Iron overload-induced ferroptosis in CD8+ T cells leads to functional abnormalities that promote endometriosis progression

All procedures were approved by the Ethics Committee of the Medical Integration and Practice Center of Shandong University (SDULCLL2022-2-17; approval date: 4 November 2022). The study was conducted in strict compliance with the requirements of the “Animals (Scientific Procedures) Act 1986” and adhered to the 3R principles (Refinement, Replacement, Reduction). Specific measures included: (i) estimating sample size based on degrees of freedom (E) in analysis of variance to minimise experimental animal use; (ii) employing in vitro cell models to pre-verify the effectiveness of intervention measures, thereby reducing invalid variables in animal experiments; and (iii) maintaining a specific pathogen-free (SPF) breeding environment, standardizing pain control, and using humane euthanasia to reduce animal suffering during the experiments. Studies involving human subjects were reviewed and approved by the Ethics Committee of the Medical Integration and Practice Center of Shandong University (SDULCLL2022-1-21; approval date: 4 November 2022). All participants provided signed informed consent in accordance with the Declaration of Helsinki. Table 1 presents the clinical characteristics of the patients. Table 1 Characteristics of patients Item Control group Endometriosis group Statistical significance Number of participants 31 57 Age (year) Mean = 38.5 Mean = 35.75 Median = 38 Median = 35 NS Range = 21–49 Range = 25–49 Age 20–40 (year) 18 41 NS Age ≥ 40 (year) 13 16 NS BMI (kg/m 2 ) Mean = 24.19 Mean = 23.04 Median = 24.12 Median = 22.83 NS Range = 17.43–34.18 Range = 17.26–31.44 BMI < 25 (kg/m 2 ) 21 44 NS BMI ≥ 25(kg/m 2 ) 10 13 NS Menstrual period (day) 6.983 5.675 ** Menstrual cycle (day) 28.10 28.67 NS Menstrual phase NS  Proliferative 38.71% 38.60%  Secretory 61.29% 61.40% Gravidity NS  0 12.90% 26.32%  1 22.58% 29.82%  ≥ 2 64.52% 43.86% Miscarriages NS  0 58.06% 66.67%  1 25.81% 17.54%  ≥ 2 16.13% 15.79% Data presented as mean BMI body mass index, NS  no significance ** P  < 0.01 Characteristics of patients Data presented as mean BMI body mass index, NS  no significance ** P  < 0.01 Eighty-eight samples were collected and allocated for immunohistochemistry, immunofluorescence, and flow cytometry analyses. The normal control group comprised eutopic endometrial tissue specimens obtained from 31 patients who underwent hysteroscopy or combined laparoscopy and hysteroscopy to treat endometrial thickening, tubal infertility, simple ovarian cysts, and similar conditions. Eutopic and ectopic endometrial tissues were obtained from 57 patients with stage III–IV EM confirmed by ultrasound and histopathology. Sample size was calculated using G*Power (n [Eu-NC, Eu-EM, Ec-EM] = 31, 25, 32; effect size d = 0.25; α = 0.05; target power = 0.08), yielding an observed power of 89.7%. All participants were women aged 20–49 years with regular menstrual cycles who had not received steroid hormone therapy within 6 months before surgery. Exclusion criteria included chronic systemic diseases (e.g., hypertension, diabetes), autoimmune disorders (e.g., systemic lupus erythematosus, Hashimoto’s thyroiditis), and malignancies. Patients were recruited from the Department of Obstetrics and Gynaecology at the Affiliated Hospital of Shandong University. Female C57BL/6N mice, aged 6–8 weeks, were purchased from Beijing Vital River Laboratory Animal Technology Co., LTD. (Beijing, China). They were housed in the SPF Laboratory Animal Centre of Shandong University (Jinan, China) for 1 week to acclimatise. During acclimation, mice were maintained under a 12 h light/dark cycle at a controlled temperature (25 ± 1 °C), relative humidity (55% ± 10%), and had free access to food and water.​ All mice were confirmed to be in good health. Eutopic and ectopic tissues were collected from 43 patients. Immediately after aseptic sampling, tissues were placed in complete culture medium and transferred to the laboratory within 1 h of surgery. Tissue-surface blood was removed using cold phosphate buffered saline (PBS), and the tissue was cut into < 1 mm 3 pieces and digested with a 0.25% (w/v) collagenase II and 0.125% (w/v) collagenase IV (Worthington, Biochemical Corp., Lakewood, NJ, USA) mixture at 37 °C for 1 h. The mixture was sequentially filtered through 100- and 70-μm cell filters, and cell precipitates were obtained by centrifugation at 400 g for 10 min. Precipitates were resuspended in a red blood cell lysis buffer (R1010, Solarbio, China) and placed on ice for 10 min. After centrifugation, cells were resuspended in 1 × Hank's Balanced Salt Solution (HBSS) and used for subsequent flow cytometry analysis. The abdominal skin of mice was peeled to expose the peritoneum, 5 mL of cold PBS containing 2% fetal bovine serum was injected into the peritoneal cavity, the abdomen was gently massaged for 1 min to resuspend the peritoneal immune cells in the PBS, and the lavage fluid was withdrawn with a 5 mL syringe; the process was repeated once. The lavage fluid was centrifuged, resuspended in red blood cell lysis buffer (R1010, Solarbio, China), and placed on ice for 5 min. After centrifuging, the cells were resuspended in HBSS and used for subsequent experiments. CD8 + T cells were isolated from mouse spleens using a negative selection kit (#19853, STEMCELL, Canada) and activated with mouse CD3/CD28 Dynabeads (11453D, Invitrogen, USA) plus IL-2 (100 U/mL), with or without indicated treatments. Purity reached 99% (Additional file 1: Fig. S1). Cells were cultured in RPMI-1640 medium (R8758, Sigma, Germany) supplemented with 10% FBS (FSD500, ExCell Bio, China), 1 M Hepes (H1095, Solarbio, China), 2% sodium pyruvate (SP0100, Solarbio, China), 1 × non—essential amino acids (N1250, Solarbio, China), 55 μM β-mercaptoethanol (M8210, Solarbio, China), 1% penicillin–streptomycin (15,140,122, Gibco, USA), and 2 mM L-glutamine (25,030,149, Gibco, USA). Mice received subcutaneous (s.c.) injections of 3 µg oestradiol benzoate (E408424, Aladdin, China) at the nape of the neck and were euthanised two days later. The abdomen was opened, and uteri were excised. Uteri were digested in 0.25% (w/v) collagenase II and 0.125% (w/v) collagenase IV at 37 °C for 1 h. The resulting suspension was filtered through a 70 µm sieve and centrifuged at 300 g for 10 min. Cells were resuspended in DMEM/F12 (D8437, Sigma, Germany) containing 10% FBS and 1% penicillin–streptomycin and used within two passages. CD8 + T cells were treated with ammonium iron citrate (AIC, A100170, Aladdin, China) at 0–500 µM for 24 h. Because 500 µM AIC caused a marked decrease in CD8⁺ T cell viability, this concentration was used for subsequent 24 h stimulations, followed by a 72 h treatment before qRT-PCR or Western blot analysis. Ferrostatin-1 (Fer-1; HY-100579, MCE, China) is a catalytic ferroptosis inhibitor that traps lipid-associated free radicals without affecting apoptosis or necrosis [ 23 ] and preferentially targets lipid peroxidation [ 24 ]. Fer-1 is not consumed in the process of inhibiting iron-dependent lipid peroxidation and functions in a catalytic manner [ 25 ]. We applied Fer-1 at 10 µM to verify ferroptosis in AIC-treated