Study on the Screening of Anti‐Osteoporosis Components in Processed Epimedium‐Based on Spectrum‐Effect Analysis

ABSTRACT

Epimedium (EP) is a traditional Chinese medicine that has been used to treat osteoporosis for several years, but its mechanism of action is unclear. We used ultrahigh‐performance liquid chromatography coupled with triple–quadrupole mass spectrometry (UPLC–QQQ–MS/MS) to detect the components of raw and processed EP samples. The antioxidant activities of the various tested compounds were evaluated using 1,1‐diphenyl‐2‐trinitrophenylhydrazine (DPPH), and their antiosteoporosis effects were assessed in H₂O₂‐stimulated MC3T3 cells and zebrafish embryos. The correlation between the chemical ingredients and efficacy indicators was determined using Pearson correlation analysis. Furthermore, the efficacy of the active ingredients in treating osteoporosis was verified through the implementation of network pharmacology and the application of molecular docking techniques, providing a more comprehensive understanding of their therapeutic potential. The pharmacologically active ingredients were quercetin, Epimedium B, Baohuoside I, and anhydroicaritin‐7‐O‐glucoside. Moreover, network pharmacology and molecular docking studies demonstrated that the four active components could have antiosteoporosis effects by influencing AKT1 and PTGS2. This study provides a basis for future research on the pharmacokinetics of EP and serves as a useful guide for further exploration of EP.

Abbreviations

BPI

base peak intensity

EP

Epimedium

KGB

kidney governing bone

KYDS

Kidney‐Yang deficiency syndrome

REP

raw Epimedium

SEP

stir‐frying Epimedium

SOEP

suet oil processed Epimedium

TCM

traditional Chinese medicine

TFC

total flavonoid content

UPLC–QQQ–MS/MS

ultrahigh‐performance liquid chromatography coupled with triple–quadrupole mass spectrometry

WEP

wine‐processed Epimedium

WSOEP

wine and suet oil processed Epimedium

1. Introduction

Epimedium (EP), a traditional Chinese herb, was first recorded in Shennong's Classic of Materia Medica. EP is a valuable tonic herb with a special medical value [1, 2, 3]. Recent studies have indicated that EP is composed mainly of flavones, polysaccharides, essential oils, phytosterols, phenolic acids, and alkaloids [4], which regulate libido, cardiovascular disease, and immunity and have antiosteoporosis, anticancer, antioxidant, and antiaging effects [5, 6]. According to traditional Chinese medicine (TCM) theory, processing is necessary to improve the clinical efficacy and/or decrease the toxicity of an herb [7]. EP is frequently used in both its raw and processed forms in clinical practice. The processing of EP facilitates drug penetration into the kidney, nourishes the liver and kidney, and enhances muscular and skeletal strength [8]. Flavonoids are a group of polyphenols widely distributed among vascular plants that appear to be essential for the prevention and treatment of osteoporosis [9]. Icariin is one of the major active flavonoid glucosides isolated from EP, can promote the proliferation, differentiation, and mineralization of osteoblasts, and inhibits osteoblast apoptosis [10]. The C‐8 position of isopentenyl in the molecular structure of EP is believed to be the cause of the superior osteogenic activity of icariin compared to that of other flavonoids [11, 12]. Currently, research on the mechanism underlying the enhanced efficacy of processed EP is limited.

TCM defines Kidney‐Yang deficiency syndrome (KYDS) as a condition characterized by fluctuations in Qi levels. Studies have shown that KYDS may result in various pathological conditions, including osteoporosis [13]. In TCM, there is no specific disease called POP. The TCM syndromes “Guwei,” “Guku,” “Guji,” “Gukong,” “Gusuo,” and “Gubi” include osteoporosis. The “kidney governing bone” (KGB) theory is a classical theory in TCM [14, 15]. EP is known to strengthen the kidneys and warm yang, as well as strengthen the muscles and bones, through traditional techniques [16, 17]. Research has indicated that EP species possess properties that combat osteoporosis and promote kidney tonification [18, 19, 20]. However, the effective ingredients of EP are unknown and further research is needed.

It is crucial to standardize the treatment of Chinese herbs to ensure the quality and safety of the product. In recent years, the study of the composition and material basis of the medicinal effect of TCM has become a hot topic in TCM circles. However, the current TCM quality assessment system mainly focuses on the analysis of one chemical composition, ignoring the emphasis on the formulation of the substance foundation. Therefore, researching the spectrum–effect relationship offers a means of identifying the material basis for the effectiveness of TCM by exploring the correlation between the changes in chemical components and treatment efficacy. Using this method, the interdependence between the chemical composition and efficacy of TCM has been revealed [21, 22].

