Neutrophil gelatinase-associated lipocalin (NGAL) has recently become established as an important contributor to the pathophysiology of cardiovascular disease. Accordingly, it is now viewed as an attractive candidate as a biomarker for various disease states, and in particular has recently become regarded as one of the best diagnostic biomarkers available for acute kidney injury. Nevertheless, the precise physiological effects of NGAL on the heart and the significance of their alterations during the development of heart failure are only now beginning to be characterized. Furthermore, the mechanisms via which NGAL mediates its effects are unclear because there is no conventional receptor signalling pathway. Instead, previous work suggests that regulation of iron metabolism could represent an important mechanism of NGAL action, with wide-ranging consequences spanning metabolic and cardiovascular diseases to host defence against bacterial infection. In the present review, we summarize rapidly emerging evidence for the role of NGAL in regulating heart failure. In particular, we focus on iron transport as a mechanism of NGAL action and discuss this in the context of the existing strong associations between iron overload and iron deficiency with cardiomyopathy.
Iron is an essential micronutrient and its crucial role in many physiological functions is often underestimated . Altered iron metabolism is implicated in a vast array of diseases, including neurodegenerative diseases , cardiovascular diseases , cancer , osteoporosis  and many more. In particular, both iron deficiency (ID) and iron overload have been associated with cardiomyopathy [6,7]. With the heart being a highly metabolically active organ, optimal iron homoeostasis is especially vital. Iron is a vital structural component of haemoglobin, myoglobin, oxidative enzymes and respiratory chain proteins that are collectively responsible for oxygen transport, storage and energy metabolism . Although epidemiological studies investigating the role of iron in various diseases are often inconsistent, this is not entirely surprising given the differences in experimental criteria and numerous methods used for assessment of iron status . In the present review, we aim to assess the altered iron status in cardiomyopathy, to discuss the possible cellular mechanisms involved, and to highlight the importance of regulation of iron metabolism by neutrophil gelatinase-associated lipocalin (NGAL).
SYSTEMIC AND MYOCARDIAL IRON METABOLISM
Essential concepts of systemic iron metabolism are briefly reviewed (see Figure 1), although readers are referred to some excellent review articles for further details [9,10]. Iron homoeostasis is essentially a closed system: iron is acquired from food as inorganic or haem iron, which is primarily absorbed in the duodenum via processes mediated by divalent metal ion transporter 1 (DMT-1) and haem carrier protein 1 (HCP-1), respectively. Iron in the cytoplasm can be either stored as ferritin or released to the bloodstream via ferroportin (FPN), where ferrous iron (Fe2+) can be oxidized to ferric iron (Fe3+) by hephaestin to facilitate its binding to transferrin (Tf) and transport in the circulation. Most cells express transferrin receptor 1 (TfR1) such that holo-Tf can be endocytosed to acquire iron, in which the ferric iron is reduced to ferrous iron by the metalloreductase STEAP3. It is then transported across the cell membrane by DMT-1. Hepcidin is a peptide hormone that induces the intracellular degradation of FPN, the only known iron exporter, and therefore it is a vital and major iron-regulatory hormone controlling plasma iron concentration and tissue iron distribution by inhibition of intestinal iron absorption, iron recycling by macrophages and iron mobilization from hepatic stores .
Schematic overview of cellular iron transport in cardiomyocytes
Various processes mediate iron transport in the cardiovascular system. Iron deposition in the heart is a gradual process, and has been suggested to occur in the ventricular myocardium before the atrial myocardium . Sequential appearance has been further documented, starting initially in the epicardium, then the myocardium and eventually the endocardium. Myocardial iron levels are normally regulated through Tf-mediated uptake mechanisms, mainly through TfR1 , although in the case of iron overload, when Tf-mediated transport becomes saturated, non-Tf-bound iron in the circulation increases and can also enter cardiac myocytes through DMT-1, T-type calcium channels (TTCCs), L-type calcium channels (LTCCs) [13,14], zinc-regulated transporter (ZRT)/iron-regulated transporter (IRT)-like protein 14 (Zip14 or Slc39a14)  and also NGAL receptor, which facilitates the entry of NGAL-bound iron . FPN1 is expressed in cardiomyocytes as an exporter of iron into the circulation .