CD8⁺ T cells. Prior to co-culture with ESCs, CD8⁺ T cells were washed and maintained in standard AIC-free medium to eliminate iron overload effects. Total RNA was extracted from the cultured cells using TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s protocol, and cDNA was synthesised using Evo M-MLV RT Premix for qPCR (AG11706, Accurate Biotechnology, China). Quantitative PCR was performed using Quant Studio 5 machine (Thermo Fisher Scientific, Waltham, MA, USA). The results were normalized based on Beta Actin mRNA level. Specific primers sequences used in this study are listed in Table  2 . Table 2 Primer sequences used in qRT-PCR analysis Gene Primer Sequence (5’–3') β-actin Forward CATCCGTAAAGACCTCTATGCCAAC Reverse ATGGAGCCACCGATCCACA Acsl4 Forward CTCACCATTATATTGCTGCCTGT Reverse TCTCTTTGCCATAGCGTTTTTCT p53 Forward ATCGCCTTCGACATCATCGC Reverse CCCCATGCGTACTCCATGAG xCT Forward ATACTCCAGAACACGGGCA Reverse AGGGCTCCAAAAAGTGACAGT Ho-1 Forward CTGGAGATGACACCTGAGGTCAA Reverse CTGACGAAGTGACGCCATCTG Gpx4 Forward GTGTAAATGGGGACGATGCC Reverse ACCACGCAGCCGTTCTTATC Fth1 Forward CAGAACTACCACCAGGACGC Reverse TCAGAGCCACATCATCTCGG Primer sequences used in qRT-PCR analysis The treated cells were lysed by RIPA lysis buffer (P0013B, Beyotime, China) supplemented with phenylmethanesulfonyl fluoride (PMSF; ST507, Beyotime, China). The protein concentration was determined using a BCA protein assay kit (P0010S, Beyotime, China). A total of 20 μg of protein was loaded per lane, and the signal intensity showed a linear correlation with the loaded protein amount. Then, the protein samples were prepared by treating with SDS-PAGE loading buffer and then separated via 10% SDS-PAGE and transferred onto 0.22-μm PVDF membranes (Millipore, Germany). Different primary antibody dilutions and corresponding HRP-labelled secondary antibodies were used to incubate the membranes. The antibodies in the experiment were GAPDH Recombinant antibody (81640-5-RR, Proteintech, Chicago, USA, 1:50000), Beta Actin Polyclonal antibody (20536-1-AP, Proteintech, Chicago, USA, 1:5000), p53 Rabbit Monoclonal antibody (A25915, ABclonal, China, 1:6000), SLC7A11/xCT Rabbit Monoclonal antibody (A2413, ABclonal, China, 1:1000), Heme Oxygenase 1 Rabbit Monoclonal antibody (A27713, ABclonal, China, 1:6000), ACSL4/FACL4 Recombinant antibody (83516–6-RR, Proteintech, Chicago, USA, 1:10,000), anti-glutathione peroxidase 4 antibody (ab125066, Abcam, UK, 1:1000), and Ferritin Heavy Chain Rabbit Monoclonal antibody (A19544, ABclonal, China, 1:1000). The signal of blot was visualized by chemiluminescence (ECL) and was detected using CHAMPCHEMI 910 (Beijing Sage Creation Science Co., Ltd.). Three independent experiments were performed. Relative protein levels were quantified using ImageJ software (version 1.5, Wayne Raband, National Institutes of Health, Bethesda, MD, USA) and normalised to GAPDH or Beta Actin expression. All human and murine specimens were fixed in 4% paraformaldehyde (P1110, Solarbio, China) for 24 h, and paraffin slides were prepared, dewaxed, and hydrated. For Prussian blue staining, slides were placed in a mixture of equal volumes of Perls working liquid A and B (G1422, Solarbio, China), stained at room temperature (RT) for 15–60 min, washed in distilled water, stained with Nuclear Fast Red solution for 10 min, and rinsed with water. Immunohistochemistry slides were placed in 1 × Tris–EDTA antigen repair solution (C1038, Solarbio, China) and microwaved on low for 15 min for antigen repair. Slides were then incubated in 3% hydrogen peroxide at 37 °C for 30 min to remove endogenous peroxidases. After blocking with 5% BSA for 1 h, slides were incubated overnight with CD8a Monoclonal antibodies (66868–1-lg, Proteintech, Chicago, USA, 1:10,000) at 4 °C. Next, slides were washed three times with PBS and incubated with secondary antibodies at 37 °C for 1 h. Finally, slides were incubated with 3,3-diaminobenzidine tetrachloride substrate and counterstained with hematoxylin. Slides were photographed using an EVOS M5000 imaging system (Thermo Scientific, Waltham, MA, USA). Five randomly selected images in the field of view were analyzed using image J (NIH, Bethesda, Maryland, USA). Slides were incubated overnight at 4 °C with primary antibodies: CD8a Monoclonal antibody (66868-1-lg, Proteintech, Chicago, USA, 1:400), Granzyme B Polyclonal antibody (13588-1-AP, Proteintech, Chicago, USA, 1:50) and GPX4 Monoclonal antibody (A11243, ABclonal, China, 1:200). After three washes in TBST, slides were incubated for 1 h at 37 °C with appropriate fluorescent secondary antibodies. Nuclei were counterstained with DAPI, and images were acquired on an EVOS M5000 imaging system. Experiments were performed according to the manufacturer’s instructions. Briefly, cells were incubated with BioTracker Far-red Labile Fe2+ Dye (1:500; SCT037, Sigma, Germany) at 37 °C for 30 min. After incubation, the cells were washed and examined using flow cytometry. 1X Lipid Peroxidation Sensor (ab243377, Abcam, UK) was added to the cells and incubated at 37 °C for 30 min in the dark. Cells were washed resuspended in HBSS. Cell fluorescence was detected within 2 h of staining (to avoid fluorescence quenching) using flow cytometer. Cells were resuspended in HBSS containing 2.5 μM CellROX ® Green Reagent ( C10444 , Invitrogen, USA) and incubated at 37 °C for 30 min. After staining, cells were centrifuged and resuspended in HBSS; the fluorescence profiles were monitored using flow cytometry. The relative GSH concentration in cell lysates was assessed using a total glutathione assay kit (S0053, Beyotime, China). After washing the cells with PBS, the cells were centrifuged. A protein removal reagent was added at three times the cell volume. Lysated the cells by ultrasound, and the supernatant was collected after centrifugation at 10,000 g/10 min. Then, 150 μL of the prepared working fluid from the kit was added to the wells of a 96-well plate, 10 μL of sample was added to each well, and the plate was incubated at RT for 5 min. Then, 50 μL of NADPH was added and the plates were left at RT for 25–60 min. The results were detected as the absorbance at 412 nm using a plate reader. After the indicated treatments, CD8 + T cells were centrifuged. The cell precipitation was fixed with the specified solution and sent to Jinan Weiya Biotechnology (Shandong, China) for transmission electron microscopy (JEM1200, JEOL, Japan). CellTrace™ CFSE labelling was performed by adding 0.5 µL of CFSE stock ( C34554 , Invitrogen, USA) to 1 mL PBS for a final concentration of 2.5 µM. Cells were incubated for 20 min at 37 °C in the dark, then quenched with five volumes of complete medium for 5 min at RT. Labelled cells were washed, resuspended, and stimulated with the indicated drugs for 72 h before analysis by flow cytometry. Mouse endometrial stromal cells (MESCs; 2.5 × 10 5 ) and CD8⁺ T cells (5 × 10 6 ) were cocultured at 37 °C for 48 h. Apoptosis of MESCs was assessed by flow cytometry using the BD Pharmingen™ PE Annexin V Apoptosis Detection Kit I (559,763, BD Biosciences, USA) according to the manufacturer’s instructions. After coculture, cells were washed twice with PBS and resuspended in 1 × binding buffer. Cells were stained with 5 µL PE Annexin