A closer look at the role of iron in glioblastoma

Abstract

Glioblastoma (GBM) is among the deadliest malignancies facing modern oncology. While our understanding of certain aspects of GBM biology has significantly increased over the last decade, other aspects, such as the role of bioactive metals in GBM progression, remain understudied. Iron is the most abundant transition metal found within the earth’s crust and plays an intricate role in human physiology owing to its ability to participate in oxidation–reduction reactions. The importance of iron homeostasis in human physiology is apparent when examining the clinical consequences of iron deficiency or iron overload. Despite this, the role of iron in GBM progression has not been well described. Here, we review and synthesize the existing literature examining iron’s role in GBM progression and patient outcomes, as well as provide a survey of iron’s effects on the major cell types found within the GBM microenvironment at the molecular and cellular level. Iron represents an accessible target given the availability of already approved iron supplements and chelators. Improving our understanding of iron’s role in GBM biology may pave the way for iron-modulating approaches to improve patient outcomes.

Glioblastoma and Iron Overview

Glioblastoma

Glioblastoma (GBM) is the most common adult primary brain cancer as well as one of the deadliest malignancies facing modern oncology.¹ Over the last decade, our understanding of GBM tumor biology has significantly expanded. Despite this, one understudied area in the context of GBM biology is the role of transition metals in tumor progression, especially iron. Iron is the most abundant transition metal found in the earth’s crust, and one of the most bioactive metals found in the human body.^2–4 Given its role in oxygen transport, iron has long been recognized as central to human physiology. Over the last decade, however, additional roles for iron in cellular physiology have been uncovered with discoveries such as ferroptosis, an iron-mediated form of cell death.⁵ These discoveries are now expanding into the field of cancer research with more studies investigating iron’s role in GBM tumor biology.⁶ Nonetheless, how iron impacts GBM patient outcomes remains unclear.

This review aims to consolidate our current understanding regarding the role iron plays in GBM tumor biology. We first briefly review iron metabolism to provide readers with a refresher on iron biology after which we delve into the specific role of iron in the context of GBM. As GBMs consist of a heterogenous mixture of neoplastic and immune cells, we organize this review by first examining iron’s impact on neoplastic cells followed by discussion of its role in the most common immune cell types represented within the GBM tumor microenvironment (TME): macrophages, microglia, neutrophils, and T cells.⁷ We subsequently discuss how these effects may translate to either pro- or antitumoral outcomes in terms of patient survival. Finally, we provide an overview of the existing literature describing preclinical and clinical associations between iron and GBM patient outcomes.

Overview of Iron Biology in Human Physiology

In humans, iron uptake occurs largely in the duodenum, where dietary iron that is often found in the ferric (Fe³⁺) state is reduced into the ferrous state (Fe²⁺) by ferrireductases that line the duodenal brush border, such as duodenal cytochrome B.^8,9 Reduced iron can subsequently enter enterocytes through apically expressed divalent metal transporter 1 (DMT1), a proton-coupled metal-ion membrane transport protein.^8,9 There is additionally evidence that iron in the form of ferritin may serve as a dietary iron source, although this absorption pathway has been less studied.^10,11 Fe²⁺ that has now entered the enterocyte can be utilized for cellular processes, stored, or exported through the basolateral side into the systemic circulation.^8,9 The primary means of iron export into the systemic circulation is through ferroportin (FPN), a transmembrane iron transporter, expressed on the basolateral side of enterocytes.¹² Upon release into the systemic circulation, Fe²⁺ is quickly oxidized to Fe³⁺ by ferroxidases, such as ceruloplasmin and hephaestin, allowing it to be carried by transferrin, a 76 kDa glycoprotein that binds Fe³⁺ in a 2:1 ratio and serves as a major iron transport protein (Figure 1). Iron-laden transferrin subsequently circulates within the body where it may be taken up by cells within different organs to meet their iron requirements.^8,9 Importantly, transferrin is able to cross the blood–brain barrier through a transcytosis pathway, making it a potential iron source for cells within the GBM TME.¹³ Several groups have demonstrated transferrin-mediated iron uptake occurring in GBM¹⁴, and others have even investigated using transferrin as a vehicle to facilitate transport across the blood–brain barrier with potential for delivery of antitherapeutic agents.^15,16 Ferritin may potentially similarly be able to be used as a delivery vehicle given its ability to encapsulate various drugs and be taken up by GBM cells.^17,18

Figure 1.

Overview of gastrointestinal iron uptake and iron uptake by glioblastoma cells. Dietary iron is absorbed in the duodenum and proximal jejunum through transporters on the apical membrane of enterocytes. This absorbed iron can then be used by the enterocyte, stored in ferritin, or exported into the circulation by ferroportin, where it is subsequently oxidized by hephaestin and carried systemically by transferrin. Glioblastoma cells can take up iron from several sources including transferrin-bound iron, non-transferrin-bound iron, and through ferritin.

