Iron: an essential nutrient for Aspergillus fumigatus and a fulcrum for pathogenesis

Abstract

Purpose of review

Aspergillus fumigatus is a ubiquitous saprophytic fungus that can cause life-threatening invasive aspergillosis in immunocompromised patients. Apart from the immune status of the host only a few characterized virulence factors have been identified. In this review, we describe the role of iron in the manifestation of A. fumigatus virulence.

Recent findings

We gathered recent clinical evidence suggesting that tissue iron overload increases the risk of invasive aspergillosis occurrence. Furthermore, we summarize the mechanisms that A. fumigatus employs to achieve iron homeostasis and their importance in A. fumigatus proliferation in vitro. We describe two recent in-vivo models that clearly demonstrate the importance of iron in A. fumigatus growth and invasion.

Summary

Based on these recent findings, therapy aimed at managing A. fumigatus iron homeostasis locally could make conditions more favorable to the host.

INTRODUCTION

Aspergillus fumigatus is the most prominent fungus in the airways of cystic fibrosis patients [1], and a prominent opportunistic pathogen in neutropenic patients and in patients after lung and liver transplantation [2–6]. A. fumigatus biofilms, three-dimensional networks of cellular hyphae and extracellular matrix, are associated with antifungal resistance, resistance to host immunity, and also persistence of infection [7]. With minor exceptions, almost all organisms, including A. fumigatus, employ iron as a cofactor for fundamental biochemical activities such as oxygen transportation, energy metabolism, and DNA synthesis [8,9]. During infection, the host’s innate immune system sequesters iron, decreasing free iron to low levels as a defense mechanism to limit microbial growth [10]. Thus, competition for iron is a critical battleground that determines the outcome of the host–pathogen relationship [11]. Towards this end, pathogenic microbes have developed mechanisms to aquire iron from their hosts [12,13]. In contrast to various bacterial and fungal pathogens, A. fumigatus lacks systems for direct uptake of host iron sources and relies on other methods to adjust to the ambient iron concentration [9]. Although A. fumigatus response to low iron conditions has been well characterized [9], its adaptations to iron overload are only increasingly being appreciated. The evidence associating A. fumigatus virulence with iron is derived from: clinical studies; studies demonstrating the requirement for iron uptake mechanisms to maintain fungal virulence; and animal and in-silico studies that demonstrate a critical role of iron overload to drive fungal growth and invasion. In this review, we highlight recent findings regarding A. fumigatus side-rophore-mediated iron uptake, intermicrobial competition for iron, the role of iron in lung pathogenesis, and clinical relevance.

CLINICAL EVIDENCE

Increased virulence with iron overload is suggested by the clinical observations of a higher incidence of Aspergillus infections in persons with medical diseases that increase tissue iron. The liver plays a key role in iron homeostasis by regulating iron uptake (by producing the iron sequestration protein, hepcidin), providing a key iron store in the body, and by producing ceruloplasmin and transferrin – proteins that are vital for iron metabolism [14]. In persons after liver transplantation, stainable iron in the explanted liver was independently associated with an increased risk of invasive fungal infections [hazard ratio 3.09, 95% confidence interval (CI) 1.45–6.56, P = 0.003], with A. fumigatus infection occur ring in 25% of the patients [15]. This increased risk of invasive aspergillosis has also been observed in patients with hematologic malignancies and after autologous and allogeneic hematopoietic cell transplantation (HCT) [16–19]. In these patients, iron overload is likely due to frequent blood transfusions, ineffective erythropoiesis, and defects in iron metabolism [16]. In persons with acute myeloid leukemia and acute lymphoblastic leukemia, invasive fungal infections were associated with higher serum iron, ferritin, and transferrin saturation concentrations than in those persons who did not develop fungal infections [19]. Kontoyiannis et al.[17], in a retrospective study among patients with leukemia and after allogeneic HCT, showed that moderate to severe iron overload, measured histologically in bone marrow iron stores, was associated with an increased risk of IA in a multivariate analysis (hazard ratio 12.3, 95% CI 3.4–44.9, P < 0.0001). In another study, in patients who died after autologous or allogeneic HCT, there was a significant relationship between hepatic iron overload and invasive aspergillosis [relative risk (RR) 9, 95% CI 1.6–50.3, P = 0.012] [16]. Similar findings have been described by Ozyilmaz et al. [18] who showed in persons after HCT that serum ferritin concentration (>1000 ng/ml) increased the risk of fungal pneumonia, the majority due to Aspergillus sp. (62%) [odds ratio (OR) 3.42, 95% CI 1.06–11.42, P = 0.045).