In this study, we conducted a cell and zebrafish experiment on osteoporosis to evaluate the effects of raw and processed (EP) sucrose. Through the study of spectral effects, the material basis that influences the change in drug effects before and after EP processing was determined. This study offers a specific reference point for the rational clinical use and quality control of EP.

2. Materials and methods

2.1. Chemicals and Reagents

Twelve flavonoid reference standards, namely, hyperoside, quercetin, Epimedium A, Epimedium B, Epimedium C, icariin, Baohuoside I, Sagittatoside A, 2′‐O‐rhamnosylicariside II, icaritin, anhydroicaritin, and anhydroicaritin‐7‐O‐glucoside, were acquired from Shanghai Yuanye Biotechnology Co. Ltd. (Shanghai, China). Acetonitrile and UPLC–MS–grade formic acid was obtained from Fisher Scientific (Geel, Belgium). Pure water was obtained from Watsons Group Ltd. (Guangzhou, China). DPPH and ABTS were obtained from Nanjing Jiancheng Biology Engineering Institute Co. Ltd. (Nanjing, China). Penicillin–streptomycin solution, MEM α modification agents and pancreatic enzymes were obtained from HyClone (Logan, UT, USA). Fetal bovine serum was purchased from Clark (VA, USA). The MC3T3‐E1 cell line was acquired from Shanghai Fuheng Biotechnology Co. Ltd. (Shanghai, China). A Cell Counting Kit‐8 (CCK‐8) was obtained from Invigentech (Irvine, CA, USA).

2.2. Plant Materials

The leaves of Epimedium koreanum Nakai (Sample no. TA220572001) were collected from Jilin origin, China (41°66′ N, 125°99′ E) in 2022 and identified by Assoc. Prof. Guangzhi Cai of the School of the Pharmaceutical Sciences, Changchun University of Chinese Medicine. The specimens of plants were deposited in the Laboratory of Traditional Chinese Medicine Processing, Changchun University of Traditional Chinese Medicine.

2.3. Sample Preparation

After removing the impurities from the leaves of the EP plants, the leaves were moistened with water, and the moistened EP leaves were subsequently cut into 5–10 mm wide filaments and then dried in an oven at 45°C for 3 h to obtain an REP. The REP was heated at 150°C for 10 min to obtain an SEP, mixed with wine in a sealed container for 30 min (10:1, REP:wine, M:M) and heated at 150°C for 10 min to obtain wine‐processed EP (WEP), added to hot melted suet oil (5:1, REP:suet oil, M:M) and heated at 150°C for 10 min to obtain suet‐oil‐processed EP (SOEP), and mixed with wine in a sealed container for 30 min (10:1, REP:wine, M:M) and then added with hot melted suet oil (5:1, REP:suet oil, M:M) and heated at 150°C for 10 min to obtain wine‐and suet‐oil‐processed EP (WSOEP).

Take leaf samples, grind them and pass through a No. 3 sieve. Accurately weigh 0.2 g of the powdered material and place it in a stoppered conical flask. Precisely add 20 mL of dilute ethanol, weigh the flask again, and then subject the mixture to ultrasonic treatment (at a power of 400 W and frequency of 50 kHz) for 1 h. After cooling, reweigh the flask and compensate for any weight loss by adding more dilute ethanol. Mix thoroughly and filter the extract through a 0.22 µm microporous filter.

Twelve reference standards (hyperoside, quercetin, Epimedium A, Epimedium B, Epimedium C, icariin, Baohuoside I, Sagittatoside A, 2′‐O‐rhamnosylicariside II, icaritin, anhydroicaritin, and anhydroicaritin‐7‐O‐glucoside) were dissolved in methanol at a final concentration of 1 mg/mL as stock solutions.

2.4. Cell Culture

MC3T3‐E1 cells (CRL‐2594) were obtained from Shanghai Fuheng Biotechnology Co. Ltd. and cultured in α‐MEM supplemented with 10% fetal bovine serum at 37°C in humidified air containing 5% CO₂.