V and 5 µL 7-AAD for 15 min at RT in the dark. Staining was terminated by adding 400 µL 1 × binding buffer, and samples were analysed by flow cytometry within 1 h. MESCs (8 × 10 4 ) were seeded in the lower chamber and CD8⁺ T cells (4 × 10 5 ) in the Transwell insert’s upper chamber. After 48 h coculture at 37 °C, MESC proliferation was evaluated by EdU incorporation using the Beyotime EdU Assay Kit (C0071S, Beyotime Biotechnology, China) per the manufacturer’s protocol. Briefly, following coculture, MESCs were incubated with EdU for 12 h, then fixed and permeabilised with the BD Fixation/Permeabilisation Solution Kit (554,714, BD Biosciences, USA). The click reaction was performed at room temperature in the dark for 30 min, and proliferation was analysed by flow cytometry. Transwell migration assays were performed as follows. For MESC migration, MESCs (3 × 10 4 ) in serum-free medium were added to the upper chamber, and CD8⁺ T cells (1.5 × 10 5 ) were added to the lower chamber. After 24 h at 37 °C, membranes were fixed and stained with crystal violet. For CD8⁺ T cell chemotaxis, CD8⁺ T cells (1 × 10 6 ) in serum-free medium were placed in the upper chamber, and medium containing 20% FBS was added to the lower chamber. After 1 h, inserts were fixed, stained with crystal violet, and imaged on an EVOS M5000 system. Animal studies received approval from the Ethics Committee of the Medical Integration and Practice Center of Shandong University (SDULCLL2022-2–17; approval date: 4 November 2022). All procedures complied with ARRIVE 2.0 guidelines. Recipient and control mice received subcutaneous (s.c.) injections of oestradiol benzoate (3 µg/mouse; HY-B1192, MCE, China) five days before model induction. Donor mice were injected with the same dose two days prior to intraperitoneal (i.p.) transplantation of uterine fragments for cycle synchronisation. Donor uteri were harvested, minced into < 1 mm 3 fragments, pooled to minimise inter-individual variability, and resuspended in PBS. Fragments from each donor were randomly injected i.p. into two recipient mice; control mice (PBS, n  = 5) received equal volumes of PBS. Postoperative pain was managed via intraperitoneal ibuprofen (5 mg/kg, twice daily). All mice received oestradiol benzoate s.c. every six days, with an additional dose on day 3 post-transplantation to enhance colonisation; PBS controls were treated in parallel. Fifteen recipient mice were randomised into EM ( n  = 5), iron overload (Fe; n  = 5), and iron overload + deferoxamine mesylate (Fe + DFOM; n  = 5) groups. The Fe group received iron dextran (0.25 g/kg, i.p.); the Fe + DFOM group received 0.25 g/kg iron dextran and 0.5 g/kg DFOM (HY-B0988, MCE, China, i.p.). Drugs were administered i.p. every three days. Mice were euthanised 17 days post-fragment injection via carbon dioxide (CO₂) inhalation. They were placed in an exposure chamber, and the CO₂ input flow rate was adjusted to 6–10 L/min. The mice were continuously observed until they exhibited the following signs: cessation of spontaneous movement and transition to deep and prolonged breathing patterns. The CO₂ flow rate was then increased to 38–40 L/min and maintained for a continuous exposure of 5 min. The absence of vital signs was confirmed by respiratory arrest and pupillary dilation. Subsequently, cervical dislocation was performed to confirm death. PFT-α (HY-15484, MCE, China) was employed to specifically inhibit p53 in wild-type cells [ 26 ]. The activation of AhR is caused by PFT-β rather than PFT-α [ 27 ]. Considering the chemical instability of the compound, the working solution of PFT-α were freshly prepared before use. In vitro, CD8⁺ T cells were treated with 10 µM PFT-α to block p53 signalling. In vivo, PFT-α was administered intraperitoneally at 2 mg/kg alongside iron dextran injections. To deplete CD8 + T cells, mice were injected intraperitoneally with 200 µg anti-CD8 antibody (clone 2.43, #BE0061, Bioxcell, USA) on days –3, 0, 5, 10, and 15 [ 28 , 29 ], achieving near-complete depletion (Additional file 1: Fig. S2). In the CD8A + Fe-TCs group, mice received 1 × 10⁶ iron-overloaded CD8⁺ T cells in 200 µL PBS i.p.; the CD8A + TCs group received 1 × 10 6 untreated CD8⁺ T cells in 200 µL PBS. The EM and EM + CD8A control groups received equivalent volumes of PBS ( n  = 5). Cells were labelled using the CellVue Claret Far Red Fluorescent Cell Linker Kit (MINCLARET, Sigma, Germany). After pelleting, cells were resuspended in 200 µL Diluent C. The dye solution was prepared by mixing 0.4 µL CellVue Claret ethanol dye with 200 µL Diluent C. Cell suspension was added to the dye solution and incubated for 5 min in the dark. Staining was quenched by adding serum for 1 min. Cells were washed twice by centrifugation at 400 g for 5 min and resuspended in PBS. In the EM + CD8 +  TCs group, mice received 1 × 10 6 untreated CD8⁺ T cells in 200 µL PBS via i.p. injection; the EM + Fe-CD8 +  TCs group received 1 × 10 6 iron-overloaded CD8⁺ T cells in 200 µL PBS. The percentage of labelled cells in peripheral blood was measured at 6, 12, 24, and 48 h post-injection and in lesions at 48 h ( n  = 6). Cells were stained with Fixable Viability Dye eFluor™ 450 (1:1000; 65–0863-14, Invitrogen, USA) or BD Horizon™ Fixable Viability Stain 780 (1:1000; 565,388, BD Biosciences, USA) at 4 °C for 30 min in the dark and then washed twice. For surface molecules, including anti-mouse CD45 (FITC-conjugated, 103,107 and BV510-conjugated, 103,138, BioLegend, USA), CD3 (anti-human CD3, PerCP-Cyanine 5.5-conjugated, 317,336, BioLegend, USA; anti-mouse CD3, apc-cy7-conjugated, 560,590, BD Pharmingen, USA and PE-conjugated, 100,205, BioLegend, USA), CD8 (anti-human CD8, BV510-conjugated, 344,731, BioLegend, USA; anti-mouse CD8, BV510-conjugated, 126,631, BioLegend, USA), and cells were incubated in premixed antibody buffer at 4 °C for 30 min. For intracellular staining, cells were stimulated with phorbol 12-myristate 13-acetate (P6741, Solarbio, China), ionomycin (II2200, Solarbio, China), and brefeldin A (B8581, Solarbio, China) for 4 h. Intracellular staining for granzyme B (anti-human/mouse granzyme B, APC-conjugated, 372,204, BioLegend, USA; anti-mouse granzyme B, APC-conjugated, 17–8898-82, Invitrogen, USA) and GPX4 (anti-human/mouse GPX4, ab125066, Abcam, UK, 1:400) involved goat anti-rabbit IgG antibody (Alexa Fluor™ 488-conjugated, A27034, Invitrogen, USA, 1:400) and BD Fixation/Permeabilization Solution Kit. Primary antibodies were incubated for 1 h, and secondary antibodies were incubated for 30 min. Cells were then washed and examined using ThermoFisher Attune NxT (Thermo Scientific, Waltham, MA, USA). Data were analyzed using FlowJo 10.4.0 software (TreeStar, Ashland, OR, USA). Data were analyzed and visualised using GraphPad Prism 10.2.3 software (GraphPad, Inc., USA) and presented as mean ± standard deviation. All data were tested for normality and continuity before being compared for differences. An unpaired two-tailed Student’s t-test was used to compare the differences between two groups ( P  < 0.05). Continuous variables were analysed by one-way ANOVA (normally distributed data) or the Kruskal–Wallis test (non-parametric data), with Tukey’s post hoc test for multiple comparisons ( P  < 0.05).

