Artemisinin increases susceptibility to ferroptosis of fibroblast-like synoviocytes in rheumatoid arthritis Authors: Maoyuan Li, Jiahao Wang, Yunyuan Yu, Liangliang Wang, Feng Lu, Shishuo Li, Jiale Wang, Yimin Liu, Guangrong Yin, Chao Xu, Yuji Wang Published: 2026 ABSTRACT AbstractObjectiveRheumatoid arthritis (RA) is a chronic inflammatory disease characterized by extensive synovial hyperplasia. Artemisinin (ART), a compound extracted from the traditional Chinese herb artemisia annua, has been proven to have anti-malarial and anti-cancer effects. Recently, the ability of artemisinin to induce ferroptosis has shown inhibitory effects on tumor cell proliferation. This study intends to examine whether artemisinin can inhibit the proliferation of RA synovial cells by sensitizing them to ferroptosis.MethodThe effect of artemisinin on fibroblast-like synoviocytes (FLSs) were detected by methods such as CCK8, Western Blot, and immunofluorescence. A collagen-induced mouse model of rheumatoid arthritis was established by injecting collagen and Freund’s adjuvant, and the mice were treated with artemisinin following model induction. Synovium was extracted, and imaging examinations along with immunostaining were used to identify the progression of ferroptosis.ResultsThe results revealed that ART increases the susceptibility of FLSs to ferroptosis inducers. The expression of ferroptosis-related proteins, such as glutathione peroxidase 4 (GPX4) and solute carrier family 7 member 11 (SLC7A11), was decreased compared with controls. Meanwhile, intracellular lipid peroxides continued to accumulate after artemisinin treatment. Additionally, ferroptosis inhibitors such as deferoxamine (DFO) and Ferrostatin (Fer-1) can inhibit artemisinin-sensitized ferroptosis. The artemisinin-sensitized ferroptosis was related to the increase of P53. In vivo experiments on collagen-induced arthritis mouse models further confirmed that artemisinin relieved joint injury by sensitizing synovial ferroptosis, which was abolished by ferroptosis inhibitors.ConclusionArtemisinin can increase the sensitivity of FLSs to ferroptosis and inhibit the proliferation of synovial cells in rheumatoid arthritis. The P53-SLC7A11 pathway play the important role in this process.Keywords: Artemisinin, Rheumatoid arthritis, Ferroptosis, Fibroblast-like synoviocyte Introduction Rheumatoid arthritis is a chronic autoimmune inflammatory disease of the joints [1]. The characteristic features of RA include abnormal proliferation and inflammatory response of synovial cells. The excessive proliferation and inflammation of synovial cells lead to joint pain, swelling, and functional impairment, severely affecting patients’ quality of life [2]. FLSs contribute critically to the pathological process of RA, participating in its onset, development, and perpetuation. Activated RA-FLSs exhibit characteristics similar to tumor cells, contributing to the progression of rheumatoid arthritis and ultimately leading to joint destruction [3]. Currently, ferroptosis is considered to be involved in the progression of rheumatoid arthritis (RA) and is an emerging target for RA. This study explored the function and underlying mechanism of ART in enhancing ferroptosis in RA, aiming to provide new targets for the prevention and treatment of RA. Ferroptosis is a form of programmed cell death. Unlike apoptosis and necrosis, ferroptosis is independent of caspase activity and receptor-interacting protein kinase 1 (RIPK1) activity [4]. Ferroptosis depends on the accumulation of iron and lipid peroxides in mitochondria. This process can be antagonized by GPX4 and iron inhibitors. The morphology of ferroptosis also differs, showing diminished mitochondrial volume, decreased membrane density, and disruption of the outer mitochondrial membrane [5]. An important feature of ferroptosis is its dependence on iron, where transferrin binds to free iron (Fe3+) and transports it into the cell. Fe3+ enters the cell and is catalyzed to Fe2+, which is finally released into the labile iron pool of the cell through divalent metal transporter 1 (DMT1), inducing ferroptosis [6, 7]. Due to its dependence on iron, this type of cell death can be rescued by iron chelators such as deferoxamine (DFO). Another important