Abstract

Background

   Alzheimer’s disease (AD) is one of the most common neurodegenerative
   disorders in the world, but still lack of effective drug treatment.
   Gynostemma Pentaphyllum (Thunb.) Makino (GpM), a Chinese medicinal
   herb, plays important roles in anti-inflammation, anti-oxidative stress
   and anti-tumor, which has been reported to ameliorate cognitive
   impairment of AD. However, the neuroprotective mechanism of GpM remains
   unclear. This study aims to investigate the targets and possible
   signaling pathways of GpM in the treatment of AD.

Methods

   Active compounds of GpM and their putative target proteins were
   selected from Traditional Chinese Medicine Systems Pharmacology (TCMSP)
   Database and Analysis Platform. AD-associated targets were identified
   from GeneCards, the Online Mendelian Inheritance in Man (OMIM) database
   and the Therapeutic Target Database (TTD). The intersecting targets of
   GpM and AD were identified and Gene Ontology (GO), Kyoto Encyclopedia
   of Genes and Genomes (KEGG) analysis were carried out to analyze the
   mechanism of them. Compound-target-pathway (CTP) network and
   protein–protein interaction (PPI) network were constructed and analyzed
   to elucidate the correlation between compounds, proteins and pathways.
   Molecular docking was performed to further demonstrate the possibility
   of GpM for AD.

Results

   A total of 13 active compounds of GpM, 168 putative target proteins of
   compounds and 722 AD-associated targets were identified. Eighteen
   intersecting targets of GpM and AD were found and the epidermal growth
   factor receptor (EGFR), interleukin-1 beta (IL-1β), interleukin-6
   (IL-6), nitric oxide synthase in endothelial (NOS3) and serum
   paraoxonase/arylesterase 1 (PON1) were selected as the primary targets
   of GpM in the treatment of AD. The neuroprotective effect of GPM was
   related to a variety of pathways, including amoebiasis, HIF-1 signaling
   pathway, cytokine-cytokine receptor interaction and so on.

Conclusions

   Our findings elucidate the active compounds, targets and pathways of
   GpM involved in effects of anti-AD. The novel mechanism of GpM against
   AD provides more treatment options for AD.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12906-022-03534-z.

   Keywords: AD pathology, Molecular mechanisms, Alzheimer’s disease,
   Gynostemma pentaphyllum (Thunb.) Makino, Network pharmacology

Background

   Alzheimer’s disease (AD) is a neurodegenerative disease with
   progressive dementia which is characterized by neuronal loss,
   neuroinflammation and pronounced memory decline. Epidemiological
   findings show that the incidence rate of AD is increasing with age.
   Taking care of AD patients require a lot of time and financial
   resources, which brings huge burden and impact to society and family
   [[31]1]. Current studies have demonstrated that extracellular
   accumulation of amyloid beta (Aβ) peptide [[32]2], dysfunction of
   cholinergic system [[33]3], neurofibrillary tangles (NFTs) of
   hyper-phosphorylated tau protein [[34]4], and multiple genes including
   ABCA7, BIN1, CASS4, CD33, CD2AP, CELF1, CLU, CR1 and DSG2 are involved
   in the occurrence and progression of AD [[35]5]. Substantial studies
   and clinical trials have demonstrated that it is difficult to develop
   one medicine for the treatment of AD, because the pathogenesis is very
   complex, including neurochemicals, amyloid and tau pathological
   processes, mitochondria, inflammatory pathways and neuroglia [[36]6].
   Chinese herbal medicine has a long history and plays a great role in
   the treatment of nervous system diseases. Its neuroprotective effect in
   AD has been confirmed by previous studies [[37]7]. In a recent review,
   the authors believe that Chinese herbal medicine may be beneficial in
   improving cognitive function of AD patients. However, a largescale and
   multi-center research should be conducted to evaluate its benefits in
   the treatment of AD [[38]8]. Therefore, the development of traditional
   Chinese medicine in the treatment of AD should be a long-term
   exploration.

