Screening Traditional Chinese Medicine Combination for Cotreatment of Alzheimer’s Disease and Type 2 Diabetes Mellitus by Network Pharmacology

Abstract

Background:

In recent years, the efficacy of type 2 diabetes mellitus (T2DM) drugs in the treatment of Alzheimer’s disease (AD) has attracted extensive interest owing to the close associations between the two diseases.

Objective:

Here, we screened traditional Chinese medicine (TCM) and multi-target ingredients that may have potential therapeutic effects on both T2DM and AD from T2DM prescriptions.

Methods:

Network pharmacology and molecular docking were used.

Results:

Firstly, the top 10 frequently used herbs and corresponding 275 active ingredients were identified from 263 T2DM-related TCM prescriptions. Secondly, through the comparative analysis of 208 potential targets of ingredients, 1,740 T2DM-related targets, and 2,060 AD-related targets, 61 common targets were identified to be shared. Thirdly, by constructing pharmacological network, 26 key targets and 154 representative ingredients were identified. Further enrichment analysis showed that common targets were involved in regulating multiple pathways related to T2DM and AD, while network analysis also found that the combination of Danshen (Radix Salviae)-Gancao (Licorice)-Shanyao (Rhizoma Dioscoreae) contained the vast majority of the representative ingredients and might be potential for the cotreatment of the two diseases. Fourthly, MAPK1, PPARG, GSK3B, BACE1, and NR3C1 were selected as potential targets for virtual screening of multi-target ingredients. Further docking studies showed that multiple natural compounds, including salvianolic acid J, gancaonin H, gadelaidic acid, icos-5-enoic acid, and sigmoidin-B, exhibited high binding affinities with the five targets.

Conclusion:

To summarize, the present study provides a potential TCM combination that might possess the potential advantage of cotreatment of AD and T2DM.

Keywords

Alzheimer’s disease network pharmacology traditional Chinese medicines type 2 diabetes mellitus

INTRODUCTION

Alzheimer’s disease (AD) is a serious neurodegenerative disease that poses an important challenge to public health. More than 50 million people wor-ldwide are currently affected by this disease and this fig will grow steadily due to the aging of the population [1]. AD patients are generally companied with the symptom of memory loss, impairment of cognitive functions, and behavior and personality changes, which bring a serious burden on individuals, family, and society [2]. Despite much effort been devoted to developing drugs to combat AD, there is still an urgent need for new effective agent.

Type 2 diabetes mellitus (T2DM) is a prevalent metabolic disease, which is characterized by dysregulation of carbohydrates, impaired insulin secretion, and insulin resistance [3]. In recent years, an increasing number of studies found that T2DM was closely related to the development of AD. Clinical and epidemiological studies supported that T2DM was an important risk factor promoting the development of AD. For example, a large population-based study comprising a total of 1,396 individuals in Sweden demonstrated that patients with diabetes had an in-creased risk of developing AD by approximately 69%[4]. Moreover, further studies found that there were multiple common pathogenic mechanisms between T2DM and AD, including impaired insulin signaling, insulin resistance, inflammation, amyloid-β (Aβ) de-posits, and hyperphosphorylated tau [5 –7]. Many studies have provided evidence that impaired insulin signaling due to insulin resistance and relative insulin deficiency also occurs in the brain of AD patients [8, 9].

Owing to the similar pathogenesis between AD and T2DM, the beneficial effects of T2DM drugs in the treatment of AD have lately received great attention, and these effective drugs might be involved in multiple mechanisms [10]. Firstly, T2DM drugs might play a beneficial role in AD by decreasing the production of Aβ. For instance, glimepiride could reduce extracellular Aβ concentrations by inhibiting the activity of β-secretase 1 (BACE1) and downregulating the expression of BACE1 mRNA [11]. Secondly, T2DM drugs have been shown to ameliorate cognitive ability and other AD-related pathology by regulating insulin signaling pathways. A case of this is dulaglutide, which could reduce the hyperphosphorylation of tau and neurofilaments by regulating phosphoinositide-3-kinase/AKT/glycogen synthase kinase 3 beta (PI3K/AKT/GSK3B) signaling pathway [12]. In addition, liraglutide treatment not only significantly reduced the levels of hyperphosphorylated tau and neurofilaments by improving c-Jun N-terminal kinase (JNK) and extracellular signal-regulated kinase (ERK/MAPK) signals [13], but also decreased the load of Aβ plaques in hippocampus and caspase-3 (CASP3) level [14]. Thirdly, the improvement in inflammation-related neuropathy might also play a crucial role. It has been reported that peroxisome proliferator-activated receptor gamma (PPARG) agonists, such as pioglitazone, significantly ameliorated inflammatory responses in the brain by inhibiting proinflammatory gene expression and modulating the activation of microglia [15, 16].