feature of ferroptosis is the accumulation of lipid peroxides in the cell membrane prior to cell death. Lipid peroxides can disrupt the stability of the lipid bilayer, leading to cell membrane disintegration. Inhibition of GPX4 is considered a key step in ferroptosis, as it requires glutathione (GSH) as an electron donor to reduce intracellular lipid peroxidation. Glutathione, a tripeptide, is an important antioxidant in the cell. The synthesis of glutathione depends on the utilization of cysteine within the cell, which can be transported from outside the cell to inside via the XC- system. XC- is a dimer composed of a heavy chain (4F2, gene name SLC3A2) and a light chain (XCT, gene name SLC7A11). Interestingly, the XCT subunit has been demonstrated to be the molecular target of erastin, a classic inducer of ferroptosis, which triggers ferroptosis through cystine depletion [8, 9]. In rheumatoid arthritis, changes in iron content within the human body have become a focus of attention, especially the increased levels of free iron found in synovial joints. Additionally, the accumulation of lipid peroxides is also considered a significant cause of RA joint inflammation. These two types of manifestations are closely related to ferroptosis, indicating that ferroptosis may play a role in the pathology of RA [10]. Artemisinin and its derivatives were originally widely used as antimalarial drugs, but recent studies have found that they also show great potential in inducing iron death in cancer cells [11]. Artemisinin can make cancer cells more sensitive to iron death, and its mechanism of action is related to iron metabolism regulation, reactive oxygen species (ROS) production, and endoplasmic reticulum stress (ERS) [12]. At the same time, it can also affect the binding of iron regulatory proteins (IRP) to mRNA molecules containing iron response elements (IRE) sequences, further regulating the iron homeostasis controlled by IRP/IRE, increasing free iron in cells [13]. In vitro experiments and animal models have shown that artemisinin derivatives such as dihydroartemisinin (DHA) can significantly induce iron death in tumor cells, which is closely related to their ability to increase intracellular ROS levels and cause lipid peroxidation. More importantly, ART is one of the clinically available ferroptosis inducers [14]. FLS, as a type of tumor-like cell, has many similar characteristics to tumor cells. Considering that artemisinin may induce cancer cell death through a sensitization mechanism of iron death, it may also have a certain impact on FLS, but this requires further research to confirm [15, 16]. In summary, ferroptosis, as a way of inducing programmed cell death, has been extensively studied in many fields. Increasing evidence suggests that ferroptosis may be related to the degenerative process of rheumatoid arthritis. In addition, ART, as an inducer of ferroptosis, has also been proven useful in the treatment of autoimmune diseases. However, whether ART can treat RA through the ferroptosis pathway has not been reported yet. Therefore, our study explored the cellular effects of ART on RA-FLSS and demonstrated the relationship between this cellular effect and ferroptosis and the signaling pathways involved. Methods Isolation and culture of human RA-FLSs Synovial tissues were acquired from 6 independent female RA donors undergoing synovectomy or joint replacement surgeries. All patients met the 2010 ACR/EULAR criteria, with a disease duration of 10–20 years and rheumatoid factor (RF) levels between 150 and 200 IU/mL. Approval for this study was obtained from the Clinical Research Ethics Committee of Nanjing Medical University ([2024]KY219-01). Synovial tissue was digested with 1 mg/ml collagenase type I (sigma Aldrich, St. Louis, Mo) in serum-free Dulbecco’s modified Eagle’s medium (DMEM) for 12 h. The tissue digest was filtered through a 70 μm cell filter (BD, Durham, NC) to collect the cells. Isolated FLSs were cultured separately in DMEM (HyClone, Logan, Utah, USA) comprising 10% fetal bovine serum (Gibco, California, USA) and 100 U/mL penicillin. The culture environment was 37 °C, 5% CO2. All cells were passaged within 5 generations. Cell viability assay Cell viability assay was performed by Cell