This study was approved by the Human Ethics Committee of the Women’s Hospital, School of Medicine, Zhejiang University (No. 20160114). Written informed consent was obtained from each patient. The privacy rights of patients were observed.

Mouse mast cell line P815 from ScienCell Research Laboratories (CA, USA) was cultured with RPMI-1640 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) and 1% Penicillin-Streptomycin (PS) (Thermo Fisher). The 293T cells were purchased from ATCC and were maintained in DMEM(Gibco) supplemented with 10% FBS and 1% PS. All these cells were maintained at 37°C, in 5% CO2, with the medium being changed every 2–3 days.

The mouse ESR1-specific shRNA vectors (SH) and the negative control vectors (NC) were purchased from Obio Technology (Shanghai, China). The targeted sequences were: sh-ESR1 (5′- GCAAGCCCACTGTGTTCAA-3′), sh-NC (5′- CCTAAGGTTAAGTCGCCCTCG-3′). Lentivirus packaging and transduction were conducted according to the manufacturer’s instructions. P815 cells stably expressing sh-ESR1 or NC were enriched by puromycin selection for positive clones. The knockdown efficiency of shRNA was confirmed by quantitative PCR and western blot.

Total mRNAs of P815 cells (three duplicate samples in each NC and SH group) incubated with 100 pmol/mL β-estradiol and serum-free control for 24h were isolated. RNA sequencing using HiSeq 4000 was carried out by LC Biological Technology (Hangzhou, China) according to the manufacturer’s instructions. Primary sequencing data was produced by RNA sequencing and these were subjected to quality control. FPKM (fragments per kilobase of exon model per million mapped reads) was used to measure the abundance of gene expression. The differential expression range of |log2FC| ≥1 was used as the change threshold, and the P value < 0.05 was the standard for the differentially expressed genes between the comparison groups. Subsequently, those genes with large differences and high significance were selected, and underwent Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis. Finally, the consistency with the sequencing results was verified by real-time PCR.

Total RNA was extracted from cells using a Total RNA Isolation Kit (Vazyme, Nanjing, China) and reversed using the PrimeScript Reverse Transcription (RT) reagent kit (Takara) according to the manufacturer’s recommendations. Specific primers used for amplification were synthesized by Generay (Shanghai, China). The sequencing of primers was as follows: Mouse GAPDH: CTTCAACAGCAACTCCCACTCTTCC (F), GGTTTCTTACTCCTTGGAGGCCATG (R); Mouse ESR1: AGGTGCCCTACTACCTGGAG (F), TCTTTCCGTATGCCGCCTTT (R); and Mouse NLRP3: ATTACCCGCCCGAGAAAGG (F), TCGCAGCAAAGATCCACACAG (R). Real-time PCR was performed with an Applied Biosystems 7900HT system (Applied Biosystems) using a SYBR Premix Ex TaqTM kit (Takara). For each sample, an average cycle threshold (Ct) value was calculated from triplicate wells, expressions of genes were normalized to the expression of glyceraldehyde-3-phosphate dehydrogenase (GADPH), and the fold change was determined using the 2^-△△Ct method.

Samples were homogenized using RIPA buffer (Beyotime, Shanghai, China) with Protease and Phosphatase Inhibitor (Thermo Fisher) or Nuclear-Cytosol Extraction Kit (Fdbio science, Hangzhou, China). Proteins were separated in SDS-PAGE and then transferred to the PVDF membranes (Merck Millipore). After blocking with 5% bovine serum albumin, the membranes were incubated with antibodies raised against Estrogen Receptor alpha (1:1000, ab32063, Abcam, USA), PCNA (Proteintech, USA), NLRP3 (1:1000, ab263899, Abcam), cleaved caspase-1 (1:1000, 89332, Cell Signaling Technology), pro-caspase-1 (1:1000, ab179515, Abcam), IL-1β (1:1000, 12242, Cell Signaling Technology), ASC (1:1000, 67824, Cell Signaling Technology), and GAPDH (1:1000, Mab5465, MultiSciences, Hangzhou, China). After incubation with HRP-conjugated secondary antibody against rabbit IgG (1:10000, ab97051, Abcam) or against mouse IgG (1:10000, ab97023, Abcam), the membranes were detected using the ECL detection kit (Biological Industries) and imaged on the ImageQuant LAS 4000 Mini Biomolecular Imager (GE Healthcare Life Sciences, USA). The detected protein expression was quantified on band volume by Image J and normalized over GAPDH mean band intensity.