OBSERVATIONAL CHANGES AND DIAGNOSIS OF IRON STATUS IN CARDIOMYOPATHY
Heart failure (HF) is a highly prevalent chronic progressive condition in which the heart is incapable of pumping enough blood to meet the body's demand. The ejection fraction (EF), the measurement of the amount of blood that the left ventricle pumps out in each contraction, is an important indicator for heart function and diagnosis of HF. In patients who have HF with reduced ejection fraction (HFrEF), the EF falls from a normal range of between 55% and 70% to <40%, yet half of the patients with HF are observed to have preserved EF (HFpEF) . In the ongoing search for novel treatments of HF, strong evidence is emerging to show the significance of disturbed iron homoeostasis in HF, regardless of the degree of change in the EF , thus establishing an excellent therapeutic potential if our understanding of the mechanisms responsible for the association of iron homoeostasis and HF can be enhanced.
The normal range of circulating ferritin is from 30 μg/l to 300 μg/l. In healthy individuals, iron deficiency is defined when the circulating ferritin concentration falls to <30 μg/l. However, HF has an inflammatory component in which serum ferritin, as an acute-phase protein, is often elevated without changes in body iron stores. Therefore, it should be noted that there is a difference in the diagnostic criteria for ID between healthy individuals and those with HF (and other chronic diseases). Recently the European Society of Cardiology (ESC) guidelines for diagnosis and treatment of HF have recommended a systematic measurement of iron parameters in all patients suspected of having HF. Serum ferritin <100 μg/l is regarded as absolute iron deficiency; if serum ferritin is 100–299 μg/l and Tf saturation (TSat) <20% this is defined as functional iron deficiency . On the other hand, TSat >55% and serum ferritin >200 μg/l or >300 μg/l in women and men, respectively, is diagnosed as iron overload cardiomyopathy (IOC), as proposed by 2005 American College of Physicians guidelines . The level of serum ferritin at which iron deposition is detected in the heart has not yet been conclusively identified. Not only is it invasive to take a heart biopsy, but technical difficulty also often renders the results variable and non-definitive. Both iron overload and ID have been linked to cardiomyopathy, with the former primarily associated with an enhanced oxidative stress and the latter with mitochondrial dysfunction, impaired heart efficacy , hypercoagulable state and increased cardiac burden, and, in addition, oxidative stress due to anaemia . Cardiomyopathy associated with iron overload or ID is reviewed in more detail below.
Iron overload cardiomyopathy
IOC is defined as the presence of systolic or diastolic cardiac dysfunction secondary to increased deposition of iron in the heart independent of other concomitant processes . It is typically associated with dilated cardiomyopathy with left ventricular hypertrophy and reduced EF . Although patients may remain asymptomatic in the early disease process, severely overloaded patients can rapidly experience terminal HF. Accumulation of iron in the myocardium may occur via increased iron absorption from gastrointestinal enterocytes (haemochromatosis), excess exogenous iron intake, such as via dietary supplements, or blood transfusions (haemosiderosis). The association of IOC with haemochromatosis, an autosomal disorder involving mutation of specific genes involved in iron metabolism, leading to increased gastrointestinal absorption, is well characterized . In fact, this accounts for a third of deaths in hereditary haemochromatosis, especially in young male patients . Chronic blood transfusion is the treatment for hereditary and acquired anaemia, including thalassaemia and myelodysplastic syndromes. However, as excess body iron cannot be actively excreted, repeated blood transfusions can result in iron deposition in multiple organs, of which the heart is one of the most sensitive organs to iron toxicity . As the cardiac clinical presentations can vary widely among these patients, a recent review has provided recommendations and clinical guidelines with regard to a decision on chelation therapy by stratifying patients based on the presence or absence of heart dysfunction and heart magnetic resonance imaging T2-weighted values .