Counting KIT-8 (CCK-8). EDU incorporation assay An EDU test was conducted following the EDU guidelines. Subsequently, DNA staining was done, and the results were examined under a microscope. Immunofluorescent staining FLSs were cultured in 24-well plates preloaded with glass coverslips. Cells were fixed in 4% paraformaldehyde, followed by permeabilization with 0.5% Triton X-100 for an additional 30 min. After blocking with 5% bovine serum albumin (BSA), the cells were incubated with primary antibodies diluted in 5% BSA (1:100) at 4 °C overnight. Subsequently, the coverslips were washed and incubated with the corresponding secondary antibodies (1:100) for 1 h at room temperature in the dark. Finally, the samples were mounted using an anti-fade mounting medium containing 4′,6-diamidino-2-phenylindole (DAPI) and examined under a fluorescence microscope. ROS analysis The cells were treated as instructed for 12 h and incubated with 50 μm lipid peroxidation sensor BODIPY 581/591 C11 (thermofisher, cat#d 3861) for 1 h. The cells were washed twice with PBS buffer to remove excess bodipyc11. Labeled cells were digested by trypsin in PBS and 5% (V/V) FBS and resuspended.After filtration, the oxidation of BODIPY C11 polyunsaturated butadiene acid was analyzed by FACScan flow cytometry (BD Biosciences, San Jose, CA, USA), and this oxidation causes the fluorescence emission peak cluster to shift from 590 nm to ~ 510 nm, which is directly proportional to the formation of lipid ROS. Plasmid transfection and gene knockdown To stable Knockdown p53 expression, specific short hairpin RNA (shRNA) targeting mouse p53 (sh-p53) and a non-targeting negative control shRNA (sh-NC) were synthesized and cloned into the expression vectors (Genomeditech, Shanghai, China). The target sequence for p53 knockdown was p53-Mus-169: 5′-GGAGGAGTCACAGTCGGATAT-3′, and the sequence for the negative control was NC: 5′-TTCTCCGAACGTGTCACGT-3′. Cells were cultured to 70–80% confluence and transfected with the respective plasmids using Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. Transfection efficiency was verified by quantitative real-time PCR (RT-qPCR) and Western blot analysis after 48 h. Animal experiment CIA model development All animal experiments followed the guidelines set by the National Research Council for the Care and Use of Laboratory Animals. Eight-week-old male mice were used for CIA induction, and therapeutic treatment was initiated 4 weeks post-initial immunization (at 12 weeks of age). Animals were randomly allocated to experimental groups using a computer-generated randomization method. Cage effects were minimized by distributing animals from different groups across cages. Mice were subcutaneously immunized with collagen emulsified in complete Freund’s adjuvant near the base of the tail, followed by a booster injection with collagen in incomplete Freund’s adjuvant at a separate tail-base site on day 21. All clinical scoring and histological assessments were performed by investigators blinded to group allocation. Groups Thirty mice were randomly divided into six groups (n = 5 per group): the vehicle control (CON), collagen-induced arthritis model (CIA), CIA treated with Erastin (Erastin), CIA treated with artesunate (ART), CIA co-treated with Erastin and ART (Erastin + ART), and CIA treated with a combination of Erastin, ART, and Ferrostatin-1 (Erastin + ART+Fer-1). The administration dosages and frequencies of all drugs were selected based on previously published literature [17–22]. Specifically, treatments were administered via intraperitoneal (i.p.) injection three times per week at the following dosages: ART at 30 mg/kg, Erastin at 15 mg/kg, and Ferrostatin-1 (Fer-1) at 1 mg/kg. Throughout the experimental period, all animals were maintained on a standard laboratory diet. On day 56, the mice were euthanized, and their knee joints and paws were harvested for subsequent imaging analysis and histopathological evaluation. Immunohistochemistry Synovial tissues were collected from DBA/1 mice, paraffin-embedded, and sectioned. Immunohistochemical staining was performed following standard protocols, including deparaffinization, antigen retrieval, endogenous peroxidase blocking, antibody incubation, DAB visualization, and