Hang Gong,Yu-Cai Chen,Jia-Qi Xie,Yi-Hong Li,Li-Dan Cui,Hong-Tao Jin,Can Wang,2*

1School of Chinese Materia Medica/School of Traditional Chinese Medicine,Beijing University of Chinese Medicine,Beijing 100029,China.2State Key Laboratory of Bioactive Substance and Function of Natural Medicines,Institute of Materia Medica,Chinese Academy of Medical Sciences and Peking Union Medical College,Beijing 100050,China.3New Drug Safety Evaluation Center,Institute of Materia Medica,Chinese Academy of Medical Sciences and Peking Union Medical College,Beijing 100050,China.4Beijing Union-Genius Pharmaceutical Technology Development Co.Ltd.,Beijing 100176,China.

Abstract Background:Jinqi Jiangtang tablets(JQJT)have been approved for the treatment of type 2 diabetes mellitus (T2DM) in China formany years.Exploring the effective substances and mechanisms of JQJT is important for its clinical application and further drug research and development.This study aimed to explore the chemical basis and mechanisms of JQJT in the treatment of T2DM.Methods:With network pharmacology,we screened substances in JQJT and their possible targets,then constructed the action network and enriched the biological functions and pathways associated with the active components,and identified the potential targets and mechanisms of JQJT in the treatment of T2DM.Based on the network pharmacology data,we explored the hypoglycemic mechanisms of coptisine in JQJT through western blot and quantitative real-time polymerase chain reaction.Results:Forty-three compounds with good pharmacokinetic properties were identified in JQJT,together with 146 potential biological targets.Among these potential targets,74 were associated with treatment of T2DM.A compound-target network of the 43 compounds against T2DM was constructed.Biological process and signal pathway enrichment analysis of the network highlighted the FoxO signaling pathway.Western blot and quantitative real-time polymerase chain reaction results showed that coptisine,but not epiberberine,significantly inhibited expression of key genes involved in hepatocyte gluconeogenesis by regulating the FoxO1 signaling pathway.Conclusion:Network pharmacology analysis and cell experiments showed that coptisine regulated glucose homeostasis by inhibiting the FoxO1 signaling pathway and hepatic gluconeogenesis,which may be one of the mechanisms of JQJT in the treatment of T2DM.

Keywords: Jinqi Jiangtang tablets;FoxO1;gluconeogenesis;type 2 diabetes mellitus;network pharmacology

Background

In 2021,the International Diabetes Federation reported that 537 million adults were living with diabetes mellitus,accounting for 10%of the total worldwide population [1].This is expected to soar to 783 million by 2045 [1].In addition,diabetes mellitus can induce heart failure,cardiovascular disease,nephropathy,diabetic eye disease,diabetic ketoacidosis,nerve and/or vascular damage and other serious complications,which increases mortality and economic burden [2].Type 2 diabetes mellitus(T2DM),which accounts for the vast majority of cases,is a clinical syndrome caused by genetic and environmental factors that is characterized by glucose and lipid metabolism disorders and relative deficiency of insulin [3].Currently,the approved oral hypoglycemic drugs commonly used in clinics include sulfonylureas,thiazolidinediones,biguanides,glucagon-like peptide 1 analogs,dipeptidyl peptidase-4 inhibitors and α-glucosidase inhibitors.In clinical therapy,these single-target drugs,which effectively lower glucose,have raised concerns about side effects such as hypoglycemia,ketoacidosis,obesity and cardiovascular disease [4].The side effects of these oral hypoglycemic drugs have limited their clinical use to some extent [5],which has promoted researchers to explore new treatment strategies.Accordingly,traditional Chinese medicine(TCM)has attracted increased attention because of its good efficacy and few side effects.Chinese herbs can be beneficial in the prevention and treatment of diabetes through the additive effect of different components,the synergistic effect of multi-target activity and low toxicity.