At the cellular level, transferrin-receptor mediated endocytosis represents a major iron uptake pathway. Upon binding to transferrin receptor, iron-laden transferrin is internalized through a clathrin-mediated endocytic pathway. Subsequent acidification of the endosome leads to dissociation of ferric iron and recycling of transferrin and transferrin receptor back to the cell membrane. Iron liberated from transferrin is reduced in the endosome from Fe³⁺ to Fe²⁺ by enzymes that possess ferric reductase capacity, such as STEAP3. Reduced iron is subsequently transported into the cytosol through metal–ion transporters including DMT1.^19,20 Once in the cytosol, Fe²⁺, referred to as the labile iron pool, is available to participate in iron-dependent cellular processes. Additional sources of iron may also be available to GBM cells. For example, we recently reported uptake of ferritin by GBM stem cells.¹⁸ The presence of non-transferrin-bound iron (NTBI) has also been reported in brain extracellular fluids, potentially serving as another source of iron for GBM cells (Figure 1).²¹

Iron and Neoplastic Cells

Stem-Like Cells

Profiling of GBM tumors has revealed a cellular hierarchy with glioma stem cell-like cells (GSCs) at the top of the hierarchy. GSCs are of significant importance to patient outcomes, as they possess the capacity for self-renewal, contribute to treatment resistance, and contribute to disease progression.²² GSCs are thought to require higher amounts of iron than non-stem cells owing to their capacity for self-renewal. Indeed, Schonberg et al. demonstrated that GSCs had significantly higher uptake of radiolabeled iron compared to matched glioma non-stem cells.¹⁴ The authors also showed that knocking down expression of ferritin, a protein essential to safe iron storage, significantly reduced the ability of GSCs to proliferate. These results suggest that the ability to acquire higher amounts of iron, and more importantly, safely store that iron is a feature that distinguishes GSCs from non-stem tumor cells. Consistent with this, single-cell magnetophoresis studies comparing GSCs and non-stem tumor cells revealed that non-stem tumor cells consistently displayed increased magnetic susceptibility likely as a result of increased free iron, which has been postulated to be due to the enhanced ability of GSCs to sequester iron in ferritin.²³ GSCs may also possess the capability to manipulate their environment to secure access to iron. Tabu et al. recently reviewed a network where they discuss the potential for GSCs to recruit macrophages into the GBM where they subsequently phagocytose hemorrhaged erythrocytes and serve as iron sources for the GSCs.²⁴

Vo et al. examined how exogenously administered iron impacted GBM stem cells and discovered a layer of genetic complexity by finding that iron supplementation resulted in increased proliferation of mesenchymal stem cells but not proneural stem cells.²⁵ Interestingly, the authors also discovered that ferric iron (Fe³⁺) stabilized hypoxia-inducible factor 1-alpha (HIF1α) and pSrc in mesenchymal stem cells but not proneural stem cells.²⁵ These results highlight that responses to iron may be more complex than initially appreciated and likely depend on underlying genomic and epigenomic profiles of responding stem cells. Developing tools to predict GSC responses to iron may be essential when developing personalized iron-modulating therapies. Furthermore, while the effect of iron on proliferation in GBM stem cells has been studied, there is still much to learn about iron’s role in other aspects of GSC biology.

Non-Stem Tumor Cells

As an essential cofactor in numerous enzymes involved in replication, including DNA polymerase and ribonucleotide reductase, it is no surprise that iron has been implicated in cell cycle progression and proliferation among GBM cells.⁶ Taking advantage of the increased iron demand, several studies have demonstrated potent cytotoxicity of iron chelators toward GBM cells in vitro.^26,27 Despite years of positive in vitro data, there have been a lack of clinical trials examining iron chelation therapy in GBM. The discrepancy between promising in vitro data and a lack of human studies reporting improved survival may arise due to undesired off-target effects of iron chelation. Given the intricate link between reduced cellular iron and increased hypoxic signaling, we, and others, speculate that protumoral hypoxic responses may offset the beneficial effects of chelation. Indeed several studies have found increased expression of HIF-1α or its downstream targets such as VEGF in GBM cells exposed to iron chelators.^28,29 Many in vitro studies also fail to recapitulate the reduced oxygen tension already often found in GBM tumor cores. To illustrate the importance of this, the cytotoxic effects of at least one chelator, defersairox, have been reported to be reversed in low oxygen conditions.³⁰

Despite the necessity of iron for GBM cell proliferation, excess iron induces cytotoxicity in GBM cells. Fernández-Acosta et al. found that iron oxide nanoparticles induced toxicity in U87MG and U373MG cells at concentrations as low as 500 ng/mL.³¹ Iron oxide nanoparticles also appear to be able to induce ferroptosis of GBM cells, likely due to their ability to increase the labile iron pool and thus enhance generation of reactive oxygen species.³² Increased labile iron pool, while potentially promoting proliferation, may also offer antitumoral effects. For example, GBM cells with higher labile iron pools have been reported to exhibit greater radiosensitivity.³³ Taking advantage of this, Guerra et al. show that iron oxide nanoparticles can radiosensitize several GBM cell lines.³⁴ Increased labile iron pool may also leave GBM cells susceptible to ferroptosis, discussed in more detail later.

Iron also likely plays additional roles in neoplastic cell biology beyond cell proliferation that can influence disease progression. We reported recently that increased cellular iron reduces the ability of GBM cells to migrate in vitro, thereby implicating iron in an additional essential aspect of GBM cell biology.³⁵ Decreased cellular iron content is also associated with increased HIF1α signaling²⁹ and VEGF signaling,²⁸ which are known to be pathologic in GBM progression. Increased VEGF signaling mediates angiogenesis which allows enhanced delivery of nutrients and growth factors to GBMs, supporting their growth. For this reason, increased angiogenesis has been thought to contribute to poor outcomes in GBM and may represent a pathway by which iron may influence overall tumor phenotype.³⁶