2.5. Animals

Adult wild‐type zebrafish (AB line) were obtained from Hangzhou Huante Biotechnology Co. Ltd. (Zhejiang, China), and the license number was SCXK (Zhe) 2022‐0003. The Animal Ethics Committee of Changchun University of TCM (approval number: 2020056) reviewed and approved the study before conducting the task. The fish were cared for following the protocols outlined in the zebrafish model biological database guidelines (http://zfin.org, April 3, 2020) and under a 14 h light/10 h dark rotation.

2.6. UPLC–QQQ–MS/MS Parameters of the Analysis

A triple–quadrupole linear ion trap mass spectrometer was used to determine the flavonoid levels in the processed EP samples. To separate the flavonoids, chromatography was performed on a Waters ACQUITY UPLC HSS T3 column (100 × 2.1 mm, 1.8 µm) at 40°C and a flow rate of 0.4 mL/min. The injection volume was set to 5 µL, and the mobile phases included acetonitrile (B) and water with 0.1% formic acid (A) as follows: 0–10 min, 15%–50% A; 10–15 min, 50%–80% A; 15–20 min, 80% A; 20–20.1 min, 15% A; 20.1–23 min, 15% A.

The QQQ–MS parameters were as follows: an electrospray ionization source in a positive ion mode, full scan (m/z 130–1500), detection of gas temperature at 300°C, a gas flow rate of 5 mL/min, an auxiliary gas flow rate of 10 arb, a sheath gas flow rate of 30 arb, and a spray voltage of 3500 V.

2.7. DPPH Assay

The DPPH radical scavenging activity of the samples was determined as described above, with certain modifications [23]. The known antioxidant VC was used as the positive control. Briefly, the samples and the DPPH ethanol solution were placed in 96‐well plates, and the mixture was allowed to react for 30 min in the dark. Scavenging activity was subsequently measured using a multimode reader at a wavelength of 517 nm. The following equation was used to determine the rate of scavenging:

$$\mathrm{Scavenging}\mathrm{effect}\left(\%\right)=\left[1-\left(A_1-A_2\right)/A_0\right]\times 100\%)$$

where A ₁ is a mixture of 100 µL of the sample and 100 µL of DPPH solution, A ₂ is the blank, and A ₀ is the control.

2.8. Cell Grouping and Drug Administration

There are 17 cell groups, as follows: the control group (Con), model group (Mod, medium containing 100 µM H₂O₂), low‐dose REP group (medium containing 100 µM H₂O₂ and 6.25 µg/mL REP), moderate‐dose REP group (medium containing 100 µM H₂O₂ and 12.5 µg/mL REP), high‐dose REP group (medium containing 100 µM H₂O₂ and 25 µg/mL REP), low‐dose SEP group (medium containing 100 µM H₂O₂ and 6.25 µg/mL SEP), moderate‐dose SEP group (medium containing 100 µM H₂O₂ and 12.5 µg/mL SEP), high‐dose SEP group (medium containing 100 µM H₂O₂ and 25 µg/mL SEP), low‐dose WEP group (medium containing 100 µM H₂O₂ and 6.25 µg/mL WEP), moderate‐dose WEP group (medium containing 100 µM H₂O₂ and 12.5 µg/mL WEP), high‐dose WEP group (medium containing 100 µM H₂O₂ and 25 µg/mL WEP), low‐dose SOEP group (medium containing 100 µM H₂O₂ and 6.25 µg/mL SOEP), moderate‐dose SOEP group (medium containing 100 µM H₂O₂ and 12.5 µg/mL SOEP), high‐dose SOEP group (medium containing 100 µM H₂O₂ and 25 µg/mL SOEP), low‐dose WSOEP group (medium containing 100 µM H₂O₂ and 6.25 µg/mL WSOEP), moderate‐dose WSOEP group (medium containing 100 µM H₂O₂ and 12.5 µg/mL WSOEP), and high‐dose WSOEP group (medium containing 100 µM H₂O₂ and 25 µg/mL WSOEP).

2.9. Cell Proliferation

In accordance with the aforementioned criteria, cells were seeded in 96‐well culture plates at a concentration of 1 × 10⁵ cells per well. Once the cell density reached 80%, the drug was administered for 1 h, after which 100 µm H₂O₂ was added, and the mixture was incubated for 4 h. The cell proliferation characteristics were analyzed using a CCK‐8 kit, and the light absorption data were collected at 450 nm. Cells were processed according to the manufacturer's instructions of the CCK‐8 kit.

