THE METABOLISM OF IRON IN HOST

Dietary iron is found in haeme and ionic forms, and their absorption occurs in the proximal portion of the duodenum, carried out by enterocytes via different mechanisms (Andrews and Schmidt, Reference Andrews and Schmidt2007). Dietary non-haeme iron primarily exists in an oxidized (Fe³⁺) form that is not bioavailable and must be reduced to the ferrous Fe²⁺ form before it is transported across the intestinal epithelium. The responsible ferrireductase is a membrane bound haemoprotein called duodenal cytochrome B (DCYTB), expressed in the brush border of enterocytes (McKie et al. Reference McKie, Barrow, Latunde-Dada, Rolfs, Sager, Mudaly, Mudaly, Richardson, Barlow, Bomford, Peters, Raja, Shirali, Hediger, Farzaneh and Simpson2001). Then, a protein called divalent metal transporter 1 (DMT1) transports Fe²⁺ into enterocytes (Fig. 1A) (Morgan and Oates, Reference Morgan and Oates2002). After being transported into the enterocytes, Fe²⁺ can be accumulated inside the cell, used for cellular processes, or exported to the circulatory system by the basolateral membrane transporter ferroportin 1 (Abboud and Haile, Reference Abboud and Haile2000; Donovan et al. Reference Donovan, Brownlie, Zhou, Shepard, Pratt, Moynihan, Paw, Drejer, Barut, Zapata, Law, Brugnara, Lux, Pinkus, Pinkus, Kingsley, Palis, Fleming, Andrews and Zon2000). Once exported by ferroportin 1, iron must be transformed in process coupled by reoxidation of Fe²⁺ to Fe³⁺ by ferroxidases, such as hephaestin, and followed by the loading of Fe³⁺ onto transferrin (Tf) (Ganz and Nemeth, Reference Ganz and Nemeth2012; Musci et al. Reference Musci, Polticelli and Bonaccorsi di Patti2014). A transferrin can bind two atoms of ferric iron, which can limit iron-catalysed free radical production and facilitate transport to target cells (Fig. 1B) (Reyes-Lopez et al. Reference Reyes-Lopez, Pina-Vazquez and Serrano-Luna2015). Iron-loaded transferrin can bind with high affinity to the transferrin receptor 1 (TFR1) expressed ubiquitously at cell surfaces, and is then internalized by clathrin-dependent endocytosis (El Hage Chahine et al. Reference El Hage Chahine, Hemadi and Ha-Duong2012; Harris, Reference Harris2012). The endosomal acidification facilitates release of iron, and the TFR complex is recycled to the cell surface. Ferric iron released from transferrin is reduced by the ferrireductase six-transmembrane epithelial antigen of the prostate 3 (STEAP3) protein in erythrocytes and other STEAP proteins in non-erythroid cells (Ohgami et al. Reference Ohgami, Campagna, Greer, Antiochos, McDonald, Chen, Sharp, Fujiwara, Barker and Fleming2005, Reference Ohgami, Campagna, McDonald and Fleming2006), and subsequently transported into the cytoplasm by DMT1 (Graham et al. Reference Graham, Chua, Herbison, Olynyk and Trinder2007). From this point, iron can be incorporated into metalloproteins in complexes with haeme (e.g. catalase, cytochromes, haemoglobin and myoglobin), as mono and dinuclear iron (e.g. ribonucleotide reductase), or as Fe–S clusters (e.g. aconitase, succinate dehydrogenase) (Rouault and Tong, Reference Rouault and Tong2008; Dlouhy and Outten, Reference Dlouhy and Outten2013). Iron that is not needed for immediate use or export can be stored in ferritin, a spherical heteropolymer capable of storing up to 4500 iron atoms (Harrison and Arosio, Reference Harrison and Arosio1996).

Fig. 1. Iron transport and homeostasis in human cells. (A) Prior to transport into intestinal enterocytes, dietary ferric iron is reduced by ferric reductase duodenal cytochrome B (DCYTB), followed transported into cells by DMT1. Haeme absorption is performed by the haeme receptor FLVCR2, and the iron is extracted from haeme by haeme oxygenase. After absorption, iron can be used for cellular processes, stored in ferritin, or exported out of cells by ferroportin in partnership with hephaestin. (B) Macrophages gain iron via phagocytosis of senescent erythrocytes, TFR-mediated endocytosis of iron-loaded transferrin, or uptake of haeme–haemopexin and haemoglobin–haptoglobin complexes. Haeme in the phagolysosome is transported into the cytosol by HRG-1, where it can be degraded by HO to iron. This haeme may also be effluxed from the cell by the haeme exporter FLVCR1. Cytoplasmic iron be used for cellular processes, stored in ferritin, or exported out of macrophages by ferroportin coupled with ceruloplasmin.