Ferroptosis and GBM

Ferroptosis is a distinct form of cell death characterized by widespread lipid peroxidation. Ferroptosis and iron metabolism are intrinsically linked as iron is required for this recently discovered form of cell death.^5,37,38 Two well-described approaches to induce ferroptosis are through the inhibition of reduced glutathione (GSH) production or reduction in the ability of glutathione peroxidase 4 (GPX4) to limit lipid peroxidation.^5,37 Inhibiting production of GSH by limiting import of cystine, a precursor in the GSH synthesis pathway, compromises the ability of cells to safely manage reactive oxygen species (ROS) that would have otherwise been reduced by GSH, leading to excessive lipid peroxidation. A well-described approach to inhibiting cystine import and subsequently compromising GSH production is through the inhibition of System xc⁻, an amino acid antiporter that exchanges intracellular glutamate for extracellular cystine. Likewise, inhibiting GPX4 prevents the reduction of lipid peroxides and allows for their accumulation—eventually leading to cell death.^5,37 Several alternate pathways associated with ferroptosis induction have recently been reported and are reviewed in detail elsewhere.¹

Here, we discuss the role of iron in inducing ferroptosis specifically in the context of GBM. Iron is essential for the induction of ferroptosis as evidenced by the ability iron chelators such as deferoxamine to serve as potent ferroptosis inhibitors, including in GBM cells.³⁹ The inverse also appears to be true whereby elevated cellular labile iron pools, either through enhanced iron uptake or reduced capacity to sequester iron, promote susceptibility to ferroptosis. Demonstrating increased susceptibility to ferroptosis in cells with higher iron uptake, Tong et al. reported that overexpression of transferrin receptor 2 (TFR2), a protein responsible for cellular iron uptake, was associated with enhanced susceptibility to ferroptosis in U87 and U251 cells.³⁹ Another study reported that ferroptosis contributed to cytotoxicity exerted by iron oxide nanoparticles in GBM cell lines.³¹ Deficits in cellular capacity to safely store iron in a nonreactive state produces similar effects. As described earlier in this article, ferritin plays an essential role in safely storing iron in its redox-inactive form. Therefore, pathways that degrade ferritin and thereby compromise the ability of GBM cells to safely store iron may allow for increased ROS generation and facilitation of ferroptosis. Ferritinophagy is a well-characterized process where nuclear receptor coactivator 4 (NCOA4) mediates autophagic degradation of ferritin and mediates the release of its stored iron.⁴⁰ Inhibition of NCOA4 has been reported to inhibit ferroptosis likely due to reducing the labile iron pool.⁴¹ Consistent with this, downregulating coat complex subunit zeta 1 (COPZ1), an inhibitor of NCOA4, increased NCOA4-mediated ferritinophagy, and enhanced susceptibility to ferroptosis in several GBM cell lines.⁴² Inhibition of tripartite motif-containing protein 7 (TRIM7), another negative regulator of NCOA4, demonstrated similar findings where knockout of TRIM7 in A172 or U87 GBM cells resulted in increased NCOA4-mediated ferritinophagy, elevated labile iron pools, and ferroptosis.⁴³

Ferroptosis, and therefore iron, may also play an important role in augmenting existing standard of care treatment modalities available for GBM. Tong et al. reported that upregulation of TFR2 enhanced the cytotoxic effect of temozolomide on U87 and U251 cells and that addition of the ferroptosis inhibitors ferrostatin-1 or deferoxamine attenuated this enhanced susceptibility. These findings are consistent with temozolomide exerting toxicity at least in part through ferroptosis.³⁹ As increased iron content also lends susceptibility to ferroptosis, modulation of iron content prior to temozolomide treatment may be an approach worth investigating. Radiation may similarly exert cytotoxicity in glioma at least in part through ferroptosis.⁴⁴ In their study, Ye et al. found that addition of a ferroptosis inhibitor significantly reduced radiation-induced ROS generation in human glioma slice culture models. Their study also reported that ferroptosis inducers were able to synergize with radiation to induce enhanced ROS generation in ex vivo glioma slice cultures.⁴⁴

Iron and Immune Cells

Iron and Macrophages

Macrophages play a large role in iron homeostasis of the TME. Not only do these cells make up a large portion of the TME, with some studies estimating compositions as high as 40% of tumor volume,⁴⁵ but they also have considerable capacity to take up, store, and release iron.⁴⁶ Macrophages in the GBM TME are thought to lie along a spectrum of phenotypes with inflammatory macrophages (M1) on one end and anti-inflammatory macrophages (M2) on the other end. Recent studies have raised concerns with the validity of a binary M1/M2 model for macrophage polarization states, especially in vivo, and several studies argue that true phenotypes lie along a spectrum from proinflammatory to anti-inflammatory and that some states can even simultaneously possess features from both ends of the spectrum depending on context.^47,48 Nonetheless, macrophage polarization is known to alter iron homeostasis with proinflammatory macrophages obtaining an iron-sequestering phenotype, originally described as a means of preventing pathogens from acquiring iron, while anti-inflammatory macrophages obtain an iron-releasing phenotype, originally described as aiding in tissue repair.⁴⁶

