Macranthoidin B restrains the epithelial-mesenchymal transition through COX-2/PGE2 pathway in endometriosis

In traditional Chinese medicine, Lonicera macranthoides Hand. –Mazz shows its anti-inflammatory properties in the clinical treatment of EMS ( Liu and Shi, 2020 ; Chen and Xu, 2022 ). Previously, both its extract and its saponin metabolites downregulate COX-2 and PGE 2 in inflammatory models and macrophage cells ( Guan et al., 2014 ; Zeng et al., 2020a ). Macranthoidin B, one of the main triterpenoid saponin metabolites, exhibits limited activities in cancer, such as the proliferative reduction, apoptosis and oxidant stress enhancements ( Fan et al., 2018 ; Tan et al., 2023 ). However, the effect and mechanism of macranthoidin B had not been clarified in EMS. EMS is an estrogen-dependent inflammatory disorder of the endometrium that is characterized by the presence of functionally active endometrial tissue growing outside the uterus. High E 2 and dysregulated PROG productions are the consistently observed endocrine feature of EMS ( Chantalat et al., 2020 ; MacLean and Hayashi, 2022 ). EMS is associated with the increase in the risk of epithelial ovarian cancer ( Vercellini et al., 2014 ). Now, nonsteroidal anti-inflammatory drugs (NSAIDs) and hormonal therapies are the main therapeutic options in clinic ( Taylor et al., 2021a ). As a prevalent gynecological disorder, the activation of epithelial-mesenchymal transition promotes invasion and metastasis in EMS ( Zhang C. et al., 2021 ). The process of epithelial-mesenchymal transition is characterized by a decrease in E-cadherin and an increase in Vimentin, N-cadherin, Twist, Snail, Slug, and Zeb1/2, which allows cells to invade and metastasize ( Debnath et al., 2022 ). As a common gynecological inflammatory disease, EMS is accompanied with COX-2/PGE 2 pathway upregulation ( Taylor et al., 2021a ). In recent years, COX-2/PGE 2 pathway has been implicated in promoting epithelial-mesenchymal transition of tumors ( Zhang H. et al., 2021 ; Zhao et al., 2024 ). However, it is unknown whether COX-2/PGE 2 pathway regulates epithelial-mesenchymal transition in EMS. This study aimed to determine the therapeutic effects of macranthoidin B on EMS in rat EMS model and endometrial cells. The impacts of macranthoidin B on COX-2/PGE 2 pathway, epithelial-mesenchymal transition, invasion and metastasis were investigated both in vivo and in vitro . Additionally, the effects of macranthoidin B, combined with a COX-2/PGE 2 pathway activator or inhibitor, were assessed separately in epithelial-mesenchymal transition, invasion and metastasis.

Macranthoidin B (20200822, purity ≥98%, Nanjing Spring and Autumn Biological Engineering Co., Ltd., China) was dissolved separately in physiological saline or PBS for rat administration or endometrial cells treatment ( Figure 1A ). LPS (BS904, purity ≥98%, Sigma-Aldrich, United States) or celecoxib (EA9450, Pfizer Inc., United States) was dissolved in PBS or DMSO, respectively. Gestrinone was selected as a standard (53200201, Zizhu Pharmaceutical Co., Ltd., Beijing, China). Macranthoidin B inhibited the growth of endometriotic tissues in vivo (A) The chemical structure of macranthoidin B (B, C) The volume of isolated ectopic lesions were calculated by vernier caliper after 28 days treatment. (D) Using H&E staining, pathological changes were observed. (E, F) E 2 and PROG levels in serum were investigated by ELISA assay. $ P < 0.05 to pretreatment, # P < 0.05 to control, ## P < 0.01 to control, * P < 0.05 to EMS, ** P < 0.01 to EMS. Columns, mean (n = 5). Bars, SD. Scale bar = 100 μm. EMS, endometriosis; GTN, gestrinone. Female SD rats, weighing 180–220 g and aged 6–7 weeks, were purchased from Hunan Slake Jingda Experimental Animal Co., Ltd. [Certification No: SCXK (Xiang) 2019-0004]. The