The mechanism of dietary haeme uptake remains to be clarified. A candidate intestinal haeme transporter, haeme carrier protein 1 (HCP1), was proposed (Shayeghi et al. Reference Shayeghi, Latunde-Dada, Oakhill, Laftah, Takeuchi, Halliday, Khan, Warley, McCann, Hider, Frazer, Anderson, Vulpe, Simpson and McKie2005), but later evidence suggests it is most likely a folate transporter (Qiu et al. Reference Qiu, Jansen, Sakaris, Min, Chattopadhyay, Tsai, Sandoval, Zhao, Akabas and Goldman2006). The haeme responsive gene-1 (HRG-1), which was first identified in Caenorhabditis elegans as a haeme importer (Rajagopal et al. Reference Rajagopal, Rao, Amigo, Tian, Upadhyay, Hall, Uhm, Mathew, Fleming, Paw, Krause and Hamza2008), appears to transport haeme in human, but rather from the lysosome into the cytosol (Delaby et al. Reference Delaby, Rondeau, Pouzet, Willemetz, Pilard, Desjardins and Canonne-Hergaux2012). FLVCR2 (feline leukaemia virus, subgroup C receptor 2) was also recently reported to mediate the endocytosis of haeme by mammalian cells (Duffy et al. Reference Duffy, Shing, Saraon, Berger, Eiden, Wilde and Tailor2010). Once internalized in the enterocyte, iron is released from haeme by haeme oxygenases (HO) and then either stored or transported out of the enterocyte via mechanisms similar to that of ionic iron.

Although iron has critical functions in many cellular processes, iron overload is harmful because Fe²⁺ can generates highly reactive hydroxyl radicals in the presence of oxygen by the Fenton and Haber–Weiss reaction (Wessling-Resnick, Reference Wessling-Resnick2000; Weiss, Reference Weiss2002). For this reason, the iron metabolism in vivo is tightly regulated. Intracellular iron levels are regulated by an elegant machinery involving the iron regulatory proteins (IRPs) and the iron-responsive elements (IREs). In low iron conditions, the binding of IRPs to IREs represses mRNA translation when located in 5′-untranslated regions (e.g., ferritins) and stabilizes the mRNA when in 3′-untranslated regions (e.g, TFR1) (Torti and Torti, Reference Torti and Torti2002). In conditions of iron excess, the expression of ferritins is promoted and TFR1 is downregulated (Torti and Torti, Reference Torti and Torti2002). In addition to IRP-mediated regulation in cellular level, iron metabolism is also regulated systemically. The hormone hepcidin is regarded as the central regulator of systemic iron homeostasis. Hepcidin is a 25 amino acid peptide that is mainly produced by hepatocytes. It can bind to the iron exporter ferroportin, leading to its internalization and degradation. Thus, it reduces the iron entry into plasma after enterocyte absorption and the iron export after recycling by the reticuloendothelial system (Nicolas et al. Reference Nicolas, Viatte, Bennoun, Beaumont, Kahn and Vaulont2002).

Macrophages play a central role in the maintenance of iron homeostasis, delivering approximately 95% of the iron required for de novo haeme synthesis during erythropoiesis (Korolnek and Hamza, Reference Korolnek and Hamza2015). Apart from gaining iron from iron-loaded transferrin as described above, macrophages can engulf senescent erythrocytes via phagocytosis, and degrade their haemoglobin content in phagolysosome. Free haemoglobin and haeme in blood, which are from ruptured erythrocytes, can be trapped by haptoglobin and haemopexin, and taken up by macrophages via the CD163 receptor and CD91 receptor, respectively (Kristiansen et al. Reference Kristiansen, Graversen, Jacobsen, Sonne, Hoffman, Law and Moestrup2001; Hvidberg et al. Reference Hvidberg, Maniecki, Jacobsen, Hojrup, Moller and Moestrup2005). Labile haeme in phagolysosome is transferred to the cytoplasm by HRG-1 (Rajagopal et al. Reference Rajagopal, Rao, Amigo, Tian, Upadhyay, Hall, Uhm, Mathew, Fleming, Paw, Krause and Hamza2008; White et al. Reference White, Yuan, Schmidt, Bresciani, Samuel, Campagna, Hall, Bishop, Calicchio, Lapierre, Ward, Liu, Fleming and Hamza2013), and then can be secreted from macrophages via haeme exporters or degraded by HO, which extracts iron from the tetrapyrrole ring of haeme, resulting in the reutilization of iron (Korolnek and Hamza, Reference Korolnek and Hamza2015). The labile iron in macrophages can be excreted via ferroportin-1 or stored in macrophages by ferritin (Knutson et al. Reference Knutson, Oukka, Koss, Aydemir and Wessling-Resnick2005; Gozzelino and Soares, Reference Gozzelino and Soares2014). Iron export from macrophages is controlled systemically by hepcidin as described above. Hepcidin expression is mainly regulated at the transcriptional level, being induced or repressed in response to multiple stimuli (Nemeth et al. Reference Nemeth, Valore, Territo, Schiller, Lichtenstein and Ganz2003; Armitage et al. Reference Armitage, Eddowes, Gileadi, Cole, Spottiswoode, Selvakumar, Ho, Townsend and Drakesmith2011; Loreal et al. Reference Loreal, Cavey, Bardou-Jacquet, Guggenbuhl, Ropert and Brissot2014). The release of iron by HO induces ferritin synthesis via increasing the production of labile iron, which can block the capacity of iron regulatory proteins to repress translation of ferritin mRNA (Eisenstein et al. Reference Eisenstein, Garcia-Mayol, Pettingell and Munro1991).