In recent years, traditional Chinese medicine (TCM) prescriptions have been proven to be effective in the treatment of T2DM. Although antidiabetic drugs mentioned above have achieved encouraging outcomes in the treatment of AD, the beneficial effects of T2DM-related TCM in AD remain to be explored. More recently, the emergence of network pharmacology has transformed the research pattern from “one target, one drug” mode to “network target, multi-component” mode, which promotes the modernization of TCM research [17]. Also, molecular docking has been used as an effective tool in modern drug discovery, which can predict the binding affinity between receptor and ligand [18]. Therefore, using network pharmacology approach, the present study aims to screen TCM and multi-target ingredients that have potential effects on cotreatment of T2DM and AD from T2DM prescriptions.

MATERIAL AND METHODS

Selection of TCM prescriptions for T2DM

A systematic search was performed up to January 2020 in the China National Knowledge Infrastructure (CNKI) and PubMed for relevant studies. The main keywords were “traditional Chinese medicine” or “traditional Chinese medicine prescription” or “Chinese patent medicine” and “Type 2 diabetes mellitus”. The literature was excluded when its content related to duplicate publication, review, complication, animal experiment, and cell experiment. If the study referred to TCM prescription for clinical treatment of T2DM, the full text was analyzed to extract information of prescription. Besides, a supplementary search for available TCM prescriptions of T2DM was carried out based on the Yao Zhi Database (https://db.yaozh.com/). TCM prescriptions included in this study were sorted out and established a database of summary information, including name, composition, and type of research. Then, the frequencies of single herbs among them were calculated and common herbs were screened.

Screening active ingredients of herbs

First of all, the active ingredients of herbs were obtained through the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP, http://tcmspw.com/tcmsp.php) database and Traditional Chinese Medicines Integrated Database (TCMID, http://www.megabionet.org/tcmid/) [19]. TCMSP database described the absorption, distribution, metabolism, and excretion (ADME) parameters of the herbal ingredients, such as oral bioavailability (OB) and drug-likeness (DL) [20]. OB≥30%and DL≥0.18 were set as filter conditions to select bioactive ingredients. Besides, the herbal ingredients from TCMID database were filtered by Lipinski’s and Veber’s rules using Discovery Studio (DS) software. The compounds that meet the above screening criteria were considered as effective ingredients with therapeutic effects.

Predicting potential targets of active ingredients

To predict the potential targets of active ingredients, based on the above mol2 format files, a reverse docking was performed using the PharmaMapper server (http://www.lilab-ecust.cn/pharmmapper/), which provides a service for the identification of potential targets with a pharmacophore matching ap-proach [21]. The species of the predicted targets was set as “Homo sapiens”, and other parameters were set to default. According to the value of “fit score”, the top 10 targets of each compound were selected for further study. Subsequently, we utilized UniProt (https://www.uniprot.org/) database to convert tar-gets to corresponding gene names.

Identification of T2DM-related targets and AD-related targets

Using “Alzheimer’s disease” and “Type 2 diabetes mellitus” as the searching terms, the DrugBank da-tabase (https://www.drugbank.ca/), Therapeutic Target Database (TTD, http://db.idrblab.net/ttd/), TCMSP database, and DisGeNET (https://www.disgenet.org/) database were employed to identify T2DM-related targets and AD-related targets in February 2020.

Searching for common targets

The intersection targets of active ingredients, T2DM, and AD were identified by the Venn diagram (http://bioinfogp.cnb.csic.es/tools/venny/). The common targets were the predicted treatment targets of active ingredients against both T2DM and AD.

Construction of networks and analysis

To understand the relationship between herbs, ac-tive ingredients, disease-related targets and pathways, the herb-active ingredient-common target network and target-pathway network were respectively constr-ucted, visualized, and analyzed by Cytoscape 3.6.1 software. And the common targets were also impor-ted into the Search Tool for the Retrieval of Interacting Genes (STRING) database (https://string-db.org) to construct protein-protein interaction (PPI) network [22]. We set medium confidence scores (> 0.4) and limited the species to “Homo sapiens”. Degree centrality was a vital parameter to measure the local centrality of a node in the network. Generally, a node with a high degree centrality could be the important node of the network. In this study, the targets with a degree centrality above median value in the PPI network were selected as key targets for further analysis.

