Targeting Ferroptosis With Natural Products for the Treatment of Skeletal System Disease: An Updated Review

Iron metabolism and lipid peroxidation are key mediators of ferroptosis (Figure 1). Ferroptosis, as indicated by its name, is iron‐dependent, with intracellular iron overload being essential for its occurrence [27]. Normally, human body obtained iron primarily through heme iron and non‐heme iron sources, and non‐heme iron can be derived from either iron ions or ferritin [28, 29]. Transferrin (TF) banded with Fe³⁺, and transported Fe³⁺ into cells via its receptor transferrin receptor (TFR1) [30, 31]. Fe²⁺ enters cells via divalent metal transporter 1 (DMT1) or through ferritin receptor in the ferritin form [32]. And FPN1 and its regulator hepcidin contributed to the iron export from cells [33]. Within cells, ferrireductases mediate the transformation of Fe³⁺ to Fe²⁺, which is either utilised in various biological processes or stored in ferritin [34]. Under normal physiological conditions, the concentration of intracellular free ionic iron remains stable owing to the regulatory effects of ferritin and TFs. While, when the intracellular iron overload, intracellular Fe²⁺ excess leads to the Fenton reaction, which Fe²⁺ catalysed hydrogen peroxide, producing toxic hydroxyl radicals, thereby inducing cellular lipid peroxidation and cellular damage [35].

Lipid peroxidation is another crucial process in ferroptosis. Under normal conditions, polyunsaturated fatty acyl (PUFA) combine with phospholipids (PLs) to form PUFA‐PLs, which regulate the fluidity of cell membranes and maintain the normal physiological function of cells [36, 37]. Lysophosphatidylcholine acyltransferase 3 and acyl‐coenzyme A synthetase long‐chain family member 4 (ACSL4) are the vital enzymes governing PUFA‐PLs production [38]. The Fenton reaction generates toxic hydroxyl radicals and ROS, which promote the formation of PUFA‐PL‐OOH and lipid peroxidation, thereby disrupting the integrity of the cell membrane and leading to ferroptosis [39, 40].

Lipid peroxidation can be neutralised in time to prevent overload and inhibit ferroptosis. Glutathione peroxidase 4 (GPX4), a member of the glutathione peroxidase family, catalyses the conversion of certain lipid hydroperoxides to lipid alcohols, thereby reducing the generation of PL‐OOHs [41]. This process is glutathione (GSH)‐dependent, and is inhibited when GSH biosynthesis is reduced [42]. Furthermore, intracellular GSH biosynthesis requires glutamate, glycine, and cysteine and is reliant on the cystine/glutamate reverse transporter (System XC‐), which influences cysteine content by exchanging extracellular cystine for intracellular glutamate [43, 44]. The system is composed of the light chain subunit solute carrier family 7 member 11 (SLC7A11) and the heavy chain subunit SLC3A2 [45]. The GSH‐dependent detoxification of PL‐OOH by GPX4 prevents lipid peroxidation and ferroptosis, with the SLC7A11‐GSH‐GPX4 axis serving as the primary defense system against ferroptosis.

Osteoarthritis is the most common degenerative joint disorder, characterised by symptoms such as joint pain, stiffness, limited mobility and deformity. Its prevalence increases with age, and the incidence rate is 10% in men and 18% in women over 60 years old worldwide [7]. The pathological hallmarks of OA include articular cartilage loss, sclerosis of the subchondral bone, formation of osteophytes and synovitis [77].

Ferroptosis has been implicated in the progression of OA. Elevated iron levels have been observed in the joints of patients with OA [78]. In 2020, Guo group's study first reported that ferric ammonium citrate could induce iron overload in chondrocytes, promoting ROS production and accumulation and ferroptosis [79]. Moreover, excessive stress loading activates the Piezo1 channel, leading to increased calcium influx and accumulation of calcium ions, and then impaired GSH production, reduced GPX4 levels, leading to ferroptosis in chondrocytes [80]. Furthermore, IL‐1β enhances the expression of ACSL4, p53 and TfR1, while inhibiting the expression of GPX4 and SLC7A11, resulting in iron overload, oxidative stress and ferroptosis [81]. IGF2BP1 promoted Fe²⁺ accumulation, ROS production and bolstered chondrocytes ferroptosis of OA by targeting the m6A/MMP3 axis [82]. In addition, SCP2 transports cytoplasmic lipid hydroperoxides to the mitochondria, resulting in mitochondrial membrane damage and ROS release, thus accelerating chondrocyte ferroptosis [83].