Prussian blue staining was used to assess tissue iron content (Fig.  1 A). Ectopic lesions exhibited a significantly greater Prussian blue–positive area than eutopic endometrium, indicating local iron overload in EM lesions (Fig.  1 B). Given the role of immune cell function in EM pathogenesis, we quantified CD8⁺ T cells and found a higher percentage in ectopic versus eutopic tissues (Fig.  1 C and D). Because granzyme B is a key mediator of cytotoxic T lymphocytes cytotoxicity [ 30 , 31 ], we dissociated tissues into single-cell suspensions and measured granzyme B expression in CD8⁺ T cells by flow cytometry (Fig.  1 E). CD8⁺ T cells from ectopic lesions showed markedly reduced granzyme B levels, whereas no difference was observed in eutopic tissues between EM patients and controls (Fig.  1 E), indicating impaired cytotoxic function of CD8⁺ T cells in the ectopic lesions of EM. Fig. 1 Localized iron overload and CD8 + T cell dysfunction with ferroptosis activation. A Prussian blue staining of eutopic and ectopic endometrial tissues. Scale bar (200x) = 150 μm. B Prussian blue-positive area of eutopic and ectopic endometrial tissues (n [Eu-NC, Eu-EM, Ec-EM] = 15, 14, 19). C and D Representative immunohistochemical staining and quantitative analysis of CD8 + T cells in eutopic and ectopic endometrial tissues (n [Eu-NC, Eu-EM, Ec-EM] = 12, 10, 17). Scale bar (200x) = 150 μm. E The secretion level of granzyme B in CD8 + T cells in eutopic and ectopic tissues was determined by flow cytometry (n [Eu-NC, Eu-EM, Ec-EM] = 16, 18, 19). F and G The ferrous ion content of CD8 + T cells was determined by flow cytometry using BioTracker Far-red Labile Fe2+ Dye (n [Eu-NC, Eu-EM, Ec-EM] = 16, 18, 19). H The lipid peroxidation levels of CD8 + T cells in eutopic and ectopic tissues were determined by flow cytometry using Lipid Peroxidation Assay Kit (n [Eu-NC, Eu-EM, Ec-EM] = 16, 18, 19). I The expression level of GPX4 in CD8 + T cells in eutopic and ectopic tissues was detected by flow cytometry (n [Eu-NC, Eu-EM, Ec-EM] = 16, 18, 19). J and K Representative immunofluorescence images depicting expression of Granzyme B and GPX4 co-stained with CD8 in eutopic and ectopic tissues, Scale bars (200x) = 150 μm. ** P  < 0.01, *** P  < 0.001; Values represent the mean ± standard error. ns, not statistically significant. EM, endometriosis; Eu-NC, eutopic endometrial tissue of normal controls; Eu-EM, eutopic endometrial tissue of patient with endometriosis; Ec-EM, ectopic endometrial tissue of patient with endometriosis; GPX4, glutathione peroxidase 4 Localized iron overload and CD8 + T cell dysfunction with ferroptosis activation. A Prussian blue staining of eutopic and ectopic endometrial tissues. Scale bar (200x) = 150 μm. B Prussian blue-positive area of eutopic and ectopic endometrial tissues (n [Eu-NC, Eu-EM, Ec-EM] = 15, 14, 19). C and D Representative immunohistochemical staining and quantitative analysis of CD8 + T cells in eutopic and ectopic endometrial tissues (n [Eu-NC, Eu-EM, Ec-EM] = 12, 10, 17). Scale bar (200x) = 150 μm. E The secretion level of granzyme B in CD8 + T cells in eutopic and ectopic tissues was determined by flow cytometry (n [Eu-NC, Eu-EM, Ec-EM] = 16, 18, 19). F and G The ferrous ion content of CD8 + T cells was determined by flow cytometry using BioTracker Far-red Labile Fe2+ Dye (n [Eu-NC, Eu-EM, Ec-EM] = 16, 18, 19). H The lipid peroxidation levels of CD8 + T cells in eutopic and ectopic tissues were determined by flow cytometry using Lipid Peroxidation Assay Kit (n [Eu-NC, Eu-EM, Ec-EM] = 16, 18, 19). I The expression level of GPX4 in CD8 + T cells in eutopic and ectopic tissues was detected by flow cytometry (n [Eu-NC, Eu-EM, Ec-EM] = 16, 18, 19). J and K Representative immunofluorescence images depicting expression of Granzyme B and GPX4 co-stained with CD8 in eutopic and ectopic tissues, Scale bars (200x) = 150 μm. ** P  < 0.01, *** P  < 0.001; Values represent the mean ± standard error. ns, not statistically significant. EM, endometriosis; Eu-NC, eutopic endometrial tissue of normal controls; Eu-EM, eutopic endometrial tissue of patient with endometriosis; Ec-EM, ectopic endometrial tissue of patient with endometriosis; GPX4, glutathione peroxidase 4 To assess iron overload in CD8⁺ T cells, we quantified intracellular Fe 2 ⁺ using a ferrous ion–specific probe (Fig.  1 F). CD8⁺ T cells from ectopic lesions showed a significantly higher proportion of Fe 2 ⁺-positive cells than those from eutopic tissue (Fig.  1 G). Ferroptosis is characterised by diminished antioxidant capacity, elevated lipid peroxidation, and reduced GPX4 activity or expression [ 32 ]. As the ratio of PE/FITC drops, the level of lipid peroxidation increases. Consistent with this, the PE/FITC ratio in CD8⁺ T cells from ectopic lesions was decreased, indicating increased lipid peroxidation (Fig.  1 H). Moreover, GPX4 expression in CD8⁺ T cells from ectopic tissues was significantly lower than in eutopic counterparts, with no significant difference between EM eutopic tissues and controls (Fig.  1 I). Immunofluorescence analysis confirmed GPX4 and granzyme B co-localisation in CD8⁺ T cells, corroborating the flow cytometry findings (Fig.  1 J and K). Thus, iron overload was present in CD8 + T cells in the ectopic tissues of EM, and the abnormal immune function of CD8 + T cells might be related to ferroptosis. Next, we explored the relationship between iron overload and ferroptosis in CD8 + T cells. CD8 + T cells are affected by iron overload; therefore, we simulated iron overload by stimulating CD8 + T cells with different concentrations of AIC. With the increase of iron ion concentration, the cell activity decreased (Additional file 1: Fig. S3). At 500 µM AIC, viability declined markedly but was rescued by the ferroptosis inhibitor Fer-1 (Fig.  2 A), so this concentration was used for subsequent assays. Quantification of ROS and lipid peroxidation confirmed that excess iron elevated both parameters, which were reversed by Fer-1 (Fig.  2 B and C). To confirm ferroptosis of CD8 + T cells, we observed the mitochondrial morphological structure of CD8 + T cells using a transmission electron microscope. Iron-treated CD8⁺ T cell mitochondria were shrunken, exhibited increased membrane density, and displayed disrupted cristae architecture (Fig.  2 D). At the mRNA level, iron overload upregulated Fth1, Acsl4 and Ho-1 while downregulating xCT and Gpx4 (Fig.  2 E and Additional file 2: The raw data of qRT-PCR). Western blotting showed increased FTH1 and ACSL4 proteins, a trend toward higher HO-1, and reduced xCT and GPX4 expression (Fig.  2 F–K and Additional file 3: Fig. S1). Fig. 2 CD8 + T cell treated with excess iron show ferroptosis characteristics. A CD8 + T cells activity was analyzed by flow cytometry using Fixable Viability Dye eFluor™ 450 after treatment with 500 μM iron ion and the ferroptosis inhibitor Fer-1 ( n  = 3). B Total ROS levels of CD8 + T cells determined by flow cytometry using CellROX ® Green Reagent after treatment with 500 μM iron ion and Fer-1. C Lipid peroxidation level of CD8 + T cells evaluated by flow cytometry using Lipid Peroxidation Sensor after treatment with 500 μM