While macrophage polarization is known to alter cellular iron homeostasis, the role iron directly plays in determining cellular polarization is less well characterized. Studies examining iron’s role in determining macrophage polarization in vitro and have found varying results which likely arises due to differences in the cells being studied (RAW264.7 vs. THP-1 vs. mouse bone-marrow derived macrophages (BMDMs) vs. primary human macrophages) as well as differences in their starting polarization states at the time of iron treatment.⁴⁹ In LPS-stimulated primary human macrophages, iron chelation appeared to reduce expression of inflammatory mediators, suggesting that iron promotes proinflammatory phenotypes in these cells.⁵⁰ Similarly, in THP-1 cells, addition of iron promoted a proinflammatory phenotype by increasing secretion of IL-6,⁵¹ although it has also been reported that iron treatment of LPS and IFN-γ-stimulated THP-1-reduced expression of inflammatory markers.⁵² Studies in mouse BMDMs have found that addition of iron to resting cells in vitro promoted inflammatory cytokine secretion and that iron overload in vivo led to M1 polarization in hepatic macrophages.⁵³ Another study found that iron treatment of M2 polarized macrophages resulted in a decrease of M2 gene expression and upregulation of M1 gene expression, further supporting a proinflammatory role of iron.⁵⁴ Examining iron’s impact on LPS+IFNγ-treated BMDMs, however, found reduced inflammatory marker production.^52,55 Iron supplementation in alveolar macrophages was associated with increased NF-κΒ activity, consistent with promotion of M1 polarization.⁵⁶ Iron treatment of RAW264.7 cells, a mouse macrophage cell line, has similarly revealed a proinflammatory role of iron treatment in resting cells,^57,58 and inhibition of inflammatory cytokine production in M1 polarized cells. Iron treatment in another study of RAW264.7 cells, however, interestingly failed to induce production of proinflammatory production in resting cells.⁵⁹ Altogether, there is significant variability and inconsistency in the literature surrounding the role of iron in macrophage polarization, likely due to variability in initial polarization state, origin of the macrophages, and length of treatments. Nonetheless, the literature does lean in the direction of an M1-inducing role of iron in resting macrophages (Figure 2) and reduction of inflammatory responses in macrophages that are already M1 polarized.

Figure 2.

Antitumoral effects of iron in major immune cell types within the glioblastoma tumor microenvironment. Examination of the role of iron in modulating effector functions of the major immune cell types within glioblastoma: macrophages, microglia, neutrophils, and T cells, have revealed several effects that diminish protumoral characteristics or enhance antitumoral characteristics. These include inducing proinflammatory cytokine production and M1 polarization of macrophages and microglia, promoting proinflammatory cytokine secretion, promotion of proliferation, and upregulation of costimulatory molecules in T cells, and inhibition of neutrophil degranulation and NETosis along with increased ROS generation.

While several studies have attempted to characterize iron’s influence on macrophage polarization in vitro, few in vivo studies examine iron’s impact on macrophage polarization in the context of GBM. In other malignancies, however, iron appears to promote inflammatory phenotypes. Costa da Silva et al. reported that iron in the form of iron oxide nanoparticles or from lysed red blood cells could promote antitumor activity in tumor-associated macrophages in lung cancer models.⁶⁰ Similar findings regarding promotion of M1 macrophage phenotypes by iron oxide nanoparticles have been reported in breast cancer and metastatic lung cancer mouse models.⁵⁷ Given the in vitro data supporting an inflammatory role for iron in resting macrophages as well as in vivo data from other cancer models supporting induction of antitumor immunity, further investigating iron as an immune-activating agent in GBM tumor-associated macrophages is of interest.

Iron and Microglia

Much like bone marrow-derived macrophages, microglia have been described to obtain phenotypes along a spectrum from proinflammatory M1 microglia to anti-inflammatory M2 microglia. The role iron plays in determining microglial phenotype appears to be less well characterized than that of bone marrow-derived macrophages, possibly due to increased difficulty in culturing primary microglial cells.⁶¹

Iron treatment of an mouse microglial cell lines or primary microglial cultures have been reported to increase nuclear NF-κB and secretion of TNFα and IL-6, respectively, suggesting a proinflammatory role for iron.^62,63 Wang et al. examined how iron loading prior to LPS stimulation impacted microglial polarization and found that iron loading of primary rat microglial cultures prior to LPS stimulation resulted in significantly higher expression of M1 markers IL-1β and TNF-α.⁶⁴ Interestingly, however, iron treatment alone did not induce secretion of IL-1β or TNF-α in their study. In another model of LPS-stimulated microglia, Yauger et al. found that iron induced reactive oxygen species but did not impact microglial polarization in BV2 mouse microglial cultures and primary rat microglia.⁶⁵ The study by Yauger et al. differed from Wang et al.’s study with regard to whether the iron loading occurred before or after LPS stimulation, which may partially explain the discrepancy.

Studies examining iron depletion further support an inflammatory role for iron within microglia. Iron chelation was able to inhibit hypoxia-induced secretion of TNF-α and IL-1β production in primary rat microglia.⁶⁶ In a mouse model of Alzheimer’s disease chelation with deferoxamine resulted in lower levels of inflammatory markers such as IL-1β and iNOS and higher levels of anti-inflammatory markers Arg1 and Ym1 in hippocampal microglia.⁶⁷ Similarly, another study examining the use of iron chelators in a mouse model of intracranial hemorrhage found that microglia in mice treated with iron chelation had reduced expression of the M1 marker CD16/32 and increased expression of M2 marker CD206.⁶⁸ While the above studies seem to largely suggest an M1-promoting role for iron in microglia, it is worth noting that conflicting reports have also been published. A recent study by Kenkhuis et al. examined iron loading of human iPSC-derived microglia prior to exposure to inflammatory stimuli such as IFNγ and amyloid β.⁶⁹ Interestingly, the study found that iron treatment reduced both pro- and anti-inflammatory responses as assessed by transcriptomic profiling.⁶⁹ The discrepancy may partially be explained due to differences between species (murine vs. human).