Synovitis is a significant feature of OA. Ma et al. [84] reported that GPX4 and GSH were downregulated, while ferrous ions were upregulated in synovial tissue samples from patients with OA. Moreover, in LPS‐induced synovitis, MDA levels and iron content were elevated and GPX levels were reduced; icariin protected synoviocytes via the inhibition of ferroptosis by activating the Xc‐/GPX4 axis [85]. Knee OA synoviocytes undergo ferroptosis, contributing to the chondrocyte's matrix degradation and neuropathic pain [86]. The mechanisms and key regulatory signalling pathways of ferroptosis in OA are illustrated in Figure 3.

Given the important regulator role of ferroptosis in OA, targeting ferroptosis may be a novel therapeutic strategy for OA treatment. Deferoxamine, an effective iron‐chelating agent, alleviated OA by inhibiting chondrocyte ferroptosis [87]. Moreover, adenovirus‐mediated expression of FTH1 inhibited the MAPK pathway and ferroptosis in chondrocytes, and suppressed extracellular matrix degradation [88]. MSC‐derived exosomes prevented macrophage ferroptosis via inducing expression of GOT1/CCR2, and then rescued cartilage injury in OA [89]. In addition, Han et al. [90] reported that moderate mechanical stress could activate the Nrf2 antioxidant system, inhibited the NF‐κB signalling pathway, and suppressed chondrocyte ferroptosis and cartilage matrix degradation by regulating P53, SLC7A11 and GPX4.

RA is a multifactorial autoimmune disease of unknown aetiology, primarily characterised by inflammatory responses, joint pain and progressive cartilage destruction. It has been reported that iron metabolism is different in RA than in general health and focal iron overload is frequently observed in RA patients [91, 92]. Recently, Yang et al. found that the expressions of GSH, GPX4, Nrf2 and Keap‐1 were lower, and the ferritin was higher in RA patients. Inhibition of oxidative stress‐induced ferroptosis can alleviate RA symptoms in humans. Moreover, excess iron in synovial fluid positively correlates with RA disease severity and aggravates arthritis by inducing macrophage ferroptosis [93]. SIRT1 has been proved downregulated in RA and its function in inhibiting the ferroptosis of synoviocytes has also been suppressed [94]. LPS triggered ferroptosis via NCOA4‐mediated ferritinophagy in RA fibroblast‐like synoviocytes (FLSs) and induced inflammation under hypoxic conditions [95]. Furthermore, CD8⁺ T cells deliver IFNγ to RA‐FLS, regulating the GPX4 antioxidant signalling pathway to promote ferroptosis and inhibit the viability of RA‐FLS [96].

However, some studies reported contradictory findings. For example, ferroptosis has been found to decrease in the RA synovium and FLS compared with healthy controls [97]. Moreover, Wu et al. [98] found that TNF antagonists induced ferroptosis in fibroblasts, thereby attenuating arthritis progression in a collagen‐induced arthritis model. Recently, Cathepsin B has been found highly expressed in patients with RA, its inhibitor reduced RA‐FLS proliferation and migration through leading the ferroptosis in RA‐FLS, targeting ferroptosis may be a potential treatment for RA [99]. In addition, Ruan et al. [100] reported that iron released from Fe₃O₄ nanoparticles could induce ferroptosis in both resident inflammatory cells and proliferating FLSs, presenting a novel therapeutic approach for RA.

These conflicting results may be due to the multifactorial nature of RA, where dysregulation in FLSs, dendritic cells, chondrocytes, macrophages and lymphocytes contributed to the progression of RA. The impact of ferroptosis in RA may vary depending on the specific cell type undergoing ferroptosis. Therefore, further clinical data and experimental evidence are necessary to explore the relationship between ferroptosis and the development of RA in future.

