Targeting Iron Homeostasis in Acute Kidney Injury

Summary

Iron is an essential metal involved in several major cellular processes required to maintain life. Because of iron’s ability to cause oxidative damage, its transport, metabolism, and storage is strictly controlled in the body, especially in the small intestine, liver, and kidney. Iron plays a major role in acute kidney injury and has been a target for therapeutic intervention. However, the therapies that have been effective in animal models of acute kidney injury have not been successful in human beings. Targeting iron trafficking via ferritin, ferroportin, or hepcidin may offer new insights. This review focuses on the biology of iron, particularly in the kidney, and its implications in acute kidney injury.

Iron is an essential element that is necessary for life. The primary function of iron in the body is the transportation of oxygen. Iron is bound to the heme group in hemoglobin, which allows red bloods cells to supply oxygen to the tissues. Hemoglobin also facilitates the transfer of carbon dioxide from the tissues to the lungs for removal from the body. Myoglobin is another heme protein that uses iron to store and transport oxygen in muscle cells. More than 70% of the body’s iron is contained within hemoglobin and myoglobin. Another 12% can be found in iron storage proteins such as ferritin and transferrin. The remaining 15% of the body’s iron is in heme-containing proteins (such as cytochromes, respiratory burst enzymes, catalase, nitric oxide synthase, myeloperoxidase, and others) that are ubiquitous and imperative to the maintenance of proper functioning of cells and tissues.

Heme is a highly conjugated heterocyclic organic ring with ferrous iron in the center, which is a necessary component of several proteins in the body involved in respiration and energy metabolism. Cytochromes, such as cytochrome c, contain a heme group and are responsible for the production of adenosine triphosphate in the electron transport chain. Other heme-containing enzymes such as cytochrome P450s are monooxygenases that catalyze the metabolism of a wide variety of endogenous and exogenous compounds for either synthesis or detoxification. Although peroxidases such as catalase are protective and catalyze the reduction of hydrogen peroxide to water, cyclooxygenases convert fatty acids into vasoactive prostaglandins. In addition, iron is an essential component of several enzymes involved in the synthesis of collagen and neurotransmitters. Iron also is necessary for proper immune function.

Although iron is essential for life, it is extremely labile in certain forms and, as a result, can be highly reactive and toxic. Iron can be found in two different redox states: the ferrous state (Fe²⁺) and the ferric state (Fe³⁺). The fact that iron can bind six different ligands simultaneously accounts for the high reactivity levels of this metal and its ability to undergo Fenton chemistry. When the body insufficiently reduces oxygen to water, a superoxide radical is formed, but it is subsequently converted to hydrogen peroxide by superoxide dismutases. During the Haber-Weiss reaction, the ferrous form of iron can catalyze a reaction with the seemingly innocuous superoxide and hydrogen peroxide to form the toxic hydroxyl radical. Consequently, the superoxide radical can remove iron from iron-sulfur clusters in proteins, which can lead to certain diseases.¹ On the other hand, the deleterious effects of the hydroxyl radical are well established. Several researchers have studied the ability of the hydroxyl radical to cause oxidative DNA damage, ultimately leading to disease.² Because of the potential reactivity of iron, its movement throughout the body is strictly controlled. Table 1 shows some of the major proteins involved in regulating iron homeostasis.

Table 1.

Some Key Proteins Involved in Iron Homeostasis

Protein	Function
DMT1	The divalent metal transporter 1 imports Fe2+ into the cell
Ferritin	Iron storage protein with heavy (H)- and light (L)-chain subtypes H-ferritin has ferroxidase activity that allows for the safe incorporation of iron into ferritin for storage
Ferroportin	An iron transporter that exports iron from the cell
Hepcidin	A peptide synthesized in the liver that controls iron absorption by regulating ferroportin expression
Hephaestin	A ferroxidase localized mostly in the small intestine that is responsible for converting Fe2+ to Fe3+
NGAL	Neutrophil gelatinase- associated lipocalin (also known as lipocalin-2) sequesters iron to inhibit bacterial growth It is also a biomarker for acute kidney injury
Transferrin	A glycoprotein that can bind up to two atoms of iron in biological fluids. Binding is reversible and the iron pool in transferrin has a high turnover rate

