neurons were prepared from the hippocampus of E17-E18 Sprague-Dawley (SD) rat embryos under sterile condition. After anesthesia and sterilization, the hippocampus was isolated and cut into small pieces, then digested in 0.05% trypsin-EDTA at 37 °C for 15 min. After terminating the digestion with 10% fetal bovine serum (FBS), the digested tissue was triturated, then undisturbed for 5 min to allow the non-dissociated tissue to settle at the bottom. The upper fraction was transferred to another tube. The cells were resuspended in planting medium (45 mL MEM with HEPES 266

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

levels of TNF-α (Cat. No: ab100747), IL-6 (Cat. No: ab100712) and IL1β (Cat. No: ab100704) in the supernatants were determined using commercially available ELISA kits (Abcam, Cambridge, UK) according to the manufacturer’s instructions. The levels of TNF-α, IL-6 and IL-1β were expressed as pg/mg protein.

(glutamin (-)), 5 mL FBS, 500 µL GlutaMax and 250 µL PSN) and plated onto poly-D-lysine-coated plates at a density of 5 × 104 cells/well, followed by incubating for 45–60 min at 37 °C in a humidified atmosphere of 95% air and 5% CO2. After cells attached to the substrate, the planting medium was replaced with neuronal culture medium (49 Ml serum-free Neurobasal (TM) plus medium, 1 mL B27 (TM) plus supplement, 250 µL GlutaMax, 250 µL PSN), followed by incubation for 4 days with all the medium being changed every 2 days to ensure cell maturation. The cells were then subjected to drug treatments. To determine the cytotoxicity of Aβ42 or IRN, primary rat hippocampal neurons were treated with different concentrations of Aβ42 (1, 2.5, 5 and 10 μM) or IRN (6.5, 12.5, 25, 50 and 100 μM) for 24 or 48 h. To evaluate the preventive effect of IRN, primary rat hippocampal neurons were pre-treated with different concentrations of IRN (final concentrations: 1, 20, 40 and 80 μM) or SP600125 (final concentrations: 0.1 and 20 μM) for 2 h. The concentrations of SP600125 were chosen as per previously published protocol (Chen et al., 2018). Aβ42 at a final concentration of 5 μM was then added to the cells for an additional 24 h.

2.12. Western blotting For preparation of protein lysates, frozen brain tissues or cells were homogenized in RIPA lysis buffer (150 mM NaCl, 50 mM Tris-HCl (pH 8.0), 1% NP-40, 0.1% SDS, 0.5% sodium deoxycholate and 1% Protease/Phosphatase Inhibitor Cocktail) (Cat. 5872S, Cell Signaling Technology, USA)) for 30 min on ice. After centrifugation at 12,000 rpm at 4 °C for 15 min, the supernatants were collected. Protein concentrations were determined by Bradford method. Equal amounts of proteins of different samples were loaded. The proteins were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDSPAGE) and then transferred to PVD membranes. After being blocked with 5% (w/v) non-fat milk in TBS-T (150 mM NaCl, 50 mM Tris and 0.1% Tween-20, pH 7.4) at room temperature for 2 h, the PVD membranes were incubated at 4 °C overnight with antibodies against A Disintegrin and Metalloproteinase-10 (ADAM-10) (Cat. sc-48400, Santa Cruz, USA), insulin degrading enzyme (IDE) (Cat. sc-393887, Santa Cruz), β-site APP cleaving enzyme-1 (BACE-1) (Cat. SAB2100200, Sigma), presenilin-1 (PS-1) (Cat. sc-365450, Santa Cruz), anterior pharynx-defective-1 (APH-1) (Cat. PRS4001, Sigma), neprilysin (NEP) (Cat. AP1126-SP, R&D Systems), p-APP (Thr688) (Cat. 6986S, Cell Signaling Technology), CTFs (Cat. A8717, Sigma), p-tau (Ser396) (Cat. sc-12414, Santa Cruz), p-tau (Ser404) (Cat. sc-12952, Santa Cruz), ptau (Thr205) (Cat. sc-101817, Santa Cruz), tau (Tau 46) (Cat. sc-32274, Santa Cruz), phospho-c-Jun N-terminal kinase (p-JNK) (Cat. sc-12882, Santa Cruz), JNK (Cat. sc-7345, Santa Cruz), p-c-Jun (Cat. sc-822, Santa Cruz), c-Jun (Cat. sc-74543, Santa Cruz) and β-actin (Cat. sc-69879, Santa Cruz). After rinse with TBS-T for 5 min × 3 times, the PVD membranes were then incubated with secondary antibodies for 2 h at room temperature. After rinse again with TBS-T for 5 min × 3 times, the protein bands were visualized by the ECL western blotting detection reagents (Cat. WP20005, Amersham Biosciences, Buckinghamshire, UK). The intensity of each band was analyzed using Image J software (NIH Image, MD, USA).

2.9. Cell viability assay Cell viability in the presence of IRN was measured by quantitative colorimetric assay with the conventional 3-(4,5-dimethyl-2-thiazolyl)2,5-diphenyl-2-H-tetrazolium bromide (MTT) method. Briefly, after drug treatments, 20 μL/well of MTT solution (final concentration of 1 mg/mL) was added, and cells were incubated at 37 °C for 4 h. The supernatants were then aspirated off, and formazan crystals dissolved with 150 μL of DMSO for 15 min. The optical density of each well was determined at 492 nm using a FLUOstar OPTIMA microplate reader (BMG Labtech, Offenbury, Germany). Cell viability was expressed as percentage of the non-treated control. 2.10. Measurements of levels of Aβ40 and Aβ42 in the brain extracts The levels of Aβ40 and Aβ42 in the brain extracts were quantified using a commercial mouse Aβ40 ELISA kit (Cat. KMB3481, Invitrogen, USA) and a mouse Aβ42 ELISA kit (Cat. KMB3441, Invitrogen, USA) as per the manufacturer’s protocols. Briefly, frozen hemisphere was homogenized in 8 × volume of homogenization buffer (5 M guanidineHCl diluted in 50 mM Tris (pH 8.0), 1 × protease inhibitor cocktail with AEBSF (Cat. P2714, Sigma, USA)). The homogenate was then mixed on an orbital shaker at room temperature for 3 to 4 h. After centrifugation at 16,000 × g for 20 min at 4 °C, the supernatant was collected and diluted with standard diluent buffer to an appropriate concentration. The diluted supernatant was added into the wells that pre-coated with mAb to NH2 terminus of Aβ, and then incubated for 2 h at room temperature to bind antigen. After wash with 1× wash buffer, the mouse Aβ detection antibody solution and anti-rabbit IgG HRP solution were sequentially added according to the assay protocol. The reaction was terminated by addition of stop solution immediately before the absorbance was read at 450 nm on a microplate spectrophotometer (BMG Labtech). Unless otherwise indicated, all reagents mentioned above were provided in the kits. The levels of Aβ40 and Aβ42 in the brain extracts were calculated using the standard curves, then normalized to the weight of brain tissue and expressed as pg/mg brain tissue. These two sandwich ELISA kits exclusively recognize both natural and recombinant mouse Aβs, measuring both of the “soluble” and “insoluble” Aβs extracted by guanidine-HCl in the brain tissue.

