Abstract Background Alzheimer’s disease (AD) is the most common cause of dementia in the elderly, characterized by a progressive and irreversible loss of memory and cognitive abilities. Currently, the prevention and treatment of AD still remains a huge challenge. As a traditional Chinese medicine (TCM) prescription, Danggui-Shaoyao-san decoction (DSS) has been demonstrated to be effective for alleviating AD symptoms in animal experiments and clinical applications. However, due to the complex components and biological actions, its underlying molecular mechanism and effective substances are not yet fully elucidated. Methods In this study, we firstly systematically reviewed and summarized the molecular effects of DSS against AD based on current literatures of in vivo studies. Furthermore, an integrated systems pharmacology framework was proposed to explore the novel anti-AD mechanisms of DSS and identify the main active components. We further developed a network-based predictive model for identifying the active anti-AD components of DSS by mapping the high-quality AD disease genes into the global drug-target network. Results We constructed a global drug-target network of DSS consisting 937 unique compounds and 490 targets by incorporating experimental and computationally predicted drug–target interactions (DTIs). Multi-level systems pharmacology analyses revealed that DSS may regulate multiple biological pathways related to AD pathogenesis, such as the oxidative stress and inflammatory reaction processes. We further conducted a network-based statistical model, drug-likeness analysis, human intestinal absorption (HIA) and blood-brain barrier (BBB) penetration prediction to uncover the key ani-AD ingredients in DSS. Finally, we highlighted 9 key ingredients and validated their synergistic role against AD through a subnetwork. Conclusion Overall, this study proposed an integrative systems pharmacology approach to disclose the therapeutic mechanisms of DSS against AD, which also provides novel in silico paradigm for investigating the effective substances of complex TCM prescription. Keywords: Systems pharmacology, Danggui-Shaoyao-san, Alzheimer’s disease, Mechanism of action Background Alzheimer’s disease (AD) is a progressive neurodegenerative disease, which starts with mild memory loss and gradually results in severe impairment of broad executive and cognitive functions [[39]1–[40]3]. According to the data from World Health Organization in 2019, approximately 50 million people are suffering from dementia, while AD contributes to 60–70% of these cases [[41]4]. More alarming is a continuous rise in AD cases year by year, leading to increasing disability and high socioeconomic burden. Unfortunately, currently there are only four available anti-AD drugs (Donepezil, Rivastigmine, Galantamine, Memantine) introduced into the market, and no approved disease-modifying treatments (DMTs) exist for AD. It is worth noting that more and more scientists have turned their attention to the design of multi-targeted drugs instead of the traditional “one target-one drug” perspective, which could intervene the complex AD pathogenesis via multiple molecular targets [[42]5]. As a complementary and alternative medicine, traditional Chinese medicine (TCM), especially the TCM prescription, has been widely applied in Asian countries for the prevention and treatment of complex and degenerative diseases, including AD [[43]6]. Generally, a herbal formula usually comprises multiple components which could act on multiple targets and exert its synergistic effects on disease [[44]7]. In spite of lacking effective options for AD, lots of classical TCM prescriptions (e.g. SuHeXiang Wan, TongLuoJiuNao) have potential as therapeutic drugs for alleviating AD symptoms [[45]8–[46]10]. Danggui-Shaoyao-san (DSS), also known as Toki-shakuyaku-san in Japan and Dangguijakyakan in Korea, is a herb combination employed to improve cognitive function, which consists of six Chinese herbs including Paeoniae Radix Alba (BaiShao, BS), Angelica Sinensis Radix (DangGui, DG), Atractylodis Macrocephalae Rhizoma (BaiZhu, BZ), Chuanxiong Rhizoma (ChuanXiong, CX), Alismatis Rhizoma (ZeXie, ZX), and Poria (FuLing, FL). According to a meta-analysis conducted on randomized controlled trials (RCTs), DSS has a positive effect on the scores of Mini-Mental State Examination (MMSE) and activities of daily living (ADL) of AD patients [[47]11]. Clinical study found that the regional cerebral blood flow (rCBF) of AD patients in the posterior cingulate were significantly higher and orientation to place tended to improve after treatment with DSS [[48]12]. Besides, previous pharmacological studies have deciphered that DSS could exert potentially therapeutic effects for AD via multiple mechanisms such as disrupting the aggregation of Aβ, attenuating inflammatory reaction, and adjusting the mitochondrial membrane permeability (Table [49]1). For instance, a study in 2014 has reported that DSS could improve learning and memory capacity in female SAMP8 mice through modulating estrogen, nitric oxide, and glycine in plasma or hippocampal tissue [[50]13]. Furthermore, Kou et al. demonstrated that DSS ameliorated memory dysfunction and protected the ultrastructure of cortex changed by aging, which may be beneficial for the treatment of AD [[51]14]. Despite current reported therapeutic mechanisms, the underlying mechanism of actions (MOAs) and active components of DSS against AD still remain indistinct. Thus, it is necessary to comprehensively investigate the molecular mechanism of DSS for treatment of AD and deeply understand their synergistic effects on multiple pathways. Table 1. Current in vivo studies on the therapeutic mechanism of actions of DSS against AD Year Brief description PMID 2020 Ameliorating cognition deficits in APP/PS1 mice via increasing DHA content and regulating oxidative stress and inflammation 32084555 2014 Improving learning and memory in female SAMP8 mice via modulating estrogen, nitric oxide, and glycine in plasma or hippocampal tissue 24757492 2011 Improving cognition of the rats which might be related to attenuate inflammatory reaction and reduce cell apoptosis in the hippocampus 22375398 2010 Improving spatial learning and memory deficits in