Fa-Sheng Wu,Zhou-Qiang Li,Xiao-Yu Chen,Zu-Ping Lian,Shi-Jin Lu,En-Cun Hou＊

1Ruikang Hospital Affiliated to Guangxi University of Traditional Chinese Medicine,Nanning 530011,China;2Graduate School of Guangxi University of Traditional Chinese Medicine,Nanjing 530000,China.

Background:This study pioneers a network pharmacology-based approach and in silico molecular docking and visualization approaches to predict the potential molecular mechanisms involved in treatment of lung cancer with Lingzhi (Ganoderma lucidum).Methods:The active ingredients of Ganoderma lucidum were screened in the Traditional Chinese Medicine Systems Pharmacology database,and corresponding potential targets of these active ingredients were queried.Then potential targets for the treatment of lung cancer were separately queried using the GeneCards database.The overlapping targets of the active ingredients and lung cancer treatment were further analyzed in silico by calculating degree values and conducting literature searches.A protein-protein interaction network was constructed for these overlapping targets using the STRING database,and a “drug-ingredient-target”network was created using Cytoscape 3.7.1 software.Gene ontology and Kyoto Encyclopedia of Genes and Genomes(KEGG) enrichment analyses were performed for key targets using Metascape.Finally,the key molecules were examined using molecular docking in Autodock 3.7 software,and the representative docking results were visualized using PyMol software.Results:A total of 14 active ingredients were identified,including 3 key ingredients(ganolucidic acid E,ergosta-4,6,8(14),22-tetraene-3-one,and beta-sitosterol) and 8 key targets.Gene ontology analysis and KEGG enrichment showed that Ganoderma lucidum may act on 10 key signaling pathways,such as the neuroactive ligand-receptor interaction pathway,calcium signaling pathway,and cyclic adenosine monophosphate signaling pathway,to regulate neuroactivity,ion homeostasis,and other processes,thus inhibiting the progression of lung cancer.Molecular docking results predicted that among the 8 key targets,mu-type opioid receptor,transforming growth factor beta-1 proprotein,and apoptosis regulator BAX exhibit better docking with these ligands.Conclusion:The preliminary results of this study identified potential active ingredients of Ganoderma lucidum for the treatment of lung cancer,and putative mechanisms of action,which could provide a theoretical basis for further experimental studies on anticancer efficacy.

Key words:Ganoderma lucidum,Lung cancer,Mechanisms

Background

Lingzhi (Ganoderma lucidum(G.lucidum)),also known as “Ruicao” (meaning auspicious herb) and“Shenzhi” (meaning magicGanoderma) in China,is a dried fruiting body of the polypore fungusG.lucidum(Leyss.ex Fr.) Karst orG.sinense.As recorded inThe Classic of Herbal Medicine(written by Shennong between about 200 and 250 CE.)G.lucidumenhances the human body by nourishing,strengthening,and prolonging life.In theCompendium of Materia Medica(written by Li Shizhen during the Ming dynasty)G.lucidumwas also recorded as being able to refresh and prolong life.G.lucidumhas been used in China for more than 2,000 years,and is known for its high medicinal value,diverse healthful ingredients,and wide range of pharmacological effects.The main medicinal parts of theG.lucidumfungi include the spore powder,fruiting bodies,and hyphae.Modern pharmacological studies have revealed thatG.lucidumand its extracts display various effects,such as antiproliferation,apoptosis induction,antitumor,antiangiogenesis,prevention of multi-drug resistance,and toxicityreversing effects.G.lucidumis also involved in a variety of cellular and molecular mechanisms in antitumor protection and immune enhancement [1].Its endogenous and exogenous apoptotic pathways relate to cell cycle arrest,telomerase inhibition,autophagy,and oxidative stress,which are extensively involved in antiproliferation and apoptosis induction in cells [2].However,the effective ingredients ofG.lucidumfor its effective treatment of lung cancer and their mechanisms of action are still unclear.With the development and increasing recognition of network pharmacology-based approaches to investigate the mechanisms of action of drugs,we applied such a network pharmacology-based method to explore the primary active ingredients that displayed efficacy against lung cancer and their corresponding signaling pathways,in an effort to provide a theoretical basis for the clinical development and application ofG.lucidum.

Methods

Screening of G.lucidum ingredient-target network

Identifying the active ingredients of G.lucidum.A full list of ingredients ofG.lucidumwas obtained using the Traditional Chinese Medicine System Pharmacology (TCMSP,http://tcmspw.com/tcmsp.Php)database and analysis platform [3],and filtered with the oral bioavailability (OB) value set ≥ 30% and drug-like properties (DL) ≥ 0.18.