IRON TRAFFICKING IN SPECIFIC ORGANS

The body obtains iron from the diet. Some iron-rich foods include meats, dark green leafy vegetables, beans, dried fruits, and other iron-fortified foods such as cereal. Dietary iron is absorbed in the intestine and metabolized and stored in other tissues, including the liver and the kidneys. Iron is eliminated from the body through blood loss, urinary excretion, and by discarding mucosal and skin cells. The small intestine, particularly the duodenum, is the organ responsible for dietary iron absorption (Fig. 1). The acidic environment of the stomach increases the solubility of the iron acquired from the diet, usually as Fe³⁺, to aid with passage through the digestive tract. In the intestinal lumen, ascorbic acid and ferric reductases such as duodenal cytochrome b, a plasma membrane protein located in the brush-border membrane of enterocytes, reduces ferric iron to the ferrous form (Fe²⁺).³ Iron can enter enterocytes at the apical surface using the proton gradient powered divalent metal transporter-1 (DMT1).⁴ It is thought that iron uses transcytosis or a chaperone to travel from the apical membrane to the basolateral membrane, but the exact mechanism has not yet been elucidated.⁵ The efflux of iron from the basolateral side of the enterocyte is facilitated by ferroportin (FPN)⁶; interestingly, there also is evidence to suggest that FPN may modulate the activity of DMT1 at the apical membrane.⁷ Although it is unclear how iron navigates from the enterocyte through the intestinal mucosa to the plasma, the ferroxidases hephaestin and/or ceruloplasmin are involved, depending on the physiological state.⁸ After being oxidized to the ferric form, iron binds to transferrin in the plasma, which disseminates it to the bone marrow and all tissues and organs. In addition to inorganic iron, heme iron also is absorbed by the intestinal mucosa.⁹ Either the heme carrier protein-1 or the folate transporter is responsible for heme transport across the brush-border membrane of duodenal enterocytes.^10,11 Heme oxygenases (HOs) degrade heme to release free iron into the internal cellular iron pool. When iron is not being used actively, it can be bound to transferrin in the plasma or ferritin in the tissues.

Figure 1.

Intestinal iron absorption. This figure shows iron absorption in an enterocyte. BR, bilirubin; BV, biliverdin; CO, carbon monoxide; DCYTB, duodenal cytochrome b; HCP-1, heme carrier protein-1; HP, hephaestin; Tf, transferrin. Reprinted with permission from Zarjou et al.¹⁰⁰

The liver is primarily responsible for the storage of the body’s iron surplus, but macrophages participate as well. The liver has the capacity to store 10 times the normal amount of iron.¹² Because of its storage potential, there are several opportunities for iron uptake by the liver, including transferrin-bound iron (TBI) uptake, non–transferrin bound iron (NTBI) uptake, and the clearance of iron-containing complexes from the circulation.¹³ Environmental pH seems to be the driving force behind TBI uptake. The neutral pH of the extracellular fluid facilitates the binding of diferric transferrin, a transferrin group carrying two iron molecules, to the transferrin receptor 1 (TFR1).¹⁴ An endosome forms to encapsulate the diferric transferrin bound to TFR1 and the acidic environment of the endosome facilitates the dissociation of the iron atoms from transferrin.¹⁵ The newly minted apotransferrin, transferrin without iron, binds tightly to the TFR1 in the acidic environment of the endosome; however, when the endosome merges with the plasma membrane, apotransferrin readily dissociates from the TFR1 in the neutral extracellular environment. The TFR2 uses a similar mechanism, except the TFR2-mediated uptake becomes more relevant in the liver when serum iron transcends the transferrin binding capacity, similar to that of NTBI uptake.⁹