Microglial polarization has been associated with GBM patient outcomes.⁷⁰ As M2 microglia are characterized as secreting anti-inflammatory cytokines, they are thought to partially contribute to the immunosuppression observed in GBM tumors. Quantitative immunofluorescence studies of GBM specimens have found that increased presence of M2 microglia are correlated with poorer patient survival.⁷¹ Transcriptomic analyses have also suggested an association between M2 polarized microglia and poorer outcomes in GBM.⁷⁰ Given the pathologic association of expression of anti-inflammatory markers in microglia with patient survival, converting microglia into proinflammatory phenotypes has been hypothesized to be an approach at inducing antitumor immune responses.⁷² With the relative abundance of literature suggesting a proinflammatory role for iron in microglia (Figure 2), iron-based approaches at modulating microglial polarization may be promising.

Iron and Neutrophils

The existing literature points toward a pathologic role for neutrophils in GBM progression. Elevated neutrophil counts in GBM patients prior to treatment initiation are associated with poorer outcomes.⁷³ Examining neutrophil to lymphocyte ratios have found that increased neutrophil counts relative to lymphocyte counts also correlate with higher grade gliomas and poorer outcomes.^74–77 Further supporting a directly pathogenic role for neutrophils in gliomas, neutrophil infiltration has been associated with increased glioma grade in histological analysis.⁷⁸ Neutrophil activation within GBMs is also associated with shorter time to progression, suggesting that the effector functions of neutrophils may be protumoral.⁷⁹ Consistent with this, increased levels of circulating degranulated neutrophils are associated with immunosuppression in an analysis of GBM patient samples.⁸⁰ Providing some mechanistic insight, one study found that neutrophils promote the conversion of non-stem GBM cells to GBM stem cells.⁸¹ Later studies have implicated neutrophils in promoting malignant phenotypes in gliomas by inducing migration of glioma-initiating cells and contributing to anti-VEGF treatment resistance.⁸²

The role iron plays in the protumorigenic activities of neutrophils in GBM is largely unstudied. Several advances have been made, however, in studying iron’s role in neutrophil production, migration, and effector functions outside the context of GBM that are discussed in this section. At the bone marrow level, neutrophil production appears to be particularly dependent on iron compared to other leukocytes. Hypoferremia in mice, induced by hepcidin, results in reduced neutrophil production despite undisturbed production of monocytes.⁸³ Consistent with this, a study by Papadaki et al. found that anemia of chronic disease was the most common anemia found in individuals with chronic idiopathic neutropenia, although there are relatively few reports regarding associations between absolute iron deficiency anemia and neutropenia.⁸⁴ Nonetheless existing reports support a role for iron in neutrophil production at the bone marrow level.

Beyond production, iron likely also plays a role in the ability of neutrophils to migrate into sites of tissue injury. Garcia et al. found that neutrophil uptake of iron oxide nanoparticles appeared to inhibit the ability of neutrophils to migrate by altering expression of proteins needed for adhesion to ICAM-1, a process that is essential to effective neutrophil trans-endothelial chemotaxis.⁸⁵ Further demonstrating an inhibitory role for iron in neutrophil migration, neutrophils isolated from individuals with iron overload due to thalassemia major have significantly reduced neutrophil migratory capacity compared to neutrophils isolated from healthy controls.⁸⁶ These studies have significant implications as GBM is known to have rich neutrophil content due to trans-endothelial migration of circulating neutrophils into the TME. A role for iron in neutrophil migration gives rise to the possibility of targeting iron metabolism to modify neutrophil content in the GBM TME.

Neutrophil activation results in alterations to the microenvironment that are known to impact the progression of GBM. Examining circulating neutrophils in the blood of GBM patients revealed that increased activation of circulating neutrophils was associated with a shorter time to progression, suggesting that activation of neutrophils contributes to disease progression.⁸³ Neutrophil activation is often characterized by neutrophil degranulation, respiratory burst, and neutrophil extracellular trap formation (referred to as NETosis); therefore, we review iron’s impact on these three important neutrophil effector functions here.

Neutrophil degranulation is characterized by the secretion of granules containing cytotoxic proteins into the extracellular space or into phagosomes containing internalized pathogens. Most work has focused on studying the antimicrobial role of neutrophil degranulation in the context of infections. In GBM, however, neutrophil degranulation has been associated with increased peripheral immunosuppression.⁸⁰ Conflicting roles for iron in neutrophil degranulation have been reported. A study examining iron oxide nanoparticles and neutrophil function reported a dose-dependent inhibition of degranulation, potentially mediated by reduced IL-8 signaling.⁸⁵ Hisashi et al., however, found that iron deficiency in rats compromised neutrophil myeloperoxidase activity, an important protein found in neutrophil granules.⁸⁷ These discrepancies may partially be explained by differences between methods used to assess degranulation (expression of cell surface markers indicating degranulation vs. myeloperoxidase activity). Alternatively, these discrepant studies could indicate that iron levels need to be in a specific range for optimal neutrophil degranulation.

Neutrophil oxidative burst is characterized by the rapid production and release of reactive oxygen species by NADPH Oxidase and plays important roles in neutrophil-mediated anti-microbial responses. Neutrophil oxidative burst likely also plays a role in GBM progression given the signaling cascades induced in multiple cell types by ROS.⁸⁸ T cells, for example, are particularly sensitive and are inhibited by ROS in terms of proliferation, activation, and cytokine production.^89,90 Iron appears to largely promote oxidative burst capacity in neutrophils. Hassan et al. found that oxidative burst capacity of neutrophils in children with iron deficiency anemia was reduced.⁹¹ Importantly, this phenomenon has also been reported in adult patients where both iron deficiency anemia as well as anemia of chronic disease were associated with reduced oxidative burst capacity of neutrophils.⁹² On the other hand, neutrophils from mouse models of iron overload as well as from patients with hereditary iron overload disorders appear to have enhanced oxidative burst capacity, supporting an essential and contributory role for iron in efficient neutrophil oxidative burst.⁹³