There are numerous mechanisms via which excess iron can reduce cardiac function. Once the antioxidant capacity of cardio-myocytes has been exceeded, iron can produce excess oxidative stress via the Fenton reaction (see below) and lead to apoptosis . In addition, excess free iron in the blood is suggested as responsible for the generation of insoluble parafibrin, which is highly resistant to proteolytic dissolution and initiates inflammatory reactions on deposition on the arterial wall . The association of iron and atherosclerosis is well established in animal studies, e.g. iron accumulation is observed in atherosclerotic plaques , and decreasing tissue iron via chelating therapy, dietary iron restriction or phlebotomy showed decreased atheroma plaque size with improved stability [29–32]. Such association is also supported in clinical studies: a study involving 12033 men showed that increased ferritin concentration was associated with early coronary artery atherosclerosis, independent of traditional cardiovascular risk factors ; another study involving 196 participants showed a strong association between serum ferritin and pulse wave velocity or aortic stiffness in women . The 6-year-long Iron and Atherosclerosis STudy (FeAST) also established correlations of levels of ferritin, inflammatory biomarkers and mortality in a subset of patients with peripheral arterial disease . Iron availability may have contributed to atherosclerosis by impairing nitric oxide action, as demonstrated by improvement in nitric oxide-mediated endothelium-dependent vasodilatation in patients with coronary artery disease, by chelating iron with desferrioxamine . Although further mechanisms have yet to be defined, the Atherosclerosis Risk In Communities (ARIC) study has rejected the hypothesis that excess iron stores would promote low-density lipoprotein (LDL) oxidation . It was shown that dietary iron intake and body iron stores had no direct link to altered structure and function of large arteries in individuals free of cardiovascular disease, cancer or haemochromatosis .
Iron deficiency and cardiomyopathy
Iron deficiency (ID) is the most common nutritional deficiency worldwide . It is frequent, has a high occurrence rate of 30–50% in patients with HF and presents as an important co-morbidity . Two types of ID can be distinguished: absolute and functional. Absolute ID reflects depleted iron stores, whereas iron homoeostatic mechanisms and erythropoiesis often remain intact. Absolute ID development in humans can result from inadequate dietary iron intake, impaired gastrointestinal absorption/transport, drug interactions and gastrointestinal blood loss . On the other hand, functional ID presents a dysregulated iron homoeostasis in which cells and tissues might receive inadequate iron supplies despite normal whole body iron storage. This can be a result of elevated circulating hepcidin concentrations, and has been reported in patients with acute-phase myocardial infarction .
ID is often accompanied by anaemia, although both can exist independently and ID usually appears before the onset of anaemia. It is important to differentiate anaemia from ID; although ID is marked by the insufficiency of iron, anaemia is defined by insufficient haemoglobin (Hb). ID that is independent of anaemia was reported to have a higher risk of death than that dependent on anaemia [43,44]; it has been reported as an independent predictor of mortality and is associated with disease severity .
In recent years there have been several clinical trials to test whether administration of intravenous iron could improve functional parameters related to HF. One of the most well-known studies includes the Ferinject Assessment in patients with IRon deficiency and chronic Heart Failure (FAIR-HF). This involved 459 patients with ID and chronic heart failure of New York Heart Association (NYHA) functional class II or III. The treatment with intravenous ferric carboxymaltose over 24 weeks improved NYHA functional class, functional capacity and quality of life in terms of EuroQol-5 Dimension and Kansas City Cardiomyopathy Questionnaire with an acceptable side-effect profile . A simplified ferric Carboxymaltose evaluatioN on perFormance in patients with IRon deficiency in combination with chronic Heart Failure (CONFIRM-HF) trial, which has enrolled 304 stable symptomatic HF patients from 41 sites across nine European countries, is currently in progress to confirm the efficacy and safety of iron therapy using intravenous ferric carboxymaltose solution in chronic HF patients with iron deficiency, as in the FAIR-HF study . Other clinical trials including the FERRIC-HF (FERRIC iron sucrose in Heart Failure)  and IRON-HF  trials have also shown encouraging results with iron therapy, using intravenous iron sucrose in improving functional capacity in HF patients with ID. Thus, ID can serve in many cases as a promising therapeutic target for HF.
