Scientific Papers

Molecular mechanism of Danshenol C in reversing peritoneal fibrosis: novel network pharmacological analysis and biological validation | BMC Complementary Medicine and Therapies


Screening of active ingredients, ADME analysis, and potential Target Prediction

Given the limited availability of target information pertaining to Danshenol C, it becomes imperative to prognosticate its potential targets based on its structural formula. The structural formula of Danshenol C was retrieved through meticulous screening using the YaTCM database. Subsequently, utilizing the Swiss Target Prediction database, a comprehensive repertoire of 43 active component species associated with Danshenol C was ascertained. A total of 110 drug targets were successfully identified. To ensure accuracy and consistency, the collated active ingredients and drug targets underwent meticulous refinement through Perl software and the UniprotKB database, thereby securing their official gene names. In pursuit of a holistic understanding, the amalgamation of genes present in the aforementioned databases was undertaken (Fig. 2a). To attain a sufficiently robust dataset catering to PF, differential genes sourced from GEO and genes procured from the aforementioned databases were harmoniously integrated. The subsequent step involved a meticulous alignment of disease genes with corresponding drugs (Table 2 and Fig. 2b).

Fig. 2
figure 2

Venn diagram. a Five Disease datasets. b The targets in Disease and Danshenol C

Table 2 Core components of Danshenol C

Screening, batch-normalization correction, and analysis of disease-associated targets in PF

The GSE92453 dataset, encompassing 17 samples of Peritoneal Membrane and 21 samples of Omentum, was meticulously procured from the GEO chip database. To ensure data integrity and comparability, the “SVA” package, a powerful tool within the R language, was adeptly deployed to execute batch normalization on the two acquired datasets. Following the correction process, an assemblage of 15 DEGs was successfully identified from the two distinct groups, skillfully facilitated by the “Limma” package within the R language. Among these identified DEGs, 10 genes (66.66%) displayed up-regulation, while 5 genes (33.33%) exhibited down-regulation. A graphical representation of the differentially expressed genes is visually depicted in Fig. 3. Based on the PCA findings, patients with differing risks were separated into two groups (Fig. 4). Furthermore, an insightful depiction of the intricate interplay between 40 drugs and the genes associated with the disease phenotype was meticulously illustrated in Fig. 5.

Fig. 3
figure 3

Differential genes volcano and heat map of GEO. a Heatmap. b Volcano map

Fig. 4
figure 4
Fig. 5
figure 5

Herb-ingredients-targets (H-I-T) network

Construction of PPI network and identification of key targets

Cytoscape3.7.2 software was used to obtain a PPI network map of 40 related targets and their relationships (Fig. 5). Then, the intersection target interaction relationship calculated from String database was visualized, and the intersection of the above PPI network map was extracted with CytoNCA toolkit. After screening twice, STAT3, MAPK14, MAPK8, CASP3 were found to be the main efficacious genes (Fig. 6).

Fig. 6
figure 6

Target screening strategy diagram of Danshenol C in the treatment of PF key nodes

GO and KEGG enrichment analysis

The GO analysis performed in this study led to the identification of 948 core targets, encompassing MF, CC, and BP. The MF category predominantly involved passive transmembrane transporter activity (GO:0022803), channel activity (GO:0015267), and DNA-binding transcription activator activity (GO:0001216). Within the CC category, significant associations were found with synaptic membrane (GO:0097060), chromosomal region (GO:0098687), and membrane raft (GO:0045121). As for BP, key involvement was observed in neutrophil activation (GO:0042119), divalent inorganic cation homeostasis (GO:0072507), and neutrophil-mediated immunity (GO:0002446). Moreover, the utilization of KEGG analysis allowed us to discern the primary signaling pathways. The overexpressed genes were found to be predominantly associated with the Pathways of neurodegeneration-multiple diseases (hsa05022), Alzheimer’s disease (hsa05010), and Neuroactive ligand-receptor interaction (hsa04080) (Fig. 7). To enhance clarity and comprehension, the data results of KEGG were effectively visualized (Fig. 8).

