UPLC-Q-TOF/MSresults for ELC
In our previous study, GC-MS analysis was used to reveal the volatile oil components of ELC. A total of 14 compounds were identified, namely eucalyptol, D-camphor and isoborneol, l(−)-Borneol, α-terpineol, β-elemene, γ-elemene, α-humulene, germacra, curcumenol, β-cyclocostunolide, curcumenone, pulmonary Zedeone, and ent-kaurene. In this study, the compounds of ELC were identified by UPLC-Q-TOF/MS. According to the multi-stage MS information combined with the data obtained from the literature and databases, 26 compounds were identified in the ELC compound preparation.Representative fingerprint chromatograms of ELC are displayed in Figure 2. The identified compouds of ELC were showed in Table 4. Combined with our previous study results, we have identified a total of 40 compounds in ELeng Capsule.These results provide identified compounds in ELC for network pharmacology analysis.
Network pharmacology analysis
The compounds associated targets of ELC
Based on the UPLC-Q-TOF/MS data, we performed a network pharmacology analysis on the related compounds of ELC and predicted the targets of compounds. Based on the data obtained from the network pharmacology-related databases, we identified 27 core compounds with 194 targets based on STITCH，TCMSP and UNPD datasets. The result showed that the major targets were involved in angiogenesis, inflammation, immunity, and other modules.Supplementary Material 1 showed the major compouds and targets of ELC.In addition,we had collected 1289 endometriosis-related targets from Genecard and Genbank. A total of 75 targets of ELC capsules coincided with those of genes associated with endometriosis. Following cytoHubba analysis, VEGFA,IL6,TP53,PTGS2,AKT1,MMP9,MAPK1,JUN,etc may be core targets of ELC.
The analysis of GO terms and KEGG pathways
We analyzed the GO terms and KEGG pathways affected by targets associated with ELC. The results suggest that the BP in GO terms are related to cell death, apoptosis, and proliferation. The results of KEGG pathways were mainly enriched in cancer pathway, inflammatory response, drug reaction, etc., which were consistent with the present pathologic mechanism of endometriosis.The main pathways include neuroactive ligand-receptor interaction, TLR signaling pathway, metabolism of xenobiotics by cytochrome P450, vascular endothelial growth factor (VEGF) signaling, calcium signaling, B cell receptor signaling, among others. GO and KEGG enrichment analysis suggest that the core components of ELC may be involved in the regulation of inflammation, immunity and angiogenesis modules.The major GO terms and KEGG pathways are shown in Figure 3A and 3B. The network of major targets and compounds from database is showed in Figure 3C.
We further screened 180 endometriosis-related core genes and the core compounds of ELC for molecular docking. The results implied that tanshinone IIA, cryptotanshinone, rosmarinic acid, gallic acid, linoleic acid, γ-elemene, palmitate, hesperetin, danshensu, palmitic acid, naringin, and tanshinone I may be compounds that play a major role in endometriosis. The hub targets related to the compounds are as follows: eukaryotic translation initiation factor 2 alpha kinase 3 (EIF2AK3), actin-beta (ACTB), prostaglandin-endoperoxide synthase 2 (PTGS2), progesterone receptor (PGR), matrix metalloproteinase 2 (MMP2), chemokine (C-C motif) ligand 3 (CCL3), signal transducer and activator of transcription (STAT1), nitric oxide synthase 2 (NOS2), intercellular adhesion molecule 1 (ICAM1), MMP9, MYC, transforming growth factor-β1 (TGFβ1), and VEGFA based on molecular docking. The results of molecular docking were showed in Figure 3D and supplementary material 2.
Potential mechanism of ELC in Endometriosis rat model
Pathology of ectopic lesions
We successfully established a rat model of endometriosis. Figure 4A shows the changes in lesions after modeling. After the treatment,the average value of tissue in ELC group was lower than the model group, while the difference was not statistically significant (P＞0.05)(supplementary material 3). Compared with the model group, the lesion volumes of before and after ELC treatment in ELC middle group was change significantly (P=0.028＜0.05). These result showed that ELC may reduce the volume of ectopic lesions to a certain extent in endometriosis rats mode. The HE revealed the formation of local glands in the lesions of the model and ELC treatment rats(Figure 4B).