Osteoporosis is a prevalent metabolic bone disease, with osteocytes, osteoblasts, osteoclasts and mesenchymal stem cells involved in its pathological process. Under normal conditions, the homeostasis of bone‐forming and bone‐resorbing is maintained by the balanced activity of osteoblasts and osteoclasts. In OP, the bone homeostasis is broken, resulting in decreased bone mass, increased bone brittleness and fracture risk. Numerous studies indicate an association between ferroptosis of bone cells and OP. For example, iron overload leads to excessive free Fe²⁺, initiating the Fenton reaction and producing a substantial amount of ROS. These ROS activate a series of intracellular signalling pathways, which promote bone resorption and inhibit bone formation, thereby causing OP [101]. Osteocyte ferroptosis, induced by ATF3/TFR1, plays a role in cortical bone loss during ageing [102]. Yang et al. [103] reported that targeting ferroptosis suppresses osteocyte glucolipotoxicity and alleviates diabetic OP. Similarly, Fu et al. [104] revealed that eldecalcitol ameliorates osteocyte senescence and associated ferroptosis, contributing to bone preservation in OP.

Osteoblastic ferroptosis has also been shown to contribute significantly to OP pathogenesis. Osteoporotic bone loss from excess iron accumulation is driven by NOX4‐triggered ferroptosis in osteoblasts [105]. Epigenetic suppression of GPX4 through DNMT aberration induced osteoblastic ferroptosis and exacerbated OP [106]. Inhibiting osteoblast ferroptosis can suppress OP. YBX1 could alleviate ferroptosis via the ATF4/FSP1 axis in osteoblasts, suggesting that YBX1 could be a potential therapeutic target for OP treatment [107]. Moreover, mangiferin attenuates OP by inhibiting osteoblast ferroptosis through the Keap1/Nrf2/SLC7A11/GPX4 pathway [108]. ED‐71 ameliorates bone regeneration in type 2 diabetes by reducing osteoblast ferroptosis through the HIF1α pathway [109].

In contrast, some studies report that osteoclast ferroptosis is suppressed, and osteoclastogenesis is enhanced in OP. For example, reduced IRF9 expression in OP leads to overactive osteoclastogenesis and decreased osteoclasts ferroptosis [110]. Inducing ferroptosis in osteoclasts and suppressing osteoclast viability may present a potential strategy for OP. P. Ga‐containing calcium titanate layer has been shown to inhibit osteoclastogenesis by inducing osteoclast ferroptosis and promoting the differentiation of hMSCs into osteoblasts [111]. Moreover, Jin et al. [112] demonstrated that artesunate induces osteoclast ferroptosis, inhibits osteoclast differentiation and prevents iron overload‐induced bone loss. Furthermore, zoledronic acid has been shown to improve OP by inducing osteoclast ferroptosis through the p53 signalling pathway [113].

The osteogenic differentiation of human bone marrow‐derived mesenchymal stem cells (hBMSCs) plays a role in alleviating OP. Iron overload inhibited osteogenic commitment and differentiation of MSCs via the induction of ferritin [114]. Inhibiting ferroptosis in BMSCs can enhance their osteogenic differentiation capacity. CRYAB suppresses ferroptosis and promotes osteogenic differentiation of hBMSCs via binding and stabilising FTH1 [115]. Moreover, aucubin protected against ferroptosis and facilitated osteogenic differentiation of hBMSCs by activating the BMP2/SMADs pathway and attenuating the progression of OP [116]. Furthermore, Poliumoside protects against type 2 diabetes‐related OP by promoting osteogenic differentiation and suppressing BMSC ferroptosis via the Nrf2/GPX4 pathway [117]. Mechanisms and key regulatory signalling pathways of ferroptosis in OP are illustrated in Figure 4.