2.10. Calcein Staining

Calcein is a fluorescent marker commonly used to assess mineral growth in organisms [24]. Zebrafish embryos at 3 dpf—were randomly segregated into 12 separate groups, with each group containing 20 fish: the control group (Con), model group (Mod, medium containing 100 µM H₂O₂), etidronate disodium (ED) group (ED, medium containing 2.5 mM H₂O₂ and 30 µg/mL ED), low‐dose REP group (REP‐L, medium containing 2.5 mM H₂O₂ and 25 µg/mL REP), high‐dose REP group (REP‐H, medium containing 2.5 mM H₂O₂ and 50 µg/mL REP), low‐dose SEP group (SEP‐L, medium containing 2.5 mM H₂O₂ and 25 µg/mL SEP), high‐dose SEP group (SEP‐H, medium containing 2.5 mM H₂O₂ and 50 µg/mL SEP), low‐dose WEP group (WEP‐L, medium containing 2.5 mM H₂O₂ and 25 µg/mL WEP), high‐dose WEP group (WEP‐H, medium containing 2.5 mM H₂O₂ and 50 µg/mL WEP), low‐dose SOEP group (SOEP‐L, medium containing 2.5 mM H₂O₂ and 25 µg/mL SOEP), high‐dose SOEP group (SOEP‐H, medium containing 2.5 mM H₂O₂ and 50 µg/mL SOEP), low‐dose WSOEP group (WSOEP‐L, medium containing 2.5 mM H₂O₂ and 25 µg/mL WSOEP), and high‐dose WSOEP group (WSOEP‐H, medium containing 2.5 mM H₂O₂ and 50 µg/mL WSOEP). The 5 dpf larvae were fed twice a day. The medium was changed 1 h after each feeding, and the medications were administered. After 1 h, 2.5 mM H₂O₂ was added. The larvae were consistently nourished until they reached a density of 9 dpf. The larvae were immersed in calcein staining agent (2 mg/mL) for 10 min in the dark and subsequently washed with medium three times (for 5 min) until the staining agent was removed. Fluorescence signals corresponding to calcein‐stained spine development were observed by fluorescence microscopy.

2.11. Active Ingredient Verification

2.11.1. Compound and Disease Target Screening

The Swiss Target Prediction database (http://www.swisstargetprediction.ch/) was used to import the filtered compounds, and the predictions for each compound with a probability greater than zero were maintained. The term “osteoporosis” was used to search for disease‐related targets in GeneCards (https://genecards.org) and OMIM (https://omim.org). Using Venny 2.1 (https://bioinfogp.cnb.csic.es/tools/venny/index.html), the target genes of the substances and the target genes of osteoporosis were displayed in a Venn diagram to determine the gene overlap between the two conditions.

2.11.2. Intersection Target Network Construction and Analysis

The STRING database (https://www.string‐db.org) was used to import the common targets for Homo sapiens, and the PPI results were acquired. To construct a PPI network, the results were visualized using the Cytoscape 3.7.2 software.

2.11.3. Enrichment Pathway Analysis

The DAVID 6.8 platform (https://david.ncifcrf.gov/) was used to identify GO functions and pathways associated with the overlapping genes. GO enrichment analysis and KEGG pathway enrichment analysis were performed. The results were then graphically shown in a bubble dot diagram on the website (www.bioinformatics.com.cn).

2.11.4. Molecular Docking

Protein data were acquired from the protein database (https://www.rcsb.org/) to obtain the UniProt ID of the main target protein receptor conformation. Using the CB Dock platform (https://clab.labshare.cn/cb‐dock/), the primary target protein structure was chosen for molecular docking with each of the four chemicals.

2.12. Analysis of the Spectrum–Effect Relationships

Origin and SPSS were used to conduct Pearson correlation analysis, aiming to establish the correlation between common peaks and activity, while also providing further insights into the quality of the EPs.

2.13. Statistical Analysis

All results were expressed as mean ± SD of three independent experiments. Statistical analysis was performed using one‐way ANOVA by GraphPad Prism 8 software, with p < 0.05 considered statistically significant.

3. Results

3.1. Variations in the Total Flavonoid Content

The results showed the total flavonoid contents (TFCs) in the different processed products of EP (Figure 1). Our findings revealed that processed EP contained more TFC compared with REP. In the REP, the TFC was lower than the others. The study revealed the following trend: WEP (5.86 ± 0.08 mg/g) > WSOEP (5.69 ± 0.05 mg/g) > SOEP (5.56 ± 0.14 mg/g) > SEP (5.55 ± 0.04 mg/g) > REP (5.29 ± 0.07 mg/g).