   Gynostemma pentaphyllum (Thunb.) Makino (GpM) is one of Chinese
   medicinal herbs, mainly distributed in northeast and southeast of Asia.
   Previous studies have shown that GpM has various effects, such as
   anti-inflammation [[39]9], anti-oxidative stress [[40]10], immune
   regulation [], anti-cancer [[41]12], anti-aging [[42]13] and prevention
   of cardiovascular diseases [[43]14]. In recent years, GpM has been
   shown to alleviate brain lesion in chronic cerebral hypoperfusion
   [[44]15] and process neuroprotection against
   1-methyl-4-phenylpyridinium [[45]16]. It is worth noting that GpM has
   been shown strong neuroprotective effects in the treatment of AD
   [[46]17,[47]18]. However, the neuroprotective mechanism of GpM in AD
   has not been fully studied.

   In the present study, in order to investigate the potential
   pharmacological and molecular mechanisms of GpM in the treatment of AD,
   multiple databases and bioinformatics analysis were applied. The
   results showed that multiple pathways and targets were involved in the
   neuroprotection of GpM in the treatment of AD.

Methods

Screening for active compounds, putative target proteins and AD-associated
genes

   The active compounds of GpM and their putative target proteins were
   obtained from the Traditional Chinese Medicine Systems Pharmacology
   (TCMSP) Database and Analysis Platform
   ([48]https://old.tcmsp-e.com/tcmsp.php) [[49]19]. As a unique database
   containing a large number of herbs, active ingredients, and their
   targets, TCMSP contains pharmacokinetic properties of active compounds,
   such as oral bioavailability (OB), drug-likeness (DL) and so on. Oral
   bioavailability, as one of the most crucial pharmacokinetic parameters,
   shows the rate at which drugs enter the blood circulation. The greater
   the OB value of a compound, the more likely it is to become an
   effective drug. The drug-likeness (DL) represents the possibility of
   the compound becoming a drug. Due to poor pharmacological activity,
   most compounds in Traditional Chinese Medicine (TCM) cannot become
   effective drugs. OB ≥30% and DL ≥0.18 are considered as the criteria
   for screening clinical drugs [[50]20]. The putative target proteins of
   active compounds were collected from TCMSP database, because the
   database contains the target proteins of each active compound. All
   information about the active compounds and putative target proteins of
   GpM was obtained from TCMSP database.

   The AD-associated genes in this study were collected from 3 databases,
   which are GeneCards ([51]http://www.genecards.org/) [[52]21], Online
   Mendelian Inheritance in Man (OMIM) ([53]https://omim.org/) [[54]22]
   database and Therapeutic Target Database (TTD)
   ([55]http://db.idrblab.net/ttd/) [[56]23].

Gene ontology (GO) analysis and Kyoto encyclopedia of genes and genomes
(KEGG) pathway enrichment analysis

   Eighteen overlap targets between 168 putative compound target proteins
   and 722 AD-associated genes were considered as therapeutic targets for
   AD. In order to elucidate the pathogenesis of AD and clarify the
   mechanism and function of GpM, 18 intersecting targets were analyzed by
   Metascape ([57]http://metascape.org/) [[58]24]. Metascape is a website
   integrating gene annotation and analysis resources with more than 40
   databases, combined with GO analysis and KEGG pathway enrichment
   analysis. Results were considered significant at P < 0.01.

Compound-target-pathway network establishment

   The information of the components, targets and pathways was intuitively
   analyzed and shown a compound-target-pathway network by using Cytoscape
   3.8.2 software ([59]http://www.cytoscape.org/) [[60]25].

Protein-protein interaction (PPI) networks establishment

   PPI data were obtained from the STRING database
   ([61]https://string-db.org/) [[62]26], and the result of PPI data was
   visualized by Cytoscape software.