slide mounting. Hematoxylin-eosin (H&E) staining and safranin-fast green staining H&E staining was performed using commercial kits (Solarbio). Paraffin sections were deparaffinized, stained, dehydrated, mounted, and examined under a light microscope according to the manufacturers’ protocols. Mice arthritis index score Paw inflammation was assessed by scoring erythema and swelling in each mouse using a standardized scale (maximum score of 16 per animal). Each paw was graded from 0 to 4 based on swelling severity: 0, normal; 1, mild swelling limited to the ankle; 2, mild swelling extending to the metatarsal/metacarpal joints; 3, moderate swelling involving both ankle and metatarsal/metacarpal joints; and 4, severe swelling affecting the entire paw. The arthritis index was independently evaluated by three investigators, all of whom were blinded to the experimental group assignments. The final arthritis score was calculated as the average of the scores from the three investigators. Statistical analysis Results are reported as averages with corresponding standard deviations derived from a minimum of three independent experiments. Statistical analyses were conducted using an unpaired Student’s t-test for two-sample comparisons and one-way analysis of variance for analyses involving more than two groups. All statistical procedures were performed using two-sided tests, and differences were considered statistically meaningful at P < 0.05. Results ART inhibits FLSs proliferation We analyzed the effect of ART on the proliferation of RA FLSs. Both compounds were dissolved in DMSO, and the final concentration of DMSO did not exceed 0.1% and was kept constant across all experimental groups. The control group received an equal volume of DMSO vehicle. After treating FLSs with different concentrations of artemisinin or Erastin, CCK-8 assay results showed a dose-dependent decrease in cell viability (Figs. 1A). When 2 µM ART or 5 µM erastin was used alone, no obvious decrease in cell viability was observed; however, their combined treatment significantly reduced cell viability, with statistically significant differences (Fig. 1B). The same experimental results were obtained from EDU and crystal violet staining (Fig. 1C and D). ART inhibits FLSs proliferation. According to the group labels in the figure, detect the effect of ART and erastin on cell activity after TNF-α pretreatment. A CCK-8 was used to assess cell viability at different concentration gradients. B CCK-8 was used to assess cell viability at 3 h, 6 h, 12 h and 24 h. C Crystal violet staining was applied after a 24-hour treatment. D original magnification, ×100. After 24 h of treatment, the EDU experiment was performed, and the green spots show cells that are proliferating. ANOVA, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001 ART inhibits the migration and invasion of FLSs Subsequently, the scratch and transwell assays were used to assess how ART treatment affects the migration and invasion of FLS cells. As shown in Fig. 2, while monotherapy with either ART alone or Erastin alone exerted a negligible inhibitory effect on the migration and invasion of synoviocytes, the co-administration of ART and Erastin significantly suppressed both migratory and invasive capacities. (Figure 2A and B). ART inhibits the migration and invasion of FLSs. FLSs were divided into 4 groups: control group, 2 µmol/L ART group, 5 µmol/L Erastin group, ART+Erastin group (all groups were pretreated with TNF-α for 4 h). A Cell proliferative and migratory capacities were evaluated using scratch assay at 0 h, 6 h, 12 h and 24 h. B cell migration and invasion ability were analyzed by transwell assay at 24 h. ANOVA, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001 ART can increase the sensitivity of RA-FLSs to ferroptosis Core anti-ferroptotic proteins are known to negatively regulate the ferroptotic process in various cell types. In FLSs, our immunofluorescence staining revealed that while individual treatment with either ART or Erastin only mildly reduced the expression of the ferroptosis-defense protein GPX4, their combined application significantly downregulated its expression levels; notably, this alteration was effectively reversed by co-treatment with a ferroptosis