To assess the intracellular amount of potassium, P815 cells were first incubated for different time points with 100 pmol/mL β-estradiol and serum-free control. Cells from each treatment condition were seeded into 96-well plates (30000 cells/well, Bioland Scientific, USA) in quintuplicate and wells were loaded with 10 µM of the cell permeant potassium-binding benzofuran-isophthalate acetoxymethyl ester (PBFI-AM, fluorescent K binding probe, ab142804, Abcam) with 0.1% Pluronic F-127 (Yeasen Biotech, Shanghai, China). After incubation for 2 h at 37 °C, excess PBFI-AM was removed using PBS. P815 were plated and signal intensity was measured with Varioskan Flash (Thermo Scientific). Excitation ratios of fluorescence at 340 and 380 nm, measured at emission 500 nm were used to detemine the intracellular content of K⁺, according to the manufacturer-suggested protocol. Measurements were corrected for autofluorescence values and presented as percent of untreated control.

IL-1β levels were measured in the cell culture supernatant and peritoneal fluid of BALB/c mice using a mouse IL-1β ELISA kit (Cusabio, Wuhan, China) according to the manufacturer’s guidelines.

Sections of human paraffin-embedded ovarian endometriomas and endometrial tissues were deparaffinized and rehydrated, followed by heat-induced and enzymatic antigen retrieval. Sections were then blocked before overnight incubation of primary antibodies: mast cell tryptase (1:250, ab2378, Abcam) and NLRP3 (1:100, ab263899, Abcam). Fluorescent secondary antibodies (Alexa Fluor 488 goat anti-mouse, 1:500, ab150117; and Alexa Fluor 647 donkey anti-rabbit, 1:500, ab150075, Abcam) and 4′,6-diamidino-2-phénylindole (DAPI, ab228549, Abcam) were then used for detection. Images were acquired using an Olympus FV1200 confocal microscope (Olympus, Tokyo, Japan). In the figures, scale bars are indicated on respective magnified images for each experiment. For fluorescence immunocytochemical assays, P815 cells incubated with or without 100 pmol/mL β-estradiol (Sigma) were seeded on μ-Slide 8 well chamber slides (ibidi GmbH, Germany) in complete RPMI-1640 medium. Cells were fixed in 4% buffered formalin, blocked and permeabilized before incubation with primary NLRP3 antibody (1:100, ab263899, Abcam). Fluorescent secondary Alexa Fluor 647 donkey anti-rabbit antibody (1:500, ab150075, Abcam) was then used for detection. DAPI (Abcam) was used for cell nuclear staining.

For histochemical assays, the endometrial tissues were fixed in 4% formalin and then embedded in paraffin. The 3 µm tissue slides were incubated with diluted primary antibody against interleukin-1 receptor IL1R1 (1:500, ab106278, Abcam) and IL1R2 (1:50, DF6682, Affinity Biosciences) to detect the expression of IL-1β receptors in endometrial tissues. Reaction products were visualized using 3, 3-diaminobenzidine (DAB) solution, and the nuclei were counterstained with hematoxylin. The results were examined by two independent observers who were unaware of the background of samples and the immunohistochemical scores were graded as follows: the staining intensity (0 representing negative, 1 for weak staining, 2 for moderate staining, and 3 for strong staining) and the percentage of positive cells (0 represented staining of 0–5%, 1 for staining of 6%–25%, 2 for staining of 26%–50%, 3 for 51%–75%, and 4 for 76%–100%) were recorded, and the IRS was calculated by multiplying the two scores (24).

ChIP assay was performed using a ChIP assay kit (Abcam) according to the manufacturer’s protocol. In detail, 3 × 10⁶ P815 cells with or without estrogen treatment were fixed with 1.1% formaldehyde at room temperature for 10 min. Cells were sedimented, washed, lysed, and sonicated to reduce DNA lengths to between 200 and 1000 bp. After sonication, the chromatin was immunoprecipitated with anti‐Estrogen Receptor alpha (2µg for 1 × 10⁶ cells, ab32063, Abcam) or anti‐histone H3 (2µg for 1 × 10⁶ cells, ab1791, Abcam) antibodies at 4°C overnight. After incubating with antibody binding beads, DNA was purified using Proteinase K. PCR was performed using specific primers: forward (5′- TGCCTCATTCAGGACACT -3′), and reverse (5′- TGCTTGGCTTTCTTTACAC -3′). To confirm the PCR results, DNA sequencing of PCR products was performed by TsingKe Biological Technology (Beijing, China).