rats were housed at a temperature of 20°C ± 2°C with a 12-h light/dark cycle and had free access to food and water at the Experimental Center, College of Pharmaceutical Sciences, Southwest University. The experimental protocol was approved by the Experimental Animal Ethics Review Committee of Southwest University (Approval No. IACUC-20210130-04), ensuring compliance with anesthesia and humane methods to minimize suffering. The molding method and criteria of autograft EMS model were established according to the previous protocols ( Dai et al., 2023 ). In summary, the left uterus of female rats was ligated, excised, and placed in a saline dish. It was then cut into approximately 5 mm 2 endometrial segments, which were sutured to the right abdominal wall. After 28 days, the volume of ectopic tissue was measured by a vernier caliper with the formula (0.52 × length × width × height). The autograft section grew very well, for example, volume >8 mm 3 , blood vessels, and inflammatory encapsulation. It indicated the successfully established model. 30 female SD rats were operated for EMS establishment. The successful 25 EMS rats were randomly divided into EMS, 3, 9, 27 mg kg −1 macranthoidin B, and 0.5 mg kg −1 gestrinone groups. There were 5 rats per group. There were no significant differences in endometriotic volume among the groups before treatment. Another 5 normal female rats were treated as control group. Physiological saline was administered in control and EMS groups. All above groups were administered consecutive 28 days by gavage. For primary endometriotic stromal cells (ESCs), methods of cell acquisition and culture followed those outlined in a previous study ( Dai et al., 2023 ). The study was approved by the First Affiliated Hospital of Chongqing Medical University (Permission 2019-059). Briefly, fresh ovarian EMS tissues were minced and digested by 0.1% collagenase I (Sigma-Aldrich, United States) for 90 min at 37°C. After separated by a 400 screen mesh, primary ESCs were cultured within DMEM/F12 medium (Gibco, NY, United States) supplemented with 10% FBS (Hyclone, Logan, United States) at 37°C with 5% CO 2 . The endometrial cancer cell line, named HEC1-B, was purchased from Chinese Type Culture Collections (Wuhan, China). HEC1-B cells were cultured in MEM medium (Gibco, Grand Island, NY, United States) with 10% FBS (Hyclone, Logan, United States) at 37°C with 5% CO 2 . For H&E staining, eutopic endometrium and ectopic endometrium were collected and fixed in paraformaldehyde in the end of administration. Then sections were stained and photographed under a microscope (DFC310 FX, Leica, Germany). For ELISA assay, E 2 and PROG detection in the serum proceed according to the published paper ( Zhang C. et al., 2021 ). The PGE 2 levels of ectopic endometrial tissue or cell supernatant were examined by rat or human PGE 2 ELISA kits. All kits were used according to the instructions (SinoBestBio, Shanghai, China). For immunofluorescence staining, rabbit anti-E-cadherin or anti-Vimentin antibodies (1:200 dilution; Proteintech, China) were applied to the endometrial tissue. Afterward, the tissues were treated with Alexa Fluor 594 or FITC-labeled goat anti-rabbit IgG secondary antibody (1:100 dilution; Beyotime, China). The sections were then counterstained with DAPI (Beyotime, China) and analyzed under a Leica microscope (Germany). The previously reported method was employed for cell viability assay ( Dai et al., 2023 ). Briefly, cells in 96-well plate were treated with macranthoidin B for 24 h. Then, MTT was used for incubation, and a microplate reader was performed for measurement. The 24 h IC 50 values were calculated by GraphPad Prism Software (7.04 version, United States). The scratch wound and Transwell assays were conducted following the previous