The regulation of macrophages intracellular iron content affects innate and adaptive immune responses. Although the underlying mechanisms have not been entirely elucidated, iron perturbations regulate macrophages' response to interferon-γ (IFN-γ) (Weiss et al. Reference Weiss, Fuchs, Hausen, Reibnegger, Werner, Werner-Felmayer and Wachter1992; Oexle et al. Reference Oexle, Kaser, Most, Bellmann-Weiler, Werner, Werner-Felmayer and Weiss2003), which is a central cytokine for the regulation of antimicrobial effector mechanisms of macrophages and can promote antigen presentation (Nathan et al. Reference Nathan, Murray, Wiebe and Rubin1983; Mach et al. Reference Mach, Steimle, Martinez-Soria and Reith1996). Decreased iron availability also inhibits translation of proinflammatory cytokines TNF-α and IL-6 and reduces LPS-induced cytokine expression in macrophages (Wang et al. Reference Wang, Johnson, Shi, Walker, Wessling-Resnick and Cherayil2008, Reference Wang, Harrington, Trebicka, Shi, Kagan, Hong, Lin, Babitt and Cherayil2009). Via its inhibitory potential on IFN-γ, iron affects T-helper (Th) cell differentiation favouring the expansion of Th-2 cells that produce a number of macrophage-deactivating cytokines such as IL-4 or IL-13 (Weiss, Reference Weiss2002). Accordingly, iron overload and iron chelation have been reported to modulate the ratio of Th-1/Th-2 cells in mice infected with different microbes (Mencacci et al. Reference Mencacci, Cenci, Boelaert, Bucci, Mosci, Fe d'Ostiani, Bistoni and Romani1997; Ibrahim et al. Reference Ibrahim, Gebermariam, Fu, Lin, Husseiny, French, Schwartz, Skory, Edwards and Spellberg2007; Nairz et al. Reference Nairz, Schleicher, Schroll, Sonnweber, Theurl, Ludwiczek, Talasz, Brandacher, Moser, Muckenthaler, Fang, Bogdan and Weiss2013). Lactoferrin was demonstrated to protect the host against pathogens by influencing systemic immune responses through regulation of the Th-1/Th-2 cytokine balance (Fischer et al. Reference Fischer, Debbabi, Dubarry, Boyaka and Tome2006). Iron deficiency also significantly decreases hepatic T cell and natural killer T cell activation (Bonaccorsi-Riani et al. Reference Bonaccorsi-Riani, Danger, Lozano, Martinez-Picola, Kodela, Mas-Malavila, Bruguera, Collins, Hider, Martinez-Llordella and Sanchez-Fueyo2015). Furthermore, both NADPH oxidase (gp91phox) and inducible nitric oxide synthase (iNOS), which are critical for microbicidal activity of activated macrophages (Mastroeni et al. Reference Mastroeni, Vazquez-Torres, Fang, Xu, Khan, Hormaeche and Dougan2000; Vazquez-Torres et al. Reference Vazquez-Torres, Jones-Carson, Mastroeni, Ischiropoulos and Fang2000), are haemoproteins that require the insertion of haeme to catalyse the production of reactive oxygen species (ROS) and reactive nitrogen species (RNS) (Nathan and Shiloh, Reference Nathan and Shiloh2000). This argues that expression of these enzymes must be associated with mechanisms regulating haeme metabolism to support their activity. In support of this notion, haeme catabolism by HO-1 reduces haeme availability required to NADPH oxidase activity, acting therefore as a negative regulator of ROS production (Taille et al. Reference Taille, El-Benna, Lanone, Dang, Ogier-Denis, Aubier and Boczkowski2004). Haeme catabolism by HO-1 also inhibits cyclooxygenase-2 (COX2) expression and prostaglandin synthesis (Haider et al. Reference Haider, Olszanecki, Gryglewski, Schwartzman, Lianos, Kappas, Nasjletti and Abraham2002), which play an important role in host inflammatory responses (Rouzer and Marnett, Reference Rouzer and Marnett2009).

IRON LIMITATION IS AN INNATE IMMUNE DEFENCE

Another reason for tightly regulating iron metabolism is related to the fact that regulation of iron distribution can serves as an innate immune mechanism against invading pathogens. Successful colonization by pathogens in host requires that these must gain access to the required amount of iron. However, even in the absence of infection, the concentration of free ionic iron is very low in host (Cassat and Skaar, Reference Cassat and Skaar2013). Several facets contribute to this outcome. First, the majority of iron in humans is sequestered intracellularly, complexed within haemoglobin inside erythrocytes (Haley and Skaar, Reference Haley and Skaar2012). Second, extracellular iron is bound with iron-binding proteins transferrin, which have very high association constants for ferric iron and are normally only 30–40% saturated with iron (Bullen et al. Reference Bullen, Rogers, Spalding and Ward2005). Furthermore, even if transferrin-binding capacity is exceeded, iron also can be chelated by lower affinity molecules in plasma, including albumin, citrate and amino acids (Cassat and Skaar, Reference Cassat and Skaar2013).

During infection, additional fortification of iron-withholding defence occurs. Hepcidin is not only a major orchestrator in the integration of iron metabolism but it also provides a critical connection with immune response to infection and inflammation. In fact, hepcidin was initially characterized by an antimicrobial peptide in human urine and blood ultrafiltrate (Krause et al. Reference Krause, Neitz, Magert, Schulz, Forssmann, Schulz-Knappe and Adermann2000; Park et al. Reference Park, Valore, Waring and Ganz2001). Upon infection with pathogens, proinflammatory cytokines, TLR activation and ER stress can induce hepcidin expression and release from the liver, which results in hypoferraemia as described above (Drakesmith and Prentice, Reference Drakesmith and Prentice2012). This systemic iron retention serves to deplete circulating iron that would otherwise be available to extracellular pathogens. Furthermore, hepcidin can also be produced by myeloid cells in response to pathogens in a TLR4-dependent fashion, allowing for regulating the iron availability at the infectious focus (Peyssonnaux et al. Reference Peyssonnaux, Zinkernagel, Datta, Lauth, Johnson and Nizet2006). Some hepcindin-independent hypoferraemic mechanisms also exist, which further strengthen iron-withholding defences in response to infection (Weiss, Reference Weiss2005; Nairz et al. Reference Nairz, Schroll, Sonnweber and Weiss2010).