RELEVANCE OF IRON FOR ASPERGILLUS FUMIGATUS IN VITRO

Iron sequestration, either pharmaceutically by use of iron chelators [20], or by natural substances, for example, the Pseudomonas aeruginosa siderophore pyoverdine [21■■], or the P. aeruginosa phage Pf4 [22■■], have been shown to be detrimental for the A. fumigatus. P. aeruginosa and A. fumigatus frequently co-exist in the same micro-environment, for example, in lungs of cystic fibrosis, lung transplant, or neutropenic patients [2,5,6,23]. Because iron is their common essential nutrient, both micro-organisms have developed measures to secure iron for their own needs. A. fumigatus produces four hydroxamate-containing siderophores: ferricrocin and hydroxyferricrocin (HFC) for intracellular iron trafficking, and also fusarinine C (FsC) and its derivative triacetylfusarinine C (TAFC) for extracellular iron scavenging (Fig. 1) [24–26]. Critical for the biosynthesis of those four hydroxamate-containing siderophores is the enzyme ornithine mono-oxygenase, termed SidA [9,26], catalyzing oxygen and NADPH-dependent hydroxylation of L-ornithine to N5-L-hydroxyornithine. An A. fumigatus lacking SidA (AfΔsidA) was unable to establish invasive aspergillosis in a mouse model [24,27]. P. aeruginosa metabolites have been shown to inhibit [28,29] and induce apoptosis [30■] in A. fumigatus. In situations of chronic P. aeruginosa resistance in vivo, such as in cystic fibrosis lungs, surviving P. aeruginosa phenotypes are more powerful inhibitors of A. fumigatus than non-CF P. aeruginosa isolates [31]. We recently showed, via studies with 24 deletion mutants, that the most important P. aeruginosa molecule responsible for antifungal activity is the siderophore pyoverdine, and that the mechanism of antifungal action is sequestration of ferric iron [21■■]. Because pyoverdine cannot be used as a xenosiderophore for A. fumigatus, the fungus suffers from iron shortage, and increases expression of genes needed for production of its own siderophores, for example, sidA, hapX, and mirB, while saving energy by down-regulating genes expressed during times of iron-abundancy, for example, sreA, cccA, cycA, and acoA (as summarized in Fig. 1) [21■■]. Recently, MirC also has been shown to play an important role in bio-synthesis of the A. fumigatus intracellular siderophore ferricrocin, and to contribute to the maintenance of iron homeostasis [32]. Antifungal activity of pyoverdine seems to be exclusively based on iron denial to the fungus, because experiments with suitable mutants revealed no contribution of the pyoverdine downstream effectors ExoA or PrpL protease [21■■]. Effects of pyoverdine on A. fumigatus are independent of the bacterial strain producing the pyoverdine, the type of pyoverdine (I, II, or III), or the A. fumigatus strain used for experiments [21■■,33].

FIGURE 1.

Schematic representation of Aspergillus fumigatus siderophore biosynthesis and iron regulation. Fe: iron, FusC: fusarinine C, HMG-CoA: 3-hydroxy-3methylglutaryl-coenzyme, TAFC: triacetyl fusarinine C, SidA: ornithine monooxygenase, SidC: ferricrocin nonribosomal peptide synthetase (FC NRPS), SidF: Transacylase, SidD: Fusarinine C nonribosomal peptide synthetase (FSC NRPS), SidG: Transacetylase, Sidl: Mevalonyl-CoA ligase, SidH: Mevalonyl-CoA hydratase, RIA: reductive iron assimilation enzyme complex (involves iron reduction by metalloreductases such as FreB and uptake by a protein complex consisting of FetC and FtrA), EstB: TafC specific esterase, MirB: TAFC specific importer, CccA: iron importer to vacuole, HapX: bZip CCAAT-binding transcription factor, SreA: GATA transcription factor, FtrA: high-affinity iron permease, AmcA: putative mitochondrial ornithine transporter, FreB: iron reductase, FetC: ferroxidase, SidJ: fusarinine C esterase. Not shown: The HapX-mediated inhibition of iron-consumption in iron-dependent pathways other than CccA. The high iron activating function of HapX on iron-dependent pathways other than CccA.