Fig. 7
figure 7

GO and KEGG analysis of potential targets of Danshenol C in the treatment of PF (a): The GO barplot and bubble illustrates the scatter map of the selected gene’s logFC. b: The KEGG barplot and bubble illustrates the scatter map of the logFC of the indicated gene

Fig. 8
figure 8

KEGG analysis of the target-pathway network. The edges reflect the interactions between the targets and the paths, and the node size is proportional to the degree of connection

Molecular docking

To unravel the molecular interactions and binding affinities of the core targets extracted from the esteemed Uniprot database (namely, MAPK14 with Wogonin, MAPK8 with sitosterol, and STAT3 with STAT5), we employed the advanced SYBYL2.0 software. By doing so, we were able to thoroughly assess the binding strength and activity of these protein–ligand complexes (Table 3). In the pursuit of a deeper understanding of the intricate molecular associations, we further conducted molecular docking using the eminent AutoDock1.5.6 and PYMOL2.4 software. This elaborate process allowed us to precisely position the molecular structures of the key active components in relation to the core targets. The outcomes of these docking simulations, as showcased in Fig. 8, elegantly elucidated that Wogonin, Sitosterol, and STAT5 emerged as the principal small molecule active components and gene targets of Danshenol C in the therapeutic intervention against PF (Fig. 9).

Table 3 Binding energy of key active components of Danshenol C in the treatment of PF and key target docking
Fig. 9
figure 9

Molecular docking diagram of molecular structures and key targets (a) MAPK14-Wogonin. b MAPK8-sitosterol. c STAT3-STAT5

High glucose peritoneal dialysate inhibits HMrSV5 activity

In order to select the appropriate concentration and time for membrane construction, we selected three specifications of clinically commonly used sugar-containing peritoneal dialysate (1.5%, 2.5%, 4.25%) and performed cytotoxicity assay at different time points (24 h, 36 h, and 48 h). According to CCK-8 assay, compared with the normal group, only 4.25%PDS inhibited the cell viability at all three time points, and significantly decreased the cell viability at 48 h. (Fig. 10A, B, C, D) The results showed that 4.25%PDS could significantly inhibit the activity of HMrSV5 for 48 h, and 4.25%PDS was selected for subsequent experiments.

Fig. 10
figure 10

Effect of PDS on HMrSV5 activity A Different concentrations of PDS (1.5% Dianeal, 2.5% Dianeal, 4.25% Dianeal) were used for 24 h; B Treatment with 4 concentrations of PDS for 36 h; C Treatment with 4 concentrations of PDS for 48 h; D Comparison chart at different times; ns: P > 0.05; *: P < 0.05; **: P < 0.01; ***: P < 0.001; ****:P < 0.0001

Danshenol C reversed the inhibition of HMrSV5 activity under high glucose condition

In order to select the appropriate intervention dose of Danshenol C, we carried out cytotoxicity assay in HMrSV5 cells. We first treated cells with different concentrations of Danshenol C (5, 10, 15, 20, 25, 40 and 80 μM) alone, and found that the viability of HMrSV5 cells did not change after 48 h of intervention (Fig. 11 A). Then we cocultured with different concentrations of Danshenol C (10,20, 40, 80 and 160 μM) and 4.25% high glucose peritoneal dialysate for 48 h. According to CCK-8 assay, 10, 20, 40 and 80 μM Danshenol C promoted the cell viability after high glucose treatment, while 160 μM Danshenol C slightly decreased the cell viability (Fig. 11B). The results showed that 10 and 20 μM Danshenol C could significantly reverse the PDS-induced decrease of HMrSV5 cell viability (P < 0.05). To verify the subsequent experiments, we selected two groups of Danshenol C concentrations. The final concentration of Danshenol group C1 was 10 μM, and the final concentration of Danshenol group C2 was 20 μM.

Fig. 11
figure 11

The toxicity of Danshenol C to HMrSV5 cells. A Treatment with different concentrations of Danshenol C for 48 h; B Different concentrations of Danshenol C + 4.25%PDS for 48 h ns: P > 0.05; *: P < 0.05

Danshenol C can improve fibrosis on morphology and fibrosis markers

According to the microscopic observation, the normal cells were round and oval, and the cells were spindle shaped after 48 h of 4.25%PDS treatment, and the cell image was in the middle of the two after drug addition (Fig. 12A, B, C).

Fig. 12
figure 12

Effects of Danshenol C on morphology of cell and tissue A normal cells; B cells treated with 4.25%PDS for 48 h; C Danshenol C + 4.25%PDS for 48 h (D) normal peritoneal tissue of mice; E intraperitoneal injection 4.25%PDS for 28 days; F intraperitoneal injection Danshenol C + 4.25%PDS for 28 days

Furthermore, normal peritoneal tissue was smooth and mesothelial dense according to Massone staining. Mice treated with peritoneal dialysis fluid had deposited collagen fibers in the peritoneal interstitial layer with an enlarged submesothelial dense zone associated with increased inflammatory cells. The presentation of peritoneal histomorphology after Danshenol C intervention was intermediate between the two groups (Fig. 12D, E, F).

