Amino acid and GSH metabolism are tightly linked to the regulation of ferroptosis. The amino acid antiporter system Xc⁻ is consisted of solute carrier family 3 member 2 (SLC3A2) and the catalytic subunit solute carrier family 7 member 11 (SLC7A11) [28]. The system Xc⁻ takes charge of redox homeostasis by exchange of extracellular cystine and intracellular glutamate at the ratio of 1:1. The imported cystine can be promptly converted to cysteine and serves as substrate for GSH synthesis. Then, GPX4, a central regulator of ferroptosis, utilizes the antioxidant GSH to detoxify lipid peroxidation and modulate the initiation of ferroptosis [5]. Inhibiting system Xc⁻ with small molecular (e.g., erastin) can decrease the cellular uptake of cysteine, inhibit GSH synthesis and inactivate GPX4, which in turn, triggers excessive lipid peroxidation and ferroptotic cell death [5,29]. According to study, the alteration of GPX4 level and/or activity will immediately affect cell survival [30]. In vivo study has confirmed the core role of GPX4 in ferroptosis. Inactivation of GPX4 could cause severe acute renal failure in mice [11]. Acting as an essential cofactor of GPX4, GSH could reduce lipid peroxides, playing the ferroptosis-resistance activity [3]. The ferroptotic agonist (1S, 3R)-RSL3 can bind and irreversibly inactivate GPX4, leading to cell death [17]. Therefore, gene knockout and inactivation of GPX4 is vital way to induce ferroptosis. It is suggested that system Xc⁻-GSH-GPX4 may be effective therapeutic approach in organ damage mediated by ferroptosis.

Iron is a redox-active metal. Excessive iron contributes to the execution of ferroptosis through inducing the production of ROS by Fenton chain reaction. Thus, the sensitivity of ferroptosis is closely associated with iron homeostasis including iron uptake, export, storage and turnover. Iron homeostasis is a complex process, which relies on coordination of multiple genes such as IREB2 (iron responsive element binding protein 2), FTH1 (ferritin heavy chain 1), FTL (ferritin light chain), TF (transferrin), TFR1 (transferring receptor 1), FPN (ferropotein) and DMT1 (divalent metal-ion transporter-1). These genes encoded proteins could affect ferroptosis trigger via regulating intracellular iron. A recent study found that Tf and TfR1, mainly responsible for intestinal iron absorption, play the critical role in ferroptotic cell death [23,31]. Elevated plasma level of TfR1 increased the amount of intracellular iron, which might promote ferroptosis, e.g., leading to auditory cortex neurodegeneration [32]. The redundant intracellular iron is stored in ferritin which is composed of FTL and FTH1. RSL3-sensitive cells transformed with oncogenic RAS significantly could increase cellular iron level via increasing TfR1 and decreasing FTL and FTH1 mRNA expression [23]. The selective cargo receptor-nuclear receptor coactivator 4 (NCOA4)-regulated degradation of ferritin (referred to as ferritinophagy) influences the availability of labile iron and stimulates erastin-induced ferroptosis [33,34,35]. The bromodomain protein BRD4 inhibitor (+)-JQ1 could induce ferroptosis via ferritinophagy and regulate ferroptosis-related genes in cancer cells [36]. IREB2 is a master regulator of iron metabolism. Silencing of IREB2 by shRNA inhibits the sensitivity to erastin-induced ferroptosis in HT1080 cells [4]. Based on these findings, the cellular iron metabolism is indispensable for the induction of ferroptosis. Strategy targeting iron metabolism genes might be a promising method to treat the cancers which are drug-resistant or alleviate ferroptosis-induced diseases.