inhibitor(Fig. 3A and B). Given that a defining hallmark of ferroptosis is the pre-lytic accumulation of cell membrane lipid peroxidation, we next evaluated intracellular reactive oxygen species (ROS) levels. Following co-treatment, ROS accumulation in the cytoplasm of RA-FLSs was markedly higher than that observed in either the monotherapy groups or the vehicle control group, and this elevated ROS production was successfully abrogated by the addition of the ferroptosis inhibitor (Fig. 3C and D). Subsequently, Western blot analysis was performed to evaluate the expression levels of the ferroptosis-related proteins SLC7A11, FTL, FTH1J=, and GPX4, yielding results consistent with those obtained from GPX4 immunofluorescence staining (Fig. 3E and I). To rule out other cell death pathways and further confirm that artemisinin sensitizes RA-FLSs to ferroptosis, we examined caspase-3 and PARP cleavage, using staurosporine as a positive control. The results indicated that artemisinin exerted no significant effect on the apoptotic process (Fig. 3J and K). ART can increase the sensitivity of RA-FLSs to Ferroptosis. FLSs were divided into 6 groups: control group, 2 µmol/L ART group, 5 µmol/L Erastin group, ART+Erastin group, ART+Erastin + DFO group, ART+Erastin + Fer-1 group (all groups were pretreated with TNF-α for 4 h). A-B original magnification, x 400. After 24 h of treatment, GPX4 immunofluorescence and its quantification were performed. C-D original magnification, ×400. The BODIPY-C11 assay was used to analyze the buildup of lipid peroxidation within cells. E-I The levels and quantification of SLC7A11, FTL, FTH1, and GPX4 were determined by western blot after 24 h treatment. J-K The level and quantification of the cleavage of PARP were determined by western blot after 24 h treatment. All values are normalized to the control group, which is set as 1. Each dot signifies an individual measurement, and the absence of a dot indicates no comparison. ANOVA, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001 ART increases the sensitivity of RA-FLSs to ferroptosis through the p53-SLC7A11-GPX4 signaling pathway It has been reported that p53 sensitizes cells to ferroptosis by inhibiting the transcription of SLC7A11. SLC7A11 is a direct target of p53, which binds to the p53 response element in the promoter region of SLC7A11, leading to the inhibition of SLC7A11 expression. This enhances the cell’s sensitivity to ferroptosis inducers. SLC7A11 is overexpressed in various tumor cells, and RA-FLS exhibits characteristics similar to tumor cells. Therefore, we hypothesized that ART might also induce ferroptosis through the p53-SLC7a11-GPX4 signaling pathway. As expected, the synergistic effect of ART and Erastin increased the expression of p53 in RA-FLSs. Additionally, as a recognized p53-mediated factor, the expression of p21 in cells usually correlates with that of p53. Western blot analysis demonstrated that the expression pattern of p21 was consistent with that of p53 (Figs. 4A-C). After treatment with the p53 inhibitor PFT-α, cell death co-treated with Erastin was also alleviated, and the expression of SLC7A11, FTL, FTH1, and GPX4 was increased (Figs. 4D-J). The immunofluorescence results were consistent with those of the Western blot analysis (Figs. 4K-N). Subsequently, to further validate this pathway, we subsequently knocked down p53 using specific plasmids. The knockdown efficiency of the plasmid was verified by Western blotting (Fig. 4O-P). Western blot results showed that the levels of death-related proteins were restored after p53 knockdown (Fig. 4Q-U). Immunofluorescence staining yielded consistent results (Fig. 4V-Y). These findings indicate that ART increases the sensitivity of RA-FLSs to ferroptosis through the p53/p21-SLC7a11-GPX4 signaling pathway. ART increases the sensitivity of RA-FLSs to ferroptosis through the p53-SLC7A11-GPX4 signaling pathway. FLSs were divided into 4 groups: control group, 2 µmol/L ART group, 5 µmol/L Erastin group, ART+Erastin group (all groups were pretreated with TNF-α for 4 h). A-C western blot analysis was performed to detect and quantify the expression levels of P53 and P21 following 10 min of treatment. D-E expression level and quantification of P53 after P-α treatment. F-J