methodology ( Zhang C. et al., 2021 ). For scratch wound assay, 6-9×10 4 cells were cultured in 24-well plate until the 80% density. After being scratched by 1 mL pipette tip, cells were treated with macranthoidin B and observed under a microscope. Migration rate = (average scratch area at 0 h - average scratch area at 24 h)/average scratch area at 0 h × 100 %. For the Transwell assay, the upper chamber included 3-5×10 4 endometrial cells and serum-free medium in matrigel-coated Transwell inserts (Corning, New York, United States), while 10% FBS medium was added to the lower chamber. After 24 h, cell numbers in the bottom were fixed and counted in 3-5 random fields at ×100 magnification. RNA isolation and cDNA synthesis were adhered to the previous approach ( Zhang C. et al., 2021 ). SYBR™ Green Master Mix (Thermo Fisher, Waltham, United States) was used for gene expression with 2 −ΔΔCT calculation. Primer sequences of mRNA were produced by Shenggong Biotechnology (Shanghai, China) ( Table 1 ). Primer sequences for RT-qPCR. Briefly, the protein of cells and tissues were lysed and extracted by RIPA (Dingguo Changsheng, China). Then the proteins were separated by SDS-PAGE, and transferred to PVDF membrane (Millipore, Billerica, United States). Proteins on the membrane were bound with the primary antibodies and HRP-labeled goat anti-rabbit secondary antibody ( Table 2 ). The membranes were reacted with electrochemiluminescence (ECL) enhanced Western blotting Substrate (Bioground Biotechnology Co, Ltd., China) for chemiluminescence. Finally, the protein bands were visualized in Tanon 5200 imaging system (Tanon, China). Antibody information for Western blotting. All data were represented by mean ± SD. Two sets of data were compared with unpaired t -tests. The volume of ectopic endometrium pre- and post-treatment were compared with paired t -tests. Three or more sets of data were compared with one-way ANOVA. SPSS 26.0 statistical software was used for analysis. P < 0.05 indicated a significant difference.

After continuous gavage for 28 days, the volume of ectopic endometrium was evaluated and compared with the pretreatment. Administration of 9 or 27 mg kg −1 macranthoidin B resulted in smaller transplants with less adhesion and fewer surface blood vessels. Gestrinone diminished the volume of ectopic endometrial tissues ( Figures 1B, C ). In H&E staining, the ectopic endometrium in EMS exhibited a similar structure to the normal endometrium in control. The cortex invagination formed more pseudo glands with abundant blood vessels and inflammatory cell infiltrations. Macranthoidin B or gestrinone ameliorated the ectopic pathological morphology, with fewer and smaller pseudo glands, fewer micro vascularity and inflammatory infiltrations ( Figure 1D ). In EMS, E 2 in serum were remarkably enhanced, accompanied with the decrease of PROG. Macranthoidin B diminished E 2 and raised PROG levels ( Figures 1E, F ). Notably, macranthoidin B had no influence on rat weight during the treatment ( Supplementary Figure S1A ). The expression of the epithelial gene Cdh1 was low in the EMS group, while the mesenchymal genes were increased, such as Cdh2 , Vim , Twist , Slug , Snail , and Zeb1/2 . Macranthoidin B obviously reversed the change of gene expression ( Figures 2A–D ). Furthermore, the E-cadherin protein was remarkably reduced in the EMS group, while N-cadherin and Vimentin proteins were increased. Macranthoidin B significantly promoted the protein levels of E-cadherin, accompanied by downregulation of N-cadherin and Vimentin ( Figures 2E–G ). In immunofluorescence assay, the downregulation of E-cadherin and the upregulation of Vimentin displayed in the ectopic tissues of EMS. E-cadherin was subsequently restored following macranthoidin B treatment, accompanied with the inhibition