2.13. Immunofluorescence assay Paraffin sections (6 μm thick) containing hippocampus were prepared using a Leica radial microtome and mounted on triethoxysilanecoated glass microscope slides. After deparaffinization, the sections were immersed in 10 mM citrate buffer (pH 6.0) and heated at 95 °C for 20 min to achieve antigen retrieval. After wash with 1 × PBS for 5 min × 2 times, the sections were blocked in 5% bovine serum albumin (BSA) at room temperature for 20 min. The samples were then incubated with the following primary antibodies: biotin anti-β-amyloid 17–24 antibody (4G8, Cat. 800704, Biolegend, USA), anti-GFAP polyclonal antibody (Cat. C106874, Sigma, USA) and anti-Iba-1 antibody (Cat. 019-19741, Wako, Japan), which were used to detect Aβ plaque, astrocyte and microglia, respectively. After wash with 1 × PBS for 5 min × 3 times, the sections were then incubated with Alexa Fluor™ 647 streptavidin conjugated secondary antibody (Cat. S21374, Invitrogen, USA) or Alexa Fluor 488 Goat anti-Rabbit IgG (H + L) secondary antibody (Cat. R37116, Invitrogen, USA). BSA was used instead of primary antibody as a negative control. Cell nucleus was detected by DAPI staining if necessary. All images were acquired using a Nikon fluorescent inverted microscope with an image acquisition system (Nikon Instruments Inc. Melville, NY, USA). Images were thresholded and measured using an unbiased computer-assisted Image J software (NIH, Bethesda, MD, USA). Aβ plaque burden was quantified by the proportion of the region of interest occupied by Aβ as described

2.11. Cytokine determination The brain tissues of mice were homogenized vigorously in 0.5 mL of lysis buffer [50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.5% NP40, 1 mM Na3VO4, 1 mM NaF and 1 mM DTT]. After incubated on ice for 15 min, the homogenates were centrifuged at 10,000 × g for 30 min at 4 °C. The 267

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

previously (Josephs et al., 2008). Both astrogliosis and microgliosis were measured by cells density, which was quantified by dividing the number of cells by the total area of the region of interest as described previously (Thompson et al., 2001). 2.14. Statistical analysis All data were presented as the mean ± SEM. Multiple group comparisons were performed using one-way ANOVA followed by Post-hoc Bonferroni’s test to detect inter-group differences. GraphPad Prism software (Version 5, GraphPad Software, Inc., CA, USA) was used to perform the statistical analysis. A difference was considered statistically significant when the p < 0.05. 3. Results 3.1. Effects of IRN on the cognitive deficits in TgCRND8 transgenic mice The effects of IRN on spatial learning and memory deficits were determined using the RAM test. Fig. 3A shows that the number of total entries was markedly increased (F (4, 42) = 9.445, p < 0.001) in the Tg + vehicle group when compared with the WT group. Treatment with IRN (40 mg/kg) significantly attenuated the number of total entries (p < 0.05) when compared with the Tg + vehicle group. Donepezil (5 mg/kg) treatment also markedly reduced the number of total entries (p < 0.01) when compared with the Tg + vehicle group. However, IRN treatment (20 mg/kg) exerted no significant difference in the number of the total entries (p > 0.05) when compared with the Tg + vehicle group. The effects of IRN on the numbers of RMEs and WMEs were depicted in Fig. 3B and C, respectively. The results showed that the numbers of RMEs (F (4, 45) = 10.42, p < 0.001) and WMEs (F (4, 45) = 7.513, p < 0.001) were significantly elevated in the Tg + vehicle group when compared with the WT group. Treatment with IRN (40 mg/kg) significantly reduced the numbers of RMEs (p < 0.05) and WMEs (p < 0.05) when compared to the Tg + vehicle group. Donepezil (5 mg/kg) treatment also significantly reduced the numbers of RMEs (p < 0.01) and WMEs (p < 0.01) when compared with the Tg + vehicle group. However, IRN treatment (20 mg/kg) showed no significant differences in the numbers of RMEs (p > 0.05) or WMEs (p > 0.05) as compared to the Tg + vehicle group. 3.2. Effects of IRN on the levels of Aβ40 and Aβ42 in the brain tissues of TgCRND8 mice Fig. 4A and B show the effects of IRN on the levels of Aβ40 and Aβ42 in the brain tissues of TgCRND8 mice, respectively. The levels of Aβ40 (F (4, 25) = 50.7, p < 0.001) and Aβ42 (F (4, 25) = 65.86, p < 0.001) were significantly increased in the brain tissues of TgCRND8 mice when compared to the WT group. The ratio of Aβ42 level to Aβ40 level is about 2.7 in the TgCRND8 transgenic mice. When compared with the Tg + vehicle group, treatment with IRN (40 mg/kg) significantly reduced both Aβ40 (p < 0.05) and Aβ42 levels (p < 0.05), with reduction by 25.5% and 23.0%, respectively. On the other hand, Donepezil (5 mg/kg) treatment significantly decreased the Aβ42 level (p < 0.05), but not on the Aβ40 level (p > 0.05) when compared with the Tg + vehicle group. IRN treatment (20 mg/kg) had no significant effects on the levels of Aβ40 and Aβ42 in the brain tissues of TgCRND8 mice (p > 0.05 for both). Fig. 3. Effects of IRN on spatial learning and memory of TgCRND8 mice as evaluated by the RAM test after 7 days of training. The total entries to complete the task (A), the reference memory errors (RMEs) (B), and the working memory errors (WMEs) (C). Data were expressed as mean ± SEM (n = 10). ### p < 0.001 when compared with the WT group; * p < 0.05 and ** p < 0.01 when compared with the Tg + vehicle group.