iron ion and Fer-1 ( n  = 3). D Transmission electron microscopy analysis of mitochondrial ultrastructure in CD8 + T cells treated with 500 μM iron ion and Fer-1. The structures indicated by the arrows are the mitochondria. Scale bar = 1 μm. E The mRNA levels of ferroptosis genes of CD8 + T cells determined using RT-qPCR ( n  = 3). F – K The protein level of ferroptosis of CD8 + T cells treated with or without 500 μM iron ion determined by Western Blot ( n  = 3). L Proliferation of CD8 + T cells treated with or without 500 μM iron ion and Fer-1 was measured by flow cytometry depending on CFSE 72 h after culture start. M The secretion of granzyme B in CD8 + T cells determined by flow cytometry. N The apoptosis of MESCs was measured by flow cytometry. O and P The chemotactic ability of CD8 + T cells was evaluated by Transwell cell migration assay. Scale bar = 150 μm ( Q ) The migration ability of coMESCs was evaluated by the Transwell cell migration method. Scale bar = 300 μm ( R ) The proliferation ability of coMESCs was detected by EdU labeling assay. ** P  < 0.01, *** P  < 0.001. Values represent the mean ± standard error. ns, not statistically significant. CFSE, 5,6-carboxyfluorescein diacetate succinimidyl ester; Fer-1, Ferrostatin-1; MESCs, mouse endometrial stromal cells; ROS, reactive oxygen species; xCT, cystine/glutamate transporter; HO-1, heme oxygenase 1; GPX4, glutathione peroxidase 4 CD8 + T cell treated with excess iron show ferroptosis characteristics. A CD8 + T cells activity was analyzed by flow cytometry using Fixable Viability Dye eFluor™ 450 after treatment with 500 μM iron ion and the ferroptosis inhibitor Fer-1 ( n  = 3). B Total ROS levels of CD8 + T cells determined by flow cytometry using CellROX ® Green Reagent after treatment with 500 μM iron ion and Fer-1. C Lipid peroxidation level of CD8 + T cells evaluated by flow cytometry using Lipid Peroxidation Sensor after treatment with 500 μM iron ion and Fer-1 ( n  = 3). D Transmission electron microscopy analysis of mitochondrial ultrastructure in CD8 + T cells treated with 500 μM iron ion and Fer-1. The structures indicated by the arrows are the mitochondria. Scale bar = 1 μm. E The mRNA levels of ferroptosis genes of CD8 + T cells determined using RT-qPCR ( n  = 3). F – K The protein level of ferroptosis of CD8 + T cells treated with or without 500 μM iron ion determined by Western Blot ( n  = 3). L Proliferation of CD8 + T cells treated with or without 500 μM iron ion and Fer-1 was measured by flow cytometry depending on CFSE 72 h after culture start. M The secretion of granzyme B in CD8 + T cells determined by flow cytometry. N The apoptosis of MESCs was measured by flow cytometry. O and P The chemotactic ability of CD8 + T cells was evaluated by Transwell cell migration assay. Scale bar = 150 μm ( Q ) The migration ability of coMESCs was evaluated by the Transwell cell migration method. Scale bar = 300 μm ( R ) The proliferation ability of coMESCs was detected by EdU labeling assay. ** P  < 0.01, *** P  < 0.001. Values represent the mean ± standard error. ns, not statistically significant. CFSE, 5,6-carboxyfluorescein diacetate succinimidyl ester; Fer-1, Ferrostatin-1; MESCs, mouse endometrial stromal cells; ROS, reactive oxygen species; xCT, cystine/glutamate transporter; HO-1, heme oxygenase 1; GPX4, glutathione peroxidase 4 To assess the impact of iron overload on CD8⁺ T cell function, we evaluated their proliferation, migration, and cytotoxic capacity. Excessive iron inhibited CD8⁺ T cell proliferation, migration, and killing ability, but iron-overloaded CD8 +  T cells had no effect on the proliferation and migration of co-cultured MESCs (Fig.  2 L–R). These results indicate that iron overload–induced ferroptosis compromises CD8⁺ T cell function. To evaluate the in vivo impact of iron overload on EM progression, we established a murine endometriosis model and administered intraperitoneal PBS, iron dextran (Fe), or Fe plus DFOM (Fig.  3 A). Typical ectopic lesions with adhesions to surrounding organs developed in all groups, but the abdominal environment of the mice in the Fe group turned red (Fig.  3 B). Iron dextran significantly increased lesion volume, whereas DFOM co-treatment reduced lesion size (Fig.  3 C). Quantitative analysis showed a significant rise in lesion weight in the Fe group without altering lesion number (Fig.  3 D and E). Histological examination of excised lesions by hematoxylin and eosin (H&E) staining confirmed characteristic endometrial glandular and stromal architecture (Fig.  3 F). Prussian blue staining revealed extensive iron deposition in Fe-treated lesions, which was attenuated by DFOM (Fig.  3 G and H). These results demonstrate that iron overload exacerbates EM lesion progression. Fig. 3 Disease progression and ferroptosis of CD8 + T cells caused by iron overload can be reversed by DFOM. A Experimental treatment flowchart. B Representative visible lesions of the experimental groups. C Macroscopic aspect of implants of different groups at 17 days. D Ectopic lesion weights. E Number of ectopic lesions. F H&E staining of ectopic tissues to verify the EM model success. Scale bar (100x) = 300 μm. Scale bar (200x) = 150 μm. G Representative images of Prussian blue staining of ectopic tissues. Scale bar (200x) = 150 μm. H Prussian blue-stained areas of ectopic tissues. I Ferrous ion content of mouse peritoneal CD8 + T cells. J Total ROS levels of mouse peritoneal CD8 + T cells determined by flow cytometry using CellROX ® Green Reagent. K Lipid peroxidation level of mouse peritoneal CD8 + T cells. L GPX4 expression level of mouse primary CD8 + T cells determined by flow cytometry. M The secretion of granzyme B in CD8 + T cells determined by flow cytometry. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, n = 5. Values represent the mean ± standard error. ns, not statistically significant. DFOM, deferoxamine mesylate; EM, endometriosis; H&E, hematoxylin and eosin; ROS, reactive oxygen species; PBS, phosphate buffer saline; GPX4, glutathione peroxidase 4 Disease progression and ferroptosis of CD8 + T cells caused by iron overload can be reversed by DFOM. A Experimental treatment flowchart. B Representative visible lesions of the experimental groups. C Macroscopic aspect of implants of different groups at 17 days. D Ectopic lesion weights. E Number of ectopic lesions. F H&E staining of ectopic tissues to verify the EM model success. Scale bar (100x) = 300 μm. Scale bar (200x) = 150 μm. G Representative images of Prussian blue staining of ectopic tissues. Scale bar (200x) = 150 μm. H Prussian blue-stained areas of ectopic tissues. I Ferrous ion content of mouse peritoneal CD8 + T cells. J Total ROS levels of mouse peritoneal CD8 + T cells determined by flow cytometry using CellROX ® Green Reagent. K Lipid peroxidation level of mouse peritoneal CD8 + T cells. L GPX4 expression level of mouse primary CD8 + T cells determined by flow cytometry. M The secretion of granzyme B in CD8 + T cells determined by flow cytometry. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, n = 5. Values represent the mean ± standard error. ns, not statistically significant. DFOM, deferoxamine mesylate; EM, endometriosis; H&E, hematoxylin and eosin; ROS, reactive oxygen species; PBS, phosphate buffer saline; GPX4, glutathione peroxidase 4 The peritoneal cavity is the main site of ectopic lesions in the mouse endometriosis model; thus, we collected peritoneal cells to characterise the local immune microenvironment. CD8⁺ T cells were analysed by flow cytometry for intracellular Fe 2 ⁺, ROS, and lipid peroxidation, as well as GPX4 expression. In the Fe and Fe + DFOM groups, CD8⁺ T cells exhibited significantly higher iron and ROS levels than in the PBS and EM groups (Fig.  3 I and J). In the Fe group, the PE/FITC ratio dropped markedly, indicating elevated lipid peroxidation, whereas DFOM restored this ratio to control levels (Fig.  3 K). GPX4 expression was reduced by iron overload but rescued by DFOM treatment (Fig.  3 L). We then assessed CD8⁺ T cell function and observed that granzyme B secretion was suppressed in the Fe group and returned to baseline with DFOM (Fig.  3 M). These findings demonstrate that DFOM reverses iron overload–induced ferroptosis, functional impairment of CD8⁺ T cells and the disease progression. We established the mouse models to investigate the function of CD8 + T cells in the development of EM in vivo. As exhibited in Fig.  4 A, mouse models were peritoneally injected with CD8A antibody to inhibit CD8 + T cells, and untreated CD8 + T cells and iron-overloaded CD8 + T cells were injected. As shown in the Fig.  4 B and C, the lesions of the CD8A + Fe-TCs group were larger than all other groups. Lesion weight increased following CD8⁺ T cell depletion (Additional file 1: Fig. S4), demonstrating that CD8⁺ T cells significantly influence lesion progression. However, the lesion weight of the mouse model supplemented with normal CD8 + T cells was similar to that of normal mice. The results confirmed that iron-treated CD8 + T cells presented the status of immune functional attenuation. To investigate the effect of iron on the migration ability of CD8 + T cells, we detected the percentage of fluorescently labelled CD8 + T cells in lesion tissues. We found that the infiltration of iron-overloaded CD8 + T cells in the lesions was significantly reduced (Fig.  4 D). This was consistent with the results of in vitro experiments. Subsequently, we further examined the migration of iron-overloaded CD8 + T cells to the peripheral blood at different periods. The migration of CD8 + T was not detected in the peripheral blood at each time period (Fig.  4 E and F). Our results indicated that iron overload-mediated ferroptosis of CD8 + T cells contributed to the development of EM. Fig. 4 Iron overload-mediated ferroptosis of CD8 + T cells contributes to the development of EM. A The flowchart of establishing the mouse models with different treatments to analyse the function of CD8 + T cells in EM. B The diagram presenting the lesions collected from different mouse model groups ( n  = 5). C The weight of ectopic lesions ( n  = 5). D The percentage of Far red + cells in lesion tissues 48 h after reinfusion of untreated CD8 + T cells and iron-overloaded CD8 + T cells was determined by flow cytometry ( n  = 6). E and F The percentage of Far red + cells in peripheral blood at different time periods after reinfusion of untreated CD8 + T cells and iron-overloaded CD8 + T cells was determined by flow cytometry. * P  < 0.05, ** P  < 0.01, *** P  < 0.001. Values represent the mean ± standard error. ns, not statistically significant. EM, endometriosis; PBS, phosphate buffer saline Iron overload-mediated ferroptosis of CD8 + T cells contributes to the development of EM. A The flowchart of establishing the mouse models with different treatments to analyse the function of CD8 + T cells in EM. B The diagram presenting the lesions collected from different mouse model groups ( n  = 5). C The weight of ectopic lesions ( n  = 5). D The percentage of Far red + cells in lesion tissues 48 h after reinfusion of untreated CD8 + T cells and iron-overloaded CD8 + T cells was determined by flow cytometry ( n  = 6). E and F The percentage of Far red + cells in peripheral blood at different time periods after reinfusion of untreated CD8 + T cells and iron-overloaded CD8 + T cells was determined by flow cytometry. * P  < 0.05, ** P  < 0.01, *** P  < 0.001. Values represent the mean ± standard error. ns, not statistically significant. EM, endometriosis; PBS, phosphate buffer saline Because iron overload is not conducive to cell survival, we want to find the mechanisms that regulate ferroptosis of CD8 + T cells under the environment of iron overload. We found that iron overload led to an increase in the expression of p53 at the mRNA and protein levels in CD8 + T cells (Fig.  5 A–C, Additional file 2: The raw data of qRT-PCR and Additional file 3: Fig. S2). This might be caused by iron overload. To explore the relationship between ferroptosis and p53, we treated iron-overloaded CD8 + T cells with the p53 inhibitor PFT-α. We found that at the same iron concentration, the reduction in CD8 + T cell activity caused by iron overload was rescued by PFT-α (Fig.  5 D and E). Meanwhile, the production of ROS and the level of lipid peroxidation in CD8 + T cells decreased, while the content of GSH increased, indicating that p53 inhibition reduced the level of lipid peroxidation (Fig.  5 F–H). We confirmed that p53 in iron-overloaded environments could be inhibited by PFT-α by measuring p53 protein (Fig. 5 I and J and Additional file 3: Fig. S3). Furthermore, we measured the level of ferroptosis proteins and found that p53 inhibition increased the levels of xCT and GPX4 (Fig.  5 K and L and Additional file 3: Fig. S3). To determine the effect of p53 on CD8 + T cells, we examined the cytotoxicity of CD8 + T cells. We found that p53 inhibition reversed the inhibitory effect of iron overload on the cytotoxicity of CD8 + T cells (Fig.  5 M and N). These results indicate that inhibiting p53 can resist ferroptosis of CD8 + T cells caused by iron overload. Fig. 5 Iron overload induces ferroptosis of CD8 + T cells by the p53/xCT/GPX4 pathway. A The mRNA levels of p53 gene of CD8 + T cells determined using RT-qPCR ( n  = 3). B and C The protein level of p53 of CD8 + T cells treated with or without 500 μM iron ion determined by Western Blot ( n  = 3). D and E CD8 + T cells activity was analyzed by flow cytometry using Fixable Viability Dye eFluor™ 450 after treatment with 500 μM iron ion and PET-α ( n  = 3). F Total ROS levels of CD8 + T cells determined by flow cytometry using CellROX ® Green Reagent. G Lipid peroxidation level of CD8 + T cells evaluated by flow cytometry using Lipid Peroxidation Sensor ( n  = 3). H Relative GSH level in CD8 + T cells ( n  = 3). I – L The protein level of ferroptosis of CD8 + T cells treated with 500 μM iron ion and PET-α determined by Western Blot ( n  = 3). M and N The secretion of granzyme B in CD8 + T cells determined by flow cytometry. O PET-α could slow down the growth of ectopic lesions in vivo ( n  = 5). P Ectopic lesion weights. * P  < 0.05, ** P  < 0.01, *** P  < 0.001. Values represent the mean ± standard error. ns, not statistically significant. PET-α, pifithrin-α