In addition to systemic induction of hypoferraemia, some innate immune effectors further sequester iron locally at infectious foci. Lactoferrin is an antimicrobial peptide, which can bind two ferric ions with a high affinity (Baker and Baker, Reference Baker and Baker2012). Constitutively high levels of lactoferrin, which produced by neutrophils and epithelial cells, are found in most surface secretions, including tears, saliva, bile and breast milk (Ward et al. Reference Ward, Uribe-Luna and Conneely2002). Consequently, lactoferrin functions as a first line defence molecule against invading pathogens through its ability to sequester iron. Moreover, proinflammatory cytokines such as TNF-α also induce the release of lactoferrin from neutrophilic granules at the site of infection (Afeltra et al. Reference Afeltra, Caccavo, Ferri, Addessi, De Rosa, Amoroso and Bonomo1997). Activated phagocytes at infection foci also limit iron availability to intracellular pathogens. Several proinflammatory cytokines, including IFN-gama, downregulate the expression of TFR1 in macrophages, thus reducing the intramacrophage access of iron (Byrd and Horwitz, Reference Byrd and Horwitz1989; Zhong et al. Reference Zhong, Lafuse and Zwilling2001; Olakanmi et al. Reference Olakanmi, Schlesinger, Ahmed and Britigan2002). Besides, the activation of macrophages enhances the expression of the SLC11A1 protein (solute carrier family 11 member a1, formerly termed Nramp1) (Vidal et al. Reference Vidal, Tremblay, Govoni, Gauthier, Sebastiani, Malo, Skamene, Olivier, Jothy and Gros1995; Jabado et al. Reference Jabado, Jankowski, Dougaparsad, Picard, Grinstein and Gros2000). SLC11A1 is intracellular iron transporter, which will be recruited to the late endosomal and phagosomal membrane. It can pump iron out of phagosome into the cytosol, thereby limiting availability to intracellular pathogens in this compartment (Zwilling et al. Reference Zwilling, Kuhn, Wikoff, Brown and Lafuse1999). It has been reported that polymorphisms in SLC11A1 is associated with susceptibility to tuberculosis, suggesting that iron limitation in phagosome is an important innate immune defence (van Crevel et al. Reference van Crevel, Parwati, Sahiratmadja, Marzuki, Ottenhoff, Netea, van der Ven, Nelwan, van der Meer, Alisjahbana and van de Vosse2009).