The exact mechanism of NTBI uptake in the liver has not been elucidated. DMT1 is the preferred method of NTBI in the liver.¹⁶ Because DMT1 has reduced activity at a neutral pH, other methods of iron absorption must exist. One potential candidate is the zinc transporter 14.¹⁷ The liver also can absorb heme iron from hemoglobin when a heme-hemoglobin complex binds to a heme-hemopexin complex.^18,19 Although ferritin and lactoferrin bound to their respective receptors can be internalized by the liver and selected for lysosomal degradation after the release of the iron, ferritin is still a major storage option for iron in the liver. Iron also can be stored in hemosiderin.^9,18 Iron is exported from hepatocytes in the liver using the FPN transporter.²⁰ The rate of iron exportation is controlled by ferroxidase ceruloplasmin.¹⁹ The liver regulates the body’s iron content by producing the peptide hepcidin, which can manipulate both the distribution and the absorption of iron in other tissues. When hepcidin binds to FPN, janus kinase 2 (Jak2) is activated and subsequently phosphorylates one of two adjacent tyrosine residues.²¹ After FPN is phosphorylated it is internalized and undergoes ubiquitination.²²

KIDNEY

There are a limited number of mechanisms for iron to be eliminated from the body and one of those routes is through urinary excretion. Iron is filtered in the glomerulus and reabsorbed in the renal tubules.^13,23 Similar to the liver, the kidney can absorb iron using TBI and NTBI uptake. In fact, the glomerular filtrate contains transferrin, especially in individuals with proximal tubular dysfunction.²⁴ The kidney can absorb transferrin-bound iron using the TFR1⁹; however, the kidney can internalize ferritin when it binds to Scara5 to transport iron.²⁵ Transferrin also can undergo endocytosis in the renal proximal tubules via cubilin and megalin receptors for the uptake of iron.²⁶

In addition to TBI, there are several mechanisms that do not involve transferrin for iron absorption in the kidney. With its expression localized to the cortex, the kidney has the highest reported levels of DMT1 messenger RNA, suggesting a role in the reabsorption of iron from the lumen.²⁷ Most iron reabsorption in the kidney occurs in the loop of Henle and the collecting duct system.²⁸ Research has suggested that iron competes with copper and/or manganese for reabsorption, which is evidence for the involvement of zinc and zinclike iron-like (ZIP) transporters in kidney-mediated iron uptake.²⁹ ZIP8 and ZIP14 are the two zinc transporters that are expressed in proximal tubules.³⁰ There are several other receptors and transporters that are believed to facilitate iron uptake in the kidney including the lactoferrin receptor, neutrophil gelatinase-associated lipocalin (NGAL) receptor, proton-coupled folate transporter, and mucolipin 1 and 2.^31–34 CD163 and pro-low-density lipoprotein receptor-related protein 1 may be responsible for the uptake of heme-hemopexin complexes and haptoglobin-hemopexin complexes in the kidney, respectively.^35–38 Both heavy- and light-chain ferritins are expressed in the kidney to mediate iron storage.^23,39 Proteins expressed in renal tubules potentially handle all iron export from the kidney. These proteins include ferroportin, feline leukemia virus subgroup c receptor, and hephaestin.^23,40,41 Considering the vast machinery available to handle iron in the kidney, it is conceivable to think that altering some of these pathways may prove useful in acute kidney injury.

IRON HANDLING IN ACUTE KIDNEY INJURY

Acute kidney injury (AKI) is associated with increased mortality, length of stay, and health care costs for hospitalized patients.^42,43 The severity of the disease can range from a small decrement in glomerular filtration to complete kidney failure.⁴⁴ AKI is particularly common in critically ill patients with septic shock and in patients undergoing major cardiac surgery.^42,43,45,46 Iron may play a major role in the damage caused during AKI. Several studies have reported an increase in tissue iron content in the kidney after injury. Bleomycin detection was used to determine the increase in kidney tissue catalytic iron after glycerol-induced AKI⁴⁷ and ischemia-reperfusion injury.⁴⁵ Introduction of an iron-deficient diet decreased, but did not eliminate, the increased iron levels in the kidney after ischemia-reperfusion injury.⁴⁵ Iron also has been implicated in the pathogenesis of cisplatin-induced nephrotoxicity.^48,49 In addition to the increased free iron in kidney tissue, there is also more iron excreted in the urine with AKI. This observation is independent of the origin of the injury because it has been observed in ischemia-reperfusion injury,^45,50 transplant ischemia,⁵¹ chemotherapy-induced nephrotoxicity,^48,52 and hemoglobin/myoglobin-induced kidney injury.⁴⁷