expression levels and quantification of SLC7A11, FTL, FTH1and GPX4 after P-α treatment. K-N original magnification, ×400. Immunofluorescence staining and quantification of GPX4 and SLC7A11. O–P Western blot analysis confirmed the effectiveness of the plasmid in reducing P53 expression. Q–U Expression levels and quantification of SLC7A11, FTL, FTH1, and GPX4 following P53 knockdown. V–Y Original magnification, ×400. Immunofluorescence staining and quantification of GPX4 and SLC7A11. ANOVA, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001 Ferroptosis is involved in the ART treatment on RA in vivo The CIA mouse model was successfully established and evaluated over an eight-week experimental period. In vehicle-treated CIA mice, clinical arthritis scores, paw swelling, and toe edema peaked on day 42. Notably, monotherapy with either ART alone or Erastin alone exerted marginal therapeutic benefits on these parameters. In contrast, co-administration of ART and Erastin (ART + Erastin group) markedly attenuated these gross pathological lesions (Fig. 5A and C), demonstrating a potent synergistic effect. Ferroptosis in involved in the ART treatment on RA in vivo. DBA/1 mice were categorized into six distinct groups: control, CIA, CIA + ART, CIA+Erastin, CIA+Erastin + ART, CIA+Erastin + ART+Fer-1. A claw changes of mice in the 6 groups at 56 d. B-C index of arthritis and number of swelling of mice in 6 groups within 56 days. ART and MTX treatment significantly reduced disease severity scores compared with the model group over time (two-way repeated-measures ANOVA, p < 0.05). D, F Results and quantitative analysis of H&E staining in mouse knee joints. E, G, H imaging observation of mouse knee joint and claw. ANOVA, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001 Histological assessment via H&E staining and radiographic analysis further revealed severe synovial hyperplasia, cartilage destruction, and extensive bone erosion in the knee joints and paws of the CIA group. Consistently, the ART + Erastin combination therapy effectively alleviated these osteoarticular pathological degenerations (Fig. 5D and H). Intriguingly, these joint-protective effects were fully abrogated by the concurrent administration of the ferroptosis inhibitor Fer-1 (ART + Erastin + Fer-1 group; Fig. 5A and H). Concurring with our in vitro observations in RA-FLSs, these in vivo results demonstrate that ART operates as a ferroptosis sensitiser, and that the activation of ferroptotic cell death critically mediates its therapeutic efficacy against rheumatoid arthritis. ART enhances the susceptibility of synovial tissues to ferroptosis in CIA mice To further elucidate ART sensitizes synovial tissues to ferroptosis in vivo, Prussian blue staining was performed to monitor labile iron accumulation. While ART or Erastin monotherapy induced negligible changes, the combination of ART and Erastin markedly expanded the iron deposition area in the synovial tissues of CIA mice. This accumulation of catalytic iron was completely reversed by the addition of Fer-1 (Fig. 6A). ART induced ferroptosis in synovial of CIA mice. DBA/1 mice were categorized into six distinct groups: control, CIA, CIA + ART, CIA+Erastin, CIA+Erastin + ART, CIA+Erastin + ART+Fer-1. A Results of Prussian blue staining. B-J Results and quantitative analysis of GPX4, SLC7A11, IL-1β, and IL-6 immunohistochemical staining. ANOVA, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001 Subsequently, immunohistochemical analysis was employed to examine the expression of core ferroptosis surveillance proteins GPX4 and SLC7A11, alongside the pro-inflammatory cytokines interleukin-1β (IL-1β) and interleukin-6 (IL-6), in the synovium. Compared with the monotherapy groups, co-administration of ART and Erastin induced potent ferroptosis, characterized by the significant downregulation of GPX4 and SLC7A11; concurrently, this combination therapy exerted a profound anti-inflammatory effect by markedly suppressing IL-1β and IL-6 expression. Strikingly, the addition of Fer-1 successfully reversed these alterations, restoring the levels of all three markers back toward baseline (Fig. 6B and J). Collectively, these findings indicate that ART does not act as a potent standalone ferroptosis