IRON METABOLISM IN LEISHMANIA INFECTION

Several studies have reported an association between susceptibility to leishmaniasis and polymorphisms in the SLC11A1 gene that codes for an intracellular iron transporter as mentioned before, which suggests that the host iron status may influence the outcome of the Leishmania infection (Bucheton et al. Reference Bucheton, Abel, Kheir, Mirgani, El-Safi, Chevillard and Dessein2003; Sanchez-Robert et al. Reference Sanchez-Robert, Altet, Utzet-Sadurni, Giger, Sanchez and Francino2008; Blackwell et al. Reference Blackwell, Fakiola, Ibrahim, Jamieson, Jeronimo, Miller, Mishra, Mohamed, Peacock, Raju, Sundar and Wilson2009; Castellucci et al. Reference Castellucci, Jamieson, Miller, Menezes, Oliveira, Magalhaes, Guimaraes, Lessa, de Jesus, Carvalho and Blackwell2010). However, the results concerning the impact of iron depletion on Leishmania infection are frequently contradictory. Some studies supported that the iron deprivation could be an effective strategy to control infections, similarly to what is found for other pathogens such as Mycobacterium tuberculosis. Incubation of Leishmania major promastigotes with iron-chelating compounds significantly suppresses parasite growth in a dose-dependent manner (Soteriadou et al. Reference Soteriadou, Papavassiliou, Voyiatzaki and Boelaert1995). A study about Leishmania chagasi infection in BALB/c mice showed that the iron chelator desferrioxamine (DFO) caused significant reduction in haemoglobin concentration of treated mice and reduction in parasite load in spleen and liver (Malafaia et al. Reference Malafaia, Marcon Lde, Pereira Lde, Pedrosa and Rezende2011). Similarly, DFO has shown an inhibitory effect on the intracellular growth of Leishmania donovani (Segovia et al. Reference Segovia, Navarro and Artero1989) and Leishmania amazonensis (Borges et al. Reference Borges, Vannier-Santos and de Souza1998). DFO treatment also leads to the reduction of parasite levels of Trypanosoma cruzi, a phylogenetic relative of Leishmania, and increases the efficacy of drug benznidazol (Arantes et al. Reference Arantes, Pedrosa, Martins, Veloso, de Lana, Bahia, Tafuri and Carneiro2007; Francisco et al. Reference Francisco, de Abreu Vieira, Arantes, Pedrosa, Martins, Silva, Veloso, de Lana, Bahia, Tafuri and Carneiro2008). However, using the same iron chelator, Murray et al. found iron depletion did not affect the intracellular growth of L. donovani in unstimulated human macrophages (Murray et al. Reference Murray, Granger and Teitelbaum1991). And dietary iron deficiency did not influence Leishmania infantum proliferation in the mouse model (Vale-Costa et al. Reference Vale-Costa, Gomes-Pereira, Teixeira, Rosa, Rodrigues, Tomas, Appelberg and Gomes2013). In a L. major infection study, treatment with DFO just resulted in a very slight delay of the development of cutaneous lesions on the mouse footpad whereas systemic iron delivery significantly reduced L. major-caused pathology and parasite loads (Bisti et al. Reference Bisti, Konidou, Papageorgiou, Milon, Boelaert and Soteriadou2000). The impact of iron supplementation is also not consistent. Thus iron supplementation enhances the growth of L. donovani in macrophages and hamsters (Das et al. Reference Das, Biswas, Solanki and Mukhopadhyay2009). By contrast, the iron administration to susceptible mice clearly leads to inhibition of the growth of L. major in the skin and L. infantum in the liver and spleen (Bisti and Soteriadou, Reference Bisti and Soteriadou2006; Vale-Costa et al. Reference Vale-Costa, Gomes-Pereira, Teixeira, Rosa, Rodrigues, Tomas, Appelberg and Gomes2013). Undoubtedly, iron acquisition is essential for leishmanial pathogenicity based on its fundamental role and iron deprivation could be an effective host strategy to control leishmanial infections, which supported by the fact that these parasites are equipped with efficient iron acquisition systems and these systems are essential for their intracellular growth (Huynh et al. Reference Huynh, Sacks and Andrews2006; Renberg et al. Reference Renberg, Yuan, Samuel, Miguel, Hamza, Andrews and Flannery2015). However, on the other hand, the iron status of macrophages is also important to the host defence. Adequate concentrations of iron are required to support macrophage-killing mechanisms during infection that were correlated to oxidative burst (Collins et al. Reference Collins, Kaufmann and Schaible2002; Vale-Costa et al. Reference Vale-Costa, Gomes-Pereira, Teixeira, Rosa, Rodrigues, Tomas, Appelberg and Gomes2013). Intracellular iron accumulation is also involved in pro-inflammatory activation of macrophages and development of Th-2-type immune response as described above. Therefore, the host defence mechanism regarding macrophage iron sequestration against pathogen proliferation has to maintain a difficult balance. The apparent discrepancy regarding the role of iron in Leishmania infection may reflect the different sensitivity of Leishmania species to iron deprivation and iron-induced host killing like reactive oxygen and nitrogen species. Indeed, It has been reported that Leishmania tropica promastigotes were considerably more susceptible than L. donovani to hydrogen peroxide, showing a 50% lethal dose (LD50) of 0·5 and 1·5 nm min⁻¹ respectively (Murray, Reference Murray1981). Leishmania braziliensis and L. infantum also exhibited different susceptibility to hydrogen peroxide and antimony tartrate in a parallel study (Andrade and Murta, Reference Andrade and Murta2014). And the sensitivities of Leishmania species to nitric oxide in vitro are different as well (Lemesre et al. Reference Lemesre, Sereno, Daulouede, Veyret, Brajon and Vincendeau1997). Furthermore, another point which we should note is that the chelating/loading efficiency of different iron depletion/supplementation protocols may be different.

HOW DO LEISHMANIA ACQUIRE IRON?