of Vimentin ( Figure 2H ). Macranthoidin B downregulated epithelial-mesenchymal transition and COX-2/PGE 2 pathway in endometriotic tissues. (A–D) The mRNA levels of Cdh1 , Cdh2 , Vim , Twist , Slug , Snail , and Zeb1/2 were measured by RT-qPCR. (E–G) The protein levels of E-cadherin, N-cadherin, and Vimentin were detected by Western blotting assay. (H) Detection of E-caderin and Vimentin in ectopic tissue by immunofluorescence assay. (I) The gene expression of Ptgs2 and Ptges were tested by RT-qPCR. (J–L) The contents of COX-2 and PGE 2 were analyzed by Western blotting and ELISA assays. # P < 0.05 to control, ## P < 0.01 to control, * P < 0.05 to EMS ** P < 0.01 to EMS. Columns, mean (n = 3). Bars, SD. Scale bar = 100 μm. EMS, endometriosis. Similarly, the mRNA and protein levels of COX-2/PGE 2 pathway were elevated in EMS compared with control. Using macranthoidin B, the mRNA levels of PTGS2 and PTGES were obviously decreased ( Figure 2I ). This was accompanied by a reduction in COX-2 and PGE 2 protein contents ( Figures 2J–L ). According to the 24 h IC 50 , the primary ESCs or HEC1-B cells were treated with three concentrations of macranthoidin B ( Supplementary Figures S1B, C ). In primary ESCs, 2, 1, and 0.5 time of 24 h IC 50 were performed. HEC1-B cells were treated with 1/2, 1/4, and 1/8 of 24 h IC 50 . After 24 h treatment, cells were prevented to cicatrize ( Figures 3A, B, E, F ). Migrating cells across insert membrane were significantly decreased in macranthoidin B groups vs. control group ( Figures 3C, D, G, H ). All above data implied that cell invasion and metastasis were resisted by macranthoidin B, particularly at 250 μM in primary ESCs, at 360 μM in HEC1-B cells. The suppression of invasion and metastasis by macranthoidin B in endometrial cells. (A–D) After being treated with macranthoidin B, the primary ESCs were tested in scratch wound and Transwell assays. (E–H) The cell migration and invasion ability of HEC1-B cells were observed in scratch wound and Transwell assays. * P < 0.05 to control, ** P < 0.01 to control. Columns, mean (n = 3). Bars, SD. Scale bar = 100 μm. ESCs, endometriotic stromal cells. In primary ESCs, CDH1 gene expression was expanded at 250 μM macranthoidin B. The mRNA levels of CDH2 , VIM , TWIST , SNAIL , and ZEB2 were significantly reduced in macranthoidin B groups compared to control group ( Figures 4A–C ). Meanwhile, macranthoidin B exerted similar effect on HEC1-B cells ( Figures 4K–M ). Consequently, macranthoidin B increased the E-cadherin protein, and decreased the protein of mesenchymal markers, including N-cadherin and Vimentin in both types of endometrial cells ( Figures 4D–F, N–P ). Macranthoidin B downregulated epithelial-mesenchymal transition and COX-2/PGE 2 pathway in vitro . In primary ESCs (A–C) and HEC1-B cells (K–M) , the mRNA levels of CDH1 , CDH2 , VIM , TWIST , SNAIL , and ZEB2 were measured by RT-qPCR. (D–F, N–P) The protein levels of epithelial-mesenchymal transition were tested by Western blotting assay. (G, Q) The gene expression of PTGS2 and PTGES were investigated by RT-qPCR. (H–J , R–T) The content of COX-2 and PGE 2 were analyzed by Western blotting and ELISA assays. * P < 0.05 to control, ** P < 0.01 to control. Columns, mean (n = 3). Bars, SD. ESCs, endometriotic stromal cells. After using macranthoidin B for 24 h, there was a noticeable inhibitory effect on the gene expression of COX-2/PGE 2 pathway in two endometrial cells ( Figures 4G, Q ). Subsequently, the levels of COX-2 and PGE 2 proteins in endometrial cells were lessened in macranthoidin B groups compared with control group ( Figures 4H–J, R–T ). LPS is a common inflammatory activator, which can promote COX-2, PTGES, and PGE 2 ( Alqinyah et al., 2023 ). In two types of endometrial cells, LPS stimulated the COX-2/PGE 2 