OS is the most prevalent orthopaedic malignancy that originated from mesenchymal cells and is characterised by lung metastasis and resistance to chemotherapy. It has been reported that the absorption rate of iron in OS cells is exceptionally high, suggesting that ferroptosis may play a role in the progression of OS [15]. The dysregulation of ferroptosis related to SLC7A11, GPX4 or oncogenic genes influences ferroptosis in OS. For example, KDM4A mediated histone demethylation of SLC7A11 and then inhibited ferroptosis in OS [118]. Li et al. [119] reported that circRNA‐BLNK inhibited ferroptosis and promoted OS progression by regulating the miR‐188‐3p/GPX4 axis. Moreover, the oncogene PTPRC blocks the TFEB/FTH1 signalling pathway and inhibits ferroptosis in OS [120]. SLC38A5 is upregulated in OS and is associated with poor prognosis in patients; it suppresses ferroptosis via glutamine‐mediated activation of the PI3K/AKT/mTOR signalling pathway [121]. Elzbieta et al. [122] demonstrated that FSP1 is a predictive biomarker of OS cells' susceptibility to ferroptosis, with its expression regulated by p53.

Given the downregulation of ferroptosis in OS, its induction has emerged as a potential therapeutic strategy for OS treatment. For example, ANK1 inhibited the proliferation, migration and invasion of OS cells by promoting ferroptosis [123]. Bone‐targeting exosome nanoparticles activate the Keap1/Nrf2/GPX4 signalling pathway to induce ferroptosis in OS cells and suppress OS progression [124]. In addition, the natural compound oridonin could promote the accumulation of ROS and Fe²⁺ in OS cells, as well as reduce mitochondrial membrane potential, induce ferroptosis and suppress the growth of OS cells [125].

Furthermore, it was well known that cancer cells gradually develop resistance to chemotherapy‐induced cell death, while recent studies show that targeted induction of ferroptosis combined with chemotherapy agents brought new opportunities for OS treatment. For example, NGR (Asn‐Gly‐Arg)‐modified cancer‐associated fibroblasts‐derived exosomes target tumour vasculature to induce ferroptosis and overcome chemoresistance [126]. The induction of ferroptosis by impairing STAT3/Nrf2/GPX4 signalling enhances the sensitivity of OS cells to cisplatin [127]. The reduction of FTH1 translation dramatically increases ferroptosis and promotes the sensitivity of OS cells to chemotherapy drugs [128]. Nanoparticle‐encapsulated doxorubicin alleviates drug resistance in OS via inducing ferroptosis [129]. Inhibitors of APE1 redox and ATM synergistically sensitise OS cells to ionising radiation through ferroptosis induction [130]. In addition, the natural compound, ursolic acid, degraded ferritin by inducing intracellular overload of ferrous ions and leading to ferroptosis, enhancing the DNA‐damaging effect of cisplatin on OS cells [131].

It has been reported that iron homeostasis plays a positive role in maintaining articular cartilage health. Studies have proved that natural compounds could protect chondrocytes via regulating iron metabolism. Biochanin A, an isoflavone isolated from Huangqi, could directly reduce intracellular iron concentration by inhibiting TfR1 and promoting ferroprotein and then prevent the OA progress [133]. Moreover, Ruscogenin was reported to induce the upregulation of Ferritin and SLC7A11, regulate iron metabolism, suppress chondrocyte ferroptosis and then attenuate cartilage destruction [134]. Nrf2 regulates FTL/FTH1 expression to control intracellular free iron concentrations, induces cystine uptake and promotes the synthesis of GSH and GPX4. Gamma‐oryzanol disrupts Keap1‐NRF2 binding and then activates Nrf2 signalling, reducing chondrocyte ferroptosis and mitigating OA [135]. Furthermore, protopine, an isoquinoline alkaloid, protected chondrocytes from ferroptosis by activating the Nrf2 pathway [136]. Sappanone A alleviated OA progression by inhibiting chondrocyte ferroptosis via activating the SIRT1/Nrf2 signalling pathway [137].