In fact,according to the theory and practice of TCM,about 400 different Chinese herbal extracts or compounds have been shown to be useful for prevention and/or treatment of T2DM [6].Among them,Astragalus membranaceus(Fisch.) Bunge andCoptis chinensisFranch appear very frequently in these traditional prescriptions or Chinese herbal compounds [7].Jinqi Jiangtang tablets (JQJT) is a compound formula modified and optimized according to ‘QianJin Huanglian pills’,recorded in the TCM classic bookThousand Golden Prescriptions(Si-Miao Sun,652 C.E.) [8].It has been approved by the National Medical Products Administration as a prescription drug for the treatment of T2DM and diabetic complications [9,10].It is also the first Chinese herbal compound with a therapeutic effect on T2DM to be confirmed by randomized,double-blind,placebo-controlled clinical trials in China [9,11].Years of clinical use have verified the safety of oral JQJT.It is composed of three herbs:Coptis chinensisFranch.,Astragalus membranaceus(Fisch.) Bunge andLonicera japonicaThunb.Coptis chinensisFranch.plays an anti-inflammation role by blocking mRNA replication and transcription in inflammatory cells,and an anti-hyperglycemic role by promoting islet β-cell regeneration and insulin release,inhibiting liver glycogenesis and promoting glycolysis[12].Astragalus membranaceus(Fisch.) Bunge has been proven to lower blood glucose,improve insulin resistance,suppress the inflammatory response,and relieve pathological injuries caused by diabetic complications [13].Lonicera japonicaThunb.ameliorates hyperglycemia by decreasing the activities of α-amylase and α-glycosidase[14]and improves insulin resistance[15].The effects of JQJT are mediated by multiple components and multiple targets,but systematic elucidation of its pharmacological mechanism is difficult.Network pharmacology is based on a ‘disease-gene-target-drug’interaction network,combining systems biology,bioinformatics,multidirectional pharmacology and computational biology.The principle of network pharmacology coincides with the holistic,multi-component,multi-target and multi-pathway approach of TCM.Consequently,network pharmacology is recognized as a promising method to determine the intrinsic mechanisms of compound formulas.In this study,network pharmacology was used for preliminary screening of candidate compounds in JQJT and their potential mechanisms in the treatment of T2DM.Subsequently,cell experiments were used to verify the network pharmacology results.

Methods

Reagents and kits

According to thePharmacopoeia of the People’s Republic of China(2020 Edition),to prepare JQJT,50% ethanol extract ofCoptis chinensisFranch.(343 g),75% ethanol extract ofAstragalus membranaceus(Fisch.)Bunge(513 g)and aqueous extract ofLonicera japonicaThunb.(2058 g) are prepared,freeze-dried and mixed.Pregelatinized starch(33-87 g),microcrystalline cellulose (76 g) and cross-linked sodium carboxymethylcellulose (6.75 g) are then mixed with the powdered plant extracts and made into 1,000 tablets (Pharmacopoeia Commission of the People’s Republic of China,2020).Coptisine and epiberberine were obtained from Chengdu Herbpurify Co.,Ltd.(Chengdu,China).Rabbit anti-p-FoxO1(S256) (#9461),rabbit FoxO1(#2880) and rabbit anti-β-actin (#4970) antibodies were sourced from Cell Signaling Technology,Inc.(Danvers,MA,USA).Reagents required for cell total RNA extraction and quantitative real-time polymerase chain reaction were purchased from Qiagen (Hilden,Germany) or Applied Biosystems (Foster City,CA,USA).