inducer; rather, it restrains the hyperproliferation of synoviocytes and abrogates synovial inflammation precisely by sensitizing cells to Erastin-induced ferroptosis, thereby exerting its therapeutic efficacy against rheumatoid arthritis. Discussion RA is a common chronic autoimmune disease involving symmetry across multiple joints [23]. During the course of RA, the synovium transforms into proliferative invasive tissue, leading to the destruction of cartilage and bone. FLS forming the inner lining of joints, possess an aggressive phenotype in RA with epigenetic marks and play a significant role in these pathological processes [24]. In this study, we discovered that the combined treatment of ART and the ferroptosis inducer Erastin significantly inhibited the proliferation, migration, and invasion of FLSs. Moreover, it could regulate the P53/P21-SLC7A11-GPX4 signaling pathway, reducing the expression of iron-death-related proteins GPX4, FTH1, and FTL. This effect was much more pronounced than when ART or Erastin were used alone. This phenomenon could be reversed by ferroptosis inhibitors (FER-1 and DFO), demonstrating that artemisinin can increase the sensitivity of RA-FLSs to ferroptosis. Y. Liu proposed that the accumulation of ROS and iron overload are involved in the pathogenesis of rheumatoid arthritis [25]. In our animal model experiments, synovial inflammation and joint swelling were lessened by artemisinin in CIA mice, concluding that artemisinin may treat RA by promoting iron death in FLSs. Although ART is a traditional antimalarial drug, it has been extensively studied in various fields of disease in recent years, especially in the field of oncology, showing that it can induce death in different tumor cells through different pathways, thereby achieving therapeutic effects on cancer [26, 27]. Interestingly, studies have shown that ART and its derivatives also have therapeutic effects on RA. Research by Ma et al. demonstrated that artesunate could inhibit the migration and invasion of RA-FLSs, and its synergistic effect with MTX not only better suppresses inflammation but also allows for a reduction in the dosage of MTX, thereby reducing the side effects of antirheumatic drugs [28]. Meanwhile, Fan et al. found that dihydroartemisinin can not only improve RA joint inflammation and control joint degradation through the NRF2-P62-Keap1 pathway but also restore the Treg/Th17 balance and inhibit the occurrence of synovitis [29]. Du. et al., through the preparation of artemisinin liposomes as a novel drug for the treatment of RA, found that this drug could effectively inhibit the secretion of proinflammatory cytokines TNF-α, IL1-β, IL-6 with low toxicity [30]. Overall, the application of ART and its derivatives in RA is being extensively studied, further illustrating the potential of ART in treating RA. In recent years, with the discovery of ART’s function in inducing ferroptosis, we speculated whether ART could treat RA through the ferroptosis pathway. Our results showed that the sole use of ART had no significant cellular effect on FLSs, the main effector cells of RA, which may be related to the concentration of ART.In most studies, the concentration of ART that produces cellular effects usually has to reach 100 μm, which often requires the use of special solvents or nanomaterial encapsulation [31]. We then tried the classic ferroptosis inducer Erastin on RA-FLS and, surprisingly, Erastin could not significantly induce cell death. However, in the study by Chen et al., we found that ART could sensitize to ferroptosis, so we tried treating RA-FLSs with both ART and Erastin [11]. Surprisingly, significant cell death occurred in RA-FLSs under the combined action of only 2 μm concentration of ART and 5 μm concentration of Erastin. To further study the relationship between cell death caused by co-treatment with ART and Erastin and ferroptosis, we attempted to rescue this cell death with the ferroptosis inhibitors Fer-1 and DFO, and as expected, these two classic ferroptosis inhibitors could rescue the death of RA-FLSs, indicating that the cell death caused by the combined action of ART and Erastin is ferroptosis. The primary cause of ferroptosis is the failure of the cell’s antioxidant system, especially