GPX4 is the only known glutathione peroxidase responsible for reducing liposomal peroxides. Gossypol acetic acid, a natural phenolic compound, alleviated the ferroptosis of chondrocytes and OA progression by increasing GPX4 expression [138]. What's more, Shang et al. [138] further revealed that Gossypol Acetic Acid alleviates chondrocyte ferroptosis in OA by inhibiting GPX4 methylation. Botulinum toxin A, a macromolecular protein extracted from Clostridium Botulinum, could improve mitochondrial function and promote the activation of SLC7A11/GPX4 anti‐ferroptosis system in chondrocytes, thereby attenuating OA [139]. In addition, Vinpocetine protected against OA by inhibiting ferroptosis and extracellular matrix degradation through activation of the Nrf2/GPX4 pathway [140]. And Zhang et al. [141] reported that asperosaponin VI suppresses chondrocyte ferroptosis and ameliorates OA by modulating the Nrf2/GPX4/HO‐1 signalling pathway. Further details on the mechanisms of other natural products regulating ferroptosis in OA are summarised in Table 1.

Chronic pain is the key manifestation of RA. It has been reported that Anemoside B4, the bioactive ingredient of triterpenoid saponins in the traditional Chinese medicine pulsatilla, alleviates RA pain via suppressing ferroptosis‐mediated inflammation [162]. Moreover, the dysregulation of FLS also contributed to the progression of RA. Amentoflavone, a polyphenolic compound derived from Selaginella tamariscina, induced ferroptosis to inhibit the proliferation, invasion and inflammation in RA‐FLS [163]. JinWu JianGu capsules improved RA primarily through the integrated regulation of the SLC7A11/GSH/GPX4 pathway in M1 macrophages and attenuated ferroptosis [164]. Similarly, Jingfang Granules alleviates lipid peroxidation‐induced ferroptosis in RA by modulating gut microbiota, metabolism of short chain fatty acids and AMPK signalling [165]. Wuwei Ganlu and Myricetin alleviate RA by inhibiting M1 macrophage polarisation through modulation of SHBG/SREBP1‐mediated lipid metabolism [166]. Furthermore, Huo et al. [167] found that cupuncture suppressed synovitis of RA by repressing ferroptosis via butyric acid. Osmundacetone reduced RA‐induced swelling and bone destruction, as well as alleviating inflammation‐related factors and oxidative stress via regulating osteoclast ferroptosis [168].

However, some studies present contradictory findings. Sun et al. [169] reported that Asiatic acid induces ferroptosis in RA‐FLS via the Nrf2/HMOX1 pathway to relieve inflammation in RA. And it has been demonstrated that wasp venom could accumulate lipid ROS to induce GPX4‐mediated ferroptosis and were potential therapeutic agents for RA through modulating JAK/STAT signalling pathway [170]. The effects of ferroptosis in RA may depend on the specific cell type undergoing ferroptosis.

OP is a major clinical issue in older individuals, featured with a disorder of decreased bone mass, microarchitectural deterioration and fragility fractures. Due to abnormal osteoblasts and osteoclasts homeostasis, bone‐forming is weakened and bone‐resorbing is enhanced, resulting in decreased bone mass and increased bone brittleness. Therefore, maintaining osteoblast activity and function, and suppressing osteoclast activity are critical strategies for OP treatment. It has been reported recently that ferroptosis is involved in the pathogenesis of OP via regulating the activity of both osteoblasts and osteoclasts [171].

Research showed that osteoblast ferroptosis is a key factor in age‐related OP [172], and inhibiting ferroptosis in osteoblasts may help alleviate the progression of OP. For instance, Proanthocyanidins treatment improves trabecular bone structure, reduces bone marrow adipocytes, decreases oxidative stress and enhances the expression of key osteogenic proteins by regulating the SIRT6/Nrf2/GPX4 pathways [173]. Deng et al. [108] found that Mangiferin attenuated OP by inhibiting osteoblastic ferroptosis via the Keap1/Nrf2/SLC7A11/GPX4 pathway. Similarly, Fructus Ligustri Lucidi protects osteogenic potential by inhibiting ferroptosis through the Nrf2/HO‐1/GPX4 signalling pathway [174]. Moreover, Poliumoside protects against type 2 diabetes‐related OP by suppressing ferroptosis via activation of the Nrf2/GPX4 pathway [117]. Asperosaponin VI reduced osteoblast ferroptosis and alleviated diabetic OP by increasing the GPX4 expression [175]. Picein alleviated oxidative stress and promoted bone regeneration in osteoporotic bone defects by inhibiting ferroptosis through the Nrf2/HO‐1/GPX4 pathway [176].