FIGURE 1.

Changes in the contents of total flavonoids of Epimedium during processed. Experiments were performed in triplicate, and the data were expressed as the mean ± SD, n = 3.

3.2. Variations in the Flavonoid Contents

The changes in the chemical components of REP during processing were assessed using UPLC–QQQ–MS. The representative UPLC–QQQ–MS chromatograms of the reference compounds and samples are shown in Figure 2. The contents of the samples varied, but no difference was found between the original and processed samples (Figure 3). Processing increased the hyperoside and quercetin contents. Hyperoside and quercetin are natural flavonoids found in plants that protect cells from ultraviolet radiation induced oxidative damage and inflammation [25]. Epimedium A, Epimedium B, Epimedium C, icariin, Baohuoside I, and Sagittatoside A share a common parent nuclear structure. The contents of Epimedium A, Epimedium B, Epimedium C, icariin, and Baohuoside I increased during processing, and the contents of the processed samples were approximately two times greater than those of the raw samples. The difference in the content of Sagittatoside A was not obvious between the raw and processed samples. The contents of 2′‐O‐rhamnosylicariside II, icaritin, and anhydroicaritin decreased during processing. The anhydroicaritin‐7‐O‐glucoside concentration was the highest in the SOEP samples but was lower in REP, SEP, WEP, and WSOEP.

FIGURE 2.

MRM chromatograms of 12 compounds (A), BPI overlapping map of reference (B), and samples (C) mass spectrometry results.

FIGURE 3.

The 12 contents of raw and processed Epimedium. Experiments were performed in triplicate, and the data were expressed as the mean ± SD, n = 3.

3.3. DPPH Radical Scavenging Activity

The studies used varying concentrations (25–200 µg/mL) of raw and processed EP, and the results are shown in Figure 4. The scavenging rate of high dose raw and processed EP can reach the scavenging rate of low dose VC. The higher the sample dose was, the more effective was the DPPH free radical scavenging activity.

FIGURE 4.

The results of DPPH radical scavenging activity. Experiments were performed in triplicate, and the data were expressed as the mean ± SD, n = 3.

3.4. Determination of MC3T3‐E1 Cell Proliferation

During the induction of 100 µM H₂O₂, cell proliferation was significantly affected. Compared with those in the Mod group, the proliferation of the WEP, SOEP, and WSOEP groups was significantly different. Moreover, there was a correlation between cell proliferation and concentration (Figure 5).

FIGURE 5.

Effect of raw and processed Epimedium on H₂O₂‐induced proliferation of MC3T3‐E1 cell. Experiments were performed in triplicate, and the data were expressed as the mean ± SD, n = 5, compared with Mod group, *p < 0.05, **p < 0.01.

3.5. Raw and Processed EP Samples Ameliorate the Mineralization of H₂O₂‐Induced Osteoporotic Zebrafish Larvae

Calcein was used to stain the 9 dpf zebrafish (Figure 6). Calcein staining revealed that osteoporosis was induced after 2.5 mM H₂O₂ exposure. In contrast to that in the Mod group, green fluorescence was observed in the spinal cord in the Con group (Figure 6A). After treatment with ED, REP and processed EP samples, the intensity of green fluorescence in the spinal cord was significantly enhanced (Figure 6B).

FIGURE 6.

Effect of raw and processed Epimedium on bone tissue mineralization of H₂O₂‐induced osteoporosis zebrafish larvae. (A) Nine dpf zebrafish osteogenesis staining experiment. (B) Quantitative analysis of fluorescence intensity of vertebrate column mineralization. The experiment was conducted thrice. The mean ± SD values were used to express the data, n = 3, which were then compared to the Mod group, *p < 0.05, **p < 0.01.