mice, reverse the inhibition of Long-term potentiation in hippocampal slices, prevent aggregation of Aβ, and even disrupt the aggregated Aβ fibrils 20117199 2008 Upregulating Bcl-2 level and downregulate Bax level which might in turn adjust the mitochondrial membrane permeability, attenuate cytochrome c and its release into cytosol, following the suppression of caspase activation 18093848 2005 Ameliorating memory dysfunction, modulate metabolism of monoamine neurotransmitters and protect the ultrastructure of cortex changed by aging 15707771 2005 Reducing the Abeta25–35-induced neuronal death and the lipid peroxidation which has a protective effect against Abeta25–35-induced neuronal damage 16106382 [52]Open in a new tab Systems pharmacology is an emerging discipline which combines in silico network-based tools and experimental assays, aiming to elucidate the changes in the functions and reactions of human body induced by medicines [[53]15]. Based on the holistic principle, the herbal formulas in TCM show the characteristics of multi-component and multi-target in therapy in contrast to synthetic drugs. Due to the complexity of ingredients and targets, conventional experimental approaches are time-consuming and expensive for TCM research. Systems pharmacology offers effective strategy to explore the multi-component network target research pattern of Chinese herbal medicine and formula [[54]16–[55]18]. In this study, we attempted to systematically explore the MOAs of DSS for treatment of AD through an integrated systems pharmacology framework (Fig. [56]1). Firstly, we collected the chemical ingredients of DSS with known protein targets from our previous integrated natural products database [[57]19]. We further computationally predicted the putative targets of DSS via a network-based inference method. Subsequently, a global drug-target network of DSS was constructed by incorporating the known and predicted DTIs. Furthermore, multi-level systems pharmacology analysis methods, including molecular-function analysis, biological process analysis, target-function modules analysis, and KEGG pathway enrichment were performed to elucidate the MOAs of DSS against AD. Additionally, we developed a network-based model for identifying the active anti-AD components of DSS by mapping the high-quality AD disease genes into the global drug-target network. After integrating high-performance liquid chromatography (HPLC) analysis data, drug-likeness analysis, human intestinal absorption (HIA) and blood-brain barrier (BBB) penetration assessment, and network statistical model prediction results, we highlighted the key anti-AD ingredients in DSS and illustrated the specific synergistic mechanisms through subnetwork analysis. Fig. 1. [58]Fig. 1 [59]Open in a new tab Schematic diagram of the systems pharmacology approach for deciphering the pharmacological mechanisms of DSS for treatment of AD Methods Manual curation of genes associated with AD The genes associated with AD were collected from six authoritative databases: 1) the Malacards database ([60]https://www.malacards.org); 2) the Comparative Toxicogenomics Database (CTD) [[61]20]; 3) DisGeNET [[62]21]; 4) the GWAS Catalogue [[63]22]; 5) the Human Gene Mutation Database (HGMD) database [[64]23]; and 6) AlzBase database ([65]http://alz.big.ac.cn/alzBase/summary/Gene). Only genes labeled with “Alzheimer’s disease” were extracted from the databases mentioned above. Additionally, for the AlzBase database, the top 100 genes were preserved for further investigation. Finally, 299 AD-related genes (Supplementary Table S[66]1) were integrated after removing the duplicates. Collection of herbal ingredients and known protein targets The ingredients and their protein targets in each herb of DSS were collected from our previous integrated database [[67]24], which includes experimental validated DTIs of natural products extracting from over 2000 literatures and five authoritative compound-protein interaction databases: ChEMBL (v21) [[68]25], BindingDB [[69]26] (accessed in Sep. 2017), STITCH [[70]27], the Herbal Ingredients’ Target Database (HIT) [[71]28], and the Traditional Chinese Medicine Integrated Database (TCMID) [[72]29]. Consequently, a total of 1042 herbal ingredients in DSS were obtained. The corresponding number of the collected compounds in DG, CX, FL, ZX, BS, BZ is 549, 351, 84, 42, 125, 170, respectively. The detailed structural information of the 1042 herbal ingredients is provided in Supplementary Table S[73]2. Network-based target prediction for DSS In this study, target fishing is carried out to predict targets based on known DTIs via a balanced substructure-drug-target network-based inference (bSDTNBI) method [[74]30]. We previously developed predictive network models to identify new targets of natural products with bSDTNBI method, which prioritizes potential targets utilizing resource-diffusion processes for both known and new natural products [[75]31]. During the process, four parameters (α = β = 0.1, γ = − 0.5, and k = 2) of bSDTNBI were adopted. The first parameter α was introduced to balance the initial resource allocation of different node types, while β was utilized to adjust weighted values of different edge types. The third parameter γ was applied to balance the effect of hub nodes in resource diffusion processes, and the last parameter κ represented the number of resource-diffusion processes. We subsequently calculated four types of molecular fingerprints for each compound based on PaDEL-Descriptor (version 2.18) [[76]32], including MACCS, PubChem, Substructure (FP4) and Klekota-Roth (KR). Compared with the other three generated predictive models, bSDTNBI_KR performed best given its highest values of the area under the receiver operating characteristic curve (AUC = 0.959). Eventually, the bSDTNBI_KR predictive model was chosen to identify new targets of each natural product. In this study, the top 20 putative targets for each compound with known targets were selected (Supplementary Table S[77]3). Identification of key anti-AD ingredients in DSS The key anti-AD ingredients in DSS were determined by four steps. Firstly, we obtained the main components in DSS according to the HPLC analysis results reported in previous references [[78]33, [79]34].