Further, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis were employed based on common targets mentioned above and Metascape database (http://metascape.org/gp/index.html) [23]. The species of the targets was set as “Homo sapiens” and terms with p-value < 0.05 were saved and grouped into clusters. The most statistically significant biological functions and pathways were chosen.

Molecular docking

To discover natural compounds with potential effects on T2DM and AD, MAPK1, PPARG, BACE1, GSK3B, and NR3C1 were selected for molecular docking with the main ingredients of herbs. The CD-OCKER in DS was adopted to perform molecular docking studies. First, the conformations of proteins were obtained from the Protein Data Bank database and Autotools was used to optimize the protein str-ucture, remove water molecules, add hydrogen atoms and force field. Besides, all small molecules were processed with energy minimization. The location of the ligand in the co-crystallographic structure was defined as the binding site. CDOCKER depended on CHARMm-based force field and generated confo-rmations using the high temperature, then conformations were moved to the binding site for binding pose analysis [24]. Generally, the higher the -CDOCKER interaction energy, the greater the binding affinity between protein and ligand. The conformation with the highest -CDOCKER interaction energy was picked out for further analysis.

RESULTS

T2DM-related TCM prescriptions and common herbs

According to statistics, 263 TCM prescriptions for T2DM were obtained and the frequencies of single herbs among them were calculated. As shown in Table 1, the top 10 herbs were frequently used in TCM prescriptions for clinical treatment of T2DM, including Huangqi (Hedysarum Multijugum Ma-xim.), Huanglian (Coptidis Rhizoma), Shanyao (Rhizoma Dioscoreae), Danshen (Radix Salviae), Gegen (Radix Puerariae), Fuling (Poria Cocos (Schw.) Wolf.), Tianhuafen (Trichosanthis Radix), Gancao (Licorice), Maidong (Ophiopogonis Radix), and Shengdihuang (Rehmanniae Radix).

Table 1

Frequencies of single herbs in TCM prescriptions for treating T2DM

Number	Latin name	Chinese name	Frequency	Percentage (%)
1	Hedysarum Multijugum Maxim.	Huangqi	145	55.13
2	Coptidis Rhizoma	Huanglian	97	36.88
3	Rehmanniae Radix	Shengdihuang	96	36.50
4	Rhizoma Dioscoreae	Shanyao	90	34.22
5	Radix Salviae	Danshen	89	33.84
6	Radix Puerariae	Gegen	85	32.32
7	Poria Cocos(Schw.) Wolf.	Fuling	77	29.28
8	Ophiopogonis Radix	Maidong	63	23.95
9	Trichosanthis Radix	Tianhuafen	63	23.95
10	Licorice	Gancao	55	20.91

Active ingredients of herbs

According to the established screening criteria, we retrieved 275 active ingredients of herbs from two databases after removing the repetitive compounds. On the one hand, a total of 228 active compounds were acquired by consulting the TCMSP database, and these active compounds were primarily originated from Huangqi (20 compounds), Huanglian (14 compounds), Shanyao (16 compounds), Danshen (65 compounds), Gegen (4 compounds), Fuling (15 compounds), Tianhuafen (2 compounds), and Gancao (92 compounds). On the other hand, the rest of compounds were obtained from the TCMID database, and they were from Maidong (37 compounds) and Shengdihuang (24 compounds).

Potential targets of active ingredients

The targets of active ingredients of the top 10 frequently used herbs were predicted by PharmaMapper server. After eliminating the repetitive targets, we finally gained 208 potential targets of 275 active ingredients.

Related targets of T2DM and AD

After deleting duplicate targets, 1,740 T2DM-related targets and 2,060 AD-related targets were identified from four databases. Using a comparative analysis of two diseases related targets, a total of 614 overlapped targets were identified to be shared between T2DM and AD. These targets were closely related to the occurrence and development of T2DM and AD.

Common targets of active ingredients, T2DM, and AD

The disease-related targets regulated by active ingredients were screened via a Venn diagram. A total of 61 common targets, including BACE1, PCK1, MAOB, PTPN1, NR3C1, PPARG, CASP3, ALB, MAPK1, GSK3B, ESR1, MAPK14, INSR, and so on, were identified by the intersection of active components, T2DM, and AD. The common targets were regarded as the related targets of herbal ingredients in treating T2DM and AD, and the results were shown by the Venn diagram (Fig. 1).