Normal siderophore production, and hence sufficient iron assimilation, is of high importance for A. fumigatus. It has recently been shown that pharmacological interference with SidA production, and hence hydroxamate siderophore production, reduced fungal growth in vitro [34■].

RELEVANCE OF IRON FOR ASPERGILLUS FUMIGATUS IN VIVO

The critical role of A. fumigatus genes involved in iron acquisition, as outlined above, have been examined in vivo using animal infection models [35–37]. In addition to the in-vitro studies, the correlation of iron and A. fumigatus virulence was predicted by a computational model that showed that iron acquisition was essential for the development of invasive aspergillosis [38■■]. However, to date, few animal studies have directly examined the association between iron overload and fungal virulence. Here, we describe a fungal keratitis and a A. fumigatus tracheal transplantation model that clearly suggests the importance of iron in A. fumigatus growth and invasion.

FUNGAL KERATITIS MODEL

Aspergillus fumigatus and other filamentous fungi are major causes of fungal keratitis in developing nations, accounting for 1 million cases annually in Asia and Africa [39]. As treatments for these infections are often limited, Leal et al. examined the role of targeting iron acquisition as a novel treatment strategy for these infections. The first series of experiments examined whether iron availability would regulate A. fumigatus growth. Mice (C57BL/6) were treated with iron dextran or deferoxamine (DFO), an iron chelator, which A. fumigatus can use as a xenosiderophore in acquiring iron. Mice were then infected by direct corneal stromal injection with A. fumigatus conidia. In these experiments, the mice treated with iron dextran and DFO demonstrated higher fungal mass at 48 h after inoculation compared to control mice, vehicle-treated animals. Conversely, the addition of topical lactoferrin, an iron-chelating host protein, lessened the fungal mass and colony-forming unit (CFU) per eye compared to mice treated with vehicle control. Together, these studies demonstrate that fungal growth in the cornea is determined by bioavailable iron [38■■]. The next series of experiments examined the role of fungal siderophores, using A. fumigatus iron acquisition mutants. In these experiments, researchers examined the role of A. fumigatus mutants in the production of intracellular and extracellular siderophores. Figure 1 depicts the pathways involved in fungal siderophore production. In these studies, mutants of ΔsidA, ΔsidF, ΔsidD, ΔsidI, and ΔsidH had significantly lower fungal burden and CFU than wild-type or their respective reconstituted mutants. Whereas mutants of intracellular siderophores, including ΔsidC and ΔsidG or ΔftrA, a mutant that lacks a membrane-bound iron transport channel, making it deficient in the uptake of extra-cellular iron, were not significantly different than wild-type in the corneal infection model. Moreover, treatment of A. fumigatus with an HMG-CoA reductase inhibitor (statin class of medications), which inhibits mevalonate production, a critical step in siderophore biosynthesis (see Fig. 1), and deferi-prone (iron chelator that cannot be used by A. fumigatus as a xenosiderophore) inhibited A. fumigatus growth in the corneal infection model. Together, these data are consistent with the side-rophore data mentioned previously, suggesting that extracellular iron siderophores are essential for A. fumigatus virulence, and that blocking iron uptake or reducing bioavailable iron attenuates A. fumigatus growth.