ELeng Capsules Promoted Apoptosis in Ectopic Endometrial Tissues
Figure 4 C was showed the ELC on apoptosis of endometriotic tissues, as determined by TUNEL assay (×100, n = 4 each). The distribution of green fluorescence included glandular epithelium and mesenchymal cells. The nuclei of positive staining apoptotic cells emitted green fluorescent signals in the ectopic endometrium tissues. Compared with the model group, the number of apoptotic cells in ectopic lesions in high, medium and low dose of ELC were significantly increased. Compared with the model group, the apoptotic area in the middle dose group of ELC increased significantly(P < 0.05).And there was no statistically significant difference in the dipsion area between the high-dose group and low-dose group compared with the model group(P > 0.05). This result suggested that ELC could participate in the process of cell apoptosis.
Ultrastructural Features of Ectopic Endometrium in a Rat Model
The glandular epithelial cells in the model group were relatively neat and had microvilli.In the ectopic tissue cells in ELC group,the microvilli were reduced, and the mitochondria were swollen in the cytoplasm, and autophagy and apoptotic bodies were observed(Figure 4D). Figure 4D(c,d) showed the apoptotic body,and figure 4D(e,f) showed the autophagosome.This result suggested that ELC treatment may be related to the regulation of autophagy and apoptosis.
The effect of Eleng capsule on angiogenesis
Figure 5A showed that CD34 was positively expressed in both the model group and the ELeng capsule group. Among them, the expression of MVD in the ectopic lesion was significantly higher than that in the eutopic endometrium, and angiogenesis was observed(n=6, P=0.015＜0.05). The expression of CD34 in high, medium and low dose groups of ELeng capsule was significantly lower than the model group expression (P＜0.05).
Figure 5B showed the expression of VEGFA, VEGFB and VEGFC in the cytoplasm and membrane of glandular epithelial cells in ectopic lesion in endometriosis rat model.The expression of mesenchymal cells was weaker than that of glandular epithelial cells.The expression of VEGFA in middle dose and the low dose group of ELC were decreased compared with the model group (P < 0.05).There was no significant difference between the model group and the ELC high, middle and low doze groups of the VEGFB expression (P＞0.05).Compared with the model group, the expression of VEGFC in ectopic lesions were significantly decreased in the high, middle and low dose of ELC, and the differences were statistically significant(P < 0.05). These results suggested that ELC could inhibit angiogenesis through inhibiting VEGFA and VEGFB expression.
The Effect of Eleng Capsule in Serum VEGFA,VEGFB and IL1-β
Compared with the control group(34.838±1.403pg/ml, n=8), the VEGFA level in serum increased in the model group(38.866±2.706pg/ml, n=8), and the difference was statistically significant (P=0.008 < 0.05). Compared with the model group, the serum VEGFA level in the high-dose group(35.345±4.205pg/ml, n=8) and low-dose group(35.024±2.662pg/ml, n=8) of ELC was significantly deseased(**P=0.020, ***P=0.012).There was no significant difference in serum of VEGFB and IL1-β expression in different groups endometriosis model rats（P＞0.05）.The expression level of VEGFA, VEGFB and IL-1β are low in serum of normal SD rats and endometriosis model rats.Thus the regulation effect of ELC may be mainly localized in ectopic lesions(Figure 5C and Supplementary Material 4).
ELeng Capsules could reduce the local fibrosis in Ectopic lesions
Masson’s trichrome staining showed that the ectopic lesions were fibrotic. Compared with the model group, the positive area of fibrosis were decreased in the high, middle and low dose groups of ELC, and the difference was statistically significant (P<0.05). The result showed that ELC could reduce the degree of fibrosis in endometriosis model rats in ectopic lesions. The results of Masson’s trichrome were shown in Figure 6.
The expression of The expression of α-SMA in ectopic lesions of model model rats was significantly increased, and the difference was statistically significant (P<0.05), suggesting that ectopic endometrium α-SMA is highly expressed. After ELC treatment, the expression of α-SMA in ELC groups(high medium and low dose groups) were all reduced compared with model groups(P<0.05) (Figure 6). These results suggest that ELC may reduce the fibrosis process of ectopic lesions by inhibiting the expression of α-SMA in ectopic lesions of endometriosis model rats.