Fig. 1

Targets’ intersection of 10 herbs, T2DM, and AD (The blue circle represents targets of 10 herbs, the green circle represents T2DM-related targets, the yellow circle represents AD-related targets).

Network of herb-active ingredient-common target

To further comprehensively understand the as-sociations among herbs, active ingredients, and dise-ase-related targets, we established an herb-active ingredient-common target network as shown in Fig. 2. The network contained 10 herbs, 275 active in-gredients, and 61 common targets. It was obvious that one target was regulated by multiple compo-unds and one compound also interacted with multiple targets. This network revealed that components such as poricoic acid C (degree = 9), α-amyrin (degree = 8), dehydrotanshinone II A (degree = 8), miltionone II (degree = 8), glycyrol (degree = 8), and worenine (degree = 8) were the high-degree ingredients and that targets such as TTR (degree = 130), VDR (degree = 87), RBP4 (degree = 75), SHBG (deg-ree = 57), and BACE1 (degree = 56) were the high-degree targets. We speculated that the high-degree ingredients involved in the treatment of T2DM and AD. For this reason, the compounds with degree ≥5 were selected as the representative ingredients for further molecular docking research. Notably, net-work analysis showed that 154 representative ingredients were mainly from Danshen, Gancao, and Shanyao.

Fig. 2

Herb-active ingredient-common target network (Green triangles represent 10 herbs, blue circles represent active ingredients from 10 herbs, and red rectangles represent common targets of active ingredients, T2DM, and AD).

PPI network

To further comprehend protein-protein interactions, based on the STRING database, a PPI network comprising of 61 common targets was constrsucted and finally visualized by Cytoscape. As shown in Fig. 3, the higher the degree value, the darker the color, the larger the node. The threshold value was degree >12.9, and 26 key targets were further scr-eened, including ALB, MAPK1, IGF1, CASP3, EGFR, ESR1, PPARG, MMP9, KDR, ACE, MAPK14, AR, NR3C1, HPGDS, JAK2, GRB2, REN, GSK3B, PTPN1, RXRA, PCK1, F2, CYP2C9, ADAM17, MMP3, and CCL5 (Table 2). These 26 targets were located in the cores of the PPI network and could be regarded as valuable targets for active ingredients against T2DM and AD.

Fig. 3

The PPI network for the identification of key targets (the larger the node, the deeper the color, representing the greater the degree of the node).

Table 2

Summary information of 26 key targets, their corresponding uniprot ID, gene symbols, and degrees of correlation with proteins

Uniprot ID	Gene symbol	Protein name	Degree
P02768	ALB	Serum albumin	47
P28482	MAPK1	Mitogen-activated protein kinase 1	34
P05019	IGF1	Insulin-like growth factor I	31
P42574	CASP3	Caspase-3	30
P00533	EGFR	Epidermal growth factor receptor	29
P03372	ESR1	Estrogen receptor	26
P37231	PPARG	Peroxisome proliferator-activated receptor gamma	25
P14780	MMP9	Matrix metalloproteinase-9	22
P35968	KDR	Vascular endothelial growth factor receptor 2	20
P12821	ACE	Angiotensin-converting enzyme	20
Q16539	MAPK14	Mitogen-activated protein kinase 14	19
P10275	AR	Androgen receptor	19
P04150	NR3C1	Glucocorticoid receptor	19
O60760	HPGDS	Hematopoietic prostaglandin D synthase	19
O60674	JAK2	Tyrosine-protein kinase JAK2	18
P62993	GRB2	Growth factor receptor-bound protein 2	17
P00797	REN	Renin	16
P49841	GSK3B	Glycogen synthase kinase-3 beta	15
P18031	PTPN1	Tyrosine-protein phosphatase non-receptor type 1	14
P19793	RXRA	Retinoic acid receptor RXR-alpha	14
P35558	PCK1	Phosphoenolpyruvate carboxykinase, cytosolic [GTP]	14
P00734	F2	Prothrombin	14
P11712	CYP2C9	Cytochrome P450 2C9	14
P78536	ADAM17	Disintegrin and metalloproteinase domain-containing protein 17	13
P08254	MMP3	Stromelysin-1	13
P13501	CCL5	C-C motif chemokine 5	13