3.3. Effects of IRN on the APP processing and APP phosphorylation in the brain tissues of TgCRND8 mice In the brain tissues of TgCRND8 mice, the protein levels of BACE-1 (F (4, 10) = 8.000, p < 0.05), p-APP (T668) (F (4, 10) = 13.150, p < 0.001), APH-1 (F (4, 10) = 10.270, p < 0.001), PS-1 (F (4, 268

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

TgCRND8 were significantly elevated when compared with the WT group (Fig. 6). IRN treatment (20 and 40 mg/kg) effectively mitigated the level of p-tau (T205) (p < 0.05 for both) when compared to the Tg + vehicle control group. On the other hand, IRN treatment (20 mg/ kg) and donepezil (5 mg/kg) effectively inhibited the level of p-tau (S396) (p < 0.05 for both) when compared to the Tg + vehicle control group. However, no significant differences were found among groups in the levels of p-tau (S404) (F (4, 16) = 0.2814, p > 0.05). 3.5. Effects of IRN on Aβ deposition and Aβ plaque-associated neuroinflammation in the hippocampus and cerebral cortex in TgCRND8 mice Fig. 7 shows that the Aβ plaque burdens in the hippocampus (F (4, 20) = 10.88, p < 0.001) and the cortex (F (4, 20) = 11.86, p < 0.001) were significantly accentuated in the TgCRND8 mice when compared with the WT group. IRN treatment (40 mg/kg) significantly decreased the Aβ plaque burdens in both the hippocampus (p < 0.05) and the cortex (p < 0.01) of TgCRND8 mice when compared with the Tg + vehicle group. IRN treatment (20 mg/kg) also significantly attenuated the Aβ plaque burden in the hippocampus (p < 0.05) of TgCRND8 mice when compared with the Tg + vehicle group. Fig. 8 revealed marked increases of the astrocyte density in the hippocampus (F (4, 20) = 4.761, p < 0.01) and the cortex (F (4, 20) = 7.649, p < 0.01) of TgCRND8 mice when compared with the WT group. IRN treatment (40 mg/kg) significantly decreased the astrocyte density in the hippocampus (p < 0.05) and the cortex (p < 0.05) of TgCRND8 mice when compared with the Tg + vehicle group. As shown in Fig. 9, significant increase of microglia density was observed in the hippocampus (F (4, 20) = 12.75, p < 0.001) and the cortex (F (4, 20) = 7.358, p < 0.01) of TgCRND8 mice when compared with the WT group. IRN treatment (40 mg/kg) significantly decreased the microglia density both in the hippocampus (p < 0.01) and the cortex (p < 0.05) of TgCRND8 mice when compared with the Tg + vehicle group. 3.6. Effects of IRN on the expression of cytokines in the brain tissues of TgCRND8 mice As shown in Fig. 10, the protein levels of TNF-α (F (4, 25) = 28.08, p < 0.001), IL-6 (F (4, 25) = 104.2, p < 0.001) and IL-1β (F (4, 25) = 17.50, p < 0.001) were markedly increased in the brain tissues of TgCRND8 mice when compared with the WT group. Treatment with IRN (40 mg/kg) could significantly inhibit the levels of TNF-α (p < 0.001), IL-6 (p < 0.001) and IL-1β (p < 0.001) in the brain tissues of TgCRND8 mice when compared with Tg + vehicle group. Treatment with IRN (20 mg/kg) could also markedly suppress the protein level of IL-1β (p < 0.05).

Fig. 4. Effects of IRN on Aβ40 level (A) and Aβ42 level (B) in the brain tissues of TgCRND8 mice. Data were expressed as mean ± SEM (n = 6). ### p < 0.001 when compared with the WT group; * p < 0.05 when compared with the Tg + vehicle group.

10) = 20.700, p < 0.001) and CTFs (F (4, 10) = 20.350, p < 0.001) were significantly augmented, while the protein expression of IDE (F (4, 10) = 34.590, p < 0.001) was markedly reduced, as compared to the WT group (Fig. 5). Treatment IRN (40 mg/kg) significantly attenuated the expressions of BACE-1 (p < 0.05), p-APP (T668) (p < 0.05), APH1 (p < 0.05), when compared with the Tg + vehicle group. Treatment with IRN (20 mg/kg) and donepezil (5 mg/kg) also markedly mitigated the expressions of APH-1 (p < 0.05 for both) and PS-1 (p < 0.05 for both). However, there was no significant difference among groups in the protein expression of ADAM-10 (F (4, 16) = 0.2985, p > 0.05) and NEP (F (4, 16) = 0.1927, p > 0.05) in the brain tissues of TgCRND8 mice when compared with the WT group (Fig. 5). Treatment with IRN (40 mg/kg) obviously increased the protein expression of IDE (p < 0.001) in the brain tissues of TgCRND8 mice.

3.7. Effects of IRN on the JNK signaling pathway in the brain tissues of TgCRND8 mice Fig. 11 shows significant increases in the ratios of p-c-Jun/c-Jun (F (4, 10) = 24.410, p < 0.001) and p-JNK/JNK (F (4, 10) = 23.630, p < 0.001) in the brain tissues of TgCRND8 mice, when compared with the WT group. When compared with the Tg + vehicle group, IRN treatment (40 mg/kg) markedly attenuated the ratios of p-c-Jun/c-Jun and p-JNK/JNK (p < 0.001 for both). Treatment with IRN (20 mg/kg) also significantly decreased the ratio of p-c-Jun/c-Jun (p < 0.01).