NETosis refers to the release of structures composed of chromatin and proteins from neutrophils, referred to as neutrophil extra cellular traps (NETs) and was originally discovered in the context of neutrophil-mediated antimicrobial activity. NETosis, however, can also occur within the GBM TME and has been implicated in tumor progression.⁹⁴ Several studies have found that increased neutrophil iron content inhibits NETosis.^95–97 Studies examining iron chelation and NETosis, however, have been conflicting. Kono et al. found that iron chelation with deferoxamine inhibited NET release;^98,99 however, Kuźmicka and Volager et al. found that chelation actually induced NETosis.^100,101 These discrepancies may partially be explained by the utilization of different chelators, which may induce different degrees of chelation or promote non-chelation related effects, as well as different sources of neutrophils. Although existing studies seem to lean toward an inhibitory role for supplemented iron in NETosis, given the conflicting reports, additional studies are required. Nonetheless, it is evident that iron alters important neutrophil effector functions that are known to impact GBM tumor progression including degranulation, oxidative burst, and neutrophil NET production (Figure 2).

Iron and T Cells

T cells are essential players in mediating anti-tumor immunity in a large variety of cancers. In GBM, however, T cells fail to induce robust antitumor immune responses and instead enter into a senescent, anergic state.¹⁰² As evidence for the necessity of robust T-cell responses for efficient antitumor immunity, one study found that overall survival was significantly shorter among GBM patients expressing higher levels of markers associated T-cell senescence such as CD57 or lower levels of costimulatory marker CD28.¹⁰³ As a result of defective T-cell immunity, several T-cell dependent immunotherapies that have found success in other malignancies have failed to show as much promise in GBM.¹⁰⁴

Iron plays an important role in T-cell development peripherally. Several early studies established the necessity for iron uptake through the transferrin receptor for normal T-cell proliferation and development.^105,106 Consistent with this, iron deficiency has been reported to cause a reduction in peripheral T cells and atrophy of the thymus, the site of T-cell maturation.^107,108 Reduced peripheral T-cell counts in iron deficiency may be due to impaired thymocyte development as blocking transferrin-receptor-mediated iron uptake inhibits thymocyte proliferation and differentiation.¹⁰⁹ After thymocytes differentiate into T cells, T cells retain the ability to further proliferate which is also iron dependent. Iron deprivation leads to severely reduced CD4+ T-cell proliferation and cell cycle progression in vitro, which may also contribute to reduced circulating T-cell populations observed in iron deficiency.¹¹⁰

In addition to T-cell proliferation and development, iron plays a role in T-cell activation and effector functions (Figure 2). Iron positively regulates IL-2R signaling and mitochondrial function and is necessary for CD4+ T-cell activation.¹¹⁰ Iron deprivation during T-cell activation results in significantly reduced mitochondrial ROS which are needed for T-cell activation, and these effects can be rescued by administration of exogenous iron.^110,111 Another study found that iron deprivation induced by hepcidin disturbed CD8+ T-cell function and mitochondrial activity to the extent that hypoferrimic mice had impaired responses to immunizations.¹¹² Expression of CD28, a costimulatory receptor required for T-cell activation and survival, is also reduced upon iron deficiency.¹¹³ Further supporting a T-cell activating role for iron, Wang et al. found that iron-promoted proinflammatory cytokine secretion in T cells by altering the stability of PRCB1, an RNA-binding protein.¹¹⁴

While iron appears to be essential for T-cell activation and effector functions, iron overload may also have negative effects on effector functions. In a model of chronic infection, iron loading prior to infection appeared to reduce Th1 responses induced by CD4⁺ T cells by increasing expression of TIM-3, a negative immune checkpoint regulator.¹¹⁵ On the other hand, examination of T cells in iron overload secondary to β-thalessemia has reported reduced CD4⁺ and increased CD8⁺ T cells.¹¹⁶ Patients with iron overload secondary to hereditary hemochromatosis are also reported to exhibit reduced CD8 effector memory T cells.¹¹⁷ In vitro examinations of the effect of iron on T cells have found that iron treatment reduces expression of CD4 and CD2, possibly explaining some of the altered CD4/CD8 ratios observed in iron-overload patients.¹¹⁸

The aforementioned studies suggest that iron is necessary for T-cell activation and function but is potentially detrimental in excess. Dysregulation of iron homeostasis in GBM may contribute to the defective T-cell responses observed in these tumors. Strategies to favorably modulate T-cell iron metabolism may be promising at inducing antitumor immunity in the context of a highly immunosuppressive GBM microenvironment.

Clinical and Preclinical Associations of Iron and GBM Outcomes

There is evidence to suggest that iron may be dysregulated in GBM. A study examining ferritin in GBM found that GBM patients have increased serum ferritin levels and strong staining for ferritin in resected tumors, although the study did not examine associations with outcome or elemental iron content.¹¹⁹ GBMs are also reported to express high levels of transferrin receptor, providing a plausible mechanism by which they could accrue additional iron, although elemental iron measurements are needed.¹⁴ Given that iron appears to exert both protumoral and antitumoral effects when examined in the context of individual cell types found within the GBM TME, it is unclear how iron may contribute to overall patient outcomes. Patient outcomes are a product of effects brought about by numerous different cell types acting together and are difficult to infer from studies examining cells in isolation. For example, while iron may promote proliferation in neoplastic cells, it may inhibit their migration or potentially stimulate antitumor T-cell responses that outweigh the negative contribution of increased proliferation.

