2.1 Glucose metabolism

Glucose is the main source of cellular energy and an important substrate for the synthesis of biomolecules in cells [20]. After entering cells through transporters, extracellular glucose is consumed by cells through glycolysis and the tricarboxylic acid (TCA) cycle to produce energy and intermediate metabolites for cellular biosynthesis. Glycolysis is the first step of intracellular glucose metabolism. Glucose entering the cell is catabolized in the cytoplasm by hexokinase (HK) to yield glucose 6-phosphate (G6P), which is subsequently transformed into fructose 6-phosphate under the action of G6P isomerase, and fructose 6-phosphate participates in the next step of the glucose catabolism process, which eventually produces nicotinamide adenine dinucleotide (NADH) and pyruvate. Some of this generated pyruvate is converted to lactate under the action of lactate dehydrogenase to complete glycolysis. The other portion of the generated pyruvate enters mitochondria and is consumed in the TCA cycle, which produces energy as well as intermediate metabolic products [21]. Mitochondria are the main sources of energy production in cells. Energy production is an oxygen-consuming process, and the TCA cycle drives the transfer of electrons across the mitochondrial inner membrane, where large amounts of reactive oxygen species (ROS), including H₂O₂, OH-, ·OH, etc.) are also produced. ROS plays multifaceted roles in cells. On the one hand, ROS can be signaling molecules that stimulate cell proliferation; on the other hand, ROS at high concentrations can damage cells and cause cell death. Thus, glycolysis and the TCA cycle maintain a dynamic balance in cells [22-24]. Notably, in the 1920s, scientists discovered that in most tumor cells, energy is primarily produced through glycolysis, even in the presence of sufficient oxygen, a phenomenon known as the Warburg effect. The glycolysis reaction is rapid and can produce a large amount of energy in a short period. Moreover, a pathway that bypasses glycolysis, known as the pentose phosphate pathway (PPP pathway), produces large amounts of ribose 5-phosphate (a precursor of DNA synthesis), 4 phosphate and the reducing equivalent nicotinamide adenine dinucleotide phosphate (NADPH) to induce the rapid proliferation of tumor cells [25]. Furthermore, NADPH is involved in the synthesis of the reduced form of glutathione (GSH), an important reducing agent, which can effectively attenuate oxidative damage in tumor cells [26]. As mentioned above, the primary cause of ferroptosis is disrupted redox balance. Abnormalities in glucose metabolism in tumor cells can alter redox homeostasis, leading to the sensitivity or resistance of tumor cells to ferroptosis [27]. Therefore, the normal regulation of intracellular glucose metabolism plays an important role in tumor ferroptosis (Figure 1).