ORTHOTOPIC TRACHEAL TRANSPLANT MODEL OF ASPERGILLUS FUMIGATUS INVASION

One in three lung transplant recipients suffer from an Aspergillus-related pulmonary disease that presents as chronic colonization, airway anastomotic infections, or invasive aspergillosis [40–42]. Other than immune suppression, the factors that determine disease presentation are unknown [43,44]. To address this question, our research group developed a murine orthotopic tracheal transplant (OTT) model of A. fumigatus infection to study the transplanted host–pathogen interaction. In these studies, we performed both allogeneic (BALB/c donor transplanted into a C57BL6 recipient) and syngeneic (C57BL6 donor and recipient) to study the effects of alloimmune-mediated rejection on A. fumigatus transplant invasion, as measured histologically. In our initial studies, we found that Aspergillus became more deeply invasive as the allograft underwent progressive rejection-mediated ischemia [45]. Subsequently, we showed that up-regulating endothelial hypoxia-inducible factor (HIF)-1α (a stimulator of angiogenesis and vascular repair) limited the invasion of the mould [46]. Together, these findings suggested that alloimmune-mediated microvascular damage precipitated a more virulent A. fumigatus phenotype, and that by preventing microvascular damage, through HIF-1a up-regulation, we were able to attenuate this virulence. To further elucidate these findings, we examinedthe consequencesof microvascular damage in the OTT model. We found that damage resulted in diffuse microhemorrhage, which increased the tissue iron content, as measured histologically and by inductively coupled-mass spectroscopy. We then conducted a series of gain and loss-of-function experiments. In these experiments, the depth of A. fumigatus invasion progressively increased in: allografts beginning at day 8 post-transplantation, the onset of iron accumulation in the graft; iron rich tracheal transplants from mice with hemochromatosis, a genetic disorder that increases tissue iron content; allograft recipients treated with DFO; and syngrafts treated with a topical solution containing iron sulfate [47■■]. However, the depth of invasion decreased in allograft recipients treated with the iron chelator, deferasirox, which is not useful to A. fumigatus as a xenosiderophore [47■■]. We then posited thatA. fumigatus mutants with severe iron intolerance would be less invasive in the OTT model during high allograft iron conditions. These mutants included: ΔhapX, which lacks hapX – a gene that is required for adaptation to iron starvation through repression of iron-dependent pathways [36]; ΔsreA, which lacks the sreA gene encoding a repressor of iron uptake during high iron conditions [48]; and the double knockout iron intolerant mutant strain ΔsreA/ΔcccA, which lacks both sreA and cccA – a gene that is critical for A. fumigatus iron detoxification [49]. The wild-type strain (AF77, similar genetic background as the mutants) was studied concurrently and served as the control. In iron-overload conditions, the iron-intolerant mutant (ΔsreA/ΔcccA) was significantly (P < 0.05) less invasive than both the wild-type and the ΔsreA moulds. The data indicate that knockout of both sreA and cccA in A. fumigatus was required to lessen fungal invasion. Taken together, these data indicate that iron is a critical determinant of A. fumigatus allograft invasion and that this virulence can be mitigated by reducing tissue iron content.

CONCLUSION

The understanding of the role of iron in A. fumigatus pathogenicity has advanced enormously during the past decade. Iron is the central nutrient for A. fumigatus, and interference with iron availability is detrimental for the fungus. Inhibition of A. fumigatus siderophores has been shown to affect the fungal growth both in vitro and in vivo. The models described here suggest that iron availability is crucial not only for A. fumigatus proliferation but also plays a pivotal role in the fungal virulence. In addition, there is an increase in the clinical evidence, suggesting that immunocompromised patients with iron-overload after transplantation are at high risk of developing invasive aspergillosis. Having a better understanding of the role of iron in A. fumigatus pathogenicity will facilitate the development of treatment strategies and novel diagnostics for invasive aspergillosis. Inhibiting A. fumigatus siderophore synthesis pharmacologically, or use of certain chelators locally, could favor the host.

KEY POINTS.

• Interference with iron availability is detrimental for A. fumigatus.

• Loss of siderophores inhibits A. fumigatus proliferation.

• Iron availability is crucial for A. fumigatus invasiveness in vivo.

• Immunocompromised patients with iron overload post transplantation are at high risk of developing invasive aspergillosis.