RNA-sequencing analysis of endometriosis rats model characteristic and the treatment of ELC
The differentially expressed genes(DEGs) screening analysis
We further analyzed the potential mechanism of ELeng Capsule by RNA transcriptome.The tissues of Model_Ecto and ELC_Ecto groups were used for the mRNA microarray assays. According to the results of the principal components analysis, there was no significant difference between the endometria of rats in the endometriosis model and ELC intervention groups. The results of the principal components analysis also showed that the rat endometriosis model were no obvious eutopic endometrial changes. These suggested that ELC may not affect the eutopic endometrium in endometriosis rat models. There were a total of 1461 DEGs in control_Eu versus model_Ecto groups, 557 DEGs in model_Ecto versus ELC_Ecto groups, and 1097 DEGs in control_Eu versus ELC_Ecto groups(FC-1.5). Supplementary material 5 showed the PCA of five groups,veen analysis of five group, up- and down regulated DEGs and Heatmap illustration. In this study, there were 1048 up-regulated DEGs and 413 down-regulated DEGs between Model_Ecto versus Con_Euto groups. The transcriptome analysis of the endometriosis rat model revealed some genes among the DEGs that participate in processes, such as inflammation, cytoskeleton, EMT, and angiogenesis. In the ELC_Ecto versus Model_Ecto group analysis, a total of 66 and 491 up-regulated and down-regulated DEGs, respectively, were identified, reflecting the differential expression of related genes after treatment with ELC.
GO and KEGG enrichment in the endometriosis rat model ectopic endometrium versus control rat eutopic endometrium
As shown in GO terms, the up-regulated genes were most significantly enriched in the CC of extracellular region, the BP of muscle contraction, and MF of actin filament binding, fibronectin binding, calcium ion binding, etc. The actin-associated GO terms may relate to the development of ectopic lesions in the endometriosis rat model(Figure 7A). The KEGG analysis of genes associated with endometriosis suggested that these genes are associated with the pathway of extracellular matrix-receptor (ECM-receptor) interaction, p53 signaling pathway, pathways in cancer, endocrine resistance, interleukin-17 (IL-17) signaling pathway, chemokine signaling pathway, cytokine–cytokine receptor interaction, etc(Figure 7B).
We performed a Venn analysis of DEGs and human endometriosis-related genes using Genecard datasets. The results showed that there were 113 up-regulated and 28 down-regulated genes that were related to endometriosis. The proteins expressed by these genes are involved in EMT and may be involved in the process of fibrosis in endometriosis. The up-regulated genes associated with endometriosis in humans are related to cytokine–cytokine receptor interaction, phosphatidylinositol 3 kinase-Akt (PI3K-Akt) signaling pathway, pathway in cancer, MAPK signaling pathway, ECM-receptor interaction, Ras signaling pathway, toll-like receptor (TLR) signaling pathway, IL-17 signaling pathway, p53 signaling pathway, forkhead box protein O signaling pathway, focal adhesion, etc. Based on the above analyses, the rat model of endometriosis may be suitable for investigating the transcriptome level.
The enrichment score for this signaling pathway was positively correlated to endometriosis samples. Based on the path of GSEA enrichment, neuroactive ligand receptor interaction, cell adhesion molecules, and regulation of actin cytoskeleton are closely related to the occurrence and development of endometriosis. The development of endometriosis are related to the GO terms of “skeletal muscle fiber”, “endodermal cell differentiation”, “regulation of signaling receptor activity”, “positive regulation of myoblast differentiation”, “response to cytokine”, “chemokine-mediated signaling pathway”, “positive regulation of smooth muscle cell migration”,etc. The KEGG pathways of GSEA showed that the peroxisome proliferator-activated receptor signaling, tumor necrosis factor signaling, MAPK signaling, apelin signaling, hypoxia-inducible factor-1 signaling, PI3K-Akt signaling pathway, and focal adhesion are related to endometriosis(Figures 7C). (P<0.01, FDR<0.25).
GO and KEGG enrichment analysis of ELC_Ecto versus Model_Ecto
Based on the GO analysis, the development and drug(ELC) regulation of endometriosis animal models are related to the regulation of troponin and the cytoskeleton. The BP, CC, and MF GO terms suggest muscle- and troponin-associated regulation,which could related to ELC treatment. The major enriched GO BPs were positive regulation of fast-twitch skeletal muscle fiber contraction, muscle contraction, striated muscle contraction, etc. The major enriched GO CCs were terminal cisterna, junctional sarcoplasmic reticulum membrane, Z disc, etc. The major enriched GO MFs were actin filament binding, structural constituent of muscle, actin binding, etc. (Figure 8A).