3.4. Effects of IRN on the hyperphosphorylation of tau protein in the brain tissues of TgCRND8 mice

3.8. Effects of IRN on Aβ42-induced cytotoxicity in primary rat hippocampal neurons

The protein levels of p-tau (T205) (F (4, 16) = 9.533, p < 0.05) and p-tau (S396) (F (4, 16) = 19.860, p < 0.05) in the brain tissues of

Treatment of the primary rat hippocampal neuronal cells with 5 μM of Aβ42 for 24 h induced conspicuous cytotoxicity as the cell viability 269

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

Fig. 5. Effects of IRN on the APP processing and APP phosphorylation in the brain tissues of TgCRND8 mice. Data were expressed as mean ± SEM (n = 3). p < 0.05 and ### p < 0.001 when compared with the WT group; * p < 0.05 and *** p < 0.001 when compared with the Tg + vehicle group.

#

Fig. 6. Effects of IRN on tau hyperphosphorylation at sites of T205, S396 and S404 in the brain tissues of TgCRND8 mice. Data were expressed as mean ± SEM (n = 3). # p < 0.05 when compared with the WT group; * p < 0.05 when compared with the Tg + vehicle group.

(F (7, 40) = 31.65, p < 0.001) (Fig. 12C). Treatment with IRN (80 µM) and SP600125 (0.1 and 20 µM) effectively blocked the injury caused by Aβ42 (p < 0.001, p < 0.05 and p < 0.001, respectively) when compared with the Aβ42 group.

was reduced to 50% of the control value (100%) (F (4, 25) = 267.4, p < 0.001) (Fig. 12A). On the other hand, IRN at concentrations up to 100 µM showed no cytotoxicity on the primary rat hippocampal neurons after treatment for 24 or 48 h (Fig. 12B). Our results showed that Aβ42 (5 μM) treatment significantly reduced the cell viability in the cultured primary rat hippocampal neuronal cells 270

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

Fig. 7. Effects of IRN on Aβ deposits in the brains of TgCRND8 mice. Aβ plaque burdens were quantified by the proportion of the region of interest occupied by Aβ. The Aβ plaque burdens in hippocampus and cortex were measured respectively. Data were expressed as mean ± SEM (n = 5). ### p < 0.001when compared with the WT group; * p < 0.05 and ** p < 0.01 when compared with the Tg + vehicle group.

271

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

Fig. 8. Effects of IRN on GFAP-positive astrocytes in the hippocampus and cortex of TgCRND8 mice. Astrocyte density was quantified by dividing the number of astrocytes by the area of the region of interest (cells/mm2). The astrocyte densities in hippocampus and cortex were measured respectively. Data were expressed as mean ± SEM (n = 5). ## p < 0.01 when compared with the WT group; * p < 0.05 when compared with the Tg + vehicle group.

IRN (40 and 80 µM) and SP600125 (20 µM, a JNK inhibitor) significantly suppressed the upregulated ratios of p-JNK/JNK (p < 0.01, p < 0.001 and p < 0.001, respectively) and p-c-Jun/c-Jun (p < 0.01, p < 0.001 and p < 0.001, respectively) in the Aβ42-treated primary rat hippocampal neurons (Fig. 13).

3.9. Effects of IRN on the JNK signaling pathway in Aβ42-treated rat hippocampal neuronal cells As shown in Fig. 13, Aβ42 upregulated the ratios of p-JNK/JNK (F (5, 12) = 20.62, p < 0.001) and p-c-Jun/c-Jun (F (5, 12) = 13.39, p < 0.001) in the Aβ42-treated primary rat hippocampal neurons, when compared with vehicle control group. However, treatment with

272

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

Fig. 9. Effects of IRN on Iba1-positive microglia in the hippocampus and frontal cortex of TgCRND8 mice. Microglia density was quantified by dividing the number of microglia by the area of the region of interest (cells/mm2). The microglia densities in hippocampus and cortex were measured respectively. Data were expressed as mean ± SEM (n = 5). ## p < 0.01 and ### p < 0.001 when compared with the WT group; * p < 0.05 and ** p < 0.01 when compared with the Tg + vehicle group.

273

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

Fig. 10. Effects of IRN on the expression of cytokines in the brain tissues of TgCRND8 mice. Data were expressed as mean ± SEM (n = 6). compared with the WT group; * p < 0.05 and *** p < 0.001 when compared with the Tg + vehicle group.

4. Discussion

###

p < 0.001 when

Aβ is a key pathogenic molecule in AD. Aβ derives from abnormal processing of APP, which is an integral membrane protein processed by α-, β-, and γ-secretases to release Aβ. As one of the three α-secretases in the ADAM family, ADAM-10 is physiologically relevant and constitutive for APP α-cleavage (Kuhn et al., 2010). BACE-1 is a key β-secretase, and its cleavage of APP initiates Aβ production (Vassar et al., 1999). The γsecretase is a membrane protein complex comprising at least four proteins including PS-1 and APH-1 (Kimberly et al., 2003). Increased Aβ production through sequential cleavage of APP by the β- and γsecretases contributes to the etio-pathological basis of AD (Wilquet and De Strooper, 2004). Therefore, inhibition of the β- or γ-secretase can help to reduce the toxic Aβ production. Our results demonstrated that IRN could significantly inhibit the protein expressions of BACE-1, p-APP (T668), APH-1, and PS-1, and the protein levels of Aβ40 and Aβ42, but enhance the IDE expression and not alter the ADAM-10 expression in the brains of TgCRND8 mice. These results are indicative that IRN could modulate the APP processing through suppressing the activities of βand γ-secretases to clear the Aβ deposition in the brains of TgCRND8 mice. Microgliosis and astrogliosis contribute to the progression and severity of AD. Activated microglia and astrocyte can secrete various pro-