3.6. Pearson Correlation Analysis

Pearson's correlation coefficient was used to quantify the degree of colocalization between paired data. A positive correlation occurs when the correlation coefficient (r) is between +0.3 and +1.0, whereas a negative correlation occurs when the correlation coefficient (r) is between −0.1 and −0.9. The results are shown in Figure 7. The presence of the components hyperoside, quercetin, Epimedium B, icariin, Baohuoside I, Sagittatoside A, 2′‐O‐rhamnosylicariside II, icaritin, anhydroicaritin, anhydroicaritin‐7‐O‐glucoside in EP and the processed EP was positively correlated with the scavenging activity values of DPPH. The concentrations of quercetin, Epimedium A, Epimedium B, icariin, Baohuoside I, and anhydroicaritin‐7‐O‐glucoside in EP and the processed EP were positively correlated with cell proliferation. The fluorescence intensities of the components hyperoside, quercetin, Epimedium B, Epimedium C, Baohuoside I, and anhydroicaritin‐7‐O‐glucoside in EP and the processed EP were positively correlated with bone tissue mineralization. The screened compounds were hyperoside, quercetin, Epimedium A, Epimedium B, icariin, Baohuoside I, Sagittatoside A, 2′‐O‐rhamnosylicariside II, icaritin, anhydroicaritin, and anhydroicaritin‐7‐O‐glucoside. Taken together, the data presented in Figure 3 indicate that quercetin, Epimedium B, Baohuoside I, and anhydroicaritin‐7‐O‐glucoside are the primary bioactive components of EP and that processed EP is involved in the antioxidant and antiosteoporosis effects.

FIGURE 7.

Heatmap analysis of Pearson correlation of the chemical compounds and bioactivity.

3.7. Screening Results for Ingredients and Disease Targets

SwissTargetPrediction was used to retrieve the four ingredient targets. After eliminating multiple targets, 46 objectives that met the criteria were achieved. The GeneCards and OMIM databases were searched for “osteoporosis” to obtain the target data. To obtain a total of 3099 disease‐related targets, the information from the two databases was merged and duplicate entries were eliminated. A Venn diagram was drawn using Venny 2.1, the details of which are shown in Figure 8. A total of 16 intersection targets were obtained after the intersection of ingredient targets and disease‐related targets.

FIGURE 8.

Ingredients and disease overlap in Venn diagram.

3.8. Results of Building PPI Networks and Screening of Core Objectives

The PPI network consisted of 15 nodal points and 38 edges. The interactions between targets are shown in Figure 9. With an increasing degree of core targets, the size and color of the nodes become deeper, which means that the corresponding nodes are more crucial in the network. On the basis of the mean, proximity, and betweenness, four potential core targets, as defined in Table 1, were obtained, namely, AKT1, ABCB1, CYP19A1, and EGFR. It has been speculated that EP is mainly responsible for the antiosteoporosis effect through these targets.

FIGURE 9.

Intersection targets PPI network diagram.

TABLE 1.

Core targets of EP in treatment of osteoporosis.

No.	Core targets	Degree	Betweenness	Closeness
1	AKT1	10.0	40.71429	0.05556
2	ABCB1	7.0	38.64762	0.04545
3	CYP19A1	6.0	34.64762	0.04348
4	EGFR	8.0	18.67619	0.05000

3.9. Enrichment Analysis of Target Pathways

The overlapping targets were subjected to GO enrichment analysis, and entries with a p value of 0.05 were selected. The findings showed that 76 biological processes (BP), 14 cellular components (CC), and 26 molecular function terms were enriched. Arrange each of them according to their level of importance. Using a bar chart and bubble graph, the first 10 items were chosen to construct the bar chart, and the results are displayed in Figure 10. The size of the circle or the length of the line in the GO graph signifies the number of genes that are enriched according to the GO analysis. Hue symbolizes the concept of enrichment. Certain BP, such as the insulin receptor signaling pathway, xenobiotic metabolic process, and glucose homeostasis, may be involved in the mechanism of EP in preventing osteoporosis. Numerous CC, such as the insulin receptor complex, plasma membrane, and integral components of the membrane, are involved in this process. The relevant molecular functions were heme binding, oxidoreductase activity, and aromatase activity.

FIGURE 10.

Bar chart (A) and bubble chart (B) from GO pathway enrichment analysis.

The KEGG pathway enrichment analysis revealed that the DEGs were significant, and the top 20 enrichments are shown in Figure 11. In the KEGG graph, the number of KEGG‐rich genes is expressed by the size of the circle or the length of the line. The color indicates the level of enrichment. The analysis demonstrated that the genes could be expressed through various routes, such as chemical carcinogenesis‐reactive oxygen species, ovarian steroidogenesis, endocrine resistance, the HIF‐1 signaling pathway, and the FoxO signaling pathway, and that the targets were strongly enriched. The figure shows that the effect of EP was positively correlated with chemical carcinogenesis and reactive oxygen species levels and negatively correlated with prostate cancer incidence.