Glucose-6-phosphate dehydrogenase (G6PD) is a rate-limiting enzyme in the PPP. Abnormal expression of G6PD in tumor cells can lead to the disruption of intracellular NADPH and GSH metabolism, causing intracellular redox imbalance and contributing to tumor ferroptosis [28, 29]. Extensive glycogen accumulation and mutations in histone-modifying genes are important to the development of clear cell renal carcinoma. For example, a high frequency of histone methylase lysine demethylase 5C (KDM5C) mutations has been found in patients with clear cell renal cell carcinoma, and restoration of KDM5C expression in clear renal cell carcinoma cells led to a reduction in glycogen production, inhibition of G6PD expression and PPP activation, decreased NADPH and GSH production, and ferroptosis inhibition [28]. In addition, accumulating evidence has shown that G6PD can inhibit ferroptosis by regulating the expression of cytochrome P450 oxidoreductase. In hepatocellular carcinoma (HCC), G6PD expression is elevated and negatively correlated with the prognosis of HCC patients [29]. Further study showed that G6PD inhibited the expression of P450, thereby inhibiting the ferroptosis of HCC cells and enhancing the proliferation, invasion and metastasis of HCC cells [29]. Lactate is a metabolite of glycolysis, and mucosal melanoma cells can take up extracellular lactate through mononuclear transport carriers, thereby increasing PPP pathway signaling and upregulating ferroptosis-related GPX4 and FSP1 expression. In addition, lactate induces an increase in the intracellular levels of NADH, NADPH and GSH, leading to the resistance of tumor cells to ferroptosis [30]. In addition, lactate can activate the hydroxycarboxylic acid receptor 1 (HCAR1)/monocarboxylate transporter 1 (MCT1)-sterol regulatory element-binding protein 1 (SREBP1)-stearoyl-coenzyme A desaturase-1 (SCD1) signaling pathway to induce monounsaturated fatty acid (MUFA) generation, thereby inhibiting the ferroptosis of tumor cells [31]. Ras-selective lethal small molecule 3 (RSL3) is a ferroptosis inducer that targets GPX4 and promotes ferroptosis. Furthermore, in addition to targeting GPX4, RSL3 downregulated the expression of glycolysis-related proteins such as hexokinase 2 (HK2), phosphofructokinase (PFKP), and pyruvate kinase M2 (PKM2) and reduced the glycolysis rate in glioma cells, while the addition of sodium pyruvate greatly reduced the killing effect of RSL3 on glioma cells, reflecting the role of glycolysis in anti-ferroptosis effects [32]. AMP-activated protein kinase (AMPK) is a sensory receptor of intracellular ATP and is activated during energy stress in tumor cells, while activated AMPK can promote GPX4-dependent ferroptosis via the action of the janus kinase 2 (JAK2)/signal transducer and activator of transcription 3 (STAT3)/tumor protein p53 (p53) signaling axis in kidney cancer cells [33]. AMPK can also promote ferroptosis by promoting the phosphorylation of beclin1 (BECN1), which leads to the inactivation of the SLC7A11 protein [34]. These findings confirm that highly efficient glycolysis can inhibit ferroptosis in tumor cells because AMPK is inactivated when glycolysis is accelerated. In contrast, the activation of AMPK activates ACACA, which inhibits the synthesis of PUFAs, thereby inhibiting ferroptosis [35]. This finding suggests that AMPK plays a dual role in regulating ferroptosis in cancer cells, which mainly depends on substrate AMPK activation. Pyruvate is an important product of glycolysis and initiates the mitochondrial TCA cycle. In pancreatic cancer, upstream stimulatory factor 2 (USF2) is an upstream regulator of PKM2; specifically, USF2 binds to the promoter of PKM2 and upregulates the expression of PKM2 to increase the intracellular GSH concentration and the expression of GPX4, thereby inhibiting pancreatic cancer cell ferroptosis [36].

Mitochondria are sites of the TCA cycle, oxidative phosphorylation (OXPHOS) and ROS production. Therefore, mitochondrial metabolism exerts an important impact on ferroptosis [37]. Notably, cysteine deficiency leads to mitochondrial membrane hyperpolarization and the accumulation of lipid peroxides, leading to ferroptosis [38, 39]. Inhibition of the TCA cycle or OXPHOS can alleviate this phenomenon [38]. Pyruvate is an important raw material for the TCA cycle. Pyruvate enters the TCA cycle through the action of pyruvate dehydrogenase to form acetyl coenzyme A (CoA). In a study on pancreatic ductal adenocarcinoma, researchers found that pyruvate dehydrogenase kinase 4 (PDK4) drove resistance to ferroptosis. Specifically, PDK4 inhibited the entry of pyruvate into the TCA cycle by preventing pyruvate oxidation, thereby reducing the fatty acid synthesis rate and ultimately inhibiting ferroptosis [40]. However, the role of mitochondria in ferroptosis in tumor cells is unclear because HT-1080 human fibrosarcoma cells undergo ferroptosis even though they lack the mitochondrial electron transport chain [41]. Hence, ferroptosis signaling may differ in different tumors. Isocitrate dehydrogenase 2 (IDH2) is a key enzyme in the TCA cycle, playing an important role in NADH production and an important factor involved in mitochondrial GSH turnover. Studies have shown that the downregulation of IDH2 expression enhances the sensitivity of tumor cells to ferroptosis, which may be associated with the reduction in NADH production caused by IDH2 downregulation [42]. Fumarate hydratase is also an important metabolic enzyme in the TCA cycle, and inhibition of this enzyme activity can inhibit ferroptosis [38]. With research advancements, increasing evidence has shown that mitochondrial metabolism plays an important role in the ferroptosis of tumor cells [43, 44]. Glycolysis and mitochondrial OXPHOS are dynamic processes, and inhibition of aerobic glycolysis in tumor cells can cause metabolism in these cells to shift toward OXPHOS. In fact, inducing this metabolic shift is considered an effective strategy for tumor treatment, and the increase in OXPHOS in tumor cells may further increase the sensitivity of tumor cells to ferroptosis.