Standard naming of target proteins of G.lucidum and corresponding genes.Verified relevant protein targets of the above-identified drug ingredients were obtained through the TCMSP,and filtered using the UniProt database [4] (https://www.uniprot.org/) for human proteins,followed by conversion of the names of these target proteins to standard gene names.Target proteins that failed to be converted to standard gene names were removed,as these proteins may be expressed only in species other than human beings.

Constructing a drug-component-target network.The active ingredients and genes obtained from the screening were imported into Cytoscape 3.7.1 software to construct a drug component-target network.Ingredients with no known corresponding target proteins identified were removed.The active pharmaceutical ingredients ofG.lucidumwith satisfactory OB values,DL values,and the maximum number of target proteins were determined.

Screening predictive targets for G.lucidum treatment of lung cancer and constructing a target protein interaction network

Retrieving lung cancer target genes.The GeneCard database (https://www.genecards.org/) was used to retrieve all relevant target genes for lung cancer using the search term “lung cancer” and the time frame up to and including November 2020.

Predicting overlapping targets.All targets of both lung cancer andG.lucidumwere imported into VENNY 2.1 to produce a Venn diagram showing the intersection of these target proteins.These overlapping proteins were considered candidate predictive targets ofG.lucidumtreating lung cancer.

Constructing a protein-protein interaction network and screening the core targets.The overlapping targets ofG.lucidumand lung cancer were imported into the STRING database (http://string-db.org),and protein-protein interaction (PPI) analysis was performed for the treatment of lung cancer withG.lucidum.“Homo sapiens” was selected as the target species.Network topological features including degree,combine score,betweenness centrality,and closeness centrality were obtained,and the data were imported into Cytoscape 3.7.1 software to construct a PPI network diagram of target proteins.

Enrichment analysis of predicted targets for G.lucidum treatment of lung cancer

Gene ontology function enrichment analysis.Overlapping target proteins were imported into the Metascape database (http://metascape.org) to perform gene ontology (GO) analysis;the threshold was set atP＜ 0.01.Three gene functions,including biological process (BP),molecular function (MF),and cellular component (CC),were annotated.The histograms generated for BP,CC,and MF were exported.

Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis.The overlapping target proteins were imported into the DAVID database(https://david.ncifcrf.gov/tools.jsp) [5] for Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis.The data were imported into the BIOINFORMATICS online mapping website(http://www.bioinformatics.com.cn) to generate KEGG bubble maps.

Molecular docking and visualization

Molecular docking.The 8 key target proteins obtained above were semi-flexibly docked with the three major active components using Autodock 3.7.Detailed steps are described below:(1) the 3D structures of the three traditional Chinese Medicine small molecules were obtained from the PubChem database(https://PuBchem.ncbi.nlm.nih.gov/) and saved in mol2 format.X-ray 3D crystal structures of the core targets were downloaded from the RCSB PDB structure database (www.rcsb.org/).Excess ligands and solvents were removed from the receptor protein using PyMol 1.7 software to obtain a pure receptor protein structure.(2) AutoDock software was used to hydrogenate,calculate the charge,and identify the atomic types of the receptors and ligands.Finally,Autodock software was used to perform one-by-one semi-flexible docking on a Windows 10 operating system to predict binding energy.

Visualization of the docking results.The molecular models with high predicted binding energy were imported into PyMol software after docking for molecular model visualization using stick models and surface pocket models.The docking sites of ligands and receptors were marked,and the distances between hydrogen bonding sites were measured.

Results

Collecting and screening active ingredients in G.lucidum

A total of 242 ingredients inG.lucidumwere retrieved from the TCMSP database,among which 61 active candidates were identified with an OB value ≥ 30% and a DL value ≥ 0.18.

Putative protein targets of G.lucidum.The putative protein targets and corresponding standard gene names are listed in Table 1.Fourteen components with protein targets were retained,including primarily triterpenoids and sterols (Table 2).Three main active ingredients that had interactions with the maximal number of target proteins from this list were identified by referring to the OB and DL values in Figure 2;these included ganolucidic acid E,ergosta-4,6,8(14),22-tetraene-3-one(referred to hereafter as erg),and beta-sitosterol.