Molecule docking

   By using Discovery Studio Client v19.1.0.18287, AutoDockTools-1.5.6 and
   PyMOL, the binding sites, capacities and interactions between compounds
   and proteins were analyzed [[63]27–[64]29]. The crystal structures of
   core targets were obtained from the Protein Data Bank
   ([65]http://www.pdb.org/) [[66]30, [67]31]. The 3D chemical structural
   formulas of quercetin were obtained from PubChem
   ([68]https://pubchem.ncbi.nlm.nih.gov/) [[69]32].

Results

Active compounds and putative target proteins

   We used system pharmacology to find potential pharmacological
   mechanisms, and each step in defining the role of GpM on AD is shown in
   Fig. [70]1. Based on the TCMSP database, 13 compounds were selected as
   active compounds from GpM (Table-S[71]1), as shown in Fig. [72]2.

Fig. 1.

   [73]Fig. 1
   [74]Open in a new tab

   Network pharmacology for identifying mechanisms of Gynostemma
   pentaphyllum (Thunb.) Makino (GpM) in treating Alzheimer’s disease

Fig. 2.

   [75]Fig. 2
   [76]Open in a new tab

   The two-dimensional (2D) molecular structures of 13 active compounds of
   GpM

   The putative target proteins were screened by docking and binding
   scores. A total of 168 putative target proteins along with the 13
   active compounds were captured by using TCMSP (Table-S[77]2). The value
   of score represents the relationship between the active compound and
   the target protein. The higher the score is, the closer the
   relationship is. The 13 active compounds, 168 putative target proteins
   and their relationships were shown in Fig. [78]3.

Fig. 3.

   [79]Fig. 3
   [80]Open in a new tab

   The construction of Compound-Target network. Different active compounds
   (the circular nodes with orange) were presented in circle. The size of
   the circular node represents the targets number of compounds. The
   putative target proteins (the circular nodes with blue) were listed
   inside the circle made by active compounds

AD-associated targets and the intersecting targets between GpM and AD

   We searched three databases and obtained 722 genes entries as
   AD-associated targets. Then, 18 overlap targets between 168 putative
   target proteins of GpM and AD-associated targets were regarded as
   intersecting targets (Fig. [81]4), including ATP-binding cassette
   sub-family G member 1 (ABCG1), acetylcholinesterase (ACHE), alcohol
   dehydrogenase 1C (ADH1C), cytochrome P450 1A2 (CYP1A2), cytochrome P450
   3A4 (CYP3A4), epidermal growth factor receptor (EGFR), estrogen
   receptor beta (ESR2), heat shock protein beta-1 (HSPB1), Interleukin-10
   (IL10), interleukin-1 beta (IL-1β), interleukin-6 (IL-6), Insulin
   receptor (INSR), potassium voltage-gated channel subfamily H member 2
   (KCNH2), neutrophil cytosol factor 1 (NCF1), nitric oxide synthase,
   endothelial (NOS3), urokinase-type plasminogen activator (PLAU), serum
   paraoxonase/arylesterase 1 (PON1), peroxisome proliferator-activated
   receptor delta (PPARD), listed in Table-S[82]3.

Fig. 4.

   Fig. 4
   [83]Open in a new tab

   The Venn diagram. The Venn diagram showed the overlap of putative
   target proteins of GpM and AD-associated targets

   By screening the active components corresponding to the intersecting
   targets, three compounds, rhamnazin (GpM2), isofucosterol (GpM7) and
   quercetin (GpM9), were selected to be as the main compounds.

Enrichment analysis of intersecting targets

   GO and KEGG enrichment analyses were performed to identify biological
   functions and metabolic pathways associated with intersecting targets.
   Filtering with P value < 0.01, a minimum count of 3, 337 GO entries and
   47 KEGG pathway entries were identified (Table-S[84]4, Table-S[85]5).
   Of the 337 GO entries, there were 314 for biological processes, 8 for
   cellular components, and 15 for molecular function. Next, we used
   bubble diagram to visually display the results of GO analysis and KEGG
   pathway enrichment analysis (Figs. [86]5, [87]6, [88]7 and [89]8).