Though IRN is a well-known neuroprotective agent (Kang et al., 2002, 2004; Yuan et al., 2009; Xian et al., 2012a,b, 2013, 2014a,b, 2017; Li et al., 2018), the actual therapeutic role of IRN in Aβ pathology of AD has hitherto not been evaluated. We herein reported for the first time that chronic administration of IRN ameliorated cognitive deficits seen in TgCRND8 mice via reducing Aβ production, modulating the APP processing and hyperphosphorylation, suppressing the neuroinflammation and tau hyperphosphorylation through inhibiting the JNK signaling pathway. Our experimental results indicate that IRN possesses good potential in improving cognitive deficits owing to its multiple functions in clearing Aβ deposition, inhibiting tau protein hyperphosphorylation and suppressing neuroinflammation. In our experiments, however, donepezil exerted no anti-inflammatory action. Moreover, IRN is a safer chemical compound than donepezil, with at least 10 times less toxicity. The LD50 values of IRN and donepezil are about 510 mg/kg and 45 mg/kg respectively (by oral administration), and 80 mg/kg and 3.7 mg/kg respectively (by intravenous administration) in mice (Ozaki 1989; http://www.pfizer.com/files/products/material_safety_data/ PZ00471.pdf). 274

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

Fig. 11. Effects of IRN on JNK signaling pathway in the brain tissues of TgCRND8 mice. Data were expressed as mean ± SEM (n = 3). compared with the WT group; ** p < 0.01 and *** p < 0.001 when compared with the Tg + vehicle group.

###

p < 0.001 when

neuroinflammation (Fernandez et al., 2013; Etminan et al., 2003). Thus, agents with potent anti-inflammatory effect are deemed as a potentially useful strategy to slow down the course of AD. In this study, IRN treatment significantly inhibited the levels of TNF-α, IL-1β and IL-6 in the brain tissues of TgCRND8 mice, indicative of the association of anti-inflammatory property of IRN and its cognitive deficits improving action in TgCRND8 mice. Intracellular neurofibrillary tangles formed by aggregation of hyperphosphorylation tau protein is another pathological hallmark of AD (Blennow et al., 2006). Significant correlations exist between Aβ peptides and tau protein hyperphosphorylation at several specific sites including Ser396, Ser404 and Thr205 (Zhou et al., 2006). It has been reported that the phosphorylation of tau protein is abnormally accentuated at different sites of Thr205 (7.61 times increase), Ser396 (4.95 times increase) and Ser404 (2.97 times increase) in the brains of AD patients (Zhou et al., 2006). Our previous studies showed that pretreatment with IRN markedly inhibited tau protein hyperphosphorylation at the specific sites of Ser396, Ser404, and Thr205 in the Aβ25-35-treated rats and cultured PC12 cells (Xian et al., 2012a, 2014a). Consistent with the earlier findings, the down-regulating effect of IRN on tau protein hyperphosphorylation at the Ser396 and Thr205 sites was also observed in the brains of TgCRND8 transgenic mice. All these results indicate that the inhibitory effect of IRN on these specific hyperphosphorylation sites of tau protein is one of the important molecular mechanisms underlying its cognitive function improving effect. Increasing evidence has shown that JNK is involved in several pathologies in AD. Activation of JNK pathway has been consistently found to be upregulated in the surrounding area of the Aβ plaques in AD patients and transgenic mice. In addition, it has also been found to facilitate APP phosphorylation at Thr668 site in culture cell lines (Braithwaite et al., 2010; Zhou et al., 2015; Thakur et al., 2007). Moreover, activation of JNK pathway has been found to play a cardinal role in hyperphosphorylation of tau at many sites such as Ser396 and Thr205 (Zhou et al., 2006; Ploia et al., 2011). Furthermore, activation of JNK pathway was also shown to mediate the neuroinflammation induced by Aβ in vitro (Vukic et al., 2009). Therefore, inhibition of the JNK pathway may be a target for AD treatment. Our data revealed that

inflammatory factors such as TNF-α, IL-6 and IL-1β that are all associated with Aβ production cascade during the occurrence and development of AD (Bianca et al., 1999; Heneka et al., 1994). Microglia participate in the conversion of nonfibrillar Aβ into amyloid fibrils, thereby facilitating the foundation of amyloid fibrils within plaques (Wisniewski et al., 1989). Aβ could interact with microglia to trigger neuroinflammation and modulate its own metabolism (Huang and Mucke, 2012). Glial fibrillary acidic protein (GFAP) is a marker of dynamic astrocytic responses in brain injury, as a result of either local trauma or neurodegenerative processes (Laping et al., 1994). Evidence also showed that oligomeric and fibrillar Aβ42 dramatically increased astrocytic β-cleavage of APP, while activated astrocytes may represent sources of Aβ during neuroinflammation in AD, suggesting a potential feed-forward vicious cycle of astrocytic activation and Aβ generation (Zhao et al., 2011). Thus, inhibition of microgliosis and astrogliosis might contribute to the reduction of Aβ accumulation. In this study, we found that IRN significantly reduced amyloid deposits along with amyloid plaque-associated reactive microgliosis (Iba-1) and astrogliosis (GFAP) in the hippocampus and frontal cortex of TgCRND8 mice. Neuroinflammation has been widely considered as a possible pathological factor to AD. Neuroinflammation activates glial cells, such as astrocytes and microglia (Rojo et al., 2008), and facilitates the release of cytokines, causing neuronal damage and death (Domingues et al., 2017; McGeer and McGeer, 2013). In AD brains, Aβ may activate astrocytes and induce astrogliosis to release pro-inflammatory cytokines, such as TNF-α, IL-1β and IL-6 (Batarseh et al., 2016; Gonzalez-Reyes et al., 2017). IL-6 is mainly produced by activated microglia and has been found to be increased in the brains of AD and AD animal models (McGeer and McGeer, 2013; Meraz-Rios et al., 2013). IL-6 may also stimulate the synthesis of Aβ, and impair the function of spatial learning and memory (Dugan et al., 2009; Erta et al., 2012). Similarly, both TNF-α and IL-1β, synthesized and released by both activated astrocytes and microglia, have been considered to be major pro-inflammatory cytokines in the brains and play a critical role in the progression of AD (Wang et al., 2015). Anti-inflammatory drugs or other drugs with anti-inflammatory effects have been reported to possess neuroprotective effects in AD models via suppression of 275