Although the source of catalytic iron in AKI has not been shown clearly, several possibilities have been suggested. In 1993, Baliga et al⁴⁵ reported that the excess iron that accumulated in the kidney after AKI may originate from degraded red blood cells. Some researchers have pointed to ferritin as the culprit responsible for iron release,^53–55 whereas others have suggested the iron may originate from mitochondria rich in heme and nonheme iron.^56,57 In a model of AKI initiated by administration of the antibiotic gentamycin, only catalase was protective despite the use of a superoxide anion scavenger, hydroxyl scavengers, and iron chelators.⁵⁶ This lead Ueda et al⁵⁶ in 1993 to conclude that the iron was derived from mitochondria and resulted in oxidant-induced damage. Another potential candidate is cytochrome P450. In the kidney, in vitro and in vivo cisplatin treatment causes a reduction in P450 expression and an increase in bleomycin-detectable iron.⁴⁹ Administration of the P450 inhibitor, piperonyl butoxide, reversed these effects, in addition to conferring functional and histologic protection.⁴⁹ Even though the precise source of the iron is not completely known, the role of reactive oxygen species (ROS) in iron-induced kidney injury has been well documented in animal models.

Multiple studies have shown that iron plays a major role in ROS-induced nephrotoxicity.^45,47,48,58 Oxygen free radicals cause lipid peroxidation–induced renal injury in a rat model of ischemic AKI.⁵⁹ Moreover, in 1988 Paller⁶⁰ observed that free iron and heme iron caused lipid peroxidation in several rat models of AKI, including glycerol-, hemoglobin-, and ischemia-induced renal injury. Kirschner and Fantini⁶¹ and others^45,62 have reported that iron and its production of ROS plays a major role in the pathology of renal ischemia-reperfusion injury. In another form of AKI induced by glycerol, Shah and Walker,⁶³ in 1988, suggested that iron generates toxic hydroxyl radicals.

TARGETING IRON AS THERAPY IN AKI

Iron Removal

If the damage sustained during AKI is initiated by iron, then altering iron trafficking in the kidney should be potentially helpful. Because of the protective effect of iron chelators in several models of kidney injury, it is believed that iron causes the production of the hydroxyl radicals responsible for renal damage.⁶⁴ Several researchers have used iron chelators such as deferoxamine (DFO) to treat AKI. After administration of DFO, glycerol-induced kidney injury was attenuated significantly as determined by reduced blood urea nitrogen and creatinine levels in rats.⁶³ DFO also caused a pronounced decrease in histologic renal damage.⁶³ Paller⁶⁰ showed that DFO was beneficial in hemoglobin- and myolgobin-induced AKI. Likewise, infusion of DFO during reperfusion improved renal function and curtailed lipid peroxidation in a rat model of postischemic renal injury, whereas addition of iron exacerbated the injury.⁶⁵ These studies led to a phase 2 randomized controlled clinical trial to test the efficacy of CRMD-001 (a unique formulation of the iron chelator, deferiprone) in CKD patients with a high risk of developing AKI owing to contrast exposure during coronary angiography (clinicaltrials.gov, NCT01146925). The result of this clinical trial was inconclusive.