Fig. 5.

   [90]Fig. 5
   [91]Open in a new tab

   The molecular functions of Gene Ontology (GO) analysis. The
   significantly enrichment of molecular functions (MF) with P < 0.01

Fig. 6.

   [92]Fig. 6
   [93]Open in a new tab

   The cellar components of Gene Ontology (GO) analysis. The significantly
   enrichment of cellar components (CC) with P < 0.01

Fig. 7.

   [94]Fig. 7
   [95]Open in a new tab

   The biological processes of Gene Ontology (GO) analysis. The
   significantly enrichment of biological processes (BP) (Top 30) with
   P < 0.01

Fig. 8.

   [96]Fig. 8
   [97]Open in a new tab

   Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment
   analysis. The KEGG pathway enrichment analysis with P < 0.01 (Top 30)
   is showed in Fig. 8

   As shown in Fig. [98]5, the results showed that the 15 molecular
   functions of the intersecting targets of GpM and AD mainly concentrated
   in steroid binding (GO:0005496), monooxygenase activity (GO:0004497),
   protein homodimerization activity (GO:0042803), oxidoreductase activity
   (GO:0016491), heme binding (GO:0020037), growth factor receptor binding
   (GO:0070851), tetrapyrrole binding (GO:0046906), oxidoreductase
   activity, acting on paired donors, with incorporation or reduction of
   molecular oxygen (GO:0016705), signaling receptor regulator activity
   (GO:0030545), cytokine activity (GO:0005125), cytokine receptor binding
   (GO:0005126), protein domain specific binding (GO:0019904),
   phospholipid binding (GO:0005543), receptor ligand activity
   (GO:0048018) and signaling receptor activator activity (GO:0030546).

   As shown in Fig. [99]6, the 8 cellular components of the intersecting
   targets of GpM and AD were mainly involved in plasma membrane protein
   complex (GO:0098797), endocytic vesicle (GO:0030139), membrane raft
   (GO:0045121), membrane microdomain (GO:0098857), focal adhesion
   (GO:0005925), cell-substrate junction (GO:0030055), receptor complex
   (GO:0043235) and Golgi membrane (GO:0000139).

   As represented in Fig. [100]7, the top 30 biological processes from the
   intersecting targets of GpM and AD were principally linked to
   regulation of hormone levels (GO:0010817), diterpenoid metabolic
   process (GO:0016101), terpenoid metabolic process (GO:0006721),
   isoprenoid metabolic process (GO:0006720), regulation of protein
   secretion (GO:0050708), cellular response to oxidative stress
   (GO:0034599), cholesterol metabolic process (GO:0008203), positive
   regulation of small molecule metabolic process (GO:0062013), response
   to wounding (GO:0009611), positive regulation of cell migration
   (GO:0030335), secondary alcohol metabolic process (GO:1902652), sterol
   metabolic process (GO:0016125), response to inorganic substance
   (GO:0010035), cellular response to reactive oxygen species
   (GO:0034614), positive regulation of cell motility (GO:2000147),
   positive regulation of cellular component movement (GO:0051272),
   positive regulation of locomotion (GO:0040017), cellular response to
   chemical stress (GO:0062197), response to lipopolysaccharide
   (GO:0032496), alcohol metabolic process (GO:0006066), protein secretion
   (GO:0009306), establishment of protein localization to extracellular
   region (GO:0035592), response to molecule of bacterial origin
   (GO:0002237), protein localization to extracellular region
   (GO:0071692), regulation of lipid localization (GO:1905952), regulation
   of inflammatory response (GO:0050727), regulation of peptide hormone
   secretion (GO:0090276), response to drug (GO:0042493), regulation of
   peptide secretion (GO:0002791) and regulation of peptide transport
   (GO:0090087).