Preclinical studies examining iron and GBM outcomes appear to suggest an antitumoral role for exogenously administered iron (Table 1). Hadjipanayis et al. reported that iron oxide nanoparticle (IONP) administration using convection enhanced delivery resulted in significantly prolonged survival in an orthotopic U87ΔEGFRvIII model of GBM.¹²⁰ Similarly, Wu et al. found that administration of iron oxide nanoparticles in an immunocompetent mouse model significantly enhanced the survival benefit offered by radiation, but that even in the absence of radiation, survival was prolonged, although falling just short of significance.¹²¹ Interestingly, that study found that MDSCs within the GBM TME acquired the IONPs which subsequently contributed to MDSC repolarization from an immunosuppressive to a proinflammatory phenotype. Further supporting an immunomodulatory role for IONPs, less benefit was observed when Wu et al. administered IONPs to an immunodeficient U87 mouse model of GBM, which contrasts with the study¹²⁰ by Hadjipanayis et al. although that study utilized convection enhanced delivery and EGFRvIII targeting.¹²¹ Another study by Chiaraelli et al. similarly found that IONP administration alone in an immunodeficient GBM model did not offer a meaningful increase in survival, but that it did significantly enhance survival when combined with radiation.¹²² Thus, in the absence of specific targeting to cancer cells or enhanced delivery techniques, iron may provide survival benefits either through immunomodulation or enhancement of existing treatment modalities such as radiation.

Table 1.

Studies Examining Iron Supplementation in Animal Models of Glioblastoma

Reference	Year	Cohort size	Glioblastoma model	Iron compound and dose	Median survival (treated vs. control)
Hadjipanayis et al.120	2010	N = 10 nude mice per group	U87ΔEGFRvIII (orthotopic xenograft)	2 µg amphiphilic triblock copolymer coated iron oxide nanoparticles, convection enhanced delivery	16 days vs. 11 days (P < .001)
Ohtake et al.129	2016	N = 6 nude mice per group	U251 (heterotopic left leg injection)	μ-oxo N,N’-bis(salicylidene)ethylenediamine iron nanoparticles, subcutaneous injection into area surrounding tumor	Survival not provided but iron nanoparticle treated tumors were smaller than control tumors (P < .001)
Wu et al.121	2019	N = 5 C57BL/6 mice per group	CT-2A (orthotopic syngeneic)	25 µg polyethylenimine coated zinc-doped iron oxide nanoparticles, intracranial injection	30 days vs. 25 days (P = .08)
Wu et al.121	2019	N = 5 nude mice per group	U87 MG (orthotopic)	25 µg polyethylenimine coated zinc-doped iron oxide nanoparticles, intracranial injection	35 days vs. 34 days (P = 0.27)
Chiarelli et al.122	2022	N = 17 (untreated), N = 14 (treated) Nude mice	GBM6 (orthotopic xenograft)	8.5 mg/kg iron oxide nanoparticle with chitosan-PEG copolymer coating, intravenous injection	29 days vs. 28 days (P > .1)

Clinical studies directly examining the role of iron in patient outcomes have unfortunately largely been unreported. The majority of published studies examine iron indirectly by looking at genes and proteins associated with iron metabolism. Studies examining the role of anemia in GBM appear to suggest that anemic patients have poorer outcomes compared to nonanemic patients (Table 2). While many of these studies were unable to specifically determine the underlying etiology of anemia, the most common etiology of anemia in cancer patients is known to be functional iron deficiency.¹²³ Functional iron deficiency is thought to be brought about by secretion of inflammatory cytokines from neoplastic cells as well as inflammatory responses to chemotherapy.¹²³ Production of IL-6, among other inflammatory chemokines, induces retention of iron in macrophages, depriving other cells of iron. Without sufficient availability of iron, hemoglobin synthesis is impaired and the oxygen-delivering capacity of blood is compromised.¹²⁴ While the existing literature seems to suggest that anemia is associated with poor outcomes in GBM patients, the underlying mechanisms behind this association have not been identified. A highly plausible leading theory is that iron deficiency anemia may exacerbate tumor hypoxia which may contribute to suppressed immune response, increased angiogenesis, and enhanced proliferation.¹²⁵

Table 2.

Summary of Studies Examining Association of Anemia and Glioblastoma Patient Outcomes

Reference	Year	Size of cohort	Definition of anemia	Anemic vs. non-anemic outcomes
Lutterbach et al.130	1999	N = 149 (anemic: 48, non-anemic: 101)	Hb < 13 g/dL for males; Hb < 12 g/dL for females (measured on admission for neurosurgery)	RR: 0.91 (0.76–1.09), P = .32
Lutterbach et al.131	2003	N = 318 (anemic: 89, non-anemic: 229)	Hb < 13 g/dL for males; Hb < 12 g/dL for females (measured on admission for neurosurgery)	HR: 1.47 (1.13–1.90), P = .004
Odrazka et al.132	2003	N = 85 (anemic: 30, non-anemic: 55)	Hemoglobin ≤ 12 g/dL for both sexes (measured prior to the start of radiotherapy)	RR: 2.17 (1.33–3.54), P = .002
Ausili et al.133	2011	N = 43 (anemic: 15, non-anemic: 28)	Hemoglobin ≤ 12 g/dL for both sexes (measured prior to the start of radiotherapy)	HR: 2.44 (1.19–5.00), P = .010
Kaisman-Elbaz et al.125	2020	N = 112 (anemic: 22, non-anemic: 90)	Hb < 14 g/dL for males; Hb < 12 g/dL for females (measured 1–3 days prior to surgery)	HR: 1.79 (1.06–2.99), P = .031
Shenoy et al. (Under review)	2023	N = 1346 (anemic: 509, non-anemic: 837)	Hb < 13 g/dL for males; Hb < 12 g/dL for females (median of all Hb measurements within 6 months of diagnosis)	HR (males): 1.36 (1.09–1.69), P = .0056 HR (females): 1.04 (0.82–1.31), P = .78