The NLRP3 promoter-reporter construct and the mutant NLRP3 promoter fragment, cloned into the pGL4 basic plasmids, were purchased from Obio Technology. The luciferase-encoding NLRP3-promoter or mutant NLRP3-promoter, renilla reporter constructs (Promega), and mouse ESR1 plasmid (Obio Technology) were co-transfected into 293T cells using Hieff Liposomal Transfection (Yeasen Biotech). The pGL4-Basic firefly luciferase vector served as the negative control. After 24 h, luciferase activities were measured using the Duo-Lite Luciferase Assay System (Vazyme) according to the manufacturer’s instructions. Renilla activity was used to normalize luciferase reporter activity. Assays were performed on cells in five wells for each experiment to obtain an average count, and three independent biological replicates were assessed.

Female BALB/C mice (age: 6–8 weeks), housed in an environment maintained at 21 °C ± 0.5 °C, under a 12h light/dark cycle, and with free access to food and water, were used in the experiments. This study was carried out in strict accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The protocol was approved by the Committee on the Ethics of Animal Experiments of Zhejiang University (No. ZJU20200015).

Eight mice were randomly selected as donors. The remaining mice were randomly divided into three groups (eight in each group): the sham group and two recipient groups, the control group, and the drug administration group. Bilateral ovariectomy was performed 7 days before endometriosis induction, followed by subcutaneous injection of 0.5 μg estradiol benzoate dissolved in 60 μL corn oil every 5 days. Induction of endometriosis was performed as reported previously (25). The uterine horns of donors were dissected and cut into fragments <1mm³. Then we suspended the fragments in 0.9% sterile saline and injected them intraperitoneally into the 16 recipient mice. Mice in the sham group were injected with an equal volume of 0.9% sterile saline.

Mice in the sham and control groups were injected intraperitoneally with the vehicle control (10% DMSO, 40% PEG400, 5% Tween80, and 45% water), while the drug administration group was given the proven NLRP3 inhibitor CY-09 (MCE, USA) (i.p.) at a dose of 2.5 mg/kg once a day (26). The mice were anesthetized with inhalation of CO₂ and euthanized by cervical dislocation after treatment for 14 or 21 days. Endometriotic lesions and peritoneal fluid were preserved for further experiments, and the size and weight of lesions were measured. We determined the volume of lesions according to the formula: = (length × width²)/2. Sections of mouse paraffin-embedded endometriosis lesions were first subjected to hematoxylin and eosin (H&E) staining to confirm the diagnosis of endometriosis. Masson’s trichrome staining was performed as previously described (27), and the areas of the collagen fiber (blue-colored) in proportion to the entire field of lesions were measured using ImageJ.

Mouse peritoneal cells were isolated as described previously with slight modifications (28). In short, we injected 3 ml of phosphate-buffered saline (PBS) and 2 mL of air into the peritoneal cavity. After carefully massaging the abdomen, the cell suspension was withdrawn from the peritoneal cavity using a Pasteur pipette and was collected in centrifuge tubes. The collected cell suspension was centrifuged at 300 × g for 5 min at 4°C. The pellet was resuspended in PBS and incubated with Leukocyte Activation Cocktail with GolgiPlug (2 μL/mL, 550583, BD Pharmingen, USA) for 5 h in a 37°C humidified CO₂ incubator. Following activation, peritoneal cells were harvested, washed with Flow Cytometry Staining Buffer (eBioscience), and incubated with mouse Fc Block (1µg for 1 × 10⁶ cells, 553141, BD Pharmingen). Then the cells were stained with APC-conjugated anti-mouse CD117/c-KIT (1:100, 553356, BD Pharmingen) and FITC-conjugated anti-mouse FcϵRIα (1:400, 11-5898-82, eBioscience) antibodies. MCs were identified by double-positive expression of FcϵRI and CD117. Intracellular staining of IL-1β was performed using the Transcription Factor Buffer Set (BD Pharmingen) and eFluor 450-conjugated anti-mouse IL-1 beta antibody (1:40, 48-7114-82, eBioscience) in accordance with the manufacturer’s instructions. APC-conjugated Rat IgG2b kappa (1:100, 17-4031-81, eBioscience) and FITC- conjugated Armenian Hamster IgG (1:400, 11-4888-81, eBioscience) antibody served as the isotype control. Flow cytometric data were acquired on a BD FACSVerse flow cytometer with BD FACSuite software.