Table 1 Gene names corresponding to protein targets

Table 2 Drug ingredients and corresponding genes of the target proteins

Drug-ingredient-targetnetworkdiagram constructed using Cytoscape.The active components,targets and their corresponding relationships were imported into the Cytoscape software to draw a diagram(Figure 1).

G.lucidum effect on lung cancer predicted target gene screening and target protein nteractions PPI network analysis results

Screening predictive targets for G.lucidum treatment of lung cancer.A total of 23,071 lung cancer-related target genes were obtained from the GeneCard database,and 36 overlapping targets of active ingredients inG.lucidumwere identified using the Venn diagram generated with VENNY 2.1 (Figure 2).

Screening results of PPI network and core targets.Input of the overlapping target into the STRING database returned 41 nodes and 350 edges,which were plotted in Cytoscape (Figure 3).The degree value is represented by the size and color of the nodes;larger nodes with darker colors indicate higher degree values.The thickness and color of lines are indicative of the combined score:thicker lines with darker color indicate higher combined scores.The targets with the top 5 degree values were CASP3,JUN,ADRB2,DRD1,and OPRM1 (in descending order).Additionally,3 key targets,transforming growth factor -β1 (TGF-β1),CASP8,and BAX,were also identified from literature review.

Results of enrichment analysis of predicted targets for G.lucidum against lung cancer

GO-BP results.BP results of GO function enrichment predicted that the major biologic process affected inG.lucidumtreatment of lung cancer included regulation of membrane potential,blood circulation,the cellular response to ammonium ions,the cellular response to organic cyclic compounds,the response to toxic substances,and the regulation of ion transport.The top 20 major biological processes were plotted by theirpvalues (Figure 4).

GO-CC results.CC results of GO function enrichment predicted thatG.lucidumtreatment of lung cancer mostly affected postsynaptic,membrane raft,and dendrite membrane (Figure 5).

GO-MF results.MF results of GO function enrichment showed that the following molecular functions predicted to be affected byG.lucidumtreatment of lung cancer are neurotransmitter receptor activity,neurotransmitter receptor activity involved in regulation of postsynaptic membrane potential,and ammonium ion binding (Figure 6).

Results of KEGG pathway enrichment analysis.The 36 genes imported into the Metascape database for KEGG pathway analysis returned 53 enriched signal pathways,among which 10 major signaling pathways were deemed relevant,including the neuroactive ligandreceptor interaction pathway,the calcium signaling pathway,morphine addiction,apoptosis,cancer pathways,and the cAMP signaling pathway.The genes involved in these 10 signaling pathways are shown in Table 3.A bubble chart is shown in Figure 7.

Molecular docking

AutoDock software was used to perform one-by-one semi-flexible docking simulations,and the minimum binding energy of each combination was recorded(Table 4),where a negative binding energy indicates that the binding of the receptor and ligand is energetically favored.A larger magnitude negative value indicates a stronger binding ability;therefore,it can be concluded that all three ingredients have good predicted binding with TGF-β1 and OPRM1.

After docking,the paired molecular models with highest binding energy were imported into PyMol software,and some representative interactions were selected for molecular model visualization (Figures 8-13).

Figure 1 The drug-ingredient-target network diagram.

Figure 2 Venn diagram showing overlap of G.lucidum drug targets and lung cancer-associated genes.

Figure 3 A PPI network diagram showing G.lucidum interaction with lung cancer-associated targets.

Figure 4 Histogram of the main biological processes implicated in lung cancer treatment with G.lucidum (bar color and length vary with P-value)

Figure 5 Histogram of the main cellular components implicated in G.lucidum treatment of lung cancer (bar color and length vary with P-value)

Figure 6 Histogram of the main molecular functions implicated in G.lucidum treatment of lung cancer (bar color and length vary with P-value)

Figure 7 A bubble chart showing the main signaling pathways of G.lucidum predicted to be involved in its treatment of lung cancer (darker colors indicate smaller P-values,and larger dots indicate more genes involved)

Figure 8 Stick model of Erg-OPRM1 predicted interaction.