Numerous iron supplements have long been available on the market to treat anemia (Table 3). While oral iron compounds are used far more frequently used owing to their easy access and administration, there is concern that gastrointestinal iron absorption in cancer patients may be significantly compromised due to inflammatory signaling that may limit gut reabsorption. For this reason, intravenous iron supplements have generated more interest in iron-replacement therapy in the setting of cancer. Iron oxide nanoparticles have particularly garnered interest because they offer the ability to simultaneously act as drug-delivery vehicles, magnetic resonance contrast agents, and hyperthermia sensitizers.¹²⁶ Iron modulation can also occur through reduction of iron content by approved iron chelators (Table 3). Unfortunately, there has been a lack of published positive trials examining iron chelation despite many years of data suggesting an anti-neoplastic effect of iron chelators in vitro, suggesting that chelation may not have the desired anticancer effects in vivo as it does in vitro. Clinical trials examining iron supplementation in GBM are largely non-existent. A study examining the use of IONPs for thermotherapy in GBM patients found that the IONPs were well tolerated, but unfortunately, the study was of insufficient size to draw any conclusions regarding survival.¹²⁷ A follow- up study using IONPs for thermotherapy in recurrent GBM reported prolonged survival, although it lacked randomization and a control arm.¹²⁸ Future studies are needed given the promising yet inconclusive results reported so far, and these newer studies should also examine the interaction of iron and clinically prognostic mutations, such as mutations in the gene for isocitrate dehydrogenase (IDH), given iron’s important role in cellular metabolism. There is currently at least one clinical trial studying the administration of iron in the form of iron oxide nanoparticles underway (NCT04900792), although it may take several years before results are available.

Table 3.

Summary of Commonly Used Iron Supplements and Iron Chelators With Approval Date and Indication Obtained From the Food and Drug Administration (FDA) Satabases (https://www.fda.gov/industry/fda-basics-industry/search-databases)

Compound	Approval indication Summary	FDA approval year	Trade name	Route of administration
Iron Supplements
Ferric carboxymaltose	Iron deficiency anemia in patients who have intolerance to oral iron or have had unsatisfactory response to oral iron or in those who have non-dialysis dependent chronic kidney disease	2013	Injectafer	Intravenous
Ferric citrate	Indicated for the control of serum phosphorus levels in patients with chronic kidney disease on dialysis	2014	Auryxia	Oral
Ferric derisomaltose	Iron deficiency anemia in patients who have intolerance to oral iron or have had unsatisfactory response to oral iron or in those who have non-dialysis dependent chronic kidney disease	2020	Monoferric	Intravenous
Ferric gluconate	Iron deficiency anemia in patients undergoing chronic hemodialysis who are receiving supplemental erythropoetin therapy	1999	Ferrlecit	Intravenous
Ferric maltol	Iron deficiency	2019	Accrufer	Oral
Ferric pyrophosphate	Replacement of iron to maintain hemoglobin in adult patients with hemodialysis-dependent chronic kidney disease	2015	Triferic	Dialysis
Ferrous bisglycinate	-	-	-	Oral
Ferrous fumarate	-	-	-	Oral
Ferrous gluconate	-	-	-	Oral
Ferrous sulfate	-	-	-	Oral
Ferumoxytol	Iron deficiency anemia in chronic kidney disease	2009	FeraHeme	Intravenous
Heme iron polypeptide	-	-	-	Oral
Iron dextran	Iron deficiency in patients who have intolerance to oral iron or an unsatisfactory response to oral iron.	1974	Infed	Intravenous
Iron sucrose	Iron deficiency anemia in chronic kidney disease	2000	Venofer	Intravenous
Polysaccharide-iron complex	-	-	-	Oral
Iron Chelators
Deferasirox	Chronic iron overload	2005	Exjade	Oral
Deferiprone	Tranfusional iron overload due to thalassemia syndromes	2011	Ferriprox	Oral
Deferoxamine	Acute iron intoxication; Chronic iron overload due to transfusion-dependent anemias	1968	Desferal	Intravenous, intramuscular, or subcutaneous

Conclusions

Iron is an essential nutrient that is known to play important roles in cellular metabolism. Numerous studies examining major cell types found within the GBM TME in isolation have demonstrated that iron significantly impacts phenotype. Here, we have outlined those roles and discussed potential implications in terms of GBM tumor progression. Fully understanding iron’s role within the GBM TME will require study of tumor-associated cells in their native environments as studying cells in isolation may fail to recapitulate the unique environment found within the GBM TME. Despite this, the existing literature provides strong evidence that iron plays an important and clinically significant role in GBM.

Funding

This work received support from the Penn State Department of Neurosurgery, Tara Leah Witmer Memorial Fund, and National Institutes of Health grants F30CA250193 to G.S and P01CA245705 to J.R.C. The contents of this paper are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health.

Conflict of interest statement

James R. Connor is a founder and chairman of the board of Siderobioscience LLC, a company founded on patented technology for management of iron deficiency. The product from this company was not used in the studies reported in this manuscript. Ganesh Shenoy reports no conflicts of interest.

Authorship statement

G.S. and J.R.C. conceptualized this review and prepared the manuscript.