Furthermore, the importance of ID as a marker in the context of HF and its assessment was highlighted and recommended by the ESC . As for anaemia, it has also been shown to have a relatively high prevalence (37%) in patients with HF . With less oxygen availability during anaemia, the heart compensates by increasing heart rate and stroke volume. Moreover, anaemia has been reported to be an independent risk factor for adverse outcomes in HF, in terms of both morbidity and mortality rates [51–58]. Efforts have been made to restore Hb levels as a potential therapeutic approach to HF using erythropoietin-stimulating agents (ESAs) and this resulted in improvements in exercise tolerance, peak V̇O2, N-terminus of the prohormone brain natriuretic peptide (NT-proBNP) and left ventricular performance in patients with HF . However, some major clinical trials that evaluated the effect of treating anaemia with the ESA darbepoetin α for cardiovascular events or HF have consistently suggested that darbepoetin α did not significantly improve HF outcomes. One of the earlier trials was the STudy of AneMIa in Heart Failure Trial (STAMINA-HeFT), which involved 319 patients with systemic HF, left ventricular EF <40% and Hb level between 9.0 g/dl and 12.5 g/dl, in which they found a 1-year treatment with darbepoetin α failed to associate with any significant clinical benefits . The Trial to Reduce cardiovascular Events with Aranesp (darbepoetin α) Therapy (TREAT) was initiated in 2004 to provide a robust estimate of the safety and efficacy of darbepoetin α . This event-driven study continued to grow until it reached approximately 1203 patients with Type 2 diabetes, chronic kidney disease (CKD) and anaemia, who have experienced primary events, including the composite end-point of death, cardiovascular morbidity or the need for long-term renal replacement therapy. TREAT initially showed that ESA treatment in diabetic and anaemic patients with CKD did not demonstrate clinical benefits in terms of mortality, morbidity or quality of life ; instead, cardiovascular risk was most strongly predicted by age, HF history, and several other established renal and cardiovascular biomarkers . In addition, most of these patients were able to maintain a stable Hb level without having any long-term ESA therapy , and long-term (2 years) ESA therapy to treat anaemia did not confer significant benefits . The most recent update from TREAT suggests that the cardiovascular or non-cardiovascular mortality rates, particularly those from sudden death and infection, were associated with lower baseline glomerular filtration rate and higher protein/creatinine ratio in diabetic CKD patients .
Another large-scale clinical trial, the Reduction of Events with Darbepoetin α in Heart Failure (RED-HF) trial, was also launched to evaluate the effect of darbepoetin α on mortality and morbidity, and quality of life in patients with HF and anaemia . Over 2600 patients with NYHA class II–IV, EF ≤40% and Hb between 9.0 g/dl and 12.0 g/dl were subcutaneously administered darbepoetin α or placebo until the primary end-point was met . The RED-HF study, compared with other recent clinical trials in HF, had patients who were older, with moderately to markedly symptomatic HF and extensive co-morbidity . The study showed that darbepoetin α did not significantly alter primary or secondary outcomes, concluding that darbepoetin α did not improve clinical outcomes in patients with systolic HF and mild-to-moderate anaemia . Thus, based on all available evidence, it has been suggested that anaemia may serve only as a surrogate marker rather than an end-point target in HF.