Real-time RT-PCR showed that compared with the control group, the mRNA expression of Fibronectin in the Danshenol C1 and C2 groups was decreased (Fig. 13A), and the mRNA expression of E-cadherin was significant increased (Fig. 13B), and the increase was more significant in the high concentration of Danshenol C.

Fig. 13
figure 13

Effects of Danshenol C on fibrosis markers (A)(B) mRNA expression; C western blot; D (E) protein expression;(F)(G) histochemistry ns: P > 0.05; *: P < 0.05; **: P < 0.01; ***: P < 0.001; ****: P < 0.0001

Western blot analysis showed that high glucose induced Epithelial-mesenchymal transition(EMT) in peritoneal mesothelial cells. Compared with normal cells, VEGF-A protein expression was increased and E-cadherin protein expression was decreased. The VEGF-A protein expression was decreased and E-cadherin protein expression was decreased in the two groups after Danshenol C treatment. As shown in Fig. 13C, D, E.

Immunohistochemical analysis of peritoneal tissue in mice showed that, after treated with 4.25% PF, VEGF-A positive area increased and E-cadherin-positive area decreased compared with normal peritoneal tissue. VEGF-A positive area decreased after Danshenol C intervention and increased in E-cadherin positive area. As shown in Fig. 13F, G.

Potential pathways and mechanisms of Danshenol C reversal of hyperglycemic fibrosis

Through the above network pharmacology, we found four target genes related to Danshenol C and PF, namely MAPK8 (JNK1), MAPK14 (P38), CASP3 and STAT3. Real-time RT-PCR showed that compared with the control group, the mRNA expression of MAPK8 in the C1 and C2 groups was significantly decreased (Fig. 14A), and the mRNA expression of MAPK14 was decreased (Fig. 14B), and the decrease was more significant in the low concentration of Danshenol C. Compared with the control group, the mRNA expression of STAT3 in the Danshenol C1 group was decreased, but there was no difference in the Danshenol C2 group (Fig. 14C). Compared with the control group, the mRNA expression of CASP3 of Danshenol C increased significantly. (Fig. 14D).

Fig. 14
figure 14

Relative RNA expression levels of the target genes (A) (B) Compared with the control group, the C1 C2 groups were significantly decreased; C Compared with the control group, group C1 was significantly decreased, while group C2 had no significant change; Compared with the control group, the C1 C2 group increased significantly; *:P < 0.05; **:P < 0.01; ***: P < 0.001

To further explore the genes involved and possible pathways, Western blotting was performed to measure the protein expression levels of STAT3(T721), STAT3(S727), STAT3(T705), MAPK14 (P38 (H174)), P-p38 (T180/182), Caspase3 and MAPK8 (JNK1). The blot of the protein band is shown in Fig. 15A. Compared with normal cells, the protein levels of P-P38, P38 and Caspase3 in fibrotic cells were decreased (Fig. 15E, F, G) (P < 0.05), indicating that the expression of these three genes was inhibited. Compared with normal cells, the protein expression level of STAT3 and JNK1 in fibrotic cells was increased (Fig. 15B, C, D, H) (P < 0.05). Compared with fibrotic cells, the protein levels of P-P38, P38 and Caspase3 in Danshenol C treated cells were significantly increased (Fig. 15E, F, G) (P < 0.05), and the protein expression level of STAT3 and JNK1 was significantly decreased (Fig. 15B, C, D, H) (P < 0.05).

Fig. 15
figure 15

Relative protein expression levels of the target genes (A): Western blot images; B (C)(D)(H): Compared with the normal group, the expression level of target protein in the control group was significantly increased; Compared with the control group, the Danshenol C group decreased significantly; E (F)(G): Compared with the normal group, the expression level of the target protein in the control group was significantly decreased; Compared with the control group, the Danshenol C group increased significantly; *:P < 0.05; **:P < 0.01; ***: P < 0.001; ****: P < 0.0001

Caspase3 Is closely related to apoptosis. We conducted apoptosis flow analysis and found that apoptotic cells increased compared with control and model groups, but there was no statistical difference (Fig. 16AB) (P > 0.05).

Fig. 16
figure 16

Cell apoptosis ns: P > 0.05



Source link