Pharmacologic therapy with apotransferrin inhibits oxidative stress, inflammation, and loss of function associated with renal ischemia-reperfusion injury.⁶⁶ Consequently, the L-type calcium channel blocker nifedipine regulates DMT1 activity to eliminate iron overload by increasing the amount of iron excreted in the urine.⁶⁷ In addition to free iron, free circulating heme is increased after AKI.⁶⁸ As a result, increasing circulating levels of hemopexin may decrease kidney damage because it can remove free heme.⁶⁸

As an alternative to using therapies that target the removal of iron, several free radical scavengers have proved beneficial in animal models of AKI. Cisplatin is an efficacious chemotherapy drug, but its use is limited by the significant renal toxicity it causes by direct injury to proximal tubules.⁶⁹ Hydroxyl radical scavengers attenuate the histologic and functional damage in models of cisplatin-induced nephrotoxicity.⁴⁸ To determine the level of protection afforded by hydroxyl radical scavengers, Shah and Walker⁶³ used dimethylthiourea and sodium benzoate and showed significant renal protection. Rats pretreated with dimethylthiourea along with superoxide dismutase, or the xanthine oxidase inhibitor allopurinol, also were protected in a model of AKI induced by renal artery occlusion.⁵⁹ Despite the success of these treatments in animal models of AKI, translation of these therapies to human beings has not been as fruitful as expected.⁷⁰

Because of the minimal therapy options for individuals with AKI, a focus on therapies that could be administered within the first hours of symptom manifestation would be ideal.^71,72 Implementation of such strategies has been hampered by the ability to diagnose AKI early enough before damage already has occurred. Currently, creatinine is used as a biomarker of AKI, but is a poor choice because of the time it takes for serum creatinine levels to increase after AKI.⁷³ Several biomarkers have been developed to address this limitation and this is a subject of great interest. These include NGAL, kidney injury molecule-1, interleukin-18, liver-type fatty acid-binding protein, α-1 microglobulin, N-acetyl-β-D-glucosaminidase, and others.^74,75 Recent studies have validated the use of urinary levels of insulin-like growth factor binding protein-7 and tissue inhibitor of metalloproteinase-2, both inducers of G₁ cell-cycle arrest, in human AKI.⁷⁶

Iron Regulators

Although hemojuvelin, hepicidin, and NGAL are excellent biomarkers of AKI, they potentially can act as therapeutic targets for the treatment of AKI.^77–79 Hemojuvelin regulates hepcidin expression,^80,81 and its soluble form is a potential early biomarker for AKI.⁷⁹ Furin protease inhibitors block the conversion of membrane-bound hemojuvelin to soluble hemojuvelin. Membrane-bound hemojuvelin is associated with reducing iron content in the kidney, hepcidin secretion, and ferroportin degradation in AKI.⁷⁹ As a result, increasing the expression levels of membrane-bound hemojuvelin may be a potential therapeutic option for AKI.

Urine hepcidin levels may be useful as a predictive biomarker of AKI, especially after cardiopulmonary bypass.⁸² A clinical study in patients undergoing cardiopulmonary bypass reported that postoperative urinary hepcidin levels increase when compared with preoperative levels in cardiac patients who did not develop AKI, but for patients who were diagnosed with AKI, their urinary hepcidin levels remain relatively unchanged preoperatively and postoperatively.⁸² Hepatic overexpression of hepcidin occurs in situations of iron overload.⁸³ Renal ischemia-reperfusion injury caused an increase in serum and nonheme iron levels.⁷⁸ Because AKI increases the levels of iron, one potential treatment or pretreatment may be to increase circulating hepcidin levels. Hepcidin reduces renal oxidative stress, apoptosis, inflammatory cell infiltration, and ischemia-reperfusion–induced renal injury.⁷⁸ Likewise, the damage caused by ischemia-reperfusion injury was magnified in mice deficient of hepcidin.⁷⁸ In fact, administration of exogenous hepcidin to the hepcidin-deficient mice attenuated kidney injury by inducing iron sequestration in the liver and increasing the expression of heavy-chain ferritin in the kidney.⁷⁸