CELLULAR MECHANISMS THAT UNDERLIE THE ASSOCIATION OF IRON AND CARDIOMYOPATHY
Iron and oxidative stress
Free iron is highly-redox reactive and can participate in a redox reaction that leads to the generation of reactive oxygen species (ROS). ROS include not only a range of free radicals such as superoxide radical anion (O2•−), carbonate radical anion (CO3•−), hydroperoxyl radical (HOO•), hydroxyl radical (HO•), peroxyl radical (ROO•) and alkoxyl radical (RO•), but also non-radicals such as hydrogen peroxide (H2O2), hypochlorous acid (HClO) and ozone (O3). Among them, H2O2 and O2•− are the major ROS in living organisms and are continuously produced by cells and must simultaneously be removed by antioxidant enzymes. Neither H2O2 nor O2•− is a strong oxidizing agent, but the extremely reactive hydroxyl radical HO• can be produced on reaction with iron or iron-containing molecules through the Fenton reaction. H2O2 oxidizes Fe2+ to Fe3+, producing the hydroxyl radical HO• and hydroxide ion OH− (eqn 1); Fe3+ is then reduced back to Fe2+ by another H2O2 molecule, forming a hydroperoxyl radical HOO• and a proton H+ (eqn 2), or by superoxide radical anion (O2•−) to produce oxygen (O2) (eqn 3). In this way, iron acts as a catalyst to generate plentiful amounts of ROS.
Although ROS have important physiological functions, e.g. to combat invading pathogens, excess ROS can result in oxidative stress that damages intracellular proteins, lipids and nucleic acids. Indeed, specific parts of the genome were found to be damaged by the Fenton reaction, and are termed ‘genomic sites vulnerable to the Fenton reaction’ . Ferroptosis, as the name implies, is a recently identified form of cell death that is morphologically, biochemically and genetically distinct from apoptosis and necrosis, and is found to depend on intracellular iron; it can be prevented by iron chelators and antioxidants . The use of the iron chelator desferrioxamine has demonstrated significant reduction of neutrophil-mediated free radical production and amplification of the inflammatory response during cardiopulmonary bypass in humans  and in in vivo studies [73–75]. In summary, iron can potently enhance oxidative stress and consequently contribute to cardiomyopathy.
Iron and mitochondrial dysfunction
In eukaryotic cells, mitochondria are the main consumers of intracellular iron . With mitochondria being the respiratory centre of the cell, plentiful oxygen can rapidly react with unregulated free iron to produce ROS. To avoid ROS-induced damage, mitochondrial iron level and homoeostasis are tightly regulated by different transport, storage and regulatory proteins . Through different biosynthesis pathways, iron is transferred in the mitochondria to its bioactive forms, haem and iron–sulfur cluster (ISC). MitoNEET is an ISC-containing protein tethered to the outer mitochondrial membrane that facilitates transfer of iron into the mitochondria . Not only does it play an essential role in redox signalling , but also it dictates the metabolic functions of mitochondria [78,80–82]. An increased level of mitoNEET can lead to accumulation of iron within the mitochondria, which in turn results in dysfunction [83,84], a hallmark of various diseases. MitoNEET is recognized as a target for the thiazolidinedione class of anti-diabetic drugs [79,85], and its genetic manipulation was shown to have striking anti-diabetic effects . Mitochondrial ferritin stores and supplies iron within the mitochondria. Its expression is restricted to highly metabolically active cells such as cardiomyocytes, in order to supply iron when demand is increased during active respiration or intense metabolic activities. Frataxin is another mitochondrial protein that handles iron in the mitochondrial matrix assembling ISC . It can act either as a chaperone for ferrous iron or as an iron storage protein that can mineralize iron as ferrihydrite. There is great interest in improving mitochondrial dysfunction as a potential therapeutic approach for HF [88,89], and the underappreciated contribution of iron homoeostasis is worthy of more consideration.
Iron and endoplasmic reticulum stress
Various pathophysiological situations can elevate stress in the endoplasmic reticulum (ER). One of the major functions of the ER is proper protein folding, and accumulation of misfolded proteins can normally be relieved by cellular responses such as ER-associated protein degradation (ERAD) and unfolded protein response (UPR). These ER stress responses are important defence mechanisms when the amount of unfolded protein exceeds the folding capacity of the ER . ER stress has been strongly implicated in cardiovascular disease, e.g. ER stress can lead to cardiomyocyte death in vivo and ex vivo , and in patients with HF . Interestingly, it was suggested that ER stress may be cardioprotective during constriction-induced hypertrophy , perhaps by inducing compensatory cellular mechanisms such as autophagy (see below). Similarly, ER stress induction protected cardiomyocytes from oxidative damage . Iron overload-induced ER stress was shown in vivo in hearts under acute and chronic conditions , and had been demonstrated in other tissue types, including neurons  and liver . In reverse, ER stress can modulate iron metabolism. Hepcidin, as mentioned above, degrades the iron efflux transporter FPN, thus leading to a systemic hypoferraemic environment. ER stress was found to induce hepcidin expression ; the UPR signalling pathway was further shown to increase the transcription of FPN and ferritin . Thus, based on available evidence, ER stress and iron homoeostasis appear to have a reciprocal relationship such that they can tightly regulate each other. ER stress and UPR-related proteins will serve as interesting targets for future clinical studies.