All data were included for statistical analyses using GraphPad Prism 7 (Graph Pad Software) and were presented as mean ± standard error of the mean (SEM). All experiments were performed at least three times. Data normality was evaluated using Shapiro-Wilk normality test. For comparisons between two groups, statistical analysis was performed by Unpaired Student’s t-test (two-tailed) unless otherwise indicated. For multiple groups, one-way analysis of variance (ANOVA) followed by Dunnett t-test was used for normally distributed variables and the Kruskal-Wallis test with Dunn’s multiple comparison test for variables with skewed distribution. *P values < 0.05 were considered statistically significant.

Essential roles of estrogen in orchestrating MCs activation have been established. In the quest to identify the underlying mechanisms that can coordinate MCs to participate in the inflammatory response in endometriosis, we first performed gene expression profiling to identify critical genes that responded to estrogen stimulation. We carried out KEGG enrichment pathway analysis of the differentially expressed genes (DEGs) between P815 cells treated with β-estradiol and the control. KEGG pathway analysis indicated that the DEGs were mainly involved in Herpes simplex virus 1 infection, NOD-like receptor signaling pathway, focal adhesion, allograft rejection, etc. ( Figure 1A ). Further study of DEGs revealed that NLRP3, an important molecule involved in regulating inflammation and the NOD-like receptor signaling pathway, was significantly increased in P815 after exposure to β-estradiol ( Figure 1B ). To gain insight into the molecular mechanism through which estrogen up-regulates the expression of NLRP3, we used ESR1-specific shRNA to silence ER-α in P815 cells. The efficiency of knockdown was verified by quantitative PCR, and western blot analysis confirmed that the protein level of ER-α in nuclei was significantly reduced ( Figures 1C, D ). RNA sequencing and quantitative PCR validated the elevated expression of NLRP3 in P815 cells transfected with negative control shRNA (NC) after incubation with β-estradiol, while the silencing of ER-α (SH) abrogated the effect ( Figures 1E, F ). Concurrent with the mRNA results, immunofluorescence staining of P815 cells revealed a pronounced increase of NLRP3 after β-estradiol treatment ( Figure 1G ). Collectively, these results suggested that the stimulation of estrogen may up-regulate the expression of NLRP3 in MCs.

In the search to define whether ER-α bound to the putative binding site, we analyzed the promoter sequence of the NLRP3 gene using online regulatory sequence analysis tools. Different primers were used to amplify the predicted ERE-like sequences. As shown in Figure 2A , one fragment containing the ERE-like sequence was captured using the ChIP assay after P815 cells were treated with estrogen. Specifically, after cell lysing and DNA shearing, extracts were subjected to ChIP-grade ER-α antibody and the anti-Histone H3 antibody (positive control). Following cross-linking reversal and protein digestion, PCR and agarose gel electrophoresis were carried out, and then gel extracted PCR products were subjected to gene sequencing. The DNA sequencing results confirmed the interaction of ER-α​ with the ERE (GTGACCAA) of NLRP3 positioned residues between −​422 through −​429 ( Figure 2B ). We further delved into the molecular mechanisms regulating ERE-dependent NLRP3 transcription using the dual-luciferase reporter assay. 293T cells were co-transfected with pGL4.10 reporter plasmid (Basic-Luc, NLRP3-Luc or delNLRP3-Luc), Renilla luciferase-expressing plasmid pRL-CMV, and plasmid pCMV-ESR1, as indicated ( Figure 2C ). Luciferase activities were measured after transfection to determine whether the putative ERE played a functional role in ER-α-related transcriptional activation. We found that the construct of the ERE (GTGACCAA) could interact with ER-α to activate the expression of downstream firefly luciferase, while the relative luciferase activity was significantly decreased when the construct of the partially deleted element was applied ( Figure 2D ). Our results suggested that a functional ERE motif existed in the NLRP3 gene promoter in MCs, and upregulation of NLRP3 was induced by estrogen stimulation via the interaction of ER-α with the ERE.