In general, NADPH NADH, GSH, and ROS produced by glucose metabolism affect the intracellular redox balance, and when the number of oxidized intracellular products increases, lipid peroxidation is induced, leading to cell ferroptosis. Thus, resistance to intracellular oxidation is an effective strategy to defend against ferroptosis.

2.2 Fatty acid metabolism

Thousands of lipid molecules are produced in the human body. In addition to being the main substance of biofilms, lipid molecules play important roles in energy storage and signal transduction [45]. Lipid metabolism is involved in several processes of cellular life activities, and abnormal lipid metabolism exerts an impact on the normal life activities of many cells, including ferroptosis. During lipid metabolism, peroxidation of PUFAs is a central trigger of ferroptosis [46]. We describe the role of lipid metabolism in ferroptosis from the perspective of the synthesis, regulation, storage and transport of PUFAs (Figure 2).

The fatty acyl groups of membrane phospholipids are very diverse in terms of chain length and saturation (double vs. single bonds), which determines the biophysical properties of the cell membrane, including its fluidity, curvature and subdomain structures [47]. These membrane characteristics, in turn, affect membrane-related cellular functions and determine the sensitivity of cells to ferroptosis [48]. The phospholipid molecule is made up of a number of different fatty acids, including saturated fatty acids (SFAs), MUFAs, and PUFAs. The phospholipid molecular layer, which contains many unsaturated fatty acids, is susceptible to ROS damage because the C = C bond in unsaturated organic molecules is particularly prone to chain reaction oxidation. PUFAs constitute a class of fatty acid molecules with two or more C = C bonds in the carbon chains [49]. The double bond adjacent to the methylene group weakens the hydrogen bond connecting a dienyl methyl group to the PUFA chain, thereby increasing the sensitivity of the carbon chain to hydrogen ion acquisition, eventually leading to their oxidation. In addition, when PUFAs are oxidized, they are converted into reactive free radical groups, which propagate a lipid peroxidation chain reaction [50]. Thus, the content and location of PUFAs in the phospholipid bilayer determine the sensitivity of cells to ferroptosis [51].

Phosphatidylethanolamine (PE) is an esterification product of PUFAs, mainly arachidonic acid (AA) and adrenoyl (AdA), which directly induce the development of ferroptosis after being oxidized. Ferroptosis requires acyl coenzyme synthetase long-chain family 4 (ACSL4) [52-55]. Peroxisomes are also sources of the synthesized lipids needed to initiate ferroptosis, and peroxisome-mediated plasmalogen enhances ferroptosis in tumor cells by synthesizing polyunsaturated ether phospholipids (PUFA-ePLs) in the endoplasmic reticulum, causing lipid peroxidation [56]. In contrast to PUFAs, MUFAs such as exogenous oleic acid (OA) and palmitoleic acid (POA) inhibit erastin- and RSL3-induced ferroptosis in tumor cells. Specifically, MUFAs inhibit the production of lipid ROS, especially in cells of the serous membrane, replacing PUFAs in cells [57].