FIGURE 11.

KEGG pathway enrichment.

3.10. Molecular Docking

The kinetics of the molecules and the free energy of binding to target proteins are often used to calculate the binding energy. It is generally assumed that the greater the conformational stability of ligands and receptors is, the lower is the binding energy. The results of molecular docking are displayed in Table 2, and Figure 12 shows the ideal conformations with the lowest binding energies. The findings showed that the compounds displayed robust affinity for crucial targets, demonstrating a low binding energy to both AKT1 and EGFR.

TABLE 2.

Molecular docking energy (kcal/mol).

Active ingredients	AKT1	ABCB1	CYP19A1	EGFR
Quercetin	−5.53	−4.25	−4.04	−6.68
Epimedium B	−4.69	−1.56	−0.35	−2.87
Baohuoside I	−5.17	−4.18	−3.39	−6.65
Anhydroicaritin‐7‐O‐glucoside	−5.71	−4.3	−5.77	−5.83

FIGURE 12.

The optimal conformations with low binding energies.

4. Discussion

Osteoporosis is a result of bone aging, other environmental factors, and a genetic predisposition [26, 27, 28]. Therefore, in the present study, the antioxidant capacities of REP and processed EP were studied, and the results showed that EP and processed EP suppressed MC3TE resistance to oxidative stress to promote proliferation. Furthermore, the results showed that the effect of EP improved after processing.

KGB is a classical theory in TCM that hypothesizes that function of the kidney is responsible for bone health. Ancient individuals proposed the “kidney storing essence, essence producing marrow, and marrow filling bones” through the discovery of anatomy and long‐term observation of physiological activities [29]. The OPG–RANKL–RANK pathway is considered one of the mechanisms of “KGBs” and can be used to treat osteoporosis by microadjustment of the microenvironment of the bone marrow [30]. Oxidative stress damage is the main pathogenesis of osteoporosis [28]. The accumulation of reactive oxygen species in the bone microenvironment plays a role in osteoblast and osteoclast apoptosis [31, 32]. The DPPH radical scavenging activity, cell proliferation, and intensity of green fluorescence in the spinal cord were measured in zebrafish to determine the effects of drug antioxidation.

Zebrafish experiments offer a cost‐effective and adaptable approach, making them a valuable model species for assessing the efficacy of chemotherapy for treating skeletal dysplasia in humans [33]. Similar to mammals, zebrafish are teleosts and possess numerous genetically conserved traits in terms of bone elements, ossification mechanisms, and bone matrix components [34]. Research has indicated the use of zebrafish embryos in osteoporosis models [35]. Isocoumarin A enhances the number and size of the vertebrae during zebrafish embryo development and could promote the treatment of osteoporosis [36]. The use of Xian‐ling‐gu‐bao effectively prevents osteoporosis in zebrafish embryos [37]. Drawing from the aforementioned studies, zebrafish embryos were used to demonstrate the efficacy of EP in treating osteoporosis.

In these experiments, the antioxidant effects of EP and processed EP were investigated. The results showed that all of the test samples had low to high DPPH scavenging activities, which increased with increasing doses. The proliferation of the cells in the WEP, SOEP, and WSOEP groups was significantly greater than that in the model group, and the difference was concentration dependent. The amount of bone mineralization was greater in both the REP and processed EP groups than in the Mod group, and the intensity of green fluorescence in the spinal tract was greater following treatment with REP and processed EP.

The spectrum–effect relationship has emerged as a rational and scientific approach for investigating the pharmacodynamic substance basis of TCM. The chemical composition of TCMs is connected to their therapeutic effects using mathematical models. The Pearson correlation coefficient is one of the most commonly used measures of linear correlation. The purpose of the Pearson correlation coefficient is to determine the relationship between the various subelements of the system through a certain method [38, 39]. On the basis of the results of the Pearson's correlation, we identified the material basis for the efficacy of REP in treating osteoporosis. The results showed that compared with that of resveratrol, the pharmacodynamic correlation of each component with the preparation of EP was greater. Quercetin, Epimedium B, Baohuoside I, and anhydroicaritin‐7‐O‐glucoside may be the major factors influencing the efficacy of EP before and after osteoporosis treatment. Quercetin is a major member of the naturally occurring flavonoid family, can improve bone pathology and bone‐related parameters under imaging and maximum bone loading [40], can activate the Nrf2/HO‐1 signaling pathway, and can attenuate oxidative stress damage [41]. Epimedium B, a major ingredient of the herb EP, regulates the PI3K–Akt, MAPK, and PPAR signaling pathways to treat osteoporosis [42]. Baohuoside I is the major active compound in EP, and has excellent therapeutic effects on various diseases, inhibits osteoclast differentiation, and prevents bone loss following ovariectomy by ameliorating the activation of the MAPK and NF‐κB pathways and reducing the expression of uPAR [43, 44]. Anhydroicaritin‐7‐O‐glucoside is generated by removing the rhamnose group from icariin, which enhances the regulation of the downstream genes of the BMP/Runx2/Osx signaling pathway to promote the osteogenic differentiation of BMSCs [45].