   As shown in Fig. [101]8, the top 30 KEGG signaling pathways of the
   intersecting targets were Amoebiasis (ko05146, hsa05146), HIF-1
   signaling pathway (hsa04066), foxo signaling pathway (hsa04068),
   African trypanosomiasis (ko05143, hsa05143), Malaria (ko05144,
   hsa05144), Inflammatory bowel disease (IBD) (ko05321, hsa05321),
   Retinol metabolism (hsa00830, ko00830), Drug metabolism - cytochrome
   P450 (hsa00982, ko00982), Leishmania infection (hsa05140, ko05140),
   Metabolism of xenobiotics by cytochrome P450 (hsa00980, ko00980),
   Pertussis (ko05133, hsa05133), Chemical carcinogenesis (ko05204),
   Cytokine-cytokine receptor interaction (hsa04060, ko04060), Chemical
   carcinogenesis (hsa05204), Estrogen signaling pathway (ko04915),
   AGE-RAGE signaling pathway in diabetic complications (ko04933,
   hsa04933), Chagas disease (American trypanosomiasis) (ko05142,
   hsa05142) and insulin resistance (hsa04931).

The compound-target-pathway (CTP) and protein-protein interaction (PPI)
networks of the intersecting targets

   The compound-target-pathway (CTP) and protein–protein interaction (PPI)
   networks were performed to demonstrate the potential mechanism of GpM
   on AD. The CTP network showed that 18 proteins and 28 related KEGG
   pathways (after deleting redundancy) (Fig. [102]9). The PPI network
   demonstrated the interaction among the 15 intersecting target proteins
   (three stray nodes were hidden) (Fig. [103]10). According to the
   degree, epidermal growth factor receptor (EGFR), interleukin-1 beta
   (IL-1β), interleukin-6 (IL-6), nitric oxide synthase, endothelial
   (NOS3), serum paraoxonase/arylesterase 1 (PON1) were selected as the
   core target proteins to determine binding energy by means of a
   molecular docking analysis.

Fig. 9.

   [104]Fig. 9
   [105]Open in a new tab

   The compound-target-pathway (CTP) network. The 3 active compounds were
   presented by blue diamond nodes in the center. The target proteins
   (orange round nodes) which are associated with AD were situated as a
   round enclosing the 3 compounds. The pathways were listed at the
   outermost and presented by green round rectangles

Fig. 10.

   [106]Fig. 10
   [107]Open in a new tab

   The Protein-Protein Interaction (PPI) network. The top 15 intersecting
   target proteins were shown in circular nodes with different gradations
   of red. The gradation of red and size of the nodes expressed the
   interaction degree on the network

Molecular docking

   To further verifying the function of GpM on AD, we selected quercetin,
   one of the three compounds, as the significant compound for the all 5
   core targets are part of its putative target proteins. With the
   utilization of AutoDockTools-1.5.6, OpenBabel-2.4.1 and Pymol-2.2.0
   software, the interactions between quercetin and the 5 core targets
   were analyzed. Figures [108]11 and [109]12 showed the molecule docking
   of 5 core targets with quercetin. The binding energy of each protein
   with quercetin was shown in Table [110]1. The lower binding energy is,
   the higher the affinity between quercetin and the target (protein) is.
   The hydrogen bonds were found between the active site of proteins and
   quercetin.

Fig. 11.

   [111]Fig. 11
   [112]Open in a new tab

   The molecular docking diagrams of quercetin and IL-6, IL-1β using
   AutoDockTools. The active sites of the proteins, the binding distances,
   molecular docking model between the proteins and the quercetin were
   presented in Fig. 11. (A, B) quercetin and the protein IL-6
   (− 7.14 kcal/mol); (C), (D) quercetin and the protein IL-1β
   (− 6.21 kcal/mol)

Fig. 12.