Like many other intracellular pathogens, Leishmania are also equipped with diverse iron acquisition mechanisms and are capable of utilizing various iron sources in order to thrive. Initial studies reported that Leishmania expressed a 70 kDa protein acting as TFR (Voyiatzaki and Soteriadou, Reference Voyiatzaki and Soteriadou1992), but subsequent studies suggested that the binding between this protein and transferrin is not specific (Wilson et al. Reference Wilson, Lewis, Miller, McCormick and Britigan2002). In the same report, it was reported that L. chagasi exhibits a NADPH-dependent iron reductase activity, capable of converting oxidized Fe³⁺ into the Fe²⁺ (Wilson et al. Reference Wilson, Lewis, Miller, McCormick and Britigan2002). This discovery strongly suggested the potential existence of a ferrous iron transport mechanism in Leishmania because there is extensive evidence that ferric iron reduction is usually coupled to ferrous iron transport in prokaryotic and eukaryotic cells (Kosman, Reference Kosman2010; Caza and Kronstad, Reference Caza and Kronstad2013). Indeed, a membrane Fe²⁺ transporter (LIT1) in L. amazonensis was identified and characterized subsequently (Huynh et al. Reference Huynh, Sacks and Andrews2006). As with other members of ZIP (Zrt/IRT-like proteins) transporter family, LIT1 has eight predicted transmembrane domains (Huynh et al. Reference Huynh, Sacks and Andrews2006). By the year 2011, a membrane protein with ferric reductase activity in L. amazonensis (LFR1) was identified (Flannery et al. Reference Flannery, Huynh, Mittra, Mortara and Andrews2011). LFR1 is a predicted 11 transmembrane protein with a molecular mass of 119 kDa, including an apparent signal peptide in N-terminal and the FAD- and NADPH-binding sites in tis C-terminal (Flannery et al. Reference Flannery, Huynh, Mittra, Mortara and Andrews2011). Together, LIT1 and LFR1 provide Leishmania with an inorganic iron acquisition pathway (Fig. 2A). These two proteins are localized to the plasma membrane reflecting that L. amazonensis can obtain iron directly from the lumen of the phagosome. Both LFR1 and LIT1 null mutants lost the ability to replicate in macrophages, indicating that this pathway of iron acquisition is essential for the ability of Leishmania to cause disease (Huynh et al. Reference Huynh, Sacks and Andrews2006; Flannery et al. Reference Flannery, Huynh, Mittra, Mortara and Andrews2011). Moreover, the overexpression of LFR1 was found to have toxic effects on the wild-type parasites, but was not toxic for LIT1 null mutants, which suggested that the toxicity of LFR1 overexpression is likely a direct consequence of increased intracellular Fe²⁺ concentration (Flannery et al. Reference Flannery, Huynh, Mittra, Mortara and Andrews2011). Interestingly, LFR1 overexpression also rescues the growth defect of LIT1 null mutants in macrophages by likely increasing available Fe²⁺ concentration (Flannery et al. Reference Flannery, Huynh, Mittra, Mortara and Andrews2011). This iron uptake mechanism is also a major trigger for the differentiation of L. amazonensis because both LIT1 and LFR1 null mutants are defective in differentiation of promastigotes into infective amastigotes after inducing by iron depletion, unlike the wild-type parasite (Flannery et al. Reference Flannery, Huynh, Mittra, Mortara and Andrews2011; Mittra et al. Reference Mittra, Cortez, Haydock, Ramasamy, Myler and Andrews2013). Iron depletion leads to ROS production in parasites by affecting Fe–S clusters and impairing the function of the mitochondrial electron transport chain. Followed up-expression of LFR1 and LIT1 induced by low iron conditions results in an increase of intracellular iron concentration and activation of Fe-SOD activity that converts the superoxide into H₂O₂, a molecule which is sufficient to trigger Leishmania differentiation signal (Mittra and Andrews, Reference Mittra and Andrews2013). Recently, a mitochondrial iron transporter LMIT1 was identified and was demonstrated to be crucial for Leishmania mitochondrial function and virulence (Mittra et al. Reference Mittra, Laranjeira-Silva, Perrone Bezerra de Menezes, Jensen, Michailowsky and Andrews2016). This study also supported that iron-dependent ROS signals generated in the mitochondria regulate differentiation of Leishmania amastigotes (Mittra et al. Reference Mittra, Laranjeira-Silva, Perrone Bezerra de Menezes, Jensen, Michailowsky and Andrews2016). It was reported that the procyclic form of Trypanosoma brucei, whose bloodstream form acquires iron from transferrin by receptor-mediated endocytosis, takes up ionic iron via a reductive mechanism, perhaps similar with LFR1-LIT1 system in Leishmania (Mach et al. Reference Mach, Tachezy and Sutak2013). This study suggests that parasites may activate different iron acquisition systems in their different life cycle stages.

Fig. 2. Iron acquisition pathways and its role in Leishmania. (A) Iron acquisition in Leishmania. The host transporter SLC11A1 (Nramp1) transports iron into cytoplasm from PVs, thereby limiting iron availability to Leishmania. To compete with host, intracellular leishmanial parasites upregulate the expression of ferric iron reductase LFR1 on their plasma membrane, converting Fe³⁺ into Fe²⁺. Fe²⁺ is then transported into leishmanial cytosol by LIT1. Iron-containing haeme can be transported into leishmanial cytosol by hexokinase-mediated endocytosis pathway or by the haeme transporter LHR1 directly. When high concentrations of iron are available, low-affinity iron transporters contribute to the iron acquisition of Leishmania. (B) The role of iron in Leishmania. Cytochrome b5 that is small haeme-binding protein acts as an electron-transfer component in the desaturation reaction, which catalysed by fatty acid desaturases. An important step in the synthesis of ergosterol, an essential component of parasite membranes, is carried out by a sterol-14-α-demethylase CYP51. Haeme-binding protein cytochrome p450 is as cofactor of CYP51 accepting electrons. Mitochondrial cytochrome c is a haemeprotein and plays a role in the electron transfer during oxidative phosphorylation. Iron-dependent superoxide dismutases (Fe-SODs) in Leishmania help to protect parasites from oxidative stress by catalysing the dismutation of superoxide radicals into H₂O₂ and O₂, the H₂O₂ then being metabolized by peroxidases. Iron is also a cofactor of ribonucleotide reductases in Leishmania, which catalyse de novo biosynthesis of deoxynucleotides that are used in the synthesis of DNA.