Compounds present in JQJT

Traditional Chinese Medicine Systems Pharmacology (TCMSP) [16] is a database of systemic pharmacological data on Chinese herbal medicines.Compounds present in JQJT were identified by inputting“huangqi (Astragalus membranaceus(Fisch.) Bunge)”,“huanglian(Coptis chinensisFranch.)” and “jinyinhua (Lonicera japonicaThunb.)”as keywords into TCMSP and screening compounds for pharmacokinetic parameters,including oral bioavailability (OB),drug-likeness (DL) and Caco-2 permeability.OB is the degree of drug absorption into the circulation after an oral dose.DL refers to the similarity between a compound and known drugs.Drug-like molecules are generally considered high-quality lead compounds.To assess the suitability of compounds for oral administration,Caco-2 permeability is used to predict human intestinal permeability and to study drug efflux.Compounds with OB ≥30%,DL ≥0.18 and Caco-2 ≥-0.4 were regarded as significant.

Biological targets of compounds in JQJT

PharmMapper [17] is a platform used for target fishing by uploading known small molecule compounds.The compounds identified in JQJT were imported into PharmMapper to obtain their corresponding targets.Moreover,PharmMapper provides a Norm Fit score for each target.In this study,only the top 30 targets according to Norm Fit score were selected.The selected targets were entered into UniProt to identify the corresponding gene symbols,which were convenient for follow-up analysis,and duplicate targets were removed.

Biological targets associated with diabetes mellitus

We gathered diabetes mellitus targets from two databases.GeneCards[18] is an integrated bioinformatics database that provides detailed information on all annotated and predicted human genes.We used“diabetes mellitus”as a search term and targets with scores ＞30 were collected [19].GenCLiP3 [20],a net service platform based on document mining,can be used to obtain disease-related genes from PubMed abstracts and text.We used “diabetes mellitus” as the keyword to acquire targets.Duplicate target genes retrieved from the two databases were removed.

Network construction

We constructed three networks to visually analyze the therapeutic mechanism of JQJT.All of the networks were constructed using Cytoscape 3.8.0 [21],which is software that is frequently used in network pharmacology to visualize molecular interactions and biological pathways.Furthermore,the software provides corresponding data to assist in the analysis.Nodes represent biological molecules such as compounds and targets,edges show the interactions between biological molecules,and degree value indicates the number of edges between a node and other nodes.The greater the degree value is,the more significant the node is in the network.Compounds and compound targets of JQJT were first imported into Cytoscape 3.8.0 to build the compound-target network.Potential targets of JQJT against diabetes were then obtained by taking the intersection of compound targets and disease targets,followed by extraction of the compounds involved in the intersection targets as the bioactive components of JQJT against diabetes.Subsequently,the intersection targets were submitted to the STRING [22] database,which contains known information on protein-protein interactions (PPI).Free targets were removed and the PPI network of intersection targets was constructed for visualization and further analysis.The node with degree value greater than twice that of the median was taken as the hub target [23].

Enrichment analysis

To elucidate the roles of potential targets of JQJT against T2DM,we analyzed biological processes,molecular function,cell composition and pathways of the targets through Gene Ontology(GO)analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis with the Database for Annotation,Visualization,and Integrated Discovery(DAVID,https://david.ncifcrf.gov/home.jsp,version DAVID 2021)[24].The GO database is used to annotate the biological functions of genes and gene products [25],and the KEGG database is a knowledge base of gene function system analysis,which is used to research the relationships between genes and biological pathways[21].

Cell culture and drug treatment

HepG2 cells were cultured as described previously [26].In brief,HepG2 cells were cultured in Dulbecco’s modified Eagle medium with 10% fetal bovine serum and 1% penicillin and streptomycin.In formal experiments,cells were seeded on 6-well plates and then grown to 70%-80% confluence.Before treatment,the cells were starved in Dulbecco’s modified Eagle medium with 0.5% fetal bovine serum.

Western blot

Western blotting was performed as described previously [27].Briefly,10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis was used to separate the total proteins of HepG2 cells.The blots were transferred onto polyvinylidene difluoride membranes.After blocking,the levels of p-FoxO1 (Ser256),FoxO1 and β-actin were detected with specific primary antibodies,horse radish peroxidase-linked secondary antibodies and an enhanced chemiluminescence kit.After scanning and quantification,the levels of p-FoxO1(Ser256) were normalized to the level of FoxO1 as indicated.

RNA extraction and RT-PCR

RNA extraction and RT-PCR were performed as described previously[27].Briefly,after treatment,1 μg of total RNA from the HepG2 cells was reverse transcribed into cDNAs.The mRNA levels of glucose-6-phosphatase (G6Pase),phosphoenolpyruvate carboxykinase(PEPCK),and glyceraldehyde-3-phosphate dehydrogenase were analyzed using the cDNAs and specific primers (PEPCK forward primer:gctctgaggaggagaatgg,PEPCK reverse primer:tgctcttgggtgacgataac,glucose-6-phosphatase forward primer:gtgaattaccaagactcccag,glucose-6-phosphatase reverse primer:gcccatggcatggccagaggg,glyceraldehyde-3-phosphate dehydrogenase forward primer: agccacatcgctcagacac,glyceraldehyde-3-phosphate dehydrogenase reverse primer: gcccaatacgaccaaatcc).Quantitative RT-PCR was performed using an ABI Prism 7900 High-Throughput Real-Time PCR System (Applied Biosystems,Foster City,CA,USA).For the relative quantification of target genes,the comparative threshold cycle (CT)method was used with glyceraldehyde-3-phosphate dehydrogenase as an internal control.The mRNA expression levels of target genes were plotted as fold of the dimethyl sulfoxide group,which were designated as 1.

Statistical analysis.

For in vitro verification,the data were mean±SD of 3-4 independent repeated experiments and analyzed using GraphPad Prism 7(GraphPad Software Inc.,San Diego,CA,USA).After validation of the test for homogeneity of variance,differences among study groups were examined by one-way ANOVA followed by the Newman-Keuls test formultiple comparisons.P＜ 0.05 was considered to be statistically significant.

Results

JQJT compound-target network

Of the 43 compounds filtered from the TCMSP database,17 were fromLonicera japonicaThunb.,16 were fromAstragalus membranaceus(Fisch.) Bunge and 13 were fromCoptis chinensisFranch.Notably,all three herbs contained quercetin,and kaempferol was common to bothLonicera japonicaThunb.andAstragalus membranaceus(Fisch.) Bunge.The specific data are provided in Table 1.Using PharmMapper,146 targets of these compounds were obtained.In the compound-target network(Figure 1),there were 192 nodes(including 3 herbs nodes,43 compounds nodes,and 146 targets nodes)and 1,318 edges.The larger the node is,the larger the degree value is.

Compound-target network of JQJT against diabetes

2,287 diabetes targets were screened from GeneCards and the GenCLiP3 database.After intersection,74 targets related to 43 compounds were regarded as potential targets of JQJT against diabetes.In the compound-target network of JQJT against diabetes(Figure 2),a deeper green color and larger ellipse area indicate that the compound has a higher degree value and stronger activity.Compounds were ranked by degree value and the top 10 compounds are shown in Table 2.

PPI network of potential targets

After importing the targets into STRING database,three targets(CDK5R1,FAP,RORA) that did not interact with other targets were eliminated.As shown in Figure 3,the network was comprised of 71 nodes and 465 edges.The median degree value of the targets was 10.There were 14 targets considered to be hub targets larger than twice the median,their gene symbols were ALB,AKT1,EGFR,MAPK1,CASP3,MAPK8,SRC,ESR1,MAPK14,HSP90AA1,KDR,NOS3,MAM2 and ANXA5.Simultaneously,the darker the color is and the larger the area is,indicating that the degree value is larger.The detailed information on hub targets was presented in Table 3.

Table 1 Information of 43 compounds in JQJT

Table 2 The top 10 compounds ranked by degree value

Table 3 Information of hub targets

Figure 1 Compound-target network of JQJT. Rectangular nodes represent herbs,circular nodes represent compounds and diamond nodes represent biological targets.JQJT,Jinqi Jiangtang tablets;JYH,jinyinhua,Lonicera japonica Thunb.;HQ,huangqi,Astragalus membranaceus(Fisch.)Bunge;HL,huanglian, Coptis chinensis Franch.

Figure 2 Compound-target network of JQJT against diabetes.Yellow rectangular nodes represent intersection targets;green oval nodes represent compounds involved in intersection targets.JQJT,Jinqi Jiangtang tablets.

Figure 3 PPI network of potential targets.Red nodes represent hub targets.

GO and KEGG enrichment analysis

The GO analysis indicated that the targets were mostly enriched in negative regulation of apoptotic process (GO: 0043066),proteolysis(GO: 0006508),response to drug (GO: 0042493),protein phosphorylation (GO: 0006468) and serine phosphorylation (GO:0018105) biological processes.In respect of molecular function,the targets were enriched in ATP binding,enzyme binding,zinc ion binding,identical protein binding and serine-type endopeptidase activity.With regards to cellular components,enrichment results included cytosol,nucleus,extracellular region,extracellular exosome and extracellular space.The enrichment processes were related to signaling pathways that mainly included pathways in cancer(hsa05200),proteoglycans in cancer (hsa05205),MAPK signaling pathway (hsa04010),prostate cancer (hsa05215),FoxO signaling pathway (hsa04068),Rap1 signaling pathway (hsa04015),and Ras signaling pathway (hsa04014).The top 20 GO and KEGG enrichment analysis results are shown in Figure 4.

We searched and analyzed the relationship between these signaling pathways and T2DM,and found that the FoxO signaling pathway is closely related to T2DM because it regulates hepatic gluconeogenesis,as described in a previous review [28].We therefore speculated that the FoxO pathway was one of the mechanisms mediating the hypoglycemic effect of JQJT.The corresponding KEGG map is shown in Figure 5.

Figure 4 GO and KEGG enrichment analysis of the 74 potential targets of JQJT against diabetes.(A)The top 20 biological processes;(B)The top 20 molecular functions;(C)The top 20 cellular components;(D)The top 20 signal pathways.JQJT,Jinqi Jiangtang tablets;GO,Gene Ontology;KEGG,Kyoto Encyclopedia of Genes and Genomes.

Figure 5 Reprint of FoxO signaling pathway with permission from KEGG.KEGG,Kyoto Encyclopedia of Genes and Genomes.

Coptisine inhibits hepatic gluconeogenesis by regulating the FoxO1 signaling pathway

Coptisine (Figure 6A) but not epiberberine (Figure 6B),significantly increased phosphorylation of FoxO1 (Ser256) (Figure 6C),which is consistent with the results of network pharmacology.Quantitative results showed that 10 μM coptisine could increase the phosphorylation of FoxO1 (Ser256) to 2.5-fold that of the dimethyl sulfoxide group (Figure 6D),while different concentrations of epiberberine had no effect.Considering the role of FoxO1 in regulation of hepatic gluconeogenesis,G6P and PEPCK mRNA levels were quantified by quantitative real-time polymerase chain reaction.As shown in Figures 6E and 6F,coptisine inhibited the expression of G6P and PEPCK mRNA levels in a dose-dependent manner.

Discussion

TCMs contain a variety of components,often exhibiting cooperative pharmacological effects through multiple targets and multiple signaling pathways.Consequently,TCM may have unique advantages in the treatment of chronic diseases,especially chronic metabolic diseases.JQJT was derived from TCM and has been approved for the treatment of T2DM.In this study,network pharmacology was used to explore the potential pharmacological components and possible mechanisms of JQJT in the treatment of T2DM,which helped guide our subsequent biological verification.We have also carried out preliminary verification by cell experiments.Combining the network pharmacology and cell experiments,we report for the first time that coptisine significantly inhibited hepatocyte gluconeogenesis by regulating the FoxO1 signaling pathway,which may play an important role in the treatment of T2DM with JQJT.It will be of great significance to further explore whether coptisine could reduce fasting blood glucose by inhibiting liver gluconeogenesis.This is the beginning of our studies on the material basis and mechanism of JQJT in the treatment of T2DM.