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

A

Control ,51ȝ0 ,51ȝ0 ,51ȝ0 ,51ȝ0 ,51ȝ0

B 150

Cell viability (% of control)

impairments on TgCRND8 transgenic mouse model of AD. We also revealed that IRN reduced Aβ production by inhibiting BACE-1, PS-1 and suppressing phosphorylation of APP at Thr668. IRN also facilitated Aβ degradation by upregulating IDE. Besides, IRN inhibited neuroinflammation through reducing activated microglia and astrocytes. These multi-target effects of IRN against AD were mediated, at least in part, via inhibiting the activation of JNK signaling pathway. Fig. 14 is a schematic drawing depicting the mechanisms underlying the actions of IRN in the TgCRND8 mice. Based on the safety and brain bioavailability of IRN, and its ability to alleviate AD pathology, we believe IRN is a promising naturally-occurring alkaloid worthy of further development for the prevention and/or treatment of AD. The findings from our present study will lay a solid foundation for further investigation of IRN-like alkaloids as candidates for Aβ and tau-based therapeutics to modify or delay the onset of Aβ and tau pathologies in AD. If the pathology in this transgenic model is representative of disease pathology in the clinical syndrome, then IRN administration to the early AD patients might be an effective prophylactic strategy for this most common and devastating dementia. 6. Declarations 6.1. Animal experimentation ethics approval

100

The experimental procedures were approved by the Department of Health, The Government of the Hong Kong Special Administrative Region (Ref. No. (15-853) in DH/HA&P/8/2/1Pt.54), and the Animal Experimentation Ethics Committee of The Chinese University of Hong Kong (Ref No. 15/018/MIS-5-C). The experimental procedures were conformed to the Guidelines of the Principles of Laboratory Animal Care (NIH publication No. 80-23, revised 1996).

50

0

24h

48h

C

7. Authors’ contributions LZX, XYF, ZGQ, TWKT, YFH and GL conceived the research idea and designed the experimental protocols. LHQ performed the in vivo and in vitro studies and collected the experimental data. ISP and TTXD procured and authenticated IRN. YQJ conducted the genotyping of the transgenic mice. LHQ, XYF and YL performed the data analysis. LHQ and XYF drafted the manuscript. LZX, XYF, ZGQ, YL, KWKT, TTXD, YFH and GL revised the manuscript. All authors read and approved the final manuscript. Funding This work was supported by the General Research Fund from Research Grants Council of Hong Kong (project no. 14110814) and the CUHK Direct Grant (project no. 2015.1.081 and 2017.076).

Fig. 12. Effects of IRN on Aβ42-induced cytotoxicity in primary rat hippocampal neurons. Primary rat hippocampal neurons cells were incubated with different concentrations of Aβ42 (1, 2.5, 5 and 10 μM), IRN (6.5, 12.5, 25, 50 and 100 μM) or SP600125 (0.1 and 20 μM) for 24 or 48 h. Data were expressed as mean ± SEM (n = 6). ### p < 0.001 when compared with the control group; * p < 0.05 and ***p < 0.001 when compared with the group treated with Aβ42 alone.

Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

IRN significantly reduced the ratios of p-JNK/JNK and p-c-Jun/c-Jun in the brain tissues of TgCRND8 mice and in the Aβ-treated primary rat hippocampal neuronal cells, indicating that inhibiting the activation of JNK signaling pathway represents a key mechanism through which IRN reduces AD pathology and improves cognitive function.

Acknowledgements Not applicable.

5. Conclusions Availability of data and materials To sum, our study has for the first time unambiguously demonstrated that IRN was able to reduce cerebral Aβ levels and Aβ plaque, tau hyperphosphorylation, neuroinflammation, and cognitive

All data supporting the conclusions of this article are included with this article. 276

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

Fig. 13. Effects of IRN on JNK signaling pathway in the primary rat hippocampal neuronal cells. Data were expressed as mean ± SEM (n = 3); ### p < 0.001 when compared with the control group; ** p < 0.01 and *** p < 0.001 when compared with the group treated with Aβ42 alone.

Fig. 14. The schematic drawing depicting the molecular mechanisms associated with the cognitive deficits ameliorating actions of IRN in the TgCRND8 mice. In the TgCRND8 mice, the transmembrane APP is processed in the amyloidogenic pathway, in which APP is sequentially cleaved by β-secretase and γ-secretase, leading to the production of Aβ peptide and formation of Aβ plaque. IRN can reduce Aβ level by inhibiting the amyloidogenic process of APP, thereby reducing the Aβassociated activation of microglia and astrocytes. Hyperphosphorylation of Tau occurs in the TgCRND8 mice, and IRN can inhibit Tau hyperphosphorylation. Activation of JNK signaling pathway also occurs in the TgCRND8 mice, and IRN reduces Aβ level and inhibits Tau hyperphosphorylation via inhibiting the activation of JNK signaling pathway. These molecular actions of IRN finally contribute to therapeutic effects on AD in the TgCRND8 mice as evidenced by improvements in spatial learning and memory.

277

Brain, Behavior, and Immunity 82 (2019) 264–278

H.-Q. Li, et al.