   [113]Fig. 12
   [114]Open in a new tab

   The molecular docking diagrams of quercetin and NOS3, PON1, EGFR using
   AutoDockTools. The active sites of the proteins, the binding distances,
   molecular docking model between the proteins and the quercetin were
   presented in Fig. 12. (A) quercetin and the protein NOS3
   (− 5.59 kcal/mol); (B) quercetin and the protein PON1
   (− 6.74 kcal/mol); (C) quercetin and the protein EGFR (− 5.52 kcal/mol)

Table 1.

   The binding energy of quercetin and 5 targets
   Binding energy
   (kcal/mol)      IL-6     IL-1β NOS3  PON1   EGFR
              quercetin     −7.14 −6.21 −5.59 − 6.74 −5.52
   [115]Open in a new tab

   Note: The binding energy < − 5 kcal/mol was considered as high affinity
   between quercetin and the target [[116]33]

   The 3D model of quercetin in the active site of IL-6 was showed in the
   Ray tracing diagram in Fig. [117]11A. The binding energy of quercetin
   with IL-6 was found to be − 7.14 kcal/mol. The type, distance and
   number of the binding complexes of the active site of IL-6 and
   quercetin were shown in the enlarged drawing in Fig. [118]11B. The
   following atoms, ARG-179, SER-176, GLU-172, PHE-173, LYS-66 SER-169 and
   MET-67 formed a pocket around quercetin by hydrogen bonding. Similarly,
   the binding interaction of quercetin and IL-1β was shown in
   Fig. [119]11C. The following atoms, GLU-37, GLN-39, MET-20 and LYS-63
   were connected to quercetin by hydrogen bonding (Fig. [120]11D). The
   other three docking result were shown in Fig. [121]12.

   To further verify the results of molecular docking, we applied
   Discovery Studio Client to investigate other chemical bonds including
   Pi-Alkyl, Pi-Sigma and Pi-Pi interaction between quercetin and 5 core
   targets. The results were shown in Figs. [122]13 and [123]14.

Fig. 13.

   [124]Fig. 13
   [125]Open in a new tab

   The molecular docking of quercetin and IL-6, IL-1β using Discovery
   Studio Client. (A) the 3D model of quercetin and protein IL-6, (B) the
   2D model of quercetin and protein IL-6, (C) the 3D model of quercetin
   and protein IL-1β, (D) the 2D model of quercetin and protein IL-1β

Fig. 14.

   [126]Fig. 14
   [127]Open in a new tab

   The molecular docking of quercetin and NOS3, PON1, EGFR using Discovery
   Studio Client. (A) the 3D model of quercetin and protein NOS3, (B) the
   2D model of quercetin and protein NOS3, (C) the 3D model of quercetin
   and protein PON1, (D) the 2D model of quercetin and protein PON1, (E)
   the 3D model of quercetin and protein EGFR, (F) the 2D model of
   quercetin and protein EGFR

   Figure [128]13A showed the 3D model of quercetin and IL-6. To
   explicitly exhibit the binding environment, 2D model was shown in
   Fig. [129]13B. Several bonds were obtained between quercetin and IL-6.
   There were Pi-Alkyl interactions between quercetin and amino acid
   residues of IL-6 including LEU-165, LYS-66 and ALA-58. Pi-Sigma
   interactions exited in active sites including GLU-55 and LEU-64.
   Figure [130]13C showed the 3D model of quercetin and IL-1β. The binding
   environment between quercetin and IL-1β was shown in Fig. [131]13D.
   There were Pi-Alkyl interactions between quercetin and IL-1β at the
   active sites of LYS-209, PRO-118, ALA-117 and ARG-120. In addition,
   Pi-Pi interactions existed between quercetin and the active site of
   HIS-115.

   The other three docking results were shown in Fig. [132]14. All the
   chemical bonds we found with their bond length were shown in
   Table [133]2. Multiple interaction sites and chemical bonds indicated
   that there was great affinity between quercetin and core targets.