As mentioned above, the abundance and localization of PUFAs determine the degree of lipid peroxidation in cells and, therefore, the degree of iron toxicity. Free PUFAs are substrates for the synthesis of lipid signaling mediator substrates, but they must be esterified to membrane phospholipids and undergo oxidation [54]. Thus, the formation of CoA derivatives of these PUFAs and the incorporation of these derivatives into phospholipids are critical for the generation of death signaling induced by iron overloading. The incorporation of newly synthesized fatty acids into phospholipids requires the conversion of the long-chain fatty acid stearyl-CoA catalyzed by acyl-coenzyme A synthase (ACS) and re-acylation catalyzed by lysophospholipid acyltransferases (LPLATs) [58, 59]. The acyl coenzyme A (acyl-CoA) synthetase long-chain family is expressed mainly in the cytoplasmic matrix and mitochondrial outer membrane and catalyzes the conversion of fatty acids to acyl-CoAs [60]. These acyl-CoAs are intermediate products of lipid metabolism and participate in fatty acid metabolism and biofilm modifications [61]. Among the ACSL family, the most closely related to ferroptosis are ACSL4 and ACSL3. ACSL4 attaches CoA to free fatty acids via esterification in an ATP-dependent manner, and its preferred substrates are AA and AdA, and its downregulation in tumors leads to the resistance of tumor cells to ferroptosis [5, 52, 62]. Elongation of very long-chain fatty acid protein 5 (ELOVL5) and fatty acid desaturase 1 (FADS1) may also increase the sensitivity of cells to ferroptosis by promoting the synthesis of PUFAs [63]. In contrast, ACSL3, for which the preferred substrate is OA, can activate MUFAs, which replace the PUFAs in the plasma membrane, thereby reducing the sensitivity of the plasma membrane to peroxidation and inhibiting PUFA-induced lipotoxicity [57, 62]. Familiar to ACSL3, stearoyl CoA desaturase (SCD/SCD1) is a multifunctional enzyme involved in lipid metabolism, and it can catalyze the further desaturation of unsaturated fatty acids and inhibit ferroptosis [64]. In addition, SCD1 also upregulates the expression of CoQ₁₀ (an antioxidant that inhibits ferroptosis), and its anti-ferroptotic activity includes both monosaturated fatty acid-dependent and non-dependent functions [65]. In addition, a number of other factors can inhibit lipid peroxidation. Phospholipid-modifying enzymes (MBOAT1/2) can inhibit ferroptosis by remodeling the cellular phospholipid profile, and strikingly, their ferroptosis surveillance function is independent of GPX4 or FSP1 [66]. Ca²⁺ independent phospholipase A2β (iPLA2β) is a calcium-independent phospholipase, iPLA2β-mediated detoxification of peroxidized lipids is sufficient to suppress p53-driven ferroptosis upon ROS-induced stress, even in GPX4-null cells [67, 68].

Esterification is an important step in the peroxidation of phospholipids, carried out mainly by lysophosphatidyl transferases (LPCATs). LPCATs constitute a group of enzymes critical for phospholipid remodeling, mainly in the endoplasmic reticulum. They catalyze the attachment of the fatty acyl chain at the sn-2 site of phosphatidylcholine, thereby regulating the fatty acyl composition of phospholipids [48, 69]. After linking CoA via ACSL4, long-chain PUFAs esterify lysophospholipids via LPCAT; this in turn triggers peroxidation, leading to ferroptosis [51]. Lysophosphatidylcholine acyltransferase 3 (LPCAT3) regulates lipid metabolism mainly by regulating lipid uptake, lipoprotein secretion and de novo lipogenesis [48].