Consent for publication

isorhynchophylline on in vitro ischemia-induced neuronal damage in the hippocampus: putative neurotransmitter receptors involved in their action. Life sci. 76, 331–343. Kimberly, W.T., LaVoie, M.J., Ostaszewski, B.L., 2003. Gamma-secretase is a membrane protein complex comprised of presenilin, nicastrin, Aph-1, and Pen-2. Proc. Natl. Acad. Sci. U.S.A. 100, 6382–6387. Kuhn, P.H., Wang, H., Dislich, B., 2010. ADAM10 is the physiologically relevant, constitutive alpha-secretase of the amyloid precursor protein in primary neurons. EMBO J. 29, 3020–3032. Laping, N.J., Teter, B., Nichols, N.R., 1994. Glial fibrillary acidic protein: regulation by hormones, cytokines, and growth factors. Brain Pathol. 4, 259–275. Lee, V.M., Goedert, M., Trojanowski, J.Q., 2001. Neurodegenerative tauopathies. Annu. Rev. Neurosci. 24, 1121–1159. Li, H.Q., Ip, S.P., Zheng, G.Q., 2018. Isorhynchophylline alleviates learning and memory impairments induced by aluminum chloride in mice. Chin. Med. 13, 29. McGeer, P.L., McGeer, E.G., 2013. The amyloid cascade-inflammatory hypothesis of Alzheimer disease: implications for therapy. Acta Neuropathol. 126, 479–497. Meraz-Rios, M.A., Toral-Rios, D., Franco-Bocanegra, D., 2013. Inflammatory process in Alzheimer’s disease. Front. Integr. Neurosci. 7, 59. Ozaki, Y., 1989. Pharmacological studies of indole alkaloids obtained from domestic plants, Uncaria rhynchophylla Miq. and Amsonia elliptica Roem. et Schult. Nihon yakurigaku zasshi Folia pharmacologica Japonica 94, 17–26. Ploia, C., Antoniou, X., Sclip, A., 2011. JNK plays a key role in tau hyperphosphorylation in Alzheimer’s disease models. J. Alzheimers Dis. 26, 315–329. Rojo, L.E., Fernandez, J.A., Maccioni, A.A., 2008. Neuroinflammation: implications for the pathogenesis and molecular diagnosis of Alzheimer’s disease. Arch. Med. Res. 39, 1–16. Shi, J.S., Yu, J.X., Chen, X.P., 2003. Pharmacological actions of Uncaria alkaloids, rhynchophylline and isorhynchophylline. Acta Pharmacol. Sin. 24, 97–101. Terasawa, K., Shimada, Y., Kita, T., 1997. Choto-san in the treatment of vascular dementia: a double-blind, placebo-controlled study. Phytomedicine 4, 15–22. Thakur, A., Wang, X., Siedlak, S.L., 2007. c-Jun phosphorylation in Alzheimer disease. J. Neurosci. Res. 85, 1668–1673. Thompson, K.A., McArthur, J.C., Wesselingh, S.L., 2001. Correlation between neurological progression and astrocyte apoptosis in HIV-associated dementia. Ann. Neurol. 49, 745–752. Vassar, R., Bennett, B.D., Babu-Khan, S., 1999. β-Secretase cleavage of Alzheimer’s amyloid precursor protein by the transmembrane aspartic protease BACE. Science 286, 735–741. Vukic, V., Callaghan, D., Walker, D., 2009. Expression of inflammatory genes induced by beta-amyloid peptides in human brain endothelial cells and in Alzheimer’s brain is mediated by the JNK-AP1 signaling pathway. Neurobiol. Dis. 34, 95–106. Wang, W.Y., Tan, M.S., Yu, J.T., 2015. Role of pro-inflammatory cytokines released from microglia in Alzheimer's disease. Ann. Transl. Med. 3, 136. Wilquet, V., De Strooper, B., 2004. Amyloid-beta precursor protein processing in neurodegeneration. Curr. Opin. Neurobiol. 14, 582–588. Wisniewski, H.M., Wegiel, J., Wang, K.C., 1989. Ultrastructural studies of the cells forming amyloid fibers in classical plaques. Can. J. Neurol. Sci. 16, 535–542. Woodhouse, A., Vickers, J.C., Adlard, P.A., 2009. Dystrophic neurites in TgCRND8 and Tg2576 mice mimic human pathological brain aging. Neurobiol. Aging 30, 864–874. Xian, Y.F., Fan, D., Ip, S.P., 2017. Antidepressant-like effect of isorhynchophylline in mice. Neurochem. Res. 42, 678–685. Xian, Y.F., Lin, Z.X., Mao, Q.Q., 2012a. Bioassay-guided isolation of neuroprotective compounds from Uncaria rhynchophylla against beta-amyloid-induced neurotoxicity. Evid Based Compl. Alternat. Med. 2012 802625. Xian, Y.F., Lin, Z.X., Mao, Q.Q., 2012b. Protective effect of isorhynchophylline against βamyloid-induced neurotoxicity in PC12 cells. Cell Mol. Neurobiol. 32, 353–360. Xian, Y.F., Lin, Z.X., Mao, Q.Q., 2013. Isorhynchophylline protects PC12 cells against beta-amyloid-induced apoptosis via PI3K/Akt signaling pathway. Evid. Based Compl. Alternat. Med. 2013 163057. Xian, Y.F., Mao, Q.Q., Wu, J.C., 2014a. Isorhynchophylline treatment improves the amyloid-β-induced cognitive impairment in rats via inhibition of neuronal apoptosis and tau protein hyperphosphorylation. J. Alzheimers Dis. 39, 331–346. Xian, Y.F., Su, Z.R., Chen, J.N., 2014b. Isorhynchophylline improves learning and memory impairments induced by D-galactose in mice. Neurochem. Int. 76, 42–49. Yuan, D., Ma, B., Yang, J.Y., 2009. Anti-inflammatory effects of rhynchophylline and isorhynchophylline in mouse N9 microglial cells and the molecular mechanism. Int. Immunopharmacol. 9, 1549–1554. Zhang, Y.N., Yang, Y.F., Xu, W., 2017. The blood-brain barrier permeability of six indole alkaloids from Uncariae Ramulus Cum Uncis in the MDCK-pHaMDR cell monolayer model. Molecules 22, 1944. Zhang, C., Wu, X., Xian, Y., Zhu, L., Lin, G., Lin, Z.X., 2019. Evidence on integrating pharmacokinetics to find truly therapeutic agent for Alzheimer’s disease: comparative pharmacokinetics and disposition kinetics profiles of stereoisomers isorhynchophylline and rhynchophylline in rats. Evid. Based Complement. Alternat. Med., 4016323. Zhao, J., O’Connor, T., Vassar, R., 2011. The contribution of activated astrocytes to Aβ production: implications for Alzheimer’s disease pathogenesis. J. Neuroinflamm. 8, 150. Zhou, Q., Wang, M., Du, Y., 2015. Inhibition of c-J un N-terminal kinase activation reverses Alzheimer disease phenotypes in APP swe/PS 1dE 9 mice. Ann. Neurol. 77, 637–654. Zhou, X.W., Li, X., Bjorkdahl, C., 2006. Assessments of the accumulation severities of amyloid β-protein and hyperphosphorylated tau in the medial temporal cortex of control and Alzheimer’s brains. Neurobiol. Dis. 22, 657–668.