hydrobromide; ROS, reactive oxygen species; GPX4, glutathione peroxidase 4 Iron overload induces ferroptosis of CD8 + T cells by the p53/xCT/GPX4 pathway. A The mRNA levels of p53 gene of CD8 + T cells determined using RT-qPCR ( n  = 3). B and C The protein level of p53 of CD8 + T cells treated with or without 500 μM iron ion determined by Western Blot ( n  = 3). D and E CD8 + T cells activity was analyzed by flow cytometry using Fixable Viability Dye eFluor™ 450 after treatment with 500 μM iron ion and PET-α ( n  = 3). F Total ROS levels of CD8 + T cells determined by flow cytometry using CellROX ® Green Reagent. G Lipid peroxidation level of CD8 + T cells evaluated by flow cytometry using Lipid Peroxidation Sensor ( n  = 3). H Relative GSH level in CD8 + T cells ( n  = 3). I – L The protein level of ferroptosis of CD8 + T cells treated with 500 μM iron ion and PET-α determined by Western Blot ( n  = 3). M and N The secretion of granzyme B in CD8 + T cells determined by flow cytometry. O PET-α could slow down the growth of ectopic lesions in vivo ( n  = 5). P Ectopic lesion weights. * P  < 0.05, ** P  < 0.01, *** P  < 0.001. Values represent the mean ± standard error. ns, not statistically significant. PET-α, pifithrin-α hydrobromide; ROS, reactive oxygen species; GPX4, glutathione peroxidase 4 To assess the impact of p53 inhibition on EM progression in vivo, mice bearing ectopic lesions received intraperitoneal PBS, iron dextran, or iron dextran with PFT-α. Compared with iron dextran alone, PFT-α co-treatment significantly reduced both lesion volume and weight (Fig.  5 O and P). These results demonstrate that p53 inhibition attenuates EM lesion development.

Endometriosis (EM) is a complex clinical syndrome that is common in women of reproductive age, and is defined by endometrial-like tissue outside the uterine cavity [ 1 – 3 ]. Retrograde menstruation is the leading pathogenic hypothesis [ 4 ], and ectopic lesions periodically shed and bleed, resulting in localised iron overload [ 5 – 7 ]. Excess iron fosters an inflammatory microenvironment and adhesions, which contribute to chronic pain and infertility in EM patients [ 8 ]. However, the precise roles and mechanisms of iron overload in EM pathogenesis remain undefined. Although iron is vital for cellular metabolism, surplus ferrous ions catalyse reactive oxygen species (ROS) generation via the Fenton reaction, triggering lipid peroxidation and ferroptosis [ 9 , 10 ]. Ferroptosis is related to the pathogenesis of diseases, such as cancer, cardiovascular disease, diabetes, and neurodegeneration [ 11 ]. Ferroptosis can be initiated by inhibiting cystine–glutamate antiporter (xCT) function or by impairing intracellular antioxidant defences such as glutathione peroxidase 4 (GPX4) [ 12 ]. Although iron overload and ferroptosis can affect the occurrence and development of EM, the key cell types and the underlying mechanisms remain unclear [ 8 , 13 – 15 ]. The immune milieu critically regulates ectopic colonization, migration, and proliferation of endometrial stromal cells (ESC) [ 16 ]. T cells dominate the immune infiltrate in endometrial tissue [ 17 ], with CD8⁺ T cells serving as central effectors of adaptive immunity, homeostasis maintenance, and pathological surveillance [ 18 ]. Paradoxically, EM lesions are enriched in CD8⁺ T cells that display diminished activation and cytotoxicity [ 19 , 20 ], suggesting defective lesion clearance. In mouse EM models, anti-CD8A–mediated CD8⁺ T cell depletion increases lesion burden, whereas adoptive CD8⁺ T cell transfer reverses this effect [ 21 ], highlighting their protective role. In oncology, ferroptosis of tumour-infiltrating CD8⁺ T cells dampens antitumour immunity [ 22 ], but the regulation of ferroptosis in CD8⁺ T cells in EM has not been explored. Here, we investigate how iron overload in EM lesions alters CD8⁺ T cell function via ferroptosis and evaluate potential interventions both in vitro and in vivo.

This study clarifies how localised iron overload induces ferroptosis in CD8⁺ T cells, driving their dysfunction and immune tolerance in EM. We demonstrate that iron accumulation in EM lesions triggers p53-mediated inhibition of xCT/GPX4, leading to reduced CD8⁺ T cell activity, impaired lesion clearance, and accelerated EM progression. Chronic bleeding adjacent to endometrioid tissues is a surgical‐pathological criterion for EM [ 33 ], and our previous work confirmed iron overload in cystic fluid and lesion sites [ 34 ]. The immune system is a key determinant of the progression of EM. Ferroptosis is a form of cell death regulated by iron and lipid peroxidation dependence [ 35 ]. The prevalent iron overload in ectopic tissues promotes the occurrence of ferroptosis, which may affect the survival of immune cells in endometrial tissues. Although CD8 + T cells are closely associated with the progression of EM, current research on CD8 + T cells in EM remains relatively scarce. Therefore, this study focuses on CD8 + T cells. Our research results confirmed that CD8⁺ T cells were enriched in EM lesions, and their cytotoxic function was significantly reduced. The results are consistent with the findings of Ma et al. [ 20 ]. Here, we found that CD8 + T cells from ectopic sites harbour higher iron levels than those from eutopic endometrium. However, the effect of iron overload on CD8 + T cells in EM remains an unknown area. This study aims to determine the role of the effect of excessive iron on CD8 + T cells in the occurrence and development of EM. Iron reversibly gains or loses individual electrons through transitions between different oxidation states and is a basic element in important enzymatic redox reactions in cells [ 36 ]. Excessive Fe 2+ forms an unstable iron pool in the cytoplasm, reacts with hydrogen peroxide and generates a large number of hydroxyl radicals, leading to lipid peroxidation and cell membrane damage in cells, and ultimately triggering ferroptosis [ 15 , 37 , 38 ]. We found that the lipid peroxidation level of CD8 + T cells in ectopic lesions increased, and we speculated that ferroptosis was activated in CD8 + T cells. Rapid progress has been made in the research of ferroptosis mechanisms, and the cystine/glutamate transporter (also known as the xCT system)—glutathione (GSH)—GPX4 pathway is a recognised regulatory system for ferroptosis lipid peroxidation [ 39 , 40 ]. xCT is responsible for transporting cystine into cells and preventing ferroptosis by promoting the production of glutathione [ 41 , 42 ]. GSH is a key cofactor for GPX4 to exert its antioxidant function, inhibiting ferroptosis by eliminating lipid peroxides. If the function of xCT is inhibited, the synthesis of GSH is blocked, which indirectly leads to the loss of GPX4 activity [ 43 ]. GPX4 maintains membrane integrity by reducing glutathione-dependent lipid peroxides (e.g., PE-OOH). Its activation may counteract iron overload-induced oxidative damage by enhancing substrate binding efficiency [ 23 , 44 , 45 ]. GPX4 is a central regulator of ferroptosis [ 46 – 48 ]. Therefore, we evaluated the protein levels of GPX4 in CD8 + T cells in eutopic and ectopic tissues. We found that