Peroxidation of PUFAs ultimately requires the incorporation of oxygen into lipids, and arachidonic acid lipoxygenases (ALOXs) are among these iron-containing dioxygenases [70]. ALOX5 is the key trigger of lipid peroxidation, and it can oxidize AA to form the unstable intermediate 5-hydrperoxyeicosatetraenoic acid (5-HPETE) [71]. In addition, ALOX5, in combination with the scaffolding protein PE-binding protein 1 (PEBP1), forms lipid peroxides and promotes ferroptosis [72, 73]. Cytochrome p450 oxidoreductase (POR) is an oxidoreductase that promotes the peroxidation of polyunsaturated phospholipids independent of ALOX action. It transfers an H molecule from NADPH to an O molecule to form H₂O₂, which induces the Fenton reaction with Fe²⁺ in cells to promote the peroxidation of PUFAs, thereby inducing ferroptosis [74, 75].

Lipids in cells are stored in the form of lipid droplets, and the outcome of this storage mechanism is the inhibition of ferroptosis by reducing the intracellular concentration of free unsaturated fatty acids [76, 77]. Tumor protein D52 (TD52) and Perilipin2 (PLIN2) promote the formation of lipid droplets and inhibit ferroptosis in tumor cells [78, 79]. In contrast, the degradation of lipid droplets promotes ferroptosis. Lipolysis and lipophagy are the two main modes of lipid droplet degradation, and RAS oncogene family member 7A (RAB7A)- and patatin like phospholipase domain containing 2 (PNPLA2)-mediated lipid droplet degradation promotes ferroptosis [79, 80].

Fatty acid metabolism is most closely related to ferroptosis, and peroxidation of PUFAs is the key factor in cellular ferroptosis; increased production of PUFAs and disruption of the PUFA regulatory system will lead to peroxidation of PUFAs, resulting in cellular ferroptosis. MUFAs, which are more stable than PUFAs, can replace PUFAs to reduce the occurrence of lipid peroxidation. Thus, for tumor cells, promoting the levels of intracellular PUFAs, increasing the activity of the PUFA production and regulation system, or attenuating the synthesis of MUFAs are all feasible solutions to promote ferroptosis and increase the efficacy of antitumor therapy.

2.3 Amino acid metabolism

Amino acids are essential nutrients for the human body and are the basic raw materials for protein synthesis. In addition, amino acids are involved in energy metabolism, the synthesis of biomolecules, signal transduction and the maintenance of intracellular redox homeostasis [81, 82]. Insufficient intake of amino acids can seriously endanger human health. Studies have shown that amino acid metabolism is closely related to tumor development, and abnormal amino acid metabolism has been found in many tumors [83]. Accumulating evidence suggests that the abnormal metabolism of amino acids is related to ferroptosis in tumor cells and the regulation of tumor sensitivity to ferroptosis inducers. Among those needed by the human body, some amino acids must be ingested through food; these are called essential amino acids; other amino acids can be produced intracellularly through a synthetic transformation pathway, and they are called nonessential amino acids. Both types of amino acids play important roles in regulating tumor cell ferroptosis (Figure 3) [84].

Cysteine is a nonessential amino acid that exists mainly in the form of cystine in the extracellular space. It is transported into cells by cystine transporters and, after being reduced to cysteine, participates in the synthesis of GSH, hydrogen sulfide (H₂S), 3-thiol pyruvate, homocysteine, and other organic compounds, thereby regulating the growth, metastasis, and drug resistance of tumors [85]. In addition, cysteine is generated via the transsulfuration pathway [86]. In recent years, many studies have reported that cysteine deprivation not only inhibits the proliferation of tumor cells but also promotes ROS production by inhibiting GSH synthesis and reducing intracellular NADPH content, which ultimately induces tumor ferroptosis [5, 85]. Cysteine starvation therapy significantly inhibited the growth of tumors transplanted into mice [87]. Studies have shown that the anabolism of cysteine in tumor cells consists of 2 pathways: direct uptake of extracellular cysteine/cystine or generation of cysteine through other reactions (mainly via the transsulfuration pathway) [88]. Direct absorption of extracellular cysteine is mediated through the excitatory amino acid transporter 3 (EAAT3) and alanine/serine/cysteine/threonine transporter 1 (ASCT1), and the absorption of cystine into the cell is mediated through system Xc- (cystine/glutamate antiporter); in cells, cystine is reduced to yield cysteine [89^–91]. The transsulfuration pathway involves the conversion of methionine to S-adenosylmethionine by the rate-limiting enzyme methionine adenylyltransferase 2A (MAT2A), and then a series of biochemical reactions culminates in the formation of cysteine [92]. In addition, cysteine can be formed by the autophagy-related breakdown of intracellular proteins and GSH [14]. We know that cysteine is a raw material in the production of reduced GSH, which is triggered in response to glutamate-cysteine ligase (GCL). GSH is an important intracellular reducing agent in cells that neutralizes intracellular ROS, thereby inhibiting cellular peroxidation. In cells, GSH exists in 2 forms: the reduced form (GSH) and the oxidized form (glutathione disulfide [GSSG]). The interconversion of the 2 forms maintains the redox balance in cells [93]. The abnormal metabolism of cysteine leads to the inhibition of GSH synthesis, causing disrupted intracellular oxidative homeostasis and tumor cell ferroptosis [94, 95].