In this study, network pharmacology was employ used to analyze and investigate potential targets and pathways for the treatment of osteoporosis using EP, which is found in four active ingredients. Four protein targets were identified through screening, and on the basis of the ingredient–target relationships, the related biological activities and possible routes were investigated and evaluated. Subsequently, molecular docking was used to initially validate the binding of the compound–protein, interactions and the compound molecules exhibited a strong affinity for the anticipated protein receptors. The AKT1 protein is predominantly expressed in skeletal muscle, and studies have shown that regulation of AKT1 can enhance osteoblast differentiation and alleviate osteoporosis [46], the ABCB1 protein is primarily expressed in biomembrane formation, and its overexpression can lead to drug resistance. Research indicates that ABCB1 significantly influences the pathogenesis of bone diseases, though there are no reports specifically linking it to osteoporosis [47, 48], CYP19A1 is a key microsomal enzyme that catalyzes the conversion of androgens to estrogens, which play a crucial role in bone metabolism [49], it has also been shown to interact with flavonoid components [50], and studies suggest that regulation of CYP19A1 can be used in the treatment of osteoporosis [51], EGFR regulates epithelial tissue development and homeostasis, and research has demonstrated that modulation of EGFR can slow the aging of bone progenitor cells and increase bone formation on the endosteal surface of cortical bone [52].

Osteoporosis is becoming a major concern in the field of public health, and oxidative stress is a common pathological state in populations at a high risk of osteoporosis [53]. Flavonoids with anti‐inflammatory effects and improved oxidative stress may regulate bone homeostasis and could be a new method for the treatment of osteoporosis. There are several limitations to this study. First, we did not perform pharmacodynamic studies of the EP‐predicted monomeric components of migraine via network pharmacology. However, additional experiments related to signaling pathways remain to be conducted.

5. Conclusion

In this study, we compared the differences between REP and processed EP using the spectrum–effect relationship to screen substances with enhanced potency after processing. The results showed that both REP and processed EP had antioxidant and antiosteoporosis effects. The antiosteoporosis effects of WEP, SOEP, and WSOEP were greater than those of REP, and the main difference was reflected in cell proliferation. Four components were selected as potential substances for osteoporosis treatment through spectrum–effect relationship studies. After processing, the content of Epimedium B decreased, which may be due to the breakage of glycosidic bonds and conversion to other components. Subsequently, the transformation law will be confirmed by simulated processing. Molecular docking and network pharmacology studies have confirmed that these four active components can have an antiosteoporosis effect on various targets and pathways. The findings of this study emphasize the importance of TCM as a multicomponent, multifunction system. We will subsequently conduct activity studies on the ratios of these four components in each processed EP. Such research can provide further information and theoretical support for understanding the anti‐osteoporotic mechanisms of EP and its processed EP.

Author Contributions

Jinling Liang and Jia Liu: writing – original draft preparation, experiment, data processing. Weixia Sun: writing – review and editing. Yulin Dai: writing – review and editing. Pan Li: writing – review and editing. Huaizhu Sun: writing – review and editing. Wenxuan Cao: writing – review and editing. Peng Yu: writing – review and editing, funding acquisition. Tianyang Xu: writing – review and editing. All authors have read and agreed to the published version of the manuscript.

Conflicts of Interest

The author declares no conflicts of interest.

Liang J., Liu J., Sun W., et al. “Study on the Screening of Anti‐Osteoporosis Components in Processed Epimedium‐Based on Spectrum‐Effect Analysis.” Chemistry & Biodiversity 22, no. 11 (2025): e02936. 10.1002/cbdv.202402936

Data Availability Statement

Data will be made available on request.