GO and KEGG pathway enrichment analysis

GO and KEGG enrichment analyses were performed to verify the effectiveness of common targets in the treatment of T2DM and AD. The most notable biological functions of the targets were nuclear receptor activity, regulation of hormone levels, muscle cell proliferation, monocarboxylic acid metabolic process, regulation of phosphatidylinositol 3-kinase signaling, response to lipopolysaccharide, response to peptide, reproductive structure develop-ment, regulation of cytokine-mediated signaling pat-hway, striated muscle cell proliferation, response to acid chemical, lipid transport, reactive oxygen spe-cies metabolic process, alcohol metabolic process, epithelial cell proliferation, monocarboxylic acid binding, cofactor binding, glucose metabolic process, response to toxic substance, amyloid-beta metabolic process, and so on (Fig. 4A). To further analyze the relationships between the terms, a subset of enriched terms was selected and presented as a network map. Enriched terms with similarities > 0.3 were regarded as a cluster and connected by edges. The nodes represented enriched terms that were colored by cluster ID as shown in Fig. 4B. And enriched terms were also colored by p-value (Fig. 4C), the deeper the color, the smaller the p-value.

Fig. 4

GO terms enrichment analysis of common targets. A) The most 20 notable biological functions of common targets, B) Enriched terms with similarity > 0.3 are regarded as a cluster and colored by cluster-ID, C) Nodes represent enriched terms that are colored by p-value. The deeper the color, the smaller the p-value).

The active ingredients of the 10 herbs might exert therapeutic effects on T2DM and AD through multiple signaling pathways according to the KEGG en-richment analysis. The most significant biological pathways were pathways in cancer, proteoglycans in cancer, insulin signaling pathway, endocrine res-istance, PPAR signaling pathway, influenza A, drug metabolism-cytochrome P450, fluid shear stress and atherosclerosis, HIF-1 signaling pathway, Alzhei-mer’s disease, serotonergic synapse, renin-angiot-ensin system, amyotrophic lateral sclerosis (ALS), endocytosis, and so on (Fig. 5). The targets involved in the regulation of these signaling pathways were shown in Fig. 6.

Fig. 5

The most 14 notable pathways of KEGG terms enrichment analysis (the y-axis shows significantly enriched pathways of common targets and the x-axis shows the -log10(p) values of these pathways).

Fig. 6

Target-pathway network (green circles represent the targets, purple diamonds represent significantly enriched pathways, nodes sizes are proportional to their degrees).

Virtual screening of ingredients

MAPK1, PPARG, BACE1, GSK3B, and NR3C1 were selected for molecular docking with 154 representative ingredients based on the key targets, pat-hways analysis, and research of T2DM drugs against AD. Corresponding protein receptors were constr-ucted to screen multi-target therapeutic constituents. Several natural compounds with high binding affinity to multiple targets have been discovered, including salvianolic acid J, gancaonin H, gadelaidic acid, icos-5-enoic acid, sigmoidin-B, and so on, as shown in Table 3. As an example, salvianolic acid J was successfully docked with the five targets and possessed high docking scores (Fig. 7).

Table 3

Docking scores of top 5 natural compounds binding with BACE1, GSK3B, PPARG, MAPK1 and NR3C1

Binding energy (kcal·mol^-1)	BACE1	GSK3B	PPARG	MAPK1	NR3C1
salvianolic acid J	58.68	55.29	63.22	66.64	57.80
gancaonin H	48.95	44.30	53.61	55.33	43.12
gadelaidic acid	50.41	48.08	59.51	49.20	50.72
icos-5-enoic acid	49.68	47.04	59.07	51.07	50.17
sigmoidin-B	44.93	53.55	50.63	52.83	45.11

Fig. 7

Molecular docking modes of salvianolic acid J with five potential targets. Salvianolic acid J binds to MAPK1 (A), PPARG (B), BACE1 (C), GSK3B (D), and NR3C1 (E).

DISCUSSION

As two prevalent diseases threatening public health worldwide, the screening of drugs for the cotreatment of AD and T2DM has gained much attention. In this work, we collected 263 T2DM-related TCM prescriptions and analyzed the 10 most frequently used single herbs. Also, 61 common targets were identified to be shared between targets of active ingredients, T2DM, and AD, and 154 compounds with high relevance to these targets were defined as representative ingredients with therapeutic effects. The common targets might be targets for herbs to exert therapeutic effects, which were involved in regulating multiple pathways related to T2DM and AD. The results of pathway enrichment analysis showed that these common targets were enriched in multiple biological pathways related to T2DM and AD, including insulin signaling pathway, PPAR signaling pathway, AD, and so on. Previous studies have reported that impaired insulin signaling and insulin resistance were common pathological features of T2DM and AD [25]. It is well known that one of the major downstream pathways of insulin receptor substrate proteins is the PI3K/AKT cascade, which regulates multiple downstream pathways, including mTORC1, GSK3B, and the FoxO family of transcription factors [26]. Impaired PI3K/AKT signaling pathway led to the activation of GSK3B activity, while excessive activation of GSK3B would increase the production of Aβ and the phosphorylation of tau [27]. In addition, the activation of the PPARG signaling pathway improved the spatial learning and recognition impairments in AD and T2DM mice [28].