The growth of Leishmania also can be supported by iron derived from heme. Heme also functions an essential group for many enzymes, involved in a variety of critical cellular functions. However, Leishmania lack a complete haeme biosynthetic pathway and only can perform the last three steps catalysed by coproporphyrinogen oxidase, protoporphyrinogen oxidase and ferrochelatase, localized in the mitochondria (Alves et al. Reference Alves, Voegtly, Matveyev, Lara, da Silva, Serrano, Buck, Teixeira and Camargo2011). To survive, these parasites must acquire haeme from the environment. The data available to date indicates that Leishmania can acquire haeme by two mechanisms, haemoglobin receptor-mediated endocytosis and direct transmembrane transport. An initial report revealed that haemoglobin endocytosis in Leishmania is mediated through a 46-kDa protein located in the flagellar pocket, which was identified as a hexokinase subsequently (Krishnamurthy et al. Reference Krishnamurthy, Vikram, Singh, Patel, Agarwal, Mukhopadhyay, Basu and Mukhopadhyay2005). But how this glycolytic enzyme is exported to cell surface remained elusive until more recently. After initial binding with hexokinase, the haemoglobin is internalized into discrete intracellular compartment and subsequently targeted to lysosome for degradation. Haemoglobin endocytosis in Leishmania is a clathrin dependent process (Agarwal et al. Reference Agarwal, Rastogi, Gupta, Patel, Raje and Mukhopadhyay2013). Its early step is regulated by a Rab5 homologue whereas transport of haemoglobin to the lysosomal compartment is controlled by Rab7 (Singh et al. Reference Singh, Tandon, Krishnamurthy, Vikram, Sharma, Basu and Mukhopadhyay2003; Patel et al. Reference Patel, Singh, Basu and Mukhopadhyay2008). The L. donovani ATP-Binding Cassette G5 (LABCG5) was proposed to be involved in haeme salvage after breakdown of haemoglobin in lysosome (Campos-Salinas et al. Reference Campos-Salinas, Cabello-Donayre, Garcia-Hernandez, Perez-Victoria, Castanys, Gamarro and Perez-Victoria2011). The direct transport of extracellular haeme into the parasites's cytosol is mediated by LHR1 protein in L. amazonensis, which shares homology with HRG-4, a plasma membrane haeme importer in C. elegans (Huynh et al. Reference Huynh, Yuan, Miguel, Renberg, Protchenko, Philpott, Hamza and Andrews2012). The LHR1 protein, with a molecular mass of about 20 kDa, localizes dually to the plasma membrane and lysosomes, regulating the intracellular pool of haeme in the parasites (Huynh et al. Reference Huynh, Yuan, Miguel, Renberg, Protchenko, Philpott, Hamza and Andrews2012). The LHR1 null strain could not be generated, suggesting that this transporter is essential for the survival of promastigote forms of L. amazonensis (Huynh et al. Reference Huynh, Yuan, Miguel, Renberg, Protchenko, Philpott, Hamza and Andrews2012). And the LHR1-single-knockout strain was not able to replicate in macrophages and was severely defective in the development of cutaneous lesion in mice (Miguel et al. Reference Miguel, Flannery, Mittra and Andrews2013). Recently, Renberg et al. demonstrated that three key tyrosine residues within predicted transmembrane domains of LHR1 are required for efficient haeme uptake and L. amazonensis virulence (Renberg et al. Reference Renberg, Yuan, Samuel, Miguel, Hamza, Andrews and Flannery2015). As these tyrosine residues are unique to LHR1, not existing in its corresponding homologue in human, this study also implied that LHR1 could be an effective target for developing anti-leishmanial drugs. A recent study demonstrated that a mitochondrial ABC half-transporter LmABCB3 in L. major, which is essential for Leishmania, is involved in mitochondrial haeme biosynthesis from cytosolic protoporphyrin IX and required for the maturation of cytosolic iron/sulphur clusters (Martinez-Garcia et al. Reference Martinez-Garcia, Campos-Salinas, Cabello-Donayre, Pineda-Molina, Galvez, Orrego, Sanchez-Canete, Malagarie-Cazenave, Koeller and Perez-Victoria2016).

As mentioned above, high iron concentrations can generate highly toxic hydroxyl radicals, which suggest that the iron uptake machinery should be tightly regulated. Indeed, the expression of LFR1 ferric reductase and LIT1 ferrous iron transporter is very low in promastigotes exposed to iron-rich condition and just upregulated in the response to iron deprivation although how Leishmania parasites sense extracellular iron concentrations remains to be clarified (Mittra et al. Reference Mittra, Cortez, Haydock, Ramasamy, Myler and Andrews2013). Moreover, overexpression of LFR1 is toxic for parasites when LIT1 is functional (Flannery et al. Reference Flannery, Huynh, Mittra, Mortara and Andrews2011). However, how Leishmania regulate their iron responsive genes is unclear. Trypanosomatid parasites do not regulate these genes expression by classic transcriptional promoters, utilizing instead post-transcriptional strategies such as regulation of mRNA stability or of translation initiation, which is thought to be mediated by control elements in the 5′ and 3′-untranslated regions (Flannery et al. Reference Flannery, Renberg and Andrews2013). A protein which binds mammalian iron response elements (IREs) was identified in L. tarentolae (Meehan et al. Reference Meehan, Lundberg and Connell2000). However, whether the mechanism of regulating iron-responsive genes in Leishmania is similar with the mammalian IRE/IRP paradigm remains elusive and future studies are needed. Another interesting question is how Leishmania parasites store iron. Although ferritins are found in bacteria, higher eukaryotes, but it has failed to yield a putative ferritin orthologue in published genomes of various Leishmania species. Yeast cells also lack ferritin but contain frataxins, which can bind and supply iron for assembly of Fe-S clusters (Adamec et al. Reference Adamec, Rusnak, Owen, Naylor, Benson, Gacy and Isaya2000). Genes encoding frataxin-like proteins are present in all trypanosomatids including Leishmania. However, although frataxin is essential for Fe-S cluster biogenesis in Trypansoma brucei, it seems not play a role in iron storage as T. brucei frataxin does not form large complexes and knockdown of frataxin did not cause iron accumulation in mitochondria (Long et al. Reference Long, Jirku, Mach, Ginger, Sutak, Richardson, Tachezy and Lukes2008). The role of frataxin in Leishmania still needs to be investigated. Another possibility is that Leishmania may store iron in a mineralized form like the Ectocarpus siliculosus that also lack ferritin (Bottger et al. Reference Bottger, Miller, Andresen, Matzanke, Kupper and Carrano2012). Furthermore, it was reported that moderate amount of ionic iron can inhibit the growth or even kill leishimanial promastigotes and amastigotes (Mittra et al. Reference Mittra, Cortez, Haydock, Ramasamy, Myler and Andrews2013; Vale-Costa et al. Reference Vale-Costa, Gomes-Pereira, Teixeira, Rosa, Rodrigues, Tomas, Appelberg and Gomes2013). Thus, the possibility that Leishmania have not an efficient iron storage mechanism may also exist. Anyway, additional studies are needed to clarify the alternative mechanism used by Leishmania for iron storage. These mechanisms may also contribute to differential sensitivity to environment iron concentrations reported for different Leishmania species.