On the other hand, some natural compounds contributed to bone homeostasis in OP via regulating osteoclastogenesis. It has been reported that artesunate inhibits osteoclast differentiation by inducing ferroptosis and preventing iron overload‐induced bone loss [112]. Moreover, Saikosaponin A promoted osteoclast ferroptosis, attenuated osteoclastogenesis and bone loss by inhibiting the Nrf2/SLC7A11/GPX4 axis [177]. Icariin regulated systemic iron metabolism and promoted osteoclast ferroptosis and thus inhibits bone loss and iron overload‐induced OP [178].

Moreover, Mesenchymal stem cells (MSCs), especially bone marrow mesenchymal stem cells, play a vital role in bone metabolism and tissue repair, with their ability to differentiate into osteoblasts being crucial in the treatment of OP. For example, Crocin can inhibit ferroptosis via the Nrf2/GPX4 pathway and promote the osteogenic function of BMSCs [179]. Li et al. [180] reported that 4‐Octyl Itaconate attenuates postmenopausal OP by inhibiting ferroptosis and enhancing osteogenesis via the Nrf2 pathway. Wang et al. [181] found that Vaccarin not only suppressed erastin‐induced ferroptosis but also enhanced the osteogenic differentiation of BMSCs, thereby alleviating OP.

However, some studies suggest that inhibiting osteoclast ferroptosis may benefit OP treatment. For instance, Xue et al. [182] demonstrated that aconine attenuates osteoclast‐mediated bone resorption and improves OP via inhibiting NF‐κB signalling and osteoclast ferroptosis. A summary of other natural compounds regulating ferroptosis in OP is provided in Table 2.

Various studies have highlighted the potential value of natural products in cancer treatment, including OS. For example, Huang et al. [197] demonstrated that Artesunate exhibits anti‐OS properties by inducing ferroptosis through NCOA4‐mediated ferritinophagy. Shikonin was reported to effectively suppress OS growth with favourable biosafety through regulating the mitochondrial ROS‐regulated HIF‐1α/HO‐1 axis, thereby inducing ferroptosis [198]. Marsdenia tenacissima extract accelerates ferroptosis in OS cells by upregulating HO‐1 and activating mitophagy [199]. Curcumin induces both ferroptosis and apoptosis in OS cells by modulating the Nrf2/GPX4 signalling pathway [200]. Casticin, a natural flavonoid constituent, promotes ferroptosis and inhibits OS growth and metastasis through Fe²⁺ overload and ROS production mediated by HMOX1 and LC3‐NCOA4 [201]. Moreover, Hydroxysafflor yellow A, a natural quinochalcone C‐glycosides compound from Carthamus tinctorius L, induces ferroptosis via the HIF‐1α/HK2 and SLC7A11 pathways, suppressing OS progression [202]. Baicalin, a biologically active flavonoid compound isolated from Radix Scutellariae, induces ferroptosis and inhibits OS through a novel Nrf2/xCT/GPX4 regulatory axis [203].

In addition, some compounds could improve the sensitivity of cancer cells to chemotherapy agents. Luo et al. [204] reported that Eriodictyol‐cisplatin coated nanomedicine synergistically promotes ferroptosis and chemosensitivity in OS cells. Formononetin improves cisplatin chemotherapy sensitivity in OS by inducing ferroptosis and reconstructing the immune microenvironment [205]. Shikonin overcomes cisplatin resistance of cancer cells by inducing ferroptosis via upregulation of HMOX1 [206]. Eicosapentaenoic acid enhances the sensitivity of OS to cisplatin by inducing ferroptosis through the NRF2 pathway [207]. Brusatol induces ferroptosis in OS by modulating the Keap1/Nrf2/SLC7A11 signalling pathway, and its combination with doxorubicin (DOX) significantly enhances DOX's anti‐OS efficacy [208]. A summary of other natural compounds regulating ferroptosis in OS is provided in Table 3.