According to the topological analysis of PPI, we further found the 26 key targets from the 61 common targets for subsequent study. Based on the key targets, pathways analysis, and related researches of T2DM drugs against AD, five potential therapeutic targets were selected for virtual screening, including MAPK1, PPARG, GSK3B, BACE1, and NR3C1. MAPK1 is a significant element of the ERK signaling pathway. Inhibition of ERK activity reduced the production of microglial neuroinflammatory molecules, such as interleukin (IL)-1beta [29]. Meanwhile, reduced activation of the ERK signaling pathway was also a novel therapeutic strategy against insulin resistance and T2DM [30]. PPARG agonists such as thiazolidinediones were widely used to treat T2DM and could effectively control blood glucose, enhance insulin sensitivity, and restore pancreatic β-cell function [31]. Moreover, clinical studies have indicated that PPARG agonists decreased AD-related pathological changes, such as lipid metabolism disorders, inflammation, and oxidative stress [32]. GSK3B is a key enzyme involved in many cellular functions, and the overexpression of GSK3B contributed to the increase of blood glucose, insulin deficiency and insulin resistance [33]. Besides, excessive activation of GSK3B in mice promoted hyperphosphorylation of tau protein, neuronal death, reactive astrocytosis, and cognitive disorder [34]. BACE1 inhibitors have also been shown to possess the potential against AD by reducing Aβ burden in the brain [35]. NR3C1 encodes glucocorticoid receptor, the excessive activation of which would induce AD-like changes in the brain. Moreover, increased glucocorticoid levels have been observed in both AD and T2DM patients [36, 37]. Further studies in AD mice demonstrated that targeting glucocorticoid receptors could prevent amyloid lesions, ameliorate inflammatory responses and cognitive function, indicating that glucocorticoid receptor was a promising target for AD treatment [38, 39].

Certainly, the limitations of the present study need to be noted. Only five targets were selected as the screening models, which meant that we might miss some candidate targets. Besides, the inhibitory effects of multi-target ingredients were estimated by virtual screening and were worthy of further experimental studies.

CONCLUSIONS

In summary, by using network pharmacology and virtual screening technology, the present study ado-pted a novel approach for discovering Chinese herbal medicines and multi-target therapeutic ingredients against T2DM and AD from T2DM prescriptions. Ultimately, several natural compounds were identi-fied as potential drug candidates against T2DM and AD, including salvianolic acid J, gancaonin H, gadelaidic acid, icos-5-enoic acid, and sigmoidin-B. Notably, representative ingredients with high relevance to the overlapped targets between T2DM and AD, and natural compounds with high binding affinities to five therapeutic targets were both chiefly originated from Danshen, Gancao, and Sha-nyao. Hence, the present study suggested that Dan-shen-Gancao-Shanyao might be a promising TCM combination for the cotreatment of T2DM and AD, and encourage further studies in the future.

Footnotes

ACKNOWLEDGMENTS

This work was supported by the University Youth Innovation Team of Shandong Province (Grant No. 2019KJK017) and Talent Program of Zibo.

Authors’ disclosures available online ().

References

Patterson

(2018) World Alzheimer Report 2018. The State of the Art of Dementia Research: New Frontiers. Alzheimer’s Disease International, London.

Lane

, Hardy

, Schott

(2018) Alzheimer’s disease. Eur J Neurol 25, 59–70.

DeFronzo

, Ferrannini

, Groop

, Henry

, Herman

, Holst

, Hu

, Kahn

, Raz

, Shulman

, Simonson

, Testa

, Weiss

(2015) Type 2 diabetes mellitus. Nat Rev Dis Primers 1, 15019.

, Qiu

, Gatz

, Pedersen

, Johansson

, Fratiglioni

(2009) Mid- and late-life diabetes in relation to the risk of dementia: A population-based twin study. Diabetes 58, 71–77.