Lipid metabolism is closely associated with the process of ferroptosis. The accumulation of lipid peroxide could directly damage cellular and organelle membrane, initiating ferroptotic cell death. Polyunsaturated fatty acids (PUFAs) serve as the main substrates of lipid peroxidation during ferroptosis [37]. Thus, the extent of lipid peroxidation and the process of ferroptosis in cells largely depend on the content and localization of PUFAs. According to some reports, PUFA- phosphatidylethanolamines (PEs) esterified with arachidonoyl (AA) and adrenoyl (AdA) acyl chains are the key phospholipids in triggering ferroptosis [26,38]. The free AA/AdA could bind to coenzyme A (CoA) to form AA/AdA-CoA under the action of Acyl-CoA synthetase long-chain family member 4 (ACSL4), facilitating their esterification into phospholipids. Subsequently, lysophosphatidylcholine acyltransferase-3 (LPCAT3) catalyzes AA/AdA-CoA esterified into PEs and then the formed AA/AdA-PEs would be oxidized into lipid hydroperoxides by lipoxygenase (LOX), inducing ferroptosis [39]. Therefore, ACSL4 and LPCAT3, two enzymes involved in the biosynthesis and remodeling of PUFA-PEs in cellular membranes, are critical determinants of ferroptosis sensitivity [3]. Genetic disruption of ACSL4 and LPCAT3 depletes the substrates for lipid peroxidation and decreases ferroptosis sensitivity [38,39,40]. It is reported that cells treated with arachidonic acid or other PUFA are sensitized to ferroptosis [26]. Peroxidation of n-3 and n-6 PUFA exerted selective cytotoxic effects in acidic cancer cells, leading to ferroptosis-mediated antitumor effects [41]. In addition, LOXs also contribute to ferroptosis. The suppression of LOX by ferroptosis inhibitors, including vitamin E, zileuton and baicalein, can relieve ferroptosis [42,43,44]. However, some LOXs are essential for normal embryonic development in vertebrates. Knockdown of 12S-LOX in zeberafish could severely impair embryonic phenotype, characterized by abnormal brain, eyes, tails as well as yolk sac and pericardial edema [45]. It is suggested that lipid metabolism is required for both ferroptosis and normal physiological function. Abnormal lipid metabolism could be a crucial trigger for ferroptosis.

FSP1 was previously called apoptosis inducing factor mitochondria associated 2 (AIFM2). Recently, two momentous back-to-back research by Doll et al. and Bersuker et al. disclosed that FSP1 is a novel inhibitor of ferroptosis and could counteract ferroptosis independently even in the absence of GPX4 [46,47]. AIFM2 is proposed before the concept of ferroptosis, whose activation could suppress tumor proliferation and enhance apoptosis [48]. In human lung cancer cell lines, the activation of AIFM2 markedly contributed to cancer cells apoptosis undergoing exogenous toxicological stimulus [49]. Recently, Bersuker and his colleagues identified that FSP1 was a potent ferroptosis-resistance factor through using a synthetic lethal CRISPR-Cas9 knockout screen [47]. Almost simultaneously, Doll et al. also executed relevant studies on FSP1 [46]. By using an overexpression screen, their group demonstrated that FSP1 complemented the loss of GPX4 or inhibition of GPX4 in human cancer cells. The inhibition of ferroptosis by FSP1 was mediated by antioxidant CoQ10, while FSP1 could catalyze the regeneration of CoQ10 by NAD(P)H [46,47]. Therefore, FSP1-CoQ10-NAD(P)H axis is an independent parallel system, cooperating with GPX4 and GSH/GSSG to inhibit lipid peroxidation and control the process of ferroptosis. Both studies disclosed that the expression of FSP1 was negatively correlated with the ferroptosis sensitivity in many cancer cell lines. In conclusion, the discovery of FSP1 supplemented and perfected the ferrroptosis pathway. FSP1 can be used as a valuable biomarker of ferroptosis resistance. Exploiting FSP1 inhibitors can be effective strategy to overcome ferroptosis resistance in cancer cells.

There are some regulators and key components that act vital roles in the ferroptosis cascade (Figure 1).