A vacuolar iron-transporter homologue acts as a detoxifier in Plasmodium

Abstract

Iron is an essential micronutrient but is also highly toxic. In yeast and plant cells, a key detoxifying mechanism involves iron sequestration into intracellular storage compartments, mediated by members of the vacuolar iron-transporter (VIT) family of proteins. Here we study the VIT homologue from the malaria parasites Plasmodium falciparum (PfVIT) and Plasmodium berghei (PbVIT). PfVIT-mediated iron transport in a yeast heterologous expression system is saturable (K_m ∼14.7 μM), and selective for Fe²⁺ over other divalent cations. PbVIT-deficient P. berghei lines (Pbvit^- ) show a reduction in parasite load in both liver and blood stages of infection in mice. Moreover, Pbvit^- parasites have higher levels of labile iron in blood stages and are more sensitive to increased iron levels in liver stages, when compared with wild-type parasites. Our data are consistent with Plasmodium VITs playing a major role in iron detoxification and, thus, normal development of malaria parasites in their mammalian host.

ARTICLE

Received 16 May 2015 | Accepted 7 Dec 2015 | Published 20 Jan 2016

Ksenija Slavic1, Sanjeev Krishna2, Aparajita Lahree1, Guillaume Bouyer2,3, Kirsten K. Hanson1,w, Iset Vera1, Jon K. Pittman4, Henry M. Staines2 & Maria M. Mota1

Iron is an essential micronutrient but is also highly toxic. In yeast and plant cells, a key detoxifying mechanism involves iron sequestration into intracellular storage compartments, mediated by members of the vacuolar iron-transporter (VIT) family of proteins. Here we study the VIT homologue from the malaria parasites Plasmodium falciparum (PfVIT) and Plasmodium berghei (PbVIT). PfVIT-mediated iron transport in a yeast heterologous expression system is saturable (KmB14.7 mM), and selective for Fe2 over other divalent cations. PbVIT-decient P. berghei lines (Pbvit ) show a reduction in parasite load in both liver and blood stages of infection in mice. Moreover, Pbvit parasites have higher levels of labile iron in blood stages and are more sensitive to increased iron levels in liver stages, when compared with wild-type parasites. Our data are consistent with Plasmodium VITs playing a major role in iron detoxication and, thus, normal development of malaria parasites in their mammalian host.

DOI: 10.1038/ncomms10403 OPEN

A vacuolar iron-transporter homologue acts as a detoxier in Plasmodium

1 Instituto de Medicina Molecular, Faculdade de Medicina Universidade de Lisboa, 1649-028 Lisbon, Portugal. 2 Institute for Infection & Immunity,St. Georges, University of London, London SW17 0RE, UK. 3 Sorbonne Universits, UPMC Univ Paris 6, CNRS, UMR 8227, Comparative Physiology of Erythrocytes, Station Biologique de Roscoff, CS 90074, 29688 Roscoff, France. 4 Faculty of Life Sciences, University of Manchester, Manchester M13 9PT, UK.

w Present address: University of Texas at San Antonio, Department of Biology and STCEID, San Antonio, Texas 78249, USA. Correspondence and requests for materials should be addressed to H.M.S. (email: mailto:[email protected]

Web End [email protected] ) or to M.M.M. (email: mailto:[email protected]

Web End [email protected] ).

NATURE COMMUNICATIONS | 7:10403 | DOI: 10.1038/ncomms10403 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 1

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10403

Malaria imposes a massive global health burden, with a current WHO estimate of around 600,000 deaths annually1, although this gure could rise sharply if

treatment failures associated with artemisinin-combination therapies becomes widespread. Perturbation of iron homeostasis is an attractive strategy to target malaria parasites as Plasmodium, like all cells, requires iron to survive2,3. Indeed, iron is an essential micronutrient, necessary for fundamental cellular processes such as ATP and DNA synthesis. While irons redox-active nature makes it essential for many catalytic processes, it also underlies its toxicity when present at high concentrations in the cytosol or other sites. As such, all organisms have evolved a wide range of strategies to acquire necessary iron and to detoxify any excesses.

Iron enters cells through the activity of membrane transporters or receptor-mediated endocytosis. While most of the intracellular iron gets safely stored either in complex with the iron-storage protein ferritin (for example, plant and mammalian cells) or inside organelles (for example, yeast and plant cells), a small amount of total cellular iron is present as a metabolically available pool in the cytoplasm, termed the labile iron pool (LIP). The LIP consists of iron loosely bound to small negatively charged molecules and proteins4 and provides iron for cellular processes including haem and FeS cluster biosyntheses5.

Appropriate storage of any excess of iron, which is not used metabolically, is essential to prevent cellular toxicity due to engagement of iron in Fenton-type chemistry in the presence of oxygen and production of potentially damaging reactive oxygen species6. Ferritin represents the most common and ancient mechanism of iron storage and homeostasis in nature, as it is found in most bacteria, archaea, plants and animals, but not in yeast7. In the absence of ferritin, the yeast vacuole serves as the main iron-storage/sequestration organelle. In response to demands, iron moves to and from the yeast vacuole through the activity of iron transporters in the yeast vacuolar membrane; CCC1 (Ca2 -sensitive cross complementer 1) is proposed to import iron, while a complex constituted by Smf3p and

Fet5p-Fth1p exports iron810. Thus, vacuolar sequestration by CCC1 in yeast is likely to be the primary mechanism for detoxication of excess iron in this organism. In addition to ferritin found in plastids, plants also have several homologues of CCC1, named vacuolar iron transporters (VITs), which are likely to transport not only iron but also other divalent cations such as manganese and zinc into the vacuole for storage and detoxication1113.

Plasmodium parasites do not contain a homologue of ferritin or any other known iron-storage protein14. However, Plasmodium spp. genomes contain one orthologue of plant VIT and yeast CCC1 proteins. Thus, we hypothesized that mechanisms regulating intracellular iron storage in malaria parasites may resemble those of yeast and plants. To test this hypothesis, we assessed the ability of Plasmodium falciparum VIT, PfVIT, to transport iron and sought to reveal the role of this protein in the establishment and course of a malaria infection by generating Plasmodium berghei parasites decient in VIT (Pbvit ). We now show that Plasmodium VITs, expressed throughout the parasites life cycle, transport iron and play a major role in iron detoxication in the parasite.

ResultsPlasmodium genomes contain a VIT homologue. Bioinformatic analysis shows that all Plasmodium spp. genomes encode one gene with homology to yeast CCC1 and plant VIT proteins PF3D7_1223700 in P. falciparum and PBANKA_143860 inP. berghei. PfVIT shares 28% amino-acid sequence identity with

Arabidopsis thaliana VIT, AtVIT1, and 19% with Saccharomyces cerevisiae CCC1 (ClustalW2 multiple protