Antagonistic Effect of Jiawei Shengjiang San on a Rat Model of Diabetic Nephropathy: Related to EGFR/MAPK3/1 Signaling Pathway

Abstract

We aimed to delve into the mechanisms underpinning Jiawei Shengjiang San's (JWSJS) action in treating diabetic nephropathy and deploying network pharmacology. Employing network pharmacology and molecular docking techniques, we predicted the active components and targets of JWSJS and constructed a meticulous "drug-component-target" network. Gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) enrichment analyses were utilized to discern the therapeutic pathways and targets of JWSJS. Autodock Vina 1.2.0 was deployed for molecular docking verification, and a 100-ns molecular dynamics simulation was conducted to affirm the docking results, followed by in vivo animal verification. The findings revealed that JWSJS shared 227 intersecting targets with diabetic nephropathy, constructing a protein-protein interaction network topology. KEGG enrichment analysis denoted that JWSJS mitigates diabetic nephropathy by modulating lipids and atherosclerosis, the PI3K-Akt signaling pathway, apoptosis, and the HIF-1 signaling pathway, with mitogen-activated protein kinase 1 (MAPK1), MAPK3, epidermal growth factor receptor (EGFR), and serine/threonine-protein kinase 1 (AKT1) identified as collective targets of multiple pathways. Molecular docking asserted that the core components of JWSJS (quercetin, palmitoleic acid, and luteolin) could stabilize conformation with three pivotal targets (MAPK1, MAPK3, and EGFR) through hydrogen bonding. In vivo examinations indicated notable augmentation in body weight and reductions in glycated serum protein (GSP), low-density lipoprotein cholesterol (LDL-C), uridine triphosphate (UTP), and fasting blood glucose (FBG) levels due to JWSJS. Electron microscopy coupled with hematoxylin and eosin (HE) and Periodic acid-Schiff (PAS) staining highlighted the potential of each treatment group in alleviating kidney damage to diverse extents, exhibiting varied declines in p-EGFR, p-MAPK3/1, and BAX, and increments in BCL-2 expression in the kidney tissues of the treated rats. Conclusively, these insights suggest that the protective efficacy of JWSJS on diabetic nephropathy might be associated with suppressing the activation of the EGFR/MAPK3/1 signaling pathway and alleviating renal cell apoptosis.

Introduction

Diabetes mellitus (DM) is a chronic disease that affects multiple systems and can cause various complications due to continuous hyperglycemia, such as diabetic nephropathy (DN), retinopathy, and neuropathy¹. DN is a serious complication of DM, accounting for about 30%-50% of end-stage renal disease (ESRD)². Its clinical manifestation is microalbuminuria, which can progress to ESRD characterized by increased glomerular volume, mesangial stromal hyperplasia, and thickened glomerular basement membrane³. The pathogenesis of DN is complex and has not been fully elucidated. Clinical methods such as lowering blood glucose, regulating blood pressure, and reducing proteinuria are mostly used to delay its progress, but the effect is general.

Currently, no specific drug has been found to treat DN⁴. For centuries, however, Chinese herbal medicines have been widely used in treating DM and its complications⁵ and have improved patients' clinical symptoms and delayed disease progression. Due to the advantages of multi-component, multi-target, and multi-pathway effects, Chinese herbal medicines are expected to be an innovative drug source for the treatment of DN⁶.

"Shengjiang san" originated from the "Wanbing Huichun" by the Ming Dynasty medical doctor Gong Tingxian. The book "Neifu Xianfang" describes the use of Bombyx Batryticatus, Cicadae Periostracum, Curcumaelongae Rhizoma, and Radix Rhei et Rhizome. Based on this, after adding Hedysarum Multijugum Maxim, Epimrdii Herba, and Smilacis Glabrae Rhixoma, it exerts the function of shengjiang san of increasing lucidity, decreasing turbidity, releasing stagnant "heat," and harmonizing "qi" and the blood^7,8. It also increases the effect of strengthening the spleen and tonifying the kidneys. Its efficacy is consistent with the pathogenesis of DN's "qi" to rise and fall out of order due to deficiency of "vital energy," excessive dryness and "heat," and stagnation of "heat" caused by a triple energizer^7,8.

Previous clinical studies have shown Chinese herbal medicines have been used to treat DM and its complications, and jiawei shengjiang san (JWSJS) has been shown to regulate blood glucose and lipids, reduce proteinuria, and significantly improve the clinical efficacy of patients with early DN⁷. The ability of JWSJS to reduce urinary protein and blood glucose levels in DN rats has been confirmed by previous studies. This probably happens by inhibiting the TXNIP/NLRP3 and RIP1/RIP3/MLKL signaling pathways, reducing podocyte pyroptosis, and preventing necrotic apoptosis in renal tissues of DN rats, thus achieving renal protection⁹. JWSJS can upregulate nephrin and podocin protein expression and reduce podocyte injury in DN rats, thus suggesting that JWSJS has an inhibitory effect on podocyte injury. JWSJS has a certain anti-DN effect with good safety profiles, but there is little research on it, and this work mostly focuses on pyroptosis and necrotic apoptosis. The literature is not sufficiently deep or systematic¹⁰. Our previous findings have confirmed that JWSJS can reduce proteinuria and alleviate kidney damage in DN rats⁷. However, there are only a few studies on the mechanism of JWSJS for DN treatment, and these lack depth and systematization. Thus, this study aims to analyze the molecular substances and mechanisms of action of JWSJS for DN treatment using network pharmacology and provide a solid foundation for future research.

Network pharmacology is an emerging method to study the mechanism of drug action, including cheminformatics, network biology, bioinformatics, and pharmacology^11,12. Network pharmacology research design is quite similar to the holistic concept of traditional Chinese medicine^13,14, and it is an important method to study the mechanism of Chinese herbal medicines. Molecular docking can study interactions between molecules and predict their binding patterns and affinity. Molecular docking has emerged as a critical technique in the field of computer-aided drug research¹⁵. Therefore, this study constructed a JWSJS-DN-target interaction network through network pharmacology and molecular docking methods that offers a reliable and theoretical basis for further exploration of DN treatment with JWSJS.

Protocol

All animals were maintained and used in accordance with the US National Research Council Guide for the Care and Use of Laboratory Animals, 8^th Edition, and were reported as recommended in the ARRIVE guidelines^16,17. The study was conducted in accordance with the China National Research Council Guide for the Care and Use of Laboratory Animals and was approved by the Animal Ethics Committee of Hebei University of Chinese Medicine (DWLL2019030).

1. JWSJS active ingredients and target collection

1. Enter the medicinal composition JWSJS (Hedysarum Multijugum Maxim, Epimrdii Herba, Smilacis Glabrae Rhixoma, Radix Rhei et Rhizome, and Curcumaelongae Rhizoma)into the traditional Chinese medicine systems pharmacology database and analysis platform (TCMSP) database¹⁸ to retrieve all chemical components.  According to absorption, distribution, metabolism, and excretion (ADME) screen oral bioavailability (OB) ≥ 30% and drug-like properties (DL) ≥ 0.18 chemical ingredients^19,20.
2. Use the TCMSP database (https://old.tcmsp-e.com/ tcmsp.php, Version. 2.3) to retrieve the corresponding targets of chemical ingredients.
3. Obtain the active ingredients of Bombyx Batryticatus and Cicadae Periostracum by searching the Traditional Chinese Medicine and Chemical Composition databases²¹.
4. For experimental credibility, select compounds with chemical abstracts service (CAS) numbers for this study. Then download 2D structure diagrams (.sdf format) of active ingredients from PubChem (https://pubchem.ncbi.nlm.nih.gov/, updated in 2021)²².
5. Use SwissADME (www.swissadme.ch, updated in 2021)²³to screen the components with high gastrointestinal (GI) absorption as drug similarity (Lipinski, Ghose, Veber, Egan, Muegge) whose ≥ 2 items were 'Yes'. This led to the active components of Bombyx Batryticatus and Cicadae Periostracum.
6. Import these into the TCMSP database for target protein prediction (https://old.tcmsp-e.com/tcmsp.php, Version. 2.3)¹⁸.
7. Standardize the targets in the UniProt database (https://www.uniprot.org, updated in 2021) with the status set as Reviewed and species set as Human²⁴.

2. DN corresponding target collection

1. Search Diabetic nephropathy through GeneCards (https://www.genecards.org/, Version. 5.1)²⁵, OMIM (https://omim.org/, updated in 2021)²⁶, TTD (http://db.idrblab.net/ttd/, updated in 2021)²⁷, PharmGKB (https://www.pharmgkb.org/, updated in 2021)²⁸,and DrugBank databases (https://www.drugbank.ca/, updated in 2021)​²⁹.Obtain the targets of DN after combining and de-duplicating.
2. Screening of common targets of JWSJS and DN and network construction
   1. Screen the common targets of JWSJS and DN using R 4.2.0 software and draw Venn diagrams. Import the active ingredients and potential targets of JWSJS into Cytoscape 3.8.0 software³⁰ to build a Drug-Ingredient-Target network diagram that visualizes the connection between drugs, ingredients, targets, and diseases.
   2. Let the size of the node reflect the size of the degree value, where a higher degree value indicates that the node is more important in the network.
3. Construction of PPI protein-protein interaction network and screening of core targets
   1. Analyze the intersecting genes using the online platform STRING (Version:11.5) to construct a protein-protein interaction (PPI) network³¹. Construct the network using a Multiple Protein analysis mode, set the species to Homo sapiens, and set the minimum required interaction score to >0.9³⁰.
   2. Use Cytoscape 3.8.0 to analyze the network topologically, calculate the betweenness centrality (BC), closeness centrality (CC), degree centrality (DC), Eigenvector centrality (EC), local average connectivity-based method (LAC), and network centrality (NC) values for each node, and screen out the six nodes with all values greater than the median³². Repeat this process several times to identify the core targets of JWSJS for DN therapy.
4. GO functional analyses and KEGG pathway enrichment analysis
   1. Perform GO and KEGG enrichment analyses to identify the biological processes, molecular functions, cellular components, and pathways associated with the common targets of JWSJS and DN.
   2. Use the org.Hs.eg.db package to obtain the IDs of the intersection targets, and use the clusterProfiler, org.Hs.eg.db, enrichplot, and ggplot2 packages for enrichment analyses³³.
   3. Screen for functional GO enrichment analysis on the top 10 biological hits with a corrected P-value < 0.05, and select top-30 pathways with the highest enrichment for KEGG analysis.

3. Molecular docking

1. Use AutoDock Vina software to perform molecular docking between the JWSJS core components and the core targets for DN therapy³⁴.
2. Search in the PubChem database to obtain the sdf file of the 2D structure of JWSJS core components. Use ChemBio 3D Ultra14.0 software to generate and optimize its 3D structure, save it in mol2 format, and use it as a ligand file.
3. Find and download the pdb format of the 3D structure of the core targets from the Research Collaboratory for Structural Bioinformatics (RCSB) Protein Data Bank (PDB) database. Use Pymol 2.4.0 software to remove water molecules and ligands from the protein structure, save it as pdb format, and use it as a receptor file.
4. Import the receptor pdb file into AutoDockTools 1.5.7 software for hydrogenation, convert both receptor protein and small molecule ligand into pdbqt format, set active pocket for receptor protein with Spacing coefficient set at 1.
5. Use Autodock vina 1.2.0 for molecular docking and calculate binding energy; if Affinity < 0, consider that molecule spontaneously binds with protein, greater absolute value of Affinity indicates more stable binding between them. Finally, visualize docking results using Pymol 2.3.0 and LigPlot 2.2.5 software.