Glutamine is a nonessential amino acid that is found in large quantities in tumor cells. It enters cells from the extracellular environment and can be produced in tumor cells through a series of reactions. After extracellular glutamine enters tumor cells through the transporter solute carrier family 1 member 5 (SLC1A5; known as ASCT2), it can be converted into glutamate under the catalytic effect of glutaminase. Glutamine contributes to the formation of GSH and NADPH together with Gly and Cys synthesis and regulates redox homeostasis in cells [96, 97]. Moreover, glutamine can be a raw material and intermediate metabolite in the synthesis of nucleic acids, other amino acids and biomolecules [98, 99]. Since the discovery of aerobic glycolysis in tumor cells, some tumor cells have been shown to produce energy independent of glycolysis; in addition, glucose deprivation does not cause cell death immediately. Studies have reported that excessive glutamine provides energy for tumor cells and thus maintains tumor cell survival [100]. Moreover, glutamine was shown to be consumed almost 10-fold faster in HCC and fibrosarcoma cells than in normal cells [84]. This rapid consumption is due to glutamine entering mitochondria and being converted into α-ketoglutarate (α-KG) under the action of glutaminase (GLS) and glutamate dehydrogenase (GDH), after which it enters the TCA cycle and produces ATP and TCA cycle intermediate metabolites needed for cell survival [99, 101, 102]. Glutamine metabolism is closely associated with ferroptosis, which may be related to the catabolic metabolism of glutamine [90]. Glutamine is converted into glutamate, aspartate, pyruvate, lactate, arginine, citrate, and other amino acids through a series of reactions catalyzed by metabolic enzymes such as GLS, GDH, and glutamic-oxaloacetic transaminase 2(GOT2), and these amino acids are subsequently consumed as raw materials in the TCA cycle and during lipid synthesis, thus reducing the ferroptosis rate [84] Furthermore, α-KG, a metabolite of glutamine, and a series of downstream α-KG products, including succinic acid, ferredoxin, and malic acid, can substitute for glutamine and promote the accumulation of lipid ROS [103]. Thus, inhibition of glutamine metabolic signaling, such as that caused by inhibition of glutamine transport carriers, GLS activity, or glutamic-oxaloacetic transaminase 1 (GOT1) activity, can profoundly inhibit ferroptosis [104-106]. However, due to the specificity of glutamine and its role as a nonessential amino acid, a compensatory mechanism is triggered in tumor cells, and the simple inhibition of glutamine uptake by inhibiting glutamine transport carriers is not sufficient to inhibit tumor growth [83]. Therefore, the combination of glutamine intake and catabolism inhibition is necessary to achieve a better therapeutic effect via glutamine-targeted tumor therapy.