Table 2.

   Interaction parameters of core targets with quercetin
   Targets Hydrogen bonding Pi-Alkyl Pi-Sigma Pi-Pi
   IL-6

   ARG-179(2.4, 1.9)

   SER-176(2.4)

   GLU-172(1.8)

   PHE-173(2.7)

   LYS-66(1.8)

   MET-67(2.1)

   SER-169(2.0)

   LEU-165(4.4)

   LYS-66(5.3)

   ALA-58(4.6)

   GLU-55(3.3, 4.1, 4.9)

   LEU-64(2.8)
   –
   IL-1B

   GLU-37(3.0, 2.1)

   GLN-39(3.1)

   MET-20(2.0, 2.2, 2.5)

   LYS-63(1.8)

   LYS-209(4.9)

   PRO-118(5.2)

   ALA-117(3.3, 4.0)

   ARG-120(3.8, 4.4, 4.8, 5.4)
   – HIS-115(4.2)
   NOS3

   ARG-70(2.0)

   TRY-475(2.2, 3.3)

   ARG-474(2.3, 2.7)

   ARG-107(2.2, 2.4)

   LEU-431(5.2, 4.9, 4.9)

   VAL-71(5.5)
   VAL-71(2.7) –
   PON1

   GLU-56(3.4)

   ASN-227(2.0, 2.0, 2.5)

   PRO-230(2.7)

   SER-272(2.2)

   VAL-336(2.2)

   VAL-346(5.2)

   ILE-291(4.4)
   –

   PHE-292(4.5)

   TYR-71(4.9)
   EGFR

   LEU-1017(2.1)

   GLU-1015(1.9, 2.7)

   ASP-1009(1.8, 2.1)

   GLU-736(2.0)

   LEU-844(4.2, 4.9)

   ALA-743(4.7)

   VAL-726(5.3)

   LYS-745(5.0, 4.9)

   LEU-788(5.3)
   – –
   [134]Open in a new tab

Discussion

   With the development of components analysis of Chinese herbs, TCM shows
   the characteristics of effective multi-target and multi-component
   [[135]34]. GpM or some of its components have been shown to possess
   clinical effects in cardiomyopathy [[136]35], diabetes mellitus
   [[137]36] and cancer [[138]12]. In recent years, many clinical studies
   have shown that GpM plays a positive role in alleviating cognitive
   impairment and brain damage of AD [[139]17, [140]18]. Although GpM has
   above medicinal value, the related mechanism remains unclear.

   In this study, we explored the active compounds of GpM, and selected
   the intersecting targets between the compounds and AD. 18 targets from
   the 168 putative target proteins of compounds and 722 AD-associated
   targets were obtained. Furthermore, multiple signaling pathways,
   biological processes, cellular components and molecular functions were
   identified by GO/KEGG pathway enrichment analysis. It is considered
   that seven pathways were closely related to the pathogenesis of AD
   including amoebiasis (ko05146, hsa05146), HIF-1 signaling pathway
   (hsa04066), African trypanosomiasis (ko05143, hsa05143), foxo signaling
   pathway (hsa04068), malaria (ko05144, hsa05144), inflammatory bowel
   disease (IBD) (ko05321), retinol metabolism (ko00830, hsa00830). Five
   essential proteins (IL-6, IL-1β, NOS3, PON1 and EGFR) were screened out
   using CTP and PPI network. Due to good molecular docking results
   presented by Auto Dock Tools, it is theoretically proved they were
   target proteins for further study of the anti-AD effects of GpM.