In HepG2 cells,we found that different concentrations of coptisine significantly increased phosphorylation of serine residue 256 of FoxO1,and significantly down-regulated mRNA levels of G6P and PEPCK.In fact,some studies have shown that FoxO1 protein in the nucleus could bind to the promoter regions of G6P and PEPCK genes,which is necessary for their transcription.The increased phosphorylation of serine residue 256 of FoxO1 will increase its transport to the cytoplasm,thus inhibiting transcription of G6P and PEPCK genes,and ultimately inhibit cell gluconeogenesis [28].These studies could explain our results to some extent.

Figure 6 Coptisine inhibits hepatic gluconeogenesis by regulating the FoxO1 signaling pathway. (A,B) The chemical structures of coptisine (Cop) and epiberberine (Epi);(C) HepG2 cells were treated with different concentrations of Epi and Cop,followed by determination of the phosphorylation level of FoxO1 (Ser256);(D)After quantification,the expression levels of p-FoxO1 (S256) were normalized to those of FoxO1,and plotted as indicated;(E,F) After treatment,G6P and PEPCK mRNA levels were quantified.Data are mean ± SD of three independent experiments;statistically significant differences compared with dimethyl sulfoxide are indicated by＊P ＜0.05,＊＊P ＜0.01.PEPCK,phosphoenolpyruvate carboxykinase;DMSO,dimethyl sulfoxide;GAPDH,glyceraldehyde-3-phosphate dehydrogenase.

This study found that JQJT could inhibit gluconeogenesis of hepatocytes by regulating the FoxO1 signaling pathway.This may be one of the material bases and mechanisms of JQJT in the treatment of T2DM.However,another study has shown that coptisine significantly increased glucose consumption in peripheral tissues and improved glucose metabolism in type 2 diabetic mice,mediating a hypoglycemic effect [29].The complex mechanisms and signaling networks of coptisine in the treatment of T2DM by improving glucose metabolism remain to be further studied and elucidated.

In addition to coptisine,other chemical components of JQJT could also be absorbed into the blood and play hypoglycemic roles.A previous study showed that seven alkaloids (berberine,epiberberine,coptisine,magnoflorine,berberubine,palmatine and jatrorrhizine)and four phenolicacids (neochlorogenic acid,chlorogenic acid,cryptochlorogenic acid and ferulic acid) could be detected by liquid chromatography-tandem mass spectrometry in the serum of rats after oral administration of JQJT [30].Studying the pharmacological effects of these serum components is very significant to reveal the material basis of JQJT in treating T2DM.Among these other compounds,berberine has been the most studied.Researchers found that berberine could significantly reduce fasting blood glucose and improve insulin resistance at the cellular level,animal experimental level and in clinical research [31].Studies have shown that the hypoglycemic effect may be due to up-regulated expression level of insulin receptors and activation adenosine 5’-monophosphate-activated protein kinase [31].In addition,chlorogenic acid and isochlorogenic acid A in JQJT have also been shown to have potential hypoglycemic activity [32].Whether other components of JQJT have hypoglycemic activity still warrants further exploration.In follow-up studies of JQJT or other TCM,we will consider introducing serum pharmacochemistry into network pharmacology to explore synergistic mechanisms of different TCM components.

All in all,the synergistic pharmacological effects of JQJT in the treatment of T2DM involve multi-components and multi-targets.We are conducting related research on the synergistic pharmacological activities of two ormore components of JQJT.We believe that we will be able to present this part of the research results to readers in the future.

Conclusion

Through network pharmacology and cell experiments,we found that coptisine increased phosphorylation of FoxO1 (Ser256) and decreased mRNA levels of G6P and PEPCK,which are key genes involved in gluconeogenesis.This represents one of the potential mechanisms of JQJT in the treatment of T2DM.