Iron and autophagy
Macroautophagy (hereafter referred to as autophagy) is an intracellular degradation system that involves the sequestration of cytoplasmic components within a double-membrane vesicle termed an ‘autophagosome’, in which the cargo content is degraded by the acidic hydrolases on fusion with a lysosome . It has a wide variety of physiological and pathophysiological roles including energy homoeostasis, cell survival and host defence against pathogen invasion .
In the heart, autophagy typically occurs at low levels, yet it is nevertheless important in maintaining cellular homoeostasis under normal conditions. Autophagy is typically up-regulated in times of stress, e.g. during ischaemia/reperfusion, pressure overload and cardiac toxicity induced by chemicals such as the athracycline doxorubicin . Although increased autophagy can promote cell survival by degrading damaged organelles, such as mitochondria and protein aggregates, to recycle catabolites and maintain ATP production, either excess or lack of autophagy can result in cell death and cardiac dysfunction. Thus, the role of autophagy can often appear controversial between different studies when different degrees of autophagy, time course and pathological conditions being studied have led to variable observations. In vivo data have shown that expression of multiple autophagy-related genes was altered in IOC, possibly contributing to cardiac diastolic dysfunction .
Specifically, it is now appreciated that iron can regulate autophagy and that autophagy has an important role in iron homoeostasis. Nuclear receptor co-activator 4 (NCOA4) was recently identified using quantitative proteomics as the cargo receptor that mediates autophagy of ferritin, a process termed ‘ferritinophagy’. NCOA4 is required for the delivery of ferritin to the lysosome; without NCOA4, cells are malfunctional in ferritin degradation and this can result in a decreased bioavailability of intracellular iron . However, excessive ferritinophagy may result in insufficient ferritin, thus reducing its buffering effect on binding intralysosomal low-mass iron, and can lead to lysosomal fragility and increased sensitivity to oxidative stress . Analysis of autophagy in iron-associated cardiomyopathy is relatively new with limited mechanistic and clinical studies; however, we believe that this must be rapidly developed because it has great potential as a therapeutic target.
REGULATION OF CARDIOMYOPATHY BY NGAL
The maintenance of optimal iron levels in the body is largely controlled and influenced by endocrine regulation, and this is likely to be of major significance in cardiomyopathy. In this section, we highlight the importance of NGAL in the regulation of iron homoeostasis and other possible mechanisms in the context of cardiomyopathy (see Figure 2).
Schematic overview of pro-inflammatory actions of NGAL
Neutrophil gelatinase-associated lipocalin
Lipocalins are a diverse family that generally bind small and hydrophobic ligands, but can also bind soluble extracellular macromolecules and specific cell surface receptors [106,107]. NGAL (human orthologue), also termed ‘lipocalin 2’ (Lcn2) or ‘24p3’ (murine orthologue), is a 25-kDa secretory protein. NGAL was originally identified as a component of neutrophil granules that bound to and prevented the degradation of matrix metalloproteinase-9, and was later found to be secreted by a number of cells including macrophages, endothelial cells , epithelial cells , cardiomyocytes (Chan, Y.K., Sung, H.K. and Sweeney, G., unpublished work), hepatocytes  and adipocytes .