Table 3 Signaling pathway results from KEGG analysis

Table 4 Molecular docking simulation data

Discussion

In this study,a total of 61 active ingredients,39 predicted targets,and 36 overlapping genes were screened using a network pharmacology-based approach.Among the 61 active ingredients,14 ingredients with relevant predicted targets,mostly sterols,triterpenes,and polysaccharides,were identified as key active ingredients.In recent years,studies have found that beta-sitosterol (BS) has a strong effect against a variety of cancer cell types.BS induces G0/G1 cell cycle arrest in A549 cells,and inhibits cell proliferation without significant damage to normal human lung cells or PBMC cells.In addition,BS reduces the expression of proteins involved in the TGFβ/p-Smad2/3/c-Myc pathway.Moreover,BS has been shown to regulate disorder and p53-dependent apoptosis through reactive oxygen species (ROS)-mediated activation in mitochondria,characterized by caspase-3 and -9 activation,the presence of Annexin-V/PI positive cells,PARP inactivation,MMP loss,decrease of the Bcl-2/Bax ratio,and cytochrome c release [6].Ganodermic acid (G.lucidumtriterpene) consists of a group of metabolites with diverse structures that exhibit antiHIV,antimicrobial,antitumor,and anticancer activities [7].Ganodermic acid mainly interacts with membrane receptor tyrosine kinases (RTKs),including IR,IGFR-1,IGFR-2,VEGFR-1,VEFGR-2,and EGFR.Ganodermic acid can regulate the cell cycle to cause apoptosis by acting on NF-κB,RAS-MAPK,and PI3K/Akt/mTOR pathways [8].

Figure 9 Surface model of Erg-OPRM1 predicted interaction.

Figure 10 The ligand Ganolucidic acid E and its receptor TGF-β1.

Figure 11 The ligand Ganolucidic acid E and its receptor TGF-β1.

Figure 12 Beta-sitosterol and its receptor BAX.

Figure 13 Beta-sitosterol and its receptor BAX.

PPI network analysis found that CASP3,JUN,ADRB2,DRD1,OPRM1,TGF-β1,CASP8,and BAX are likely among the key targets forG.lucidumtreatment of lung cancer.Among these targets,CASP3,CASP8,and CASP9 are factors involved in programmed cell death.Abnormal expression of these factors can cause cancer cells to resist apoptosis.Stimuli such as cytotoxic stress and mitochondrial damage can induce granzyme B,caspase-8 and caspase-9 cleavage and activate caspase-3.Subsequently,activated caspase-3 effects protein hydrolysis required for programmed cell death [9].Abnormal expression ofCASP3andCASP9genes has previously been observed in lung cancer tissues [10].CASP3 has been shown to be related to lung cancer risk,cell metastasis,and lung metastasis [11],while CASP9 is related to the prognosis of lung cancer patients [12].Moreover,some studies have found that targeting the translation ofCASP8helps eliminate autophagy-deficient cancer cells via the CASP8-dependent apoptotic pathway [13].Caspase-8 protein is also an important regulator of cancer-related inflammation,as caspase-8 inhibitors can induce tumor proliferation or regress lung tumor development [14].TGF-β1 is a multifunctional cytokine that can promote the invasion and metastasis of lung cancer and induce epithelial-mesenchymal transition [15,16].Ganodermic acid can significantly reduce the expression of α-smooth muscle actin and vimentin,upregulate the expression of E-cadherin,and effect epithelial-mesenchymal transition by inhibiting TGFβ/Smad and MAPK (ERK,JNK,p38) signaling pathways [17].G.lucidumspore extract can inhibit the morphological changes induced by TGF-β1 and cadherin expression,and subsequently inhibit the formation of F-actin stress fibers.G.lucidumspore extract also inhibits TGF-β1-induced TFK-1 cell migration [18].Recent studies have found that increased levels of TGF-β1 can promote the expression of integrin β3 and cause acquired EGFR-TKI resistance in EGFRmutant lung cancer [19].To discover TGF-β1 inhibitors from natural compounds is now considered an important strategy to develop new therapeutic drugs[15].The relationship between OPRM1 andG.lucidumhas not been directly studied in recent years.The balance of apoptosis and survival mediated by BAX and Bcl-2 determines the fate of cells.G.lucidumpolysaccharides and peptides can reduce the expression of the Bcl-2 antiapoptotic protein in lung cancer cells and increase the expression of Bax to promote apoptosis and reduce the expression of VEGFR in angiogenesis[20].It has also been observed in colon cancer thatG.lucidumingredients promote cancer cell apoptosis by upregulating Bax [21].In normal cells,G.lucidumextract also decreases the ratio of Bax/Bcl-2 and inhibits caspase-3 activation,thereby inhibiting the apoptosis of splenic lymphocytes induced by high-fat diet [22].Moreover,GLP extracts can improve the activity of antioxidant enzymes by regulating the expression of Bcl-2 and the Bax/Bcl-2 ratio,thereby effectively reducing oxidative damage and protecting normal cells[23].