the protein levels of GPX4 in CD8 + T cells in ectopic tissues were lower than that in eutopic tissues, while there was no significant difference in the eutopic endometrium between EM patients and the normal controls. Therefore, our results confirmed that CD8 + T cells in ectopic tissues undergo ferroptosis and have abnormal effector functions. Most current studies on EM and ferroptosis have been limited to stimulation with ferroptosis inducers and have ignored the clinical features of local iron overload. Next, we verified the relationship between iron overload and ferroptosis in CD8 + T cells in vitro and in vivo. In vitro, we simulated the environment of iron overload by stimulating CD8 + T cells with excess AIC. Our results suggested that excess iron can activate ferroptosis in CD8 + T cells. The use of Fer-1 only partially restored cell activity, possibly because Fer-1 inhibits ferroptosis by reducing lipid peroxidation levels [ 49 ]. To confirm this, we further assessed partial ferroptosis protein levels in CD8 + T cells. Ferritin, an iron storage protein, is a well-known inflammatory and iron overload biomarker [ 50 ]. FTH1 has ferroxidase activity that allows the safe incorporation of iron into ferritin for storage [ 51 ]. ACSL4 is a key enzyme that catalyses the conversion of PUFA to PUFA-COA and contributes to ferroptosis [ 52 ]. HO-1 can cause harmful effects through the accumulation of excess free iron and promote the occurrence of ferroptosis [ 53 ]. We found that at the protein level, FTH1 and ACSL4 expression increased, and xCT and GPX4 expression decreased, as expected given their mRNA levels, while HO-1 expression only showed an increasing trend. This may be due to the fact that protein is an accumulation process. These results indicate that iron overload activates ferroptosis of CD8 + T cells, thereby elucidating clinical phenomena associated with changes in ferroptosis in the context of EM. Although iron-overloaded CD8 + T cells had no effect on the proliferation and migration of co-cultured MESCs, iron overload could inhibit the proliferation, migration and killing ability of CD8 + T cells; similar phenomena have also been observed in other studies on iron overload [ 54 ]. Therefore, localised iron overload in ectopic lesions of EM may induce ferroptosis of CD8 + T cells and reduce their immune potential, thereby promoting the progression of the disease. We confirmed this in vivo using mouse models. Furthermore, we found that the infiltration of iron-overloaded CD8 + T cells in the lesion decreased, a result that contradicted the phenomenon in clinical samples. The accumulation of CD8 + T cells in ectopic tissues of EM patients may result from the active recruitment of normal CD8 + T cells by inflammatory signals. In vitro stimulated iron-overloaded CD8 + T cells exhibited impaired migratory capacity, likely due to intrinsic dysfunction. Furthermore, reinfused CD8 + T cells did not reappear in peripheral blood. During peritoneal inflammation, immune activation drives CD8 + T cells homing to the affected site; as inflammation resolves, these cells may exit the peritoneal cavity and re-enter the circulation via lymphatic and vascular routes [ 55 ]. DFOM is an effective iron chelator that can chelate free and non-heme iron ions bound to ferritin and hemosiderin. Iron chelators have been explored for cancer treatment [ 56 , 57 ]. In this study, DFOM treatment inhibited the progression of the disease and alleviated ferroptosis of CD8 + T cells. Thus, iron chelators may benefit EM by reducing the iron content in the lesion microenvironment. We found that iron overload increased the expression of p53 in CD8 + T cells in vitro. The p53 protein is best understood as a stress response factor and also a DNA sequence-specific transcription factor [ 58 , 59 ]. Recent studies have shown that p53 plays a crucial role in ferroptosis. Xiao et al. found that after treating cardiomyocytes with ferric citrate, an increase in p53 at both mRNA and protein levels was observed [ 60 ]. As an upstream mediator of xCT, p53 can inhibit the uptake of cystine by suppressing the expression of xCT, thereby inducing ferroptosis to hinder the growth of lung cancer [ 61 ]. Furthermore, p53 could activate ALOX12 by inhibiting the transcription of SLC7A11. Free ALOX12 could oxidise membrane phospholipid polyunsaturated fatty acid chains and mediate ferroptosis under ROS stress [ 62 ]. These studies emphasise the important role of p53 as a potential therapeutic target for regulating ferroptosis. To this end, we treated iron-overloaded CD8 + T cells with the p53 inhibitor PFT-α and found that p53 inhibition effectively alleviated ferroptosis and changed the expression levels of ferroptosis proteins xCT and GPX4. This indicates that p53 activation mainly induces ferroptosis of CD8 + T cells by inhibiting the xCT/GPX4 axis. As reported by Jiang et al. and Yuan et al., p53-mediated systemic xCT inhibition induces ferroptosis in cells [ 63 , 64 ]. Furthermore, PFT-α reversed the functional inhibitory state of CD8 + T cells and enhanced the killing abilities of CD8 + T cells. In vivo, under the condition of iron overload, p53 inhibition slowed down the growth of the lesions, which was consistent with the in vitro experimental results. These results suggest that iron overload may promote the progression of EM through p53-dependent ferroptosis. Of course, our research also has limitations. Our research indicates that iron overload can significantly affect the immune function status of CD8 + T cells through p53/xCT/GPX4 pathway; however, it is still necessary to expand the sample size to verify our findings. Iron overload may suppress GPX4 transcription via HIF-1α or NF-κB pathways, requiring further validation [ 65 , 66 ]. Another limitation is the limited research on CD8 + T cell subsets (such as effector and memory cells) and their potential differential susceptibility to ferroptosis. In future work, we will expand the sample size and enhance our analysis of subset sensitivity to ferroptosis. Moreover, the direct interaction between iron-overloaded CD8 + T cells and ESCs, as well as the transcriptional changes behind the abnormal function of ESCs, have not been fully clarified and require more in-depth exploration in subsequent studies.

Overall, iron overload activates p53 to inhibit the xCT/GPX4 axis, triggering ferroptosis in CD8⁺ T cells and impairing their immune activity. This work establishes a novel link between iron overload, ferroptosis, and CD8⁺ T cell dysfunction in EM, filling a critical knowledge gap and advancing our understanding of EM pathogenesis. Modulating p53-mediated ferroptosis sensitivity in CD8⁺ T cells may offer a promising therapeutic strategy. However, our findings are limited by the sample size and the need to clarify differences in ferroptosis sensitivity among CD8 + T cell subsets, the direct interactions between iron-overloaded CD8⁺ T cells and ESCs, and the resulting transcriptional changes in ESCs, which warrant investigation in future studies.

Additional file 1. Additional file 2. Additional file 3. Additional file 1. Additional file 2. Additional file 3.