Epidemiology of NGAL and heart failure
In clinical settings, NGAL is now regarded as the best biomarker for acute kidney injury, and is also emerging as a promising biomarker for HF. The heart and kidney have numerous similarities and their interdependent relationship makes it understandable that renal dysfunction often accompanies cardiac failure, and that cardiac dysfunction is frequently seen with renal failure . Therefore, many biomarkers for kidney or tubular dysfunction, e.g. kidney injury molecule 1 (KIM-1) and N-acetyl-β-D-glucosaminidase, rather than just serving as a means to assess kidney function, also provide insights into the cardiac prognosis in patients with HF. However, unlike other renal biomarkers, the NGAL level was not affected by diuretic withdrawal in patients with chronic systolic HF , and administration of NGAL in an animal model of acute ischaemic renal injury actually attenuated tubular injury . Moreover, in patients with chronic HF, NGAL has been shown to be a more effective marker than creatinine for the cardiorenal syndrome; NGAL could detect renal injury earlier than creatinine, and was an independent and novel risk predictor of mortality in chronic HF . Indeed, the elevation of NGAL seen in HF, and its association with different parameters of HF, has affirmed its potential as a HF biomarker. First, serum NGAL predicted the outcome of HF, e.g. the GALLANT (NGAL evaluation Along with B-type NaTriuretic peptide in acutely decompensated heart failure) trial concluded that, at the time of discharge, plasma NGAL was a strong prognostic indicator of 30-day outcomes in patients admitted for acute HF ; it independently predicted worse short-term prognosis in patients with acute HF , and NGAL levels correlated well with HF-related functional assessment parameters, including the 6-min walk test . Secondly, the CORONA (COntrolled ROsuvastatin multiNAtional trial in heart failure) study suggested that NGAL was associated with the severity of HF , the elevated serum NGAL in patients with acute post-myocardial infarction and chronic HF was found to be associated with more adverse outcome , and the NGAL level was shown to correlate with HF severity and haemodynamic improvements after placement of a ventricular assist device . Thirdly, serum NGAL predicted severity of chronic HF in terms of NYHA classification and mortality in elderly patients , and plasma NGAL also predicted mortality in HF patients with or without CKD , and in community-dwelling older adults independent of traditional risk factors and kidney functions . Clearly, there is strong evidence for NGAL being a useful biomarker to assess severity, prognosis and mortality in HF suggested by various individual cohorts.
Possible mechanisms via which NGAL may mediate cardiomyopathy
NGAL is most well known for its participation in innate immunity to limit bacterial growth by sequestrating iron. One way to secure iron from the host by bacteria is by synthesizing and secreting siderophores to extract iron from iron-containing compounds such as Tf and lactoferrin; NGAL is secreted by the host to tightly bind to bacterial catecholate-type ferric siderophores, competing for iron and preventing such uptake . NGAL saturated with iron (holo-form) can increase intracellular iron levels by transporting and then releasing iron into the cytoplasm; in contrast, when NGAL is iron-free (apo-form), it can deplete intracellular iron and transport it to the extracellular space via its receptor NGAL-R . Bacterial infection is often associated with hypoferrinaemia  which limits iron availability to pathogens; accordingly, mice deficient in NGAL exhibit elevated intracellular labile iron and lowered circulating iron levels . Overall, NGAL, as an iron-trafficking protein, can be regarded as an alternative to a Tf-mediated iron-delivery pathway .
Although limited studies are available, it is speculated that circulating NGAL levels may reflect the body's iron status, at least in haemodialysis patients . In haemodialysis patients, it was found that plasma NGAL was significantly lower within those who had ID, with TSat <20%, and that the level of NGAL was positively correlated with circulating iron, TSat and ferritin. NGAL was significantly increased after correction of ID with intravenous iron administration . Similar results were also observed in another two studies supporting the potential use of NGAL to identify iron deficiency in haemodialysis patients [129,131]. Likewise, a lowered NGAL level was also recorded in patients with iron deficiency anaemia  and, in patients with chronic HF (both HFpEF and HFrEF), the significantly higher circulating NGAL levels also correlated with higher serum iron concentrations in the EPOCARES (ErythroPOietin in the CArdioREnal Syndrome) study . As circulating NGAL is often recorded as significantly increased in patients experiencing HF [117,120–123], and local NGAL production in the heart is also increased significantly (Chan, Y.K., Sung, H.K. and Sweeney, G., unpublished work) [120,134], we believe that it will be of great interest to elucidate the role of NGAL in iron-associated cardiomyopathy further. It is interesting that we previously identified NGAL leading to cardiomyocyte apoptosis by causing intracellular iron accumulation . Further mechanistic studies are definitely warranted.