In this study,we identified 10 key signaling pathways,20 main biological processes,7 cellular components,and 8 molecular functions by biological enrichment analyses.GO function enrichment found thatG.lucidumis mainly involved in the regulation of lung cancer via membrane potential,blood circulation,and the cellular response to ammonium ion.KEGG pathway enrichment identified affected signaling pathways such as neuroactive ligand-receptor interaction,calcium signaling pathway,morphine addiction,apoptosis,pathways in cancer,cAMP signaling pathway,nicotine addiction,TNF signaling pathway,VEGF signaling pathway,and p53 signaling pathway.The neuroactive ligand-receptor interaction signaling pathway involves a collection of all receptors and ligands on the plasma membrane that are related to intracellular and extracellular signaling pathways.Data have shown that genes involved in the neural active ligand-receptor interaction pathway are highly related to lung cancer development,and exhibit multiple functions.Specifically,CHRNA3,CHRNA5,and CHRNB4 are subunits of nicotinic acetylcholine receptors,and are known to increase the risk of lung cancer [24].Changes in the expression of Ca2+channels and Ca2+binding proteins are characteristic in lung cancer tissues,and the regulation of such cell signaling pathways can lead to cell proliferation or apoptosis.Hypoxia induces the opening of Ca2+channels and increases Ca2+influx,leading to tumor growth [25].When treating non-small cell lung cancer with EGFR targeted therapy,restricting calcium influx can maintain RTK sensitivity,and the calcium signaling pathway can play a significant role in the adaptive resistance to targeted drugs.The combined targeted drug delivery for calcium signaling and RTK suppresses the development of drug resistance and improves the therapeutic effect[26].G.lucidumplays multiple roles in mediating signaling via calcium channels.Ganodermic acid can induce cell apoptosis by increasing the intracellular Ca2+concentration through NADPH oxidase and mitochondrial activation of JNK to generate ROS [27].G.lucidumpolysaccharide treatment can reduce ROS generation and calcium accumulation,preventing oxidative damage by modifying the redox system and maintaining calcium homeostasis [28].In addition,G.lucidumpolysaccharides also target calcium-releasing Fas,caspase-3,and caspase-8 protein expression to upregulate and activate the death receptor pathway of cancer cells [29].G.lucidumpolysaccharides increase intracellular Ca2+concentration and calcineurin activity,and increase the expression of p-ERK,p-JNK,and p-38,thereby promoting the proliferation of T lymphocytes and exerting immunomodulatory effects [30].Morphine is a clinically effective drug for cancer patients to control pain.Morphine has a direct analgesic effect on the central nervous system,but it also causes side effects,such as morphine addiction and immunosuppression.Morphine abuse in cancer patients has not been widely addressed;however,a number of studies in recent years have shown that morphine has certain effects on tumor growth,angiogenesis,metastasis,inflammation,and immune regulation,which suggests that morphine abuse may promote or inhibit tumor growth [30].In human non-small cell lung cancer transplantation models,the μ-opioid receptor(MOR) regulates a series of signaling pathways induced by EGF,and its overexpression promotes tumor growth and metastasis [31,32].Inhibition of MOR can block the binding of morphine.In addition,inhibiting MORs can also attenuate the growth factor-induced phosphorylation/activation of Src,gab-1,pi3k,Akt,and STAT3,thereby inhibiting proliferation,migration,and epithelial-mesenchymal transition in human lung cancer[33].

Molecular docking simulates macromolecular interaction,and breaks down the docking process into overall energy exchange.The interaction between ligands and macromolecules is driven by van der Waals,electrical interactions,hydrogen bonding,and desolventing component energies [34].By visualizing the docking results,the molecular state is more directly displayed,and the predicted binding site,binding relationship,and hydrogen bond distance can be observed.Finally,visualization of molecular docking simulations suggested that the main ingredients ofG.lucidumcould demonstrate good binding with TGF-β1,OPRM1,and BAX.

In summary,our study used a network pharmacologybased approach to systematically study the active ingredients ofG.lucidumand their corresponding targets.Enrichment analysis was performed focusing on the biological processes,cellular components,and molecular functions likely to be affected,supplemented by KEGG enrichment analysis.Finally,molecular docking and visualization were performed to predict a multi-ingredient,multi-target,multi-pathway pharmacological mechanism to collectively intervene against lung cancer.Our results provide a theoretical basis for further experimental research onG.lucidumand its mechanism of action.