pathway, leading to epithelial mesenchymal transformation and enhancing invasive metastasis. Macranthoidin B reversed the LPS-activated COX-2/PGE 2 pathway and epithelial-mesenchymal transition ( Figures 5A, B ). At the same time, invasion and metastasis were inhibited by macranthoidin B in scratching wound and Transwell assays ( Figures 5C–J ). Therefore, macranthoidin B exhibited the antagonistic effect on COX-2/PGE 2 pathway, epithelial-mesenchymal transition, invasion and metastasis activated by LPS. Macranthoidin B reversed LPS-induced activation of invasion and metastasis. (A, B) After using LPS or macranthoidin B, the expression of COX-2/PGE 2 and epithelial-mesenchymal transition pathways were observed by Western blotting. (C–J) Cell migration and invasion ability were measured by scratch wound and Transwell assays in 24 h * P < 0.05 to control, ** P < 0.01 to control. # P < 0.01 to LPS. Columns, mean (n = 3). Bars, SD. Scale bar = 100 μm. ESCs, endometriotic stromal cells. Celecoxib is considered as a selective COX-2 inhibitor ( Porat et al., 2023 ). In our study, celecoxib suppressed COX-2/PGE 2 pathway and epithelial-mesenchymal transition in endometrial cells. Then, celecoxib restricted migration and invasion. Simultaneously, macranthoidin B not only enhanced celecoxib-mediated inhibition of COX-2/PGE 2 pathway, but also augmented its effect on E-cadherin, N-cadherin, and Vimentin ( Figures 6A, B ). Cell migration and invasion were prevented by macranthoidin B and celecoxib ( Figures 6C–J ). In brief, macranthoidin B synergistically improved the celecoxib’s downregulation effect, especially in COX-2/PGE 2 pathway and epithelial-mesenchymal transition. Macranthoidin B enhanced celecoxib-mediated attenuation of invasion and metastasis. (A, B) The protein levels of COX-2/PGE 2 and epithelial-mesenchymal transition pathways were investigated by Western blotting after celecoxib or macranthoidin B treatment. (C–J) Cell migration and invasion ability were measured by scratch wound and Transwell assays in 24 h * P < 0.05 to control, ** P < 0.01 to control. # P < 0.05 to celecoxib, ## P < 0.01 to celecoxib. Columns, mean (n = 3). Bars, SD. Scale bar = 100 μm. ESCs, endometriotic stromal cells.

In this study, macranthoidin B significantly reduced the formation and ameliorated the pathological structure of ectopic endometrium. Moreover, macranthoidin B also decreased E 2 and raised PROG levels. In addition, macranthoidin B hindered invasion and metastasis of endometrial cells. This was attributed to the blockade of epithelial-mesenchymal transition via COX-2/PGE 2 pathway. COX-2/PGE 2 pathway is an important inflammatory pathway, in which COX-2 acts with mPGES-1 to mediate PGE 2 synthesis ( Akasaka and Ruan, 2016 ). COX-2/PGE 2 pathway can influence the epithelial-mesenchymal transition in cancer. Activation of COX-2/PGE 2 pathway causes the loss of the epithelial E-cadherin while it is boosting mesenchymal N-cadherin and Vimentin. Moreover, AGR2 and PGE 2 receptor EP2 mediate the PGE 2 -induced epithelial-mesenchymal transition through regulating Snail ( Cheng et al., 2014 ; Zhang H. et al., 2021 ). These changes facilitate the occurrence of epithelial-mesenchymal transition, thereby enhance the invasion and metastasis abilities ( Che et al., 2017 ; Gómez-Valenzuela et al., 2021 ). This indicates a significant correlation between the COX-2/PGE 2 pathway and epithelial-mesenchymal transition. EMS is a chronic inflammatory gynecological disease that affects many women of reproductive age. Clinical data indicate increasing COX-2/PGE 2 expression in EMS patients ( Cho et al., 2010 ; Peng et al., 2018 ). During EMS progression, the COX-2/PGE 2 pathway indirectly promotes angiogenesis at ectopic sites ( Jana et al., 2016 ). Recently, epithelial-mesenchymal transition