   Consistent with our results, HIF-1 signaling pathway and foxo signaling
   pathway are involved in the occurrence and development of AD, which has
   been reported in previous studies [[141]37, [142]38]. Inflammatory
   bowel disease (IBD) is considered to be closely related to pathogenesis
   of AD, based on studies of gut-brain axis [[143]39]. Additionally,
   vitamin A deficiency has been demonstrated in patients with AD
   [[144]40], while vitamin A supplementation can alleviate the
   development of AD [[145]41]. More interestingly, our study suggests
   that the AD-associated pathways may relate to amoebiasis, African
   trypanosomiasis and malaria, which are all caused by parasitic
   infections and may invade the central nervous system. Existing studies
   have shown the existence of anti-parasitic infection and anti-AD drugs
   (such as GSK-3 inhibitors) [[146]42], which imply that the incidence of
   AD may relate to parasitic infection. To a certain extent, it shows the
   credibility and innovation of our conclusion.

   Previous studies have shown that inflammatory cytokines are involved in
   AD [[147]43, [148]44]. A recent study showed that elderly individuals
   with amyloid-beta deposition had higher levels of IL-1β and IL-6
   [[149]45]. In addition, it has been reported that higher level of Aβ42
   can reduce endothelial NO synthase (eNOS, NOS3), cyclic GMP (cGMP) and
   protein kinase G (PKG) activity [[150]46]. This is consistent with our
   finding that GpM may prevent against AD by regulating the activity of
   NOS3 and the levels of IL-1β and IL-6. PON1 is a new factor associated
   with impaired cognition and may play a role in the development of AD
   [[151]47]. It is reported that EGFR is related to AD, and EGFR
   inhibitors can be used as burgeoning therapeutic strategy for AD
   [[152]48]. These data suggest that targeting on IL-1β, IL-6, NOS3, PON1
   or EGFR may be effective in the treatment of AD, which supports our
   finding that proteins with good molecular docking are the important
   target proteins for neuroprotection of GpM in the treatment of AD.

   In conclusion, based on network pharmacology and bioinformatics, this
   study illustrated the key targets and molecular mechanisms of GpM,
   which will provide instructive suggestions for the further study of GpM
   in the treatment of AD.

Conclusions

   In this study, we applied network pharmacology and bioinformatics to
   analyze the therapeutic effect of GpM on AD. The active components and
   putative targets of GpM were explored and discussed systematically.
   Comparing putative targets of GpM with known AD related genes,
   constructing and analyzing CTP/PPI networks, 5 important proteins were
   identified, showing strong therapeutic potentials against AD. Through
   enrichment analysis, we further identified GpM as a promising drug with
   multiple components, targets and pathways for the treatment of AD. In
   addition, we demonstrated the feasibility of GpM in the treatment of AD
   by molecular docking. In conclusion, our findings provide a new idea
   for the neuroprotection of GpM and contribute to the development of GpM
   in the treatment of AD. However, data mining and analysis alone is not
   enough. In future, we will conduct experiments to confirm our points
   and provide more effective treatment measures for AD patients.

Supplementary Information

   [153]12906_2022_3534_MOESM1_ESM.xlsx^ (10.7KB, xlsx)

   Additional file 1: Table-S1. Result of 13 candidate compounds of GpM.
   [154]12906_2022_3534_MOESM2_ESM.xlsx^ (20.6KB, xlsx)

   Additional file 2: Table-S2. Result of 168 putative target proteins of
   GpM.
   [155]12906_2022_3534_MOESM3_ESM.xlsx^ (24.1KB, xlsx)

   Additional file 3: Table-S3. 722 AD-associated target proteins of Homo
   sapiens from OMIM, TTD and Gene Cards were regarded as the
   substantially contributing proteins.
   [156]12906_2022_3534_MOESM4_ESM.xls^ (190.5KB, xls)

   Additional file 4: Table-S4: Result of 337 GO terms enrichment of GpM3.
   [157]12906_2022_3534_MOESM5_ESM.xls^ (47.5KB, xls)

   Additional file 5: Table-S5. Result of the KEGG pathway enrichment
   analysis of GpM.

Acknowledgements