Network Pharmacology-Based Analysis of Xiao-Xu-Ming Decoction on the Treatment of Alzheimer's Disease

Abstract

Alzheimer's disease (AD) has become a worldwide disease that is harmful to human health and brings a heavy economic burden to healthcare system. Xiao-Xu-Ming Decoction (XXMD) has been widely used to treat stroke and other neurological diseases for more than 1000 years in China. However, the synergistic mechanism of the constituents in XXMD for the potential treatment of AD is still unclear. Therefore, the present study aimed to predict the potential targets and uncover the material basis of XXMD for the potential treatment of AD. A network pharmacology-based method, which combined data collection, drug-likeness filtering and absorption, distribution, metabolism, excretion and toxicity (ADME/T) properties filtering, target prediction and network analysis, was used to decipher the effect and potential targets of XXMD for the treatment of AD. Then, the acetylcholinesterase (AChE) inhibitory assay was used to screen the potential active constituents in XXMD for the treatment of AD, and the molecular docking was furtherly used to identify the binding ability of active constituents with AD-related target of AChE. Finally, three in vitro cell models were applied to evaluate the neuroprotective effects of potential lead compounds in XXMD. Through the China Natural Products Database, Traditional Chinese Medicine Systems Pharmacology (TCMSP) Database, Traditional Chinese Medicine (TCM)-Database @Taiwan and literature, a total of 1481 compounds in XXMD were finally collected. After ADME/T properties filtering, 908 compounds were used for the further study. Based on the prediction data, the constituents in XXMD formula could interact with 41 AD-related targets. Among them, cyclooxygenase-2 (COX-2), estrogen receptor α (ERα) and AChE were the major targets. The constituents in XXMD were found to have the potential to treat AD through multiple AD-related targets. 62 constituents in it were found to interact with more than or equal to 10 AD-related targets. The prediction results were further validated by in vitro biology experiment, resulting in several potential anti-AD multitarget-directed ligands (MTDLs), including two AChE inhibitors with the IC₅₀ values ranging from 4.83 to 10.22 μM. Moreover, fanchinoline was furtherly found to prevent SH-SY5Y cells from the cytotoxicities induced by sodium nitroprusside, sodium dithionate and potassium chloride. In conclusion, XXMD was found to have the potential to treat AD by targeting multiple AD-related targets and canonical pathways. Fangchinoline and dauricine might be the potential lead compounds in XXMD for the treatment of AD.

Keywords: Alzheimer's disease; Xiao-Xu-Ming decoction; machine learning; molecular docking; network pharmacology.

FIGURE 1

Prediction of blood−brain barrier (BBB) penetration and human intestinal absorption of the constituents in XXMD (A) and herb distribution of candidate constituents in XXMD (B).

FIGURE 2

Target distributions of potential active constituent in XXMD.

FIGURE 3

Target distributions of twelve kinds of herbs in XXMD.

FIGURE 4

Global constituent-target network of candidate constituents in XXMD. Blue circles correspond to the compounds, pink circles correspond to the source of compounds, and pink rectangles in the center correspond to the target. FF, Fang Feng (Saposhnikovia divaricata (Turcz. ex Ledeb.) Schischk.); SJ, Sheng Jiang (Zingiber officinale Roscoe); RS, Ren Shen (Panax ginseng C.A.Mey.); XR, Xin Ren (Prunus armeniaca L.); GC, Gan Cao (Glycyrrhiza uralensis Fisch. ex DC.); CX, Chuan Xiong (Conioselinum anthriscoides 'Chuanxiong'); MH, Ma Huang (Ephedra sinica Stapf); FZ, Fu Zi (Aconitum carmichaeli Debeaux); GZ, Gui Zhi (Cinnamomum cassia (L.) J.Presl); HQ, Huang Qin (Scutellaria baicalensis Georgi); FJ, Fan Ji (Stephania tetrandra S.Moore); SY, Shao Yao (Paeonia lactiflora Pall.).

FIGURE 5

The number of constituent of herbs in XXMD in different frequency of targets (A) and constituent-target network for multi-target (≥10) constituent in XXMD (B). In Figure 5B, rectangles in orange, red, purple, green and light blue correspond to the compounds and their sources, and dark blue rectangles correspond to the targets.

FIGURE 6

Target-target network (A) and target-function network (B) of candidate constituents in XXMD. A functional module is linked to a target if the target is involved in that biological process. In Figure 6B, red words correspond to the targets, and black words correspond to the AD-related functional modules.

FIGURE 7

The inhibitory effect of Donepezil (A), Fangchinoline (B) and Dauricine (C) on acetylcholinesterase activity. Results are presented as means ± SEM, n = 3.

FIGURE 8

Donepezil (A), Fangchinoline (B) and Dauricine (C) interacting with acetylcholinesterase (AChE), the AChE model was established by molecular docking.

FIGURE 9

The protective effects of fangchinoline against cell injury induced by sodium nitroprusside (A), potassium chloride (B) and sodium dithionate (C) in SH-SY5Y cells. Results are presented as means ± SEM, n = 3. ### P<0.001 versus each control group. * P<0.05, ** P<0.01, *** P<0.001 versus group solely treated with sodium nitroprusside, potassium chloride or sodium dithionate in (A), (B) and (C), respectively.