has been proved to be the vital step in EMS progression ( Zhang C. et al., 2021 ). Moreover, PGE 2 also stimulates migration and epithelial-mesenchymal transition in normal endometrial epithelial cells ( Kusama et al., 2021 ). Our investigation discovered overexpression of COX-2/PGE 2 pathway and epithelial-mesenchymal transition in autograft EMS model. COX-2/PGE 2 pathway was activated by LPS. Then epithelial-mesenchymal transition, invasion and metastasis were induced in endometrial cells. Conversely, using celecoxib, the downregulation of COX-2/PGE 2 pathway led to the suppression of epithelial-mesenchymal transition, invasion and metastasis. However, further research is required to confirm the relationship in human EMS tissues or other EMS animal models. Lonicera macranthoides Hand. –Mazz, also called Lonicerae Flos , has demonstrated pharmacological properties, such as antibacterial, antiviral, antipyretic, and anti-inflammatory activities ( Li et al., 2020 ). Initially, Lonicerae Flos extract exhibits the anti-inflammatory action through the COX-2 reduction in the animal inflammation model and LPS-induced cellular model ( Zeng et al., 2020a ; Zeng et al., 2020a ). The saponins are considered as the important material basis of anti-inflammatory effect. Previous studies has shown that both Lonicerae Flos and its saponins can suppress COX-2/PGE 2 pathway ( Guan et al., 2014 ; Zeng et al., 2020b ). Lonimaranthoide VI, a saponin metabolite in Lonicerae Flos , also displays anti-inflammatory activity by regulating COX-2-induced PGE 2 synthesis ( Guan et al., 2014 ). Furthermore, macranthoidin B is one of the main triterpenoid saponin metabolites in Lonicera macranthoides Hand. –Mazz. Limited pharmacological studies indicate that it exhibits the anti-tumor effects ( Fan et al., 2018 ; Tan et al., 2023 ). In our research, the effect of macranthoidin B on EMS was uncovered both in vivo and in vitro . Its mechanism involved the suppression of COX-2/PGE 2 pathway, which subsequently caused the downregulation of epithelial-mesenchymal transition, invasion and metastasis. Notably, the dose levels of macranthoidin B lacked the dose response in primary ESCs. It might be related to the high selection of doses as a limitation. It is worthwhile to find more suitable doses and explore other mechanisms of macranthoidin B in EMS. Furthermore, paeonol, imperatorin, ginsenoside Rg3, ferulic Acid, ligustrazine, and tetrahydropalmatine also show the inhibition in EMS ( Huang et al., 2020 ; Ma et al., 2021 ; Pang et al., 2021 ; Zhang C. et al., 2021 ). So it needs to screen more potential natural metabolites in the future. In addition, EMS is an estrogen-dependent disease ( Marquardt et al., 2019 ). E 2 secretion in EMS is higher, despite of lower PROG secretion ( Patel et al., 2017 ; Marquardt et al., 2019 ). Estrogen generation in ectopic endometrium can activate COX-2 and accelerate prostaglandin synthesis. This leads to a positive feedback loop, enhancing estrogen production and inflammation, which further promotes the EMS progression ( Bulun et al., 2012 ; Takaoka et al., 2018 ). In accordance with these studies, we also detected the increasing E 2 and decreasing PROG in EMS serum. The overexpression of COX-2/PGE 2 pathway was found in ectopic endometrium. Macranthoidin B treatment resulted in the opposite changes of E 2 and PROG, alongside the downregulation of COX-2/PGE 2 pathway. Therefore, the mechanism of macranthoidin B can be further investigated in the regulation of the relationship between COX-2/PGE 2 pathway and E 2 . In summary, macranthoidin B showed the constraint on EMS progression. Macranthoidin B inhibited epithelial-mesenchymal transition via COX-2/PGE 2 pathway. These findings provide a theoretical basis for the further development of macranthoidin B.