All authors have consented for publication. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bbi.2019.08.194. References Batarseh, Y.S., Duong, Q.V., Mousa, Y.M., 2016. Amyloid-beta and astrocytes interplay in amyloid-beta related disorders. Int. J. Mol. Sci. 17, 338. Bianca, V.D., Dusi, S., Bianchini, E., Dal Prà, I., Rossi, F., 1999. Beta-amyloid activates the O-2 forming NADPH oxidase in microglia, monocytes, and neutrophils. A possible inflammatory mechanism of neuronal damage in Alzheimer’s disease. J. Biol. Chem. 274, 15493–15499. Blennow, K., de Leon, M.J., Zetterberg, H., 2006. Alzheimer’s disease. Lancet 368, 387–403. Braithwaite, S.P., Schmid, R.S., He, D.N., 2010. Inhibition of c-Jun kinase provides neuroprotection in a model of Alzheimer’s disease. Neurobiol. Dis. 39, 311–317. Chen, K., Lu, Y., Liu, C., 2018. Morroniside prevents H2O2 or Aβ1-42-induced apoptosis via attenuating JNK and p38 MAPK phosphorylation. Eur. J. Pharmacol. 834, 295–304. Chen, L., Huang, C., Shentu, J., 2017. Indirubin derivative 7-bromoindirubin-3-oxime (7Bio) attenuates Aβ oligomer-induced cognitive impairments in mice. Front. Mol. Neurosci. 10, 393. Chinese Pharmacopoeia Commission, 2015. Pharmacopoeia of the People’s Republic of China. China Medical Science Press, pp. 257. Chishti, M.A., Yang, D.S., Janus, C., 2001. Early-onset amyloid deposition and cognitive deficits in transgenic mice expressing a double mutant form of amyloid precursor protein 695. J. Biol. Chem. 276, 21562–21570. Domingues, C., da Cruz, E., Silva, O.A.B., 2017. Impact of cytokines and chemokines on Alzheimer’s disease neuropathological hallmarks. Curr. Alzheimer. Res. 14, 870–882. Dugan, L.L., Ali, S.S., Shekhtman, G., 2009. IL-6 mediated degeneration of forebrain GABAergic interneurons and cognitive impairment in aged mice through activation of neuronal NADPH oxidase. PLoS One 4 e5518. Durairajan, S.S., Liu, L.F., Lu, J.H., 2012. Berberine ameliorates beta-amyloid pathology, gliosis, and cognitive impairment in an Alzheimer’s disease transgenic mouse model. Neurobiol Aging. 33, 2903–2919. Egan, M.F., Kost, J., Tariot, P.N., 2018. Randomized trial of verubecestat for mild-tomoderate Alzheimer’s disease. N. Engl. J. Med. 378, 1691–1703. Erta, M., Quintana, A., Hidalgo, J., 2012. Interleukin-6, a major cytokine in the central nervous system. Int. J. Biol. Sci. 8, 1254–1266. Etminan, M., Gill, S., Samii, A., 2003. Effect of non-steroidal anti-inflammatory drugs on risk of Alzheimer’s disease: systematic review and meta-analysis of observational studies. BMJ 327, 128. Fernandez, P.L., Britton, G.B., Rao, K.S., 2013. Potential immunotargets for Alzheimer’s disease treatment strategies. J. Alzheimers Dis. 33, 297–312. Gonzalez-Reyes, R.E., Nava-Mesa, M.O., Vargas-Sanchez, K., 2017. Involvement of astrocytes in Alzheimer’s disease from a neuroinflammatory and oxidative stress perspective. Front. Mol. Neurosci. 10, 427. Haginiwa, J., Sakai, S., Aimi, N., 1973. Studies of plants containing indole alkaloids. 2. On the alkaloids of Uncaria rhynchophylla Miq. Yakugaku Zasshi 93, 448–452. Hampel, H., Ewers, M., Bürger, K., 2009. Lithium trial in Alzheimer’s disease: a randomized, single-blind, placebo-controlled, multicenter 10-week study. J. Clin. Psychiatry 70, 922–931. Heneka, M.T., Carson, M.J., El Khoury, J., 1994. Neuroinflammation in Alzheimer’s disease. Lancet Neurol. 14, 388–405. Honig, L.S., Vellas, B., Woodward, M., 2018. Trial of solanezumab for mild dementia due to Alzheimer’s disease. N. Engl. J. Med. 378, 321–330. Howes, M.J., Perry, E., 2011. The role of phytochemicals in the treatment and prevention of dementia. Drugs Aging. 28, 439–468. http://www.pfizer.com/files/products/material_safety_data/PZ00471.pdf. Huang, Y., Mucke, L., 2012. Alzheimer mechanisms and therapeutic strategies. Cell 148, 1204–1222. Hyde, L.A., Kazdoba, T.M., Grilli, M., 2005. Age-progressing cognitive impairments and neuropathology in transgenic CRND8 mice. Behav, Brain Res. 160, 344–355. Imamura, S., Tabuchi, M., Kushida, H., 2011. The blood-brain barrier permeability of geissoschizine methyl ether in Uncaria hook, a galenical constituent of the traditional Japanese medicine yokukansan. Cell. Mol. Neurobiol. 31, 787–793. Iwasaki, K., Satoh-Nakagawa, T., Maruyama, M., 2005. A randomized, observer-blind, controlled trial of the traditional Chinese medicine Yi-Gan San for improvement of behavioral and psychological symptoms and activities of daily living in dementia. J. Clin. Psychiatry 66, 248–252. Josephs, K.A., Whitwell, J.L., Ahmed, Z., 2008. β-amyloid burden is not associated with rates of brain atrophy. Ann. Neurol. 63, 204–212. Kang, T.H., Murakami, Y., Matsumoto, K., 2002. Rhynchophylline and isorhynchophylline inhibit NMDA receptors expressed in Xenopus oocytes. Eur. J. Pharmacol. 455, 27–34. Kang, T.H., Murakami, Y., Takayama, H., 2004. Protective effect of rhynchophylline and

278