Bello-Chavolla

, Antonio-Villa

, Vargas-Vázquez

, Ávila-Funes

, Aguilar-Salinas

(2019) Pathophysiological mechanisms linking type 2 diabetes and dementia: Review of evidence from clinical, translational and epidemiological research. Curr Diabetes Rev 15, 456–470.

Miklossy

, Qing

, Radenovic

, Kis

, Vileno

, Laszlo

, Miller

, Martins

, Waeber

, Mooser

, Bosman

, Khalili

, Darbinian

, McGeer

(2010) Beta amyloid and hyperphosphorylated tau deposits in the pancreas in type 2 diabetes. Neurobiol Aging 31, 1503–1515.

Gao

, Cui

, Shen

, Ji

(2016) Shared genetic etiology between type 2 diabetes and Alzheimer’s disease identified by bioinformatics analysis. J Alzheimers Dis 50, 13–17.

Talbot

, Wang

, Kazi

, Han

, Bakshi

, Stucky

, Fuino

, Kawaguchi

, Samoyedny

, Wilson

, Arvanitakis

, Schneider

, Wolf

, Bennett

, Trojanowski

, Arnold

(2012) Demonstrated brain insulin resistance in Alzheimer’s disease patients is associated with IGF-1 resistance, IRS-1 dysregulation, and cognitive decline. J Clin Invest 122, 1316–1338.

Mullins

, Diehl

, Chia

, Kapogiannis

(2017) Insulin resistance as a link between amyloid-beta and tau pathologies in Alzheimer’s disease. Front Aging Neurosci 9, 118.

10.

Meng

, Li

, Shen

, Ji

(2020) Type 2 diabetes mellitus drugs for Alzheimer’s disease: Current evidence and therapeutic opportunities. Trends Mol Med 26, 597–614.

11.

Liu

, Wang

, Yan

, Zhang

, Pang

, Liao

(2013) Glimepiride attenuates Abeta production via suppressing BACE1 activity in cortical neurons. Neurosci Lett 557 Pt B, 90–94.

12.

Zhou

, Chen

, Peng

, Gu

, Yu

, Zhao

, Deng

(2019) Dulaglutide ameliorates STZ induced AD-like impairment of learning and memory ability by modulating hyperphosphorylation of tau and NFs through GSK3beta. Biochem Biophys Res Commun 511, 154–160.

13.

Chen

, Sun

, Zhao

, Guo

, Chen

, Fu

, Deng

(2017) Liraglutide improves water maze learning and memory performance while reduces hyperphosphorylation of tau and neurofilaments in APP/PS1/Tau triple transgenic mice. Neurochem Res 42, 2326–2335.

14.

Holubova

, Hruba

, Popelova

, Bencze

, Prazienkova

, Gengler

, Kratochvilova

, Haluzik

, Zelezna

, Kunes

, Holscher

, Maletinska

(2019) Liraglutide and a lipidized analog of prolactin-releasing peptide show neuroprotective effects in a mouse model of beta-amyloid pathology. Neuropharmacology 144, 377–387.

15.

Landreth

(2007) Therapeutic use of agonists of the nuclear receptor PPARgamma in Alzheimer’s disease. Curr Alzheimer Res 4, 159–164.

16.

Jiang

, Heneka

, Landreth

(2008) The role of peroxisome proliferator-activated receptor-gamma (PPARgamma) in Alzheimer’s disease: Therapeutic implications. CNS Drugs 22, 1–14.

17.

, Zhang

(2013) Traditional Chinese medicine network pharmacology: Theory, methodology and application. Chin J Nat Med 11, 110–120.

18.

Guedes

, de Magalhaes

, Dardenne

(2014) Receptor-ligand molecular docking. Biophys Rev 6, 75–87.

19.

Xue

, Fang

, Zhang

, Yi

, Wen

, Shi

(2013) TCMID: Traditional Chinese Medicine integrative database for herb molecular mechanism analysis. Nucleic Acids Res 41, D1089–1095.

20.

, Li

, Wang

, Zhou

, Li

, Huang

, Li

, Guo

, Tao

, Yang

, Xu

, Li

, Wang

, Yang

(2014) TCMSP: A database of systems pharmacology for drug discovery from herbal medicines. J Cheminform 6, 13.

21.

Liu

, Ouyang

, Yu

, Liu

, Huang

, Gong

, Zheng

, Li

, Jiang

(2010) PharmMapper server: A web server for potential drug target identification using pharmacophore mapping approach. Nucleic Acids Res 38, W609–614.

22.