4. Molecular dynamic simulation

1. To perform MD simulations using the Desmond software, follow these steps:
   1. Employ the 2019-1 academic version of Desmond software for assessing the stability and flexibility of protein-ligand interactions. 
   2. Conduct molecular dynamic (MD) investigations on a trio of the most promising ligand complexes derived from docking studies. Conduct simulations in a SPC aqueous environment with 0.15 M NaCl, replicating the physiological ionic conditions. The simulation box is designed to ensure that the solute remains at least 10 Å away from its boundaries. Counterions are introduced to neutralize the system's electrical charge. Additionally, periodic boundary conditions are applied, governed by the parameters of the OPLS 2005 force field.
   3. Initiate an energy minimization phase for 100 ps using Desmond default parameters.
   4. Ensure temperature and pressure stabilization at 26.85 °C (300 K) and 1.01325 bar through the Nosé-Hoover chain and Martyna-Tobias-Klein methodologies across all production MD systems. The molecular dynamics simulation is executed with a timestep of 2 fs for a total duration of 100 ns, recording the atomic coordinates every 100 ps.
   5. Open the simulation interactions diagram (SID) module. Load the simulation result data files into the module.
   6. Select the desired analysis type from the options available in the module (e.g., root-mean-square deviation [RMSD], root mean square fluctuation [RMSF], hydrogen bonding analysis, etc.). Specify any additional parameters required for the chosen analysis type.
   7. Run the analysis by clicking on the Analyze or Start button. 
   8. Once the analysis is complete, view and interpret the results within the SID module interface. Export results if needed for further use or presentation.

5. Animal experiment validation

1. Animals and experimental design
   NOTE: This study involved 75 male Sprague-Dawley rats, which were obtained from an SPF-grade animal production facility and were 8 weeks old with a body mass of 150 g ± 20 g (animal production certificate number SCXK [Hebei] 2018-004).
   1. House the rats in an SPF-level animal laboratory and randomly divide them into normal and model groups after 1 week of adaptive feeding. Provide the normal group with a normal diet and the model group with a high-sugar and high-fat diet. After 4 weeks, fast the rats but do not deprive them of water for 12 h.
   2. Inject the model group with an intraperitoneal injection of Streptozotocin (STZ; 35 mg·kg^-1). After 72 h, test the blood sugar levels through tail vein blood sampling. If the fasting blood glucose level is ≥16.7 mmol·L^-1, confirm it as a successful diabetic model.
   3. Confirm the successful DN model by observing histopathological changes in the rat kidney.
   4. Randomly divide the successfully replicated models into the model group, which received JWSJS at 4.37 g/kg, 8.73 g/kg, and 17.46 g/kg (equivalent to 3.2, 6.3, and 12.6 times the clinical equivalent dose) as well as an irbesartan group (0.014 g/kg). Each group consisted of 10 rats. Administer all drugs via oral gavage once daily. Maintain this dosing regimen consistently for 4 weeks.
   5. Anesthetize the rats using 1% pentobarbital sodium and confirm proper anesthesia by loss of pedal withdrawal reflex. Collect blood samples from the abdominal aorta prior to euthanasia.
   6. Collect the kidneys after euthanizing the ratsusing sodium pentobarbital at a dose of 150 mg/kg.
      NOTE: The animals were euthanized humanely, following the most updated American Veterinary Medical Association (AVMA) guidelines (https://www.avma.org/resources-tools/avma-policies/avma-guidelines-euthanasia-animals) for euthanasia.
2. Renal histology analyses
   NOTE: Kidney tissues were fixed in 4% paraformaldehyde and dehydrated in ethanol after 48 h; sections were embedded in paraffin. The sections were cut into thin (4-µm) slices for hematoxylin and eosin (HE) and Periodic acid-Schiff (PAS) staining, and the morphological changes of kidney histology were observed under a light microscope.
   1. HE staining:
      1. Dewaxing and hydration: Treat the tissue sections with Xylene I and II (10 min each), absolute alcohol I and II (3 min each). Then, immerse the tissue sections in 95%, 90%, 80%, and 70% ethanol (2 min each), followed by a distilled water wash for 5 min.
      2. Place the tissue sections in hematoxylin stain for 5 min and then differentiate the sections with hydrochloric alcohol for 5 s.
      3. Place the section in the eosin staining solution for 3 min.
      4. Dehydration, clearing, and mounting: Dehydrate the sections at varied times in different ethanols (70%,80%,90%,95%, and absolute ethanols), then clear them in xylene I and II, followed by mounting with neutral gum.
   2. PAS Staining:
      1. Follow the dewaxing steps described in step 5.2.1 and treat the section with the periodic acid staining solution for ~8 min.
      2. Rinse the sections in distilled water for 2 min and then stain them with the Schiff Reagent for 15 min in the dark.
      3. Wash the sections in tap water for 10 min after treating them with the Schiff reagent.
      4. Counterstain the sections using hematoxylin for 1 min and then differentiate them with HCl-alcohol for 30 s. Wash the sections again in tap water for 5 min to color the nucleus blue.
   3. Electron microscopy:
      1. Fix the samples in 2.5% glutaraldehyde for 3 h, then rinse them in PBS for 3 h. Further, treat the samples in 1% osmic acid for 3 h and then rinse again in PBS for 3 h.
      2. Dehydrate through a graduated series of alcohol baths ending in acetone (each step lasts about 2 h total).
      3. Infiltrate the samples with a mixture of epoxy propane and epoxy resin (1:1) at room temperature (RT) for 2 h, followed by pure epoxy resin at 37 °C for an additional 2 h.
      4. After this, cure and harden the samples at increasingly high temperatures over a period of 36 h before sectioning. This process ensures the tissue is thoroughly infiltrated with the resin, which is then hardened to allow thin slicing for electron microscopy.
      5. Double stain with 3% uranyl acetate and lead acid, and observe and photograph the samples by electron microscopy within 15 min.
         NOTE: The transmission electron microscopy (TEM) settings are as follows: a magnification of 7000x in high-contrast mode (Zoom-1 HC-1), with an accelerating voltage of 80.0 kV on a TEM system.
3. Immunohistochemistry
   1. Dewaxing: Place the tissue sections in xylene I and II for 10 min each.
   2. Rehydration: Rehydrate the dewaxed tissue sections by placing them successively in absolute ethanol I and II for 3 min each, followed by 95%, 90%, 80%, and 70% ethanol for 2 min each.
   3. Antigen retrieval: Immerse the tissue sections in 0.1 M citrate-citric acid buffer solution under high pressure for about 5 min and then cool to RT.
   4. Blocking: To prevent non-specific binding of the antibodies, block by incubating the tissue sections with 5% normal goat serum at approximately 37 °C for 30 min.
   5. Primary antibody incubation: Add the primary antibodies p-EGFR (diluted 1:200) and p-MAPK3/1 (diluted 1:200) onto tissue sections and incubate overnight at 4 °C inside a humidified chamber.
   6. Secondary antibody incubation: Wash the sections three times with PBS, then add biotin-labeled secondary antibodies (diluted in PBS as per the manufacturer's instructions) onto the sections and incubate at 37 °C for 30 min.
   7. Apply horseradish peroxidase-conjugated streptavidin working solution onto the tissue sections at 37 °C followed by washing thrice using phosphate-buffered saline (PBS).
   8. For DAB development, incubate the sections with DAB developing solution (3 min in the dark at RT), followed by rinsing with distilled water. DAB acts as a chromogen substrate for the enzyme horseradish peroxidase (HRP), which turns brown upon oxidation.
   9. Use hematoxylin for counterstaining for about 2 min, differentiate the sections with 1% hydrochloric alcohol for 5 s, then wash under running water to turn blue.
   10. Dehydrate by successively placing the sections in a series of ethanol from low to high concentration, clear in xylene I and II, and mount the sections with neutral gum.
       NOTE: This procedure takes approximately 18-24 h, including overnight primary antibody incubation.
   11. Randomly select three visual fields (200x) for each section and analyze the staining results of sections using image analysis software. Image Pro Plus 6.0 software was used following the steps below.
   12. First, import an image of the section into the software.
   13. Density correction: To ensure accurate results, perform a density correction.
       1. Navigate to Measure > Calibration > Intensity > New > Std. Optional Density > Options > Image. Then, click on a blank area of the image to record the background value and confirm it by clicking OK.
   14. Setting measurement parameters: Set up measurement parameters to calculate area and integrated optical density (IOD).
       1. Click on Measure > Count/Size > Measure > Select Measurements. Select Area(100) and IOD, then click OK.
       2. Go to Options, check off Dark Background on Sample, 4-Connect Pre-Filter and click on OK.
   15. Color selection: To accurately determine positivity, select colors to detect stained areas versus unstained areas.
       1. Click on Select Colors > Histogram Based > HSI. Adjust the H value according to the picture while keeping S unchanged and I ranging between zero to the background value. Then, confirm by clicking on Close.
       2. Data collection and calculation: Perform data collection and calculation by clicking on Measure > Data Collector > Layout, where Name, Area(Sum), and IOD(Sum) are selected. Then, click on Count > Collect Now > Data List to export the data for further analysis or storage.
          NOTE: This resulted in a semi-quantitative measurement of protein expression in each section in terms of IOD/Area, which indicates how much protein is present relative to the total area analyzed.
4. Western blotting
   1. Obtain the kidney tissue and cut it into pieces. Place these pieces into a 1.5 mL centrifuge tube, add the pre-cooled protein lysis solution (50 mM Tris [pH 7.4],150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate [SDS]), and homogenize.
   2. Leave it on ice for 30 min, then centrifuge at 13400 x g for 20 min at 4 °C. Store the resulting sample in a -80 °C freezer.
   3. Use the Bradford method to quantify the proteins.
      1. Prepare the Coomassie Brilliant Blue working solution by mixing reagent and distilled water at a ratio of 1:4.
      2. Establish three groups - blank, standard protein group, and sample protein group. Add 3 mL of Coomassie Brilliant Blue working solution to each group, along with physiological saline for the blank group, standard protein for the standard group, and the extracted supernatant protein for the sample group.
      3. After letting the mixtures stand for 5 min, measure their absorbance values using an ultraviolet spectrophotometer.
      4. Calculate the sample concentration using this formula: Protein concentration (mg/mL) = (Sample absorbance value/Standard tube absorbance value) x Standard protein concentration x 5.
   4. Add an equal volume of 6x loading buffer to each protein sample. Boil at 100 °Cfor 5 min before aliquoting and storing them in a -20 °Cfreezer until further use.
   5. Perform standard electrophoresis, blotting, and blocking.
      1. Perform standard SDS-polyacrylamide gel electrophoresis (PAGE) using a 12% resolving gel. Apply a constant voltage of 100 V for the stacking gel and 120 V for the resolving gel until the dye front reaches the bottom of the gel.
      2. Transfer proteins onto polyvinylidene difluoride (PVDF) membranes at 100 V for 2 h in a cold room at 4 °C using a wet transfer system.
      3. After protein transfer, block membranes with 5% non-fat milk in TBST for 1 h at RT.
   6. Dilute primary antibodies EGFR (1:2,000), p-EGFR (1:2,000), MAPK3/1 (1:2,000), p-MAPK3/1 (1:2,000), Bcl-2 (1:2,000), BAX (1:2,000), and CAPDH (1:5,000) in TBST with 5% BSA. Add these diluted primary antibodies and incubate the membranes overnight at 4 °C with gentle agitation.
   7. After washing the membrane three times with TBST, add diluted secondary antibodies (1:5,000), followed by incubation for 1 h. Post color development, perform quantitative analysis of the target bands using ImageJ software.
5. qPCR analysis
   1. Grouping: Assign samples to groups - Control, DN, Irbesartan, JWSJS-L, JWSJS-M, JWSJS-H.
   2. RNA extraction: Perform tissue homogenization and RNA extraction, followed by chloroform addition, centrifugation, and RNA precipitation per the manufacturer's instructions.
   3. RNA precipitation and washing: Precipitate RNA with isopropanol and wash with ethanol.
   4. RNA Solubilization: Dry the RNA pellet and dissolve it in DEPC-treated water.
   5. cDNA Synthesis: Perform reverse transcription using specific reaction mixtures per the manufacturer's instructions.
   6. qPCR Amplification: Use specific primers for GAPDH, MAPK3, MAPK1, and EGFR for qPCR and perform amplification per the manufacturer's instructions.
      Primer sequences were as follows:
      GAPDH: F-5'-GCAAGTTCAACGGCACAG-3', R-5'-CTCGCTCCTGGAAGATGG-3';
      MAPK3: F-5'-AGATCTGTGATTTTGGCCT-3', R-5'-TCAATGGATTTGGTGTAGC-3';
      MAPK1: F-5'-CCTCAAGCCTTCCAACCTC-3', R-5'-GCCCACAGACCAAATATCA-3';
      EGFR: F-5'-GGGGATGTGATTATTTCTG-3, R-5'-ATTTTGGTCTTTTGATTGG-3'.

6. Statistical methods

1. Perform the analysis using appropriate software (e.g., SPSS) and represent the measured data as the mean ± standard deviation(s). If the data met the criteria of normal distribution and variance homogeneity, compare between groups using one-way ANOVA and carry out multiple comparisons with the Bonferroni method.
2. Use the T2 test for multiple comparisons if the variance is not homogeneous. Consider P < 0.05 as statistically significant.

Representative Results

Following the protocol, 90 active ingredients of JWSJS were finally obtained from the analysis after screening and deduplication according to the set standards of OB and DL. These included 20 kinds of Hedysarum Multijugum Maxim, 23 kinds of Epimrdii Herba, 15 kinds of Smilacis Glabrae Rhixoma, 16 kinds of Radix Rhei et Rhizome, four kinds of Curcumaelongae Rhizoma, 15 kinds of Cicadae Periostracum, and six kinds of Bombyx Batryticatus components. Because there are many active ingredients in JWSJS, only 20 are listed (see Table 1). After deduplication, 396 targets corresponding to the active ingredients of JWSJS were finally obtained.

After searching for disease-related genes, 3035 DN-related target genes were finally obtained after merging and deduplication. The corresponding targets of DN were compared and analyzed with the targets of drugs' active ingredients. The intersection was determined, which resulted in 227 intersection targets of JWSJS and DN. A Wayne diagram was then drawn (see Figure 1A). The "drug-ingredient-target" network diagram was built using the Cytoscape 3.8.0 software (see Figure 1B), and topological analysis was performed. Each node was ranked from high to low according to the degree value. The top five drug components were quercetin, palmitoleic acid, luteolin, kaempferol, and naringenin, and the top five targets were PTGS2, PTGS1, NCoA2, AR, and ESR1.

Next, a PPI network for JWSJS and DN intersection targets was developed (196 nodes and 1714 edges with a mean degree value of 17.4). Moreover, seven key targets that may be the key targets of JWSJS for DN treatment were obtained after three screenings: MAPK14, TP53, MAPK3, MYC, HIF1A, ESR1, and MAPK1 (see Figure 1C).

Then, GO enrichment analysis was performed to speculate about the biological characteristics of the target of JWSJS in treating DN. Among BP, CC, and MF, the top ten enriched entries are plotted in Figure 1D. The results show that JWSJS could regulate cell proliferation, differentiation, and apoptosis by binding with transcription factors and protein kinases in membrane rafts and membrane microdomains, thus resulting in a therapeutic effect on DN. KEGG pathway enrichment analysis was performed on the participating targets to explore the potential pathways of action of JWSJS in DN. The top thirty enriched pathways are plotted in Figure 1E. The results show that JWSJS treatment of DN might act on 181 signaling pathways, mainly involving lipid and atherosclerosis, the PI3K-Akt signaling pathway, chemical carcinogenesis, hepatitis, prostate cancer, apoptosis, the HIF-1 signaling pathway, and the TNF signaling pathway.

To further explore the targets of the first 30 KEGG-enriched pathways, we used Cytoscape 3.8.0 software to construct a KEGG signaling pathway relationship network map for JWSJS-treated DN. MAPK1, MAPK3, EGFR, and AKT1 were found to be targets of multiple pathways acting together (see Figure 1F), suggesting that these proteins may play a central role in mediating the effects of JWSJS on DN.

To further validate the potential targets of JWSJS in treating DN, we next screened three key targets, MAPK1, MAPK3, and EGFR, with high degree values in the PPI network and the KEGG signaling pathway network. These were molecularly docked with quercetin, palmitoleic acid, and luteolin-the top three cores' components of JWSJS for DN treatment. The conditions for screening PDB ID are as follows: (1) Be derived from humans; (2) Conformational resolution ≤ 2.5Å, i.e., as small as possible; (3) Conformational sequence to be as complete as possible, i.e., the structural complex must have small molecule ligand information; and (4) Crystalline pH value should be as close as possible to the normal physiological range of the human body³⁵. Finally, the PDB IDs of MAPK1, MAPK3, and EGFR were 4QTA, 4QTB, and 7JXQ.

To verify the reliability of the AutoDock software to the docking system in this study³⁶, we first redocked the three proteins of MAPK1, Mapk3, and EGFR. The ligand conformation in the original crystal structure of the three target proteins was superposed with the ligands after docking (Figure 2A). The root-mean-square deviation (RMSD) values between the ligand conformation post-docking and the original crystal structure were then calculated. The RMSD values were 1.129, 1.201, and 1.877 Å, respectively, all ≤ 2Å. This showed that the docking method was reliable and could be used for the next molecular docking verification. To compare and declare the inhibitory potential of core components, the original ligands of MAPK1, MAPK3, and EGFR were then used as controls. The docking analysis successfully predicted the binding energy between luteolin, palmitoleic acid, and quercetin and the three core targets. These were all negative and less than -6³⁷; see Table 2. Of note, the molecular docking between luteolin and MAPK1 has the highest cavity size and the lowest binding energy. Overall, molecular docking results showed that luteolin, palmitoleic acid, and quercetin had good binding activity with three core targets (Figure 2B,C).

The RMSD is utilized to quantify the average displacement changes of a selected group of atoms in relation to a reference frame. Meanwhile, the RMSF effectively characterizes localized alterations along the protein chain. The RMSD plot (Figure 2D) shows that the RMSD values of the compound luteolin-MAPK1 complex and quercetin-MAPK3 complex stabilized within 5 ns. The luteolin-EGFR complex showed a fluctuating RMSD ranging from 0-35 ns; this value stabilized at 35 ns until the end of the 100 ns MD simulation. All complexes stably interacted at the end of the 100 ns MD simulation. The RMSF value was assessed from the MD simulation trajectory: Atoms in the active site and the main chain that fluctuated minimally implied a minimal conformational change, suggesting that the lead compound mentioned is securely anchored within the cavity of the target protein's binding pocket. The RMSF results showcased limited fluctuations in the complex structures.

Finally, we verify the above results in vivo. We first observed that the weight of rats in the model group decreased significantly, and GSP, LDL-C, UTP, and FBG increased significantly (P < 0.05). The body weight of rats increased significantly in the irbesartan group and each JWSJS group versus the model group. GSP, LDL-C, UTP, and FBG decreased to different degrees (P < 0.05; see Table 3).

Then, HE staining showed that the glomeruli in the normal group had a normal size. They were regular and clear in structure and had a smooth basement membrane, neatly arranged renal tubules, and no inflammatory cell infiltration. In the model group, the glomeruli were hypertrophic, the basement membrane was thickened, and the mesangial matrix was increased. The interstitium was infiltrated with inflammatory cells. The pathological manifestations of each administration group were alleviated to different extents compared to the model group (see Figure 3A).

PAS staining showed that the model group had increased glomerular volumes, thickened basement membrane, and increased mesangial matrix versus the normal group. These pathological manifestations were significantly alleviated by each administration group (see Figure 3B). Using transmission electron microscopy, we found that the glomerular basement membrane in the model group was thickened versus the normal group. The mesangial matrix grew, and the podocyte foot processes were extensively fused. Microscopic manifestations of rats in each administration group were relieved to varying extents versus the model group (see Figure 3C).

Finally, the expressions of total EGFR, p-EGFR, MAPK3/1, and p-MAPK3/1 were detected to see whether they changed in JWSJS-treated DN rats. There was no significant change in total EGFR and MAPK3/1 expression in the model groups versus the normal group. The expression of p-EGFR, p-MAPK3/1, and BAX increased significantly, and the BCL-2 expression decreased significantly. Versus the model group, the expression of p-EGFR, p-MAPK3/1, and BAX were significantly reduced, and BCL-2 expression was significantly upregulated to varying degrees in rat kidney tissues in each dosing group (Figure 3D,E). This result was confirmed by immunohistochemistry (Figure 3F,G). This study showed that JWSJS could protect the kidneys in DN rats by regulating the expression of p-EGFR, p-MAPK3/1, BAX, and BCL-2 proteins and reducing cell apoptosis. In addition, qPCR was utilized to assess the expression levels of MAPK3, MAPK1, and EGFR mRNA across various groups. The results revealed that the relative expression in the DN group was significantly higher compared to the control group, as well as the Irbesartan, JWSJS-L, JWSJS-M, and JWSJS-H groups. This finding further corroborates the initial results (Figure 3H).

Figure 1: Network Pharmacology clarified the JWSJS potential targets and signaling pathways in diabetic nephropathy. (A) Venn diagram of the intersecting targets of JWSJS and diabetic nephropathy. Green represents the number of therapeutic targets of JWSJS components, and purple represents the number of relevant targets of DN. The intersecting part represents the number of intersecting targets (i.e., the number of targets of JWSJS for DN). (B) "Drug-ingredient-target" network diagram of JWSJS in the treatment of DN. The light blue represents the relevant targets of JWSJS for DN treatment. Fuchsia represents the active ingredient of Bombyx Batryticatus. Green represents the active ingredients of Hedysarum Multijugum Maxim. Dark blue represents the active ingredients of Radix Rhei et Rhizome. Red represents the active ingredients of Cicadae Periostracum. Purple represents the active ingredients of Epimrdii Herba. Yellow represents the active ingredients of Smilacis Glabrae Rhixoma. Sky blue represents the active ingredients of Curcumaelongae Rhizoma. (C) PPI network of JWSJS and DN intersection targets. (D) The 10 most significant genes identified via ontology analysis of therapy target genes implicated in JWSJS treatment of DN. (E) The 30 most significant genes identified via pathway analysis of therapy target genes implicated in JWSJS treatment of DN. (F) The relationship network of the KEGG signaling pathway in the treatment of DN by JWSJS. Yellow icons represent signaling pathways. Blue icons represent action targets. A larger graph implies that more pathways are connected to it. Please click here to view a larger version of this figure.

Figure 2: Molecular docking and molecular dynamics trajectories visualization of JWSJS in the treatment of diabetic nephropathy. (A) The conformations of the ligand in the original crystal structure of the three target protein complexes 7JXQ (a), 4QTA (b), and 4QTB (c) are compared with the conformations of the ligand after docking. Green, orange, and pink are EGFR, MAPK1, and MAPK3 target proteins, respectively. Gold is the original ligand molecule, and blue is the ligand molecule after docking. (B) 3D molecular docking visualization of JWSJS in the treatment of diabetic nephropathy. Green, orange, and pink are the target proteins 7JXQ, 4QTA, and 4QTB, respectively. 1a-3a represents the interaction diagram of 7JXQ, 4QTA, and 4QTB with the original inhibitor (blue). 1b-3b represents the interaction diagram of 7JXQ, 4QTA, 4QTB, and luteolin (yellow). 1c-3c represents the interaction diagram of 7JXQ, 4QTA, 4QTB, and palmitoleic acid (fuchsia). 1d-3d represents the interaction diagram of 7JXQ, 4QTA, 4QTB, and quercetin (purple). (C) 2D molecular docking visualization of JWSJS in the treatment of diabetic nephropathy. 1a-3a represents the interaction diagram of 7JXQ, 4QTA, and 4QTB with the original inhibitor. 1b-3b represents the interaction diagram of 7JXQ, 4QTA, 4QTB, and luteolin. 1c-3c represents the interaction diagram of 7JXQ, 4QTA, 4QTB, and palmitoleic acid. 1d-3d represents the interaction diagram of 7JXQ, 4QTA, 4QTB, and quercetin. (D) RMSD and RMSF analysis of the molecular dynamics trajectories of (a) luteolin and target protein 7JXQ, (b) luteolin and target protein 4QTA, and (c) quercetin and target protein 4QTB complexes. Please click here to view a larger version of this figure.

Figure 3: JWSJS alleviates kidney injury and suppresses the EGFR/MAPK3/1 signalling pathway activation in vivo experiments. (A) HE staining and (B) periodic acid-Schiff (PAS) staining of rat kidney tissues in different groups (Magnification, 200x; n = 10). (C) Transmission electron microscopy of kidney tissue in different groups of rats (Magnification, 5000x; n = 3). (D) EGFR, p-EGFR, MAPK3/1, p-MAPK3/1, BAX, and BCL-2 in rat kidney tissue of diabetic through western blotting (n = 3). (E) Statistical analysis of p-EGFR, p-MAPK3/1, BAX, and BCL-2 expression. (F) Immunohistochemical analysis of p-EGFR AND p-MAPK3/1 expression in the six groups (n = 6). (G) Statistical analysis of p-EGFR AND p-MAPK3/1 expression. (H) qPCR analysis of MAPK3, MAPK1, and EGFR mRNA expression in six groups. *P < 0.05 vs. control. **P < 0.01 vs. control. ▲P < 0.05 vs. DN. ▲▲P < 0.01 vs. DN. Please click here to view a larger version of this figure.

Table 1: Information of some active ingredients of JWSJS. It gives details about each component and their characteristics or properties. Please click here to download this Table.

Table 2: Table of docking results of core targets and main active ingredients of JWSJS. The table presents the results from docking studies of core targets and main active components of JWSJS. This table shows how these constituents interact with their respective targets, which can provide insight into the mechanism of action of JWSJS. Please click here to download this Table.

Table 3: Effect of JWSJS on body weight and biochemical indexes in rats with diabetic nephropathy (expressed as mean ± standard deviation, n = 10). Compared with the normal group: * P < 0.05. Compared with the model group: # P < 0.05. Please click here to download this Table.

Discussion

Our study employed a combination of network pharmacology, molecular docking, and in vivo animal models. A critical step was the establishment of the "drug-component-target" network, which was crucial for identifying the potential mechanisms of JWSJS in treating DN, focusing particularly on its interaction with the EGFR/MAPK3/1 signaling pathway.

During this study, we made several modifications, particularly in the molecular docking process, to enhance the accuracy of our predictions. Troubleshooting was mainly focused on optimizing the conditions for the in vivo animal model to ensure its relevance to human DN. By identifying the key active components and targets of JWSJS, this study provides a theoretical basis for further research and development of JWSJS as a potential drug candidate for DN treatment.

The study investigated the potential mechanism of action of JWSJS in reducing renal apoptosis in DN. Molecular docking results showed that quercetin, palmitoleic acid, and luteolin have regular binding activities to key targets, including MAPK1, MAPK3, and EGFR. We selected these targets for further study due to their common action in multiple pathways. In DN rats, JWSJS treatment resulted in a significant decrease in FBG and GSP levels, improvement in UTP and LDL-C, weight gain, and improved renal tissue microscopic manifestations, indicating the potential effectiveness of JWSJS in treating DN.

Apoptosis is a critical mode of cell death, and dysregulation can increase the risk of various diseases³⁸. Previous studies demonstrated that DN rats had increased expression of apoptosis-related proteins BAX and caspase-3, and a significant increase in the number of apoptotic cells in renal tissue^39,40. The results showed that JWSJS treatment reversed these changes, suggesting that it can attenuate podocyte apoptosis under diabetic conditions.

Further investigation into the potential mechanism of action of JWSJS showed that it has a multi-target and multi-pathway effect on DN. We focused on studying MAPK1, MAPK3, and EGFR, as they had the largest number of enrichment pathways. The ERK pathway, which is involved in various physiological and pathological processes such as cell growth, proliferation, differentiation, apoptosis, and abnormal steroid secretion, is an essential pathway in the MAPK signal transduction pathway^13,41,42,43. It inferred that JWSJS rescued podocyte apoptosis by inhibiting the EGFR/MAPK3/1 signaling pathway.

We also found that JWSJS could regulate the expression of p-EGFR and p-MAPK3/1, resulting in markedly upregulated levels of Bcl-2 but a noticeable down-regulation of Bax. This, in turn, led to reduced renal cell apoptosis. Similar results have been found in previous studies. For instance, Chen reported that EGFR deletion in podocytes attenuated diabetic nephropathy⁴⁴. Some researchers found that the EGFR/ERK pathway promoted the proliferation and differentiation of porcine intestinal epithelial cells⁹. Inhibition of EGFR can down-regulate the expression of TGF-β and BAX, thus improving renal fibrosis and apoptosis⁴⁵. These findings suggest that JWSJS is a traditional Chinese medicine prescription with a significant role in the adjuvant therapy of DN by inhibiting the EGFR/MAPK3/1 signal pathway. However, further studies are needed to fully understand the potential mechanism of action of JWSJS in treating DN.

There were several limitations to this study. First, the reliance on animal models may not fully replicate the human response to JWSJS. Additionally, network pharmacology predictions, while insightful, require further validation through experimental data. Further clinical studies on human subjects are necessary to validate the findings. Second, although network pharmacology is a useful tool for analyzing drug-disease relationships, the results should be interpreted with caution as they are based on predictions rather than experimental data. Thirdly, the mechanism of action of JWSJS for treating DN remains unclear, and the study only focused on the potential targets of MAPK1, MAPK3, and EGFR.

The methods used in this study can be applied to other traditional Chinese medicine formulations, potentially leading to the discovery of new therapeutic options for DN and other complex diseases. Further, this approach lays the groundwork for integrating traditional and modern medicine in future pharmacological research. Further studies are needed to fully elucidate the molecular mechanisms underlying the therapeutic effects of JWSJS. Finally, the study did not investigate the potential side effects or toxicity of JWSJS, which are important considerations in the development of any new drug. Further research is needed to evaluate the safety of JWSJS and ensure its clinical application.

In summary, this study primarily investigates the therapeutic potential of Jiawei Shengjiang San for diabetic nephropathy utilizing network pharmacology, molecular docking, and in vivo animal validation. It emphasizes the procedural aspects, focusing on the methodology employed to understand the interactions between JWSJS and the EGFR/MAPK3/1 signaling pathway and how these interactions potentially reduce renal cell apoptosis in DN.

Materials

Name	Company	Catalog Number	Comments
2×SYBR Green qPCR Master Mix	Servicebio, Wuhan, China	G3320-05
24-h urine protein quantification (UTP)	Nanjing Jiancheng Institute of Biological Engineering	N/A
3,3'-Diaminobenzidine	Shanghai Huzheng Biotech, China	91-95-2
Automatic biochemical analysis instrument	Hitachi, Japan	7170A
Anhydrous Ethanol	Biosharp, Tianjin, China	N/A
BAX Primary antibodies	Affinity, USA	AF0120	Rat
BCL-2 Primary antibodies	Affinity, USA	AF6139	Rat
BX53 microscope	Olympus, Japan	BX53
Chloroform Substitute	ECOTOP, Guangzhou, China	ES-8522
Desmond software	New York, NY, USA	Release 2019-1
Digital Constant Temperature Water Bath	Changzhou Jintan Liangyou Instrument, China	DK-8D
EGFR Primary antibodies	Affinity, USA	AF6043	Rat
Embed-812 RESIN	Shell Chemical, USA	14900
Fasting blood glucose (FBG)	Nanjing Jiancheng Institute of Biological Engineering	N/A
FC-type full-wavelength enzyme label analyser	Multiskan; Thermo, USA	N/A
GAPDH  Primary antibodies	Affinity, USA	AF7021	Rat
Glycated serum protein (GSP)	Nanjing Jiancheng Institute of Biological Engineering	N/A
Transmission electron microscope	Hitachi, Japan	H-7650
Haematoxylin/eosin (HE) staining solution	Servicebio, USA	G1003
Image-Pro Plus	MEDIA CYBERNETICS, USA	N/A
Real-Time PCR Amplification Instrument	Applied Biosystems, USA	iQ5
Irbesartan tablets	Hangzhou Sanofi Pharmaceuticals	N/A
Isopropanol	Biosharp, Tianjin, China	N/A
JWSJS granules	Guangdong Yifang Pharmaceutical	N/A
Kodak Image Station 2000 MM imaging system	Kodak, USA	IS2000
Low-density cholesterol (LDL-C)	Nanjing Jiancheng Institute of Biological Engineering	N/A
MAPK3/1Primary antibodies	Affinity, USA	AF0155	Rat
Medical Centrifuge	Hunan Xiangyi Laboratory Instrument Development, China	TGL-16K
Mini trans-blot transfer system	Bio-Rad, USA	N/A
Mini-PROTEAN electrophoresis system	Bio-Rad, USA	N/A
NanoVue Plus Spectrophotometer	Healthcare Bio-Sciences AB, Sweden	111765
p-EGFR Primary antibodies	Affinity, USA	AF3044	Rat
Periodic acid-Schiff (PAS) staining solution	Servicebio, USA	G1008
p-MAPK3/1 Primary antibodies	Affinity, USA	AF1015	Rat
Secondary antibodies	Santa Cruz, USA	sc-2357	Rabbit
Streptozotocin	Sigma, USA	S0130
SureScript First-Strand cDNA Synthesis Kit	GeneCopeia, USA	QP056T
TriQuick Reagent	Solarbio, Beijing, China	R1100
Ultra-Clean Workbench	Suzhou Purification Equipment, China	SW-CJ-1F