SUBVERTING HOST'S IRON SYSTEMS

Apart from the mechanisms of iron uptakes, Leishmania parasites also are armed to subvert the host's iron metabolism system to their own advantage. An early study revealed that the Leishmania pifanoi infection promotes the delivery of human transferrin to the parasitophorous vacule, leading to acquisition transferrin-carried iron by parasites (Borges et al. Reference Borges, Vannier-Santos and de Souza1998). Furthermore, L. donovani can directly deplete labile iron pool of macrophages to activate iron-sensing proteins (IRPs). IRPs then increase the TFR1 expression to increase host iron uptake, so that L. donovani could use the iron for its growth (Das et al. Reference Das, Biswas, Solanki and Mukhopadhyay2009). The decrease of labile iron pool caused by L. donovani was subsequently thought to be due to the decreased expression of SLC11A1 by a secretory peroxidase Prx (Singh et al. Reference Singh, Bajpai, Kumar, Gour and Singh2013). This peroxidase also can significantly decrease reactive oxygen/nitrogen species levels and proinflammatory cytokines levels in LPS activated macrophages (Singh et al. Reference Singh, Bajpai, Kumar, Gour and Singh2013). Hepcidin is a key regulator of host iron homeostasis. Recently, a study demonstrated that L. amazonensis infection can increase the hepcidin levels and lower the ferroportin content of macrophages in a TLR4-dependent manner, and thus increased total macrophage iron content for stimulating parasite intracellular replication (Ben-Othman et al. Reference Ben-Othman, Flannery, Miguel, Ward, Kaplan and Andrews2014). The mechanism by which Leishmania regulates hepcidin levels remains unknown. The zinc metalloprotease GP63, a leishmanial pathogenicity factor, has shown a capacity of augmenting TNF and IL-6 release probably by degrading Synaptotagmin XI (Arango Duque et al. Reference Arango Duque, Fukuda, Turco, Stager and Descoteaux2014). IL-6 is the necessary and sufficient cytokine for the induction of hepcidin during inflammation (Nemeth et al. Reference Nemeth, Rivera, Gabayan, Keller, Taudorf, Pedersen and Ganz2004). Thus, the GP63-mediated IL-6 secretion possibly is responsible for the higher hepcidin levels in Leishmania-infected macrophages. Moreover, in infected host cells, Leishmania parasites are sheltered within parasitophorous vacuoles (PVs), which raises a question that how parasites make iron enter the PVs. Iron acquisition from extracellular transferrin by M. tuberculosis occurs through delivery of Fe-transferrin to the phagosome via receptor-mediated endocytosis and fusion of the early endosome with the M. tuberculosis-containing phagosome (Olakanmi et al. Reference Olakanmi, Schlesinger, Ahmed and Britigan2002). The co-localization of human transferrin with the parasitophorous vacule was observed in Leishmania-infected macrophages, which suggests that the similar mechanism may exist. It has also been known that the iron stored within cytosolic ferritin can be mobilized and released through lysosomal proteolysis following autophagy from the cytosol into lysosomes (Linder, Reference Linder2013). Whether PVs can fuse with these lysosomal vacuoles is an interesting question and need to be investigated. Leishmania has shown capabilities to manipulate intracellular vesicle trafficking. For example, Leishimania GP63 can inhibit recruiting gp91phox and thus hinder antigen cross-presentation by degrading VAMP8, which is a key protein that mediate vesicle trafficking (Matheoud et al. Reference Matheoud, Moradin, Bellemare-Pelletier, Shio, Hong, Olivier, Gagnon, Desjardins and Descoteaux2013). Leishmania surface glycolipid lipophosphoglycan (LPG), which can cause periphagosomal accumulation of F-actin and disruption of phagosomal lipid microdomains, functionally inhibited phagosome maturation by an impaired assembly of the NADH oxidase and the exclusion of the vesicular proton-ATPase from phagosome (Lodge and Descoteaux, Reference Lodge and Descoteaux2005; Lodge et al. Reference Lodge, Diallo and Descoteaux2006; Vinet et al. Reference Vinet, Fukuda, Turco and Descoteaux2009). However, whether and how Leishmania interfere with iron-relating vesicle trafficking in macrophages remain to be studied.