Tryptophan is an essential amino acid that is involved mainly in the kynurenine (Kyn) signaling pathway and is a substrate in cells; in addition, the tryptophan metabolites serotonin (5-HT) and 3-hydroxyanthranilic acid (3-HA) markedly facilitate tumor cells, enabling them to escape ferroptosis [107, 108]. In addition, tryptophan can be metabolized to produce indole-3-pyruvate (I3P), a process catalyzed by interleukin 4 induction 1 (IL4i1), and it inhibits tumor ferroptosis. Mechanistically, I3P can upregulate the expression of ferroptosis-related genes such as the SLC7A11, NQO1, CYP1B1, and AKR1C family proteins at the transcriptional level; on the other hand, I3P shows the ability to scavenge free radicals efficiently, thereby inhibiting lipid peroxidation and inducing cell resistance to ferroptosis [109]. In addition, some other amino acids play important roles in the regulation of ferroptosis in tumor cells. Serine is an important component of the one-carbon synthesis pathway, and activation of the serine synthesis pathway is directly related to the synthesis of GSH. This may be because serine is directly involved in the synthesis of Cys, and Cys and Gly are important raw materials for GSH synthesis and are directly related to ferroptosis [84, 110]. This role played by serine implies that the metabolism of serine and glycine is likely to play an important role in ferroptosis in tumor cells. Recent studies have shown that lysine also plays an important role in the regulation of ferroptosis. L-lysine α-oxidase can activate ferroptosis signaling by catalyzing the oxidative decarboxylation of lysine and ROS production in triple-negative breast cancer cells [111].

Abnormal amino acid metabolism is an important feature of tumors. The current study suggests that amino acid metabolism affects ferroptosis mainly through the production of GSH and the induced expression of some ferroptosis-related genes, such as SLC7A11 and NQO1, by amino acid metabolites. Although reports on the regulation of ferroptosis by amino acids in tumor cells are rare, as research advances, the mechanism by which amino acid metabolism regulates ferroptosis is expected to become clearer.

2.4 Iron metabolism

Iron is an important trace element in the human body and an important component of blood cells. Furthermore, iron is a cofactor of various intracellular enzymes and participates in the regulation of various intracellular metabolic pathways. Disruptions of iron metabolism can lead to a variety of pathological processes, including iron-induced cell death, which is the basis for the term ferroptosis. The metabolism of iron in cells involves three main processes: import, transport, and storage/export. All of these processes exert a direct impact on tumor ferroptosis (Figure 4). Fe³⁺ binds to ferroportin (FPN) on the surface of the cell membrane and forms a complex with membrane protein TF receptor 1 (TFR1), which enables its transport into the cell through endocytosis [14, 112]. Fe³⁺ that enters a cell is reduced to Fe²⁺ in endosomes by six-transmembrane epithelial antigens of prostate 3 (STEAP3) and is subsequently transported to the cytoplasm by the divalent metal transporter 1 (DMT1) protein [113]. Fe²⁺ (establishing the labile iron pool [LIP]) in the cytosol as well as in mitochondria shows oxidative activity and induces ferroptosis by causing lipid peroxidation mediated through the Fenton reaction. In addition, LIP iron can bind to ferritin via poly-(rC)-binding protein 1 (PCBP1) and PCBP2, thereby reducing the number of intracellular free iron ions and inhibiting ferroptosis [114]. Intracellular iron can be transported outside the cell via the FPN protein, thus reducing the amount of intracellular iron accumulation and regulating intracellular iron homeostasis [115].

The influx, storage and efflux of intracellular iron affects the homeostasis of intracellular iron levels and regulates the sensitivity of tumor cells to ferroptosis. In glioma cells, the expression of transferrin receptor (TFRC) was increased by pseudolaric acid B, which promoted the uptake of Fe³⁺ and induced glioma cell ferroptosis [116]. Ferritin is an important protein that stores Fe²⁺. The autophagic degradation of ferritin, called ferritinophagy, causes the release of Fe²⁺, thereby increasing the content of free Fe²⁺ in cells and inducing ferroptosis [117]. Similarly, inhibition of ferroportin activity leads to the intracellular accumulation of iron in the LIP, leading to ferroptosis in tumor cells [118]. Recent studies have reported that prominin-2 plays an important role in promoting the transport of iron out of cells. Prominin-2 promotes the formation of multivesicular bodies, that is, exosomes containing ferritin, thereby enhancing the resistance of breast cancer cells to ferroptosis inducers [119].