We also analyzed the KEGG pathways of the ELC ectopic endometrium(ELC_Ecto) versus the model ectopic endometrium(Model_Ecto). The results of KEGG pathway enrichment showed that down-regulated DEGs were mainly enriched in the following pathways: calcium signaling pathway, apelin signaling pathway, cyclic guanosine monophosphate-protein kinase G (cGMP-PKG) signaling pathway, 5' adenosine monophosphate-activated protein kinase signaling pathway, hypoxia-inducible factor-1 signaling pathway, MAPK signaling pathway, PI3K-Akt signaling pathway, focal adhesion, etc(Figure 8B).These pathways could be related to the inhibitory effect of ELC in endometriosis treatment.
Through the GSEA of the KEGG analysis, we also found other signaling pathways, including the Notch signaling pathway, adherens junction, Hippo signaling pathway, basal transcription factors, regulation of actin cytoskeleton, etc., which were related to the treatment of endometriosis with ELC.The The major KEGG pathways from GSEA analysis were shown in Figure 8C(FDR<0.25). ELC may inhibit fibrosis and EMT by regulating the aforementioned pathways.
The core genes established in the network may be related to the regulation by ELeng Capsule treatment. Two modules obtained using the default criteria of MCODE for ELC_Ecto versus Model_Ecto were used to construct the network. We selected major modules for module network visualization (Figure 8D). The core nodes continued to be associated with genes related to actin, cytoskeleton, and fibrosis.
STEM analysis of differential expression patterns
Temporal expression patterns and the significance of the genes were determined using the STEM software. This approach was used to profile the DEGs, select significant gene clusters with parallel expression patterns, and identify the profiles of the “up to down” model from control to model to treatment with ELC. The results of the gene cluster analysis were statistically significant in profiles_14, profiles_11, profiles_10, and profiles_4 (P<0.05)(Figure 9A). After the development of endometriosis model, actin-associated DEGs in the Model_ecto and ELC_ecto groups were up-regulated; these may be the regulatory genes for the development of the endometriosis model (profile_11 and profile_4). The series test of profile_14 and profile_10 showed that the significant clusters were considered as potential profiles that could be affected by treatment with ELC (P<0.05). Several actin-related and microfilament proteins were up-regulated in the model group and down-regulated after treatment, suggesting that the overall regulation mechanism of ELC treatment in ectopic endometriun is related to the regulation of actin cytoskeleton. The results of the STEM analysis were further reflecting the process of the endometriosis rat model and intervention with ELC(Figure 9B).
Protein-protein interaction network
We constructed network relationships between the core genes of the two groups of DEGs, and analyzed the possible network relationships through relevant pathways. We selected the calcium signaling pathway, cGMP-PKG signaling pathway, apelin signaling pathway, and the DEGs to construct the network (Figure 10A). The hub down-regulated genes closely related to treatment with ELC were as follows: Actn3, Actn2, Myom2, myoglobin, Ryr1, Myog, Myh7, Myod1, sarcalumenin, myosin light chain kinase 2, Smyd1, Map3k7, Mapk12, Myh4, Cacna1s, EEF1A2, and Cacng1.
We compared the DEGs in gene expression profiles between induced by ELC(ELC_Ecto) vs Model_Ecto groups,and used the CMAP database to better understand the therapeutic efficacy and identify potential targets of this agent. Based on the results, a network was constructed to link ELC with representative drugs and their therapeutic targets. The profile of ELC was similar to that of classical pharmacotherapies (e.g.,sulfaphenazole, carbinoxamine, esculin, emetine, zoxazolamine, and niflumic acid). The result of CMap analysis are showed in Table 5. Carbinoxamine is a first-generation antihistamine of the ethanolamine class. Zoxazolamine is a centrally acting myorelaxant used as an antispasmodic and uricosuric. Niflumic acid are used in the treatment of rheumatoid arthritis, dysmenorrhea, and osteoarthritis. Niflumic acid inhibited invasion and reduced apoptosis in CNE-2Z cells through regulation of the extracellular regulatory kinase/MAPK pathway. Thus, this agent may serve as a candidate of anticancer drug [21, 22]. Based on the results of the CMAP analysis, we consider that ELC may similarly regulate inflammation, the cytochrome C P450 pathway, and MAPK pathway.
Real-time quantitative RT-PCR
Quantiﬁcation was performed with a two-step reaction process: reverse transcription (RT) and PCR. The expression ratios of these DEGs, determined by qPCR, were consistent with those obtained from the mRNA transcriptome analysis(Figure 10B). The results for some genes exhibited a different trend; however, the differences were not statistically significant. The observed trend may be related to individual differences in the degree of gene expression in rats.The expression of verified genes in qPCR were showed in supplementary material 6.