the inactivation of the XC-/GSH/GPX4 system, leading to the accumulation of lipid peroxides, so the reduction of GPX4 is considered to be a core link in ferroptosis [32]. The human body requires iron as an essential element, which is involved in several metabolic processes such as transporting oxygen, transferring electrons, and synthesizing DNA [33]. Due to its ability to accept and provide electrons, pathological iron accumulation can lead to cellular oxidative damage and even death, which is another characteristic of ferroptosis. In this study, we demonstrated that the combined treatment of ART and Erastin could lead to the accumulation of free iron and ROS in RA-FLSs, and reduce the expression of GPX4, FTH1, and SLC7A11. This further illustrates that the death of RA-FLSs induced by ART sensitization to Erastin is ferroptosis. P53 has been reported to sensitize cells to ferroptosis by transcriptionally repressing SLC7A11 [34]. SLC7A11 is a direct target of P53, which binds to the P53 response element in the SLC7A11 promoter region, inhibiting its expression and thereby enhancing cell sensitivity to factors such as Erastin [35, 36]. SLC7A11 has been reported to be overexpressed in different types of tumor cells, and this expression can inhibit the tumor-suppressive function of P53, while RA-FLSs also exhibit characteristics similar to tumor cells [15, 37]. Therefore, we speculate that ART can also induce ferroptosis through the P53-SLC7A11 signaling pathway, and P21, as a recognized P53-mediated factor, its expression in cells usually coincides with that of P53. Our results verified this signaling pathway, showing increased expression of P53 and P21 in RA-FLS treated with artemisinin and Erastin, indicating that this sensitization effect of ART is achieved through the P53/P21-SLC7A11-GPX4 signaling pathway. In summary, our experiments demonstrate that ART can affect the main effector cells of RA, FLSs, through the ferroptosis pathway, providing a potential treatment for RA. One limitation of the present study is that a comprehensive concentration–response gradient for ART was not established in our in vivo experiments. Although the therapeutic dosage was determined based on a rigorous literature review [1–6], a detailed dose–response analysis would provide a more precise therapeutic window. To partially address this, we introduced the ferroptosis inhibitor Fer-1 in our rescue experiments, which successfully confirmed that Fer-1 reversed the sensitization effect of ART on ferroptosis, thereby substantiating its therapeutic efficacy and specific mechanism in vivo. Further investigations are warranted to systematically explore the optimal dosage and frequency of ART to maximize its clinical translation in rheumatoid arthritis. Conclusion In conclusion, artemisinin can increase the sensitivity of RA FLSs cells to ferroptosis, which provides a new understanding for the treatment of rheumatoid arthritis, especially from the perspective of inhibiting RA synovitis. Acknowledgements We appreciate the Third Affiliated Hospital of Nanjing Medical University for providing experimental instruments and technical support. Authors’ contributions YW and ML were responsible for the study’s conception and design. ML, WJ and YY performed the experiments and statistical analysis and drafted the manuscript. LW, FL, SL, and JW participated in the collection of tissue specimens and statistical analysis. YL, CX, CZ and GY assisted with revising the manuscript. All the authors reviewed the manuscript. Funding This study was supported by Changzhou Sci&Tech Program (CJ20252038). Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Written informed consent for publication has been obtained from all patients participating in this study. Experiments were approved by the ethics committee of the Changzhou Second People’s Hospital of Nanjing Medical University . Consent for publication All authors have read and approved the final version of the manuscript and consent to its submission and publication in the relevant journal. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Maoyuan Li, Jiahao Wang and Yunyuan Yu contributed equally to this work and share first authorship. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.