Irons are important influences on ferroptosis, and the accumulation of intracellular Fe²⁺ causes lipid peroxidation through the Fenton reaction and induces ferroptosis. Thus, iron chelators can inhibit the occurrence of ferroptosis by reducing free Fe²⁺ in cells. However, not all iron accumulation leads to ferroptosis, and some tumor cells contained a large amount of iron but did not undergo ferroptosis [120], suggesting that iron-induced ferroptosis is tissue- and cell-specific, and further in-depth study of its mechanism is needed.

2.5 Selenium metabolism

Selenium is an essential trace element in the human body and plays an important role in regulating the redox balance in human cells under oxidative stimulation. Selenium in the human body is obtained mainly from the diet, including selenomethionine from plant food sources and selenocysteine from animal food sources. The metabolic pathways of selenomethionine in the body mainly include the following processes: (1) Selenomethionine is randomly inserted into the methionine site of a protein during protein synthesis. When the selenomethionine-carrying protein is degraded, selenomethionine re-enters the free selenomethionine pool for recycling. (2) Selenomethionine is metabolized by selenomethionine lyase to yield methyl selenite, which is processed by a demethylase to generate selenide that enters the normal selenium metabolism pathway. (3) Selenocysteine is formed via transsulfuration [121, 122]. However, the metabolism of intracellular selenocysteine is complex; specifically, selenocysteine can be decomposed into selenide and alanine via the action of selenocysteine lyase [123]. Through selenophosphate synthase, selenide is converted to monoselenophosphate, which mediates the generation of selenocysteine-specific transfer RNA (tRNA^sec) and participates in the synthesis of selenoproteins that play important roles in the synthesis and biological function of selenoproteins. The specific pathway includes a combination of L-serine and tRNA^sec in the presence of serine-tRNA synthase to form ser-tRNA^sec. Then, ser-tRNA^sec phosphorylation is catalyzed by phosphoseryl-tRNA kinase (PSTK) to yield p-ser-tRNA^sec. Through an alternative pathway, selenide (HSe⁻) can be converted to monoselenophosphate (H₂SePO₃⁻) by selenophosphate synthetase 2 (SEPHS2) catalysis. and H₂SePO₃⁻ functions as a donor in the formation of sec-tRNA by binding to p-ser-tRNA^sec in the presence of selenocysteine synthase (SelA/SecS). In turn, tRNA^sec recognizes a UGA codon and, through a complex series of processes, participates in selenoprotein synthesis (Figure 5) [124-126].

The selenoproteins involved in ferroptosis include mainly GSH peroxidase (GPX) family proteins and thioredoxin reductase (TrxR), with each of these proteins carrying a selenocysteine (sec, U, Se-Cys) [127]. GPX4, the role of which in tumor cell ferroptosis was previously mentioned, is an important antioxidant. TrxR1 also plays an important role in ferroptosis. In pancreatic cancer cells, diaphanous related dormin 3 (DIAPH3) activated the expression of selenoprotein TrxR1, thereby reducing the content of ROS, reducing the cellular lipid peroxidation rate, inhibiting the occurrence of ferroptosis, and ultimately promoting the malignant progression of pancreatic cancer [128].

Current research has shown that selenium is primarily involved in the synthesis of selenoproteins in the form of selenocysteine, which regulates the activity of selenoproteins. Hence, a normal rate of selenium metabolism may be an important factor in maintaining the normal physiological function of cells, and interfering with the selenium metabolism pathway or inducing abnormal selenium metabolism activity in tumor cells may be a new direction for cancer treatment.