Pro-inflammatory actions of NGAL
As NGAL is involved in defending the host during bacterial infection, it comes as no surprise that NGAL is regarded as a pro-inflammatory cytokine. In the fourth Copenhagen Heart Study, which involved more than 5000 patients and a follow-up period of 10 years, it was shown that plasma NGAL strongly associated with all inflammatory markers investigated, including high-sensitivity C-reactive protein, and total leucocyte and neutrophil count; increased NGAL was also shown to correspond to an increased risk of all-cause mortality and major adverse cardiovascular events . It was suggested that NGAL expression and secretion can be induced by interferon γ (IFN-γ) and tumour necrosis α (TNF-α), and that the transcription factors, signal transducer and activator of transcription 1 (STAT1) and nuclear factor κB (NF-κB), were shown to bind to the human NGAL promoter . Likewise, in elucidating the inflammatory mechanisms of NGAL with animal models, NGAL mRNA and proteins were up-regulated on vascular injury in an NF-κB-dependent manner . NGAL can enhance cardiac inflammation by promoting polarization of the macrophage pro-inflammatory M1 phenotype . Thus, a vicious cycle exists whereby NGAL can intensify inflammation by inducing the expression of TNF-α and other pro-inflammatory mediators . It is of interest that prevention of the clearance of NGAL from the circulation was shown to promote vascular inflammation and endothelial dysfunction . In both HFpEF and advanced HFrEF, elevated systemic and local inflammation with increased circulating TNF-α have indispensable roles in disease pathogenesis . It will be interesting to explore how NGAL contributes to cardiomyopathy in an inflammation-dependent and -independent manner.
Iron is a micronutrient that is integral to the function of many proteins required in cells with high metabolic activity. It is involved in regulating various cellular mechanisms including oxidative stress, mitochondrial dysfunction, ER stress and autophagy, all of which can contribute to cardiac dysfunction when perturbed. NGAL, an adipokine that mediates iron transport through association with a siderophore, can increase or decrease intracellular iron content based on the body iron store and its iron saturation, making it a malleable factor in the maintenance of iron homoeostasis. Indeed, elevation of NGAL is well documented in various instances of HF with strong association to severity of HF and resulting morbidity and mortality. Furthermore, NGAL is an inflammatory biomarker that can be induced by endotoxaemia and myocarditis. It can also contribute to atherosclerosis and insulin resistance. Future work to fully characterize the association of NGAL with iron overload and deficient cardiomyopathy, and in particular to understand the precise mechanisms of NGAL action contributing to cardiac dysfunction, are needed. These would both validate the potential use of NGAL as a biomarker and allow the development of novel therapeutic targets for the treatment of heart failure.
We thank Hyosik Kim for help with graphic design.
Related work in our group is funded by the Canadian Institutes of Health Research, Heart and Stroke Foundation and Canadian Diabetes Association.
chronic kidney disease
divalent metal ion transporter 1
European Society of Cardiology
Ferinject Assessment in patients with IRon deficiency and chronic Heart Failure
haem carrier protein 1
heart failure with preserved ejection fraction
heart failure with reduced ejection fraction
iron overload cardiomyopathy
L-type calcium channel
nuclear receptor co-activator 4
nuclear factor κB
neutrophil gelatinase-associated lipocalin
New York Heart Association
Reduction of Events with Darbepoetin α in Heart Failure; ROS, reactive oxygen species
tumour necrosis factor α
Trial to Reduce cardiovascular Events with Aranesp (darbepoetin α) Therapy; TSat, transferrin saturation
T-type calcium channel
unfolded protein response