Szklarczyk

, Morris

, Cook

, Kuhn

, Wyder

, Simonovic

, Santos

, Doncheva

, Roth

, Bork

, Jensen

, von Mering

(2017) The STRING database in 2017: Quality-controlled protein-protein association networks, made broadly accessible. Nucleic Acids Res 45, D362–D368.

23.

Zhou

, Zhou

, Pache

, Chang

, Khodabakhshi

, Tanaseichuk

, Benner

, Chanda

(2019) Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun 10, 1523.

24.

, Robertson

, Brooks

3rd , Vieth

(2003) Detailed analysis of grid-based molecular docking: A case study of CDOCKER-A CHARMm-based MD docking algorithm. J Comput Chem 24, 1549–1562.

25.

Tumminia

, Vinciguerra

, Parisi

, Frittitta

(2018) Type 2 diabetes mellitus and Alzheimer’s disease: Role of insulin signalling and therapeutic implications. Int J Mol Sci 19, 3306.

26.

Kleinridders

, Ferris

, Cai

, Kahn

(2014) Insulin action in brain regulates systemic metabolism and brain function. Diabetes 63, 2232–2243.

27.

Chatterjee

, Mudher

(2018) Alzheimer’s disease and type 2 diabetes: A critical assessment of the shared pathological traits. Front Neurosci 12, 383.

28.

, Wu

, Zhu

, Sha

, Yang

, Wei

, Ji

, Tang

, Mao

, Cao

, Wei

, Xie

, Yang

(2018) Insulin degrading enzyme contributes to the pathology in a mixed model of Type 2 diabetes and Alzheimer’s disease: Possible mechanisms of IDE in T2D and AD. Biosci Rep 38, BSR20170862.

29.

Kim

, Smith

, Van Eldik

(2004) Importance of MAPK pathways for microglial pro-inflammatory cytokine IL-1β production. Neurobiol Aging 25, 431–439.

30.

Ozaki

, Awazu

, Tamiya

, Iwasaki

, Harada

, Kugisaki

, Tanimura

, Kohno

(2016) Targeting the ERK signaling pathway as a potential treatment for insulin resistance and type 2 diabetes. Am J Physiol Endocrinol Metab 310, E643–E651.

31.

Vasudevan

, Balasubramanyam

(2004) Thiazolidinediones: A review of their mechanisms of insulin sensitization, therapeutic potential, clinical efficacy, and tolerability. Diabetes Technol Ther 6, 850–863.

32.

Khan

, Alam

, Haque

, Ashafaq

, Khan

, Ashraf

, Ahmad

(2019) Current progress on peroxisome proliferator-activated receptor gamma agonist as an emerging therapeutic approach for the treatment of Alzheimer’s disease: An update. Curr Neuropharmacol 17, 232–246.

33.

Zhang

, Huang

, Yan

, Jin

, Zhou

, Shi

, Jin

(2018) Diabetes mellitus and Alzheimer’s disease: GSK-3beta as a potential link. Behav Brain Res 339, 57–65.

34.

Hernandez

, Lucas

, Avila

(2013) GSK3 and tau: Two convergence points in Alzheimer’s disease. J Alzheimers Dis 33 Suppl 1, S141–144.

35.

Yan

, Vassar

(2014) Targeting the β secretase BACE1 for Alzheimer’s disease therapy. Lancet Neurol 13, 319–329.

36.

Rasmuson

, Andrew

, Näsman

, Seckl

, Walker

, Olsson

(2001) Increased glucocorticoid production and altered cortisol metabolism in women with mild to moderate Alzheimer’s disease. Biol Psychiatry 49, 547–552.

37.

Strachan

, Reynolds

, Frier

, Mitchell

, Price

(2009) The role of metabolic derangements and glucocorticoid excess in the aetiology of cognitive impairment in type 2 diabetes. Implications for future therapeutic strategies. Diabetes Obes Metab 11, 407–414.

38.

Lesuis

, Weggen

, Baches

, Lucassen

, Krugers

(2018) Targeting glucocorticoid receptors prevents the effects of early life stress on amyloid pathology and cognitive performance in APP/PS1 mice. Transl Psychiatry 8, 53.

39.

Pedrazzoli

, Losurdo

, Paolone

, Medelin

, Jaupaj

, Cisterna

, Slanzi

, Malatesta

, Coco

, Buffelli

(2019) Glucocorticoid receptors modulate dendritic spine plasticity and microglia activity in an animal model of Alzheimer’s disease. Neurobiol Dis 132, 104568.