Heavy chain ferritin (FtH) is thought to be cytoprotective, especially in the kidney. In 2015, Bolisetty et al⁸⁴ reported that FtH plays a crucial role in tubular-macrophage cross-talk during renal injury. The conditional deletion of FtH in proximal tubules (FtH^PT−/−) caused a substantial increase in proinflammatory macrophages in the kidneys of mice subjected to unilateral ureteral obstruction, a model of renal inflammation and fibrosis.⁸⁴ Moreover, these transgenic mice experienced a pronounced increase in the levels of inflammatory cytokines and fibrosis after kidney injury compared with a macrophage-specific knockout of FtH. In addition, FtH^PT−/− mice showed a decrease in renal structure and function after rhabdomyolysis and cisplatin-induced kidney injury (Fig. 2A).³⁹ After injury, these mice also experienced increased mortality compared with wild-type mice (Fig. 2B).³⁹ The proximal tubule-specific deletion of FtH increased the urinary levels of several iron acceptor proteins, including hemopexin, transferrin, and, most notably, NGAL.³⁹

Figure 2.

Deletion of heavy-chain ferritin in the proximal tubules exacerbates AKI. (A) Immunohistochemical staining of heavy-chain ferritin and the proximal tubule marker lotus lectin on serial kidney sections from FtH^PT+/+ and FtH^PT−/− mice (n = 4–8 per group). (B) Glycerol was administered to FtH^PT+/+ and FtH^PT−/− mice and their survival was monitored for up to 6 days (n = 10 per group). Reproduced with permission from Zarjou et al.³⁹

NGAL is a protein that binds small iron-carrying molecules called siderophores that act as iron chelators/transporters in several diseases.⁷¹ Similar to other types of kidney injury, urine and serum levels of NGAL are a predictive early indicator of AKI after cardiopulmonary bypass surgery.⁸⁵ Moreover, microarray data show that NGAL becomes overexpressed in kidney tissues soon after ischemic injury.⁸⁶ Data suggest that increasing NGAL concentrations in the circulation before kidney injury may attenuate or potentially eliminate functional and structural damage.^86,87 Purified NGAL administered intravenously is taken up in renal proximal tubules and conserves histologic integrity, function, and cell viability in proximal tubules after an ischemic insult.⁸⁶ Interestingly, the renal protective effects of NGAL are at least in part dependent on heme oxygenase enzyme activity.⁸⁸

HO-1 breaks down heme released from heme proteins to form biliverdin, iron, and carbon monoxide. The iron released from this reaction is sequestered by ferritin. Several lines of evidence support an important role for the induction of HO-1 in the pathophysiology of AKI,^89,90 as follows: (1) HO-1 is strongly induced in the kidney in both animal models and human AKI^90,91; (2) genetic and pharmacologic manipulation of HO-1 in animal models determines the course of AKI, deficiency or inhibition worsens renal structure and function, and increased expression is protective (reviewed by Nath⁹⁰); (3) plasma and urine levels of HO-1 have been implicated as biomarkers of AKI in both animal models and human beings⁹²; (4) polymorphisms in the human HO-1 promoter correlate with HO-1 expression and outcomes in AKI^93,94; (5) many drugs/interventions that have been tested in preclinical models of AKI (eg, α-melanocyte-stimulating hormone, erythropoietin, interleukin-10, NGAL) are potent inducers of HO-1 and mediate their effects, at least in part, through HO-1 induction^95–99; (6) each of the products of the HO-1-catalyzed reaction has shown renal protective effects in animal models of AKI; and (7) several clinical trials targeting the HO-1 pathway in the kidney, heart, and other organ systems have been initiated (Clinicaltrials.gov, NCT01430156, NCT00483587, NCT02142699, and NCT00531856).

In conclusion, iron is an important element in the body. Because of the labile nature of iron, its transport, metabolism, and storage is tightly controlled. Evidence suggests that there is an excess of iron after AKI. There are several therapies that have proven to be effective in animal models to attempt to reduce the iron levels and/or protect the kidney from ROS produced from iron-catalyzed reactions. Such treatments have not yet been effective in human beings. Approaches to target iron trafficking via ferritin, ferroportin, or hepcidin may offer new insights. Because of the complex nature of AKI, it is likely that a combination of different therapies discussed in this review may be required to ascertain the desired therapeutic outcome.