Previous studies demonstrate that MCs express functional components of the NLRP3 inflammasome (22). To directly address whether NLRP3 was expressed by MCs harbored in endometriotic lesions, double immunofluorescence was performed in paraffin sections of ovarian endometriomas. Merged images show the co-localization of NLRP3 (shown as red) and the mast cell marker, tryptase (shown as green; Figure 3 ), indicating that NLRP3 was expressed in MCs in endometriosis. Compellingly, the local high estrogen environment of endometriosis may lead to the changes of NLRP3 inflammasome in MCs, and its specific role in the development of the disease was worthy of our exploration.

Since the activation of caspase-1 is a key hallmark of the activity of the NLRP3 inflammasome pathway, we therefore assessed the protein expression level of cleaved caspase-1 and its target downstream molecule, IL-1β. Cleaved caspase-1 and IL-1β levels were significantly increased in NC groups treated with 100 pmol/mL β-estradiol for 12 h compared to the baseline measurements, while elevated NLRP3 levels were observed at 3 h, 6 h, 12 h, and 24 h after estrogen stimulation ( Figures 4A, B ). No difference between SH cells with or without estrogen treatment was observed at any time. We also evaluated the protein expression levels of ASC, pro-caspase-1, and pro-IL-1β and found that their expression did not change significantly after estrogen stimulation in either group. Since the activation of NLRP3 inflammasomes requires certain molecular and cellular signaling events (20), we further determined the intracellular K⁺ concentration of estrogen incubated P815 cells using fluorescent probe PBFI-AM at the aforementioned time points. Intracellular K⁺ was reduced in NC groups by 11% compared to control at 6 h after the start of incubation, with a maximum reduction of 24% observed at 12 h ( Figure 4C ), which indicated that K⁺ efflux occurred in MCs treated with estrogen. No significant difference in K⁺ concentration among the different treatments in SH groups was observed. ELISA assay of cell culture supernatant confirmed the increased mature IL-1β secretion level in the NC group when exposed to β-estradiol for 48 h ( Figure 4D ). The production of mature IL-1β was blunted in ER-α deficient MCs. These results clearly evidenced that estrogen could fuel the expression of NLRP3, and had a potent inflammasome activation capacity via K⁺ efflux in MCs.

We have demonstrated that mature IL-1β was elevated in estrogen-stimulated MCs. Here, we tested the expression of its receptors in various types of endometrium. Immunohistochemical staining revealed a significantly elevated expression of IL1R1 in the ectopic endometrium, while the alternate receptor IL1R2 maintained low expression, consistent with previous reports (29) ( Figure 5 ). Since IL1R2 is a recognized inhibitory receptor with no downstream signal transduction, IL-1β mainly combined with its functional receptor IL1R1 to promote the proliferation and invasive ability of endometrial stromal cells in the ectopic endometrium (30–32). These findings indicated a positive role for MC-derived IL-1β in steering the construction of the endometriosis lesions.

To investigate the role of MC-derived NLRP3 inflammasome in endometriosis, we intraperitoneally injected the NLRP3 inhibitor CY-09 into the established endometriosis mouse model. The level of IL-1β in peritoneal MCs was quantified using intracellular staining via flow cytometry. First, we identified MCs by double-positive staining of FcϵRIα and CD117 ( Figure 6A ). The average IL-1β mean fluorescence intensity (MFI) values in MCs from the control group were significantly higher than the sham and CY-09 administration group after 14 days of treatment ( Figure 6B ). Accordingly, IL-1β levels in the peritoneal fluid of mice with endometriosis were elevated compared with the sham group and were reduced when exposed to the NLRP3 inhibitor ( Figure 6C ). We further characterized the impact of NLRP3 inhibitor in the pathogenesis of endometriosis. A significant reduction of the total weight of endometriotic lesions was observed both at 14 and 21 days of CY-09 treatment ( Figures 6D, E ). The total volume of lesions in the CY-09 administration group was also significantly decreased over the control group at the time of sacrifice. Ectopic endometrial lesions obtained from the abdominal cavity of mice were verified by H&E staining, and fibrosis was assessed by staining sections with Masson’s trichrome ( Figure 6F ). Compared to the control group, CY-09 administration significantly reduced collagen deposition in endometriotic tissues ( Figure 6G ). Altogether, our data demonstrated a distinct alleviating role of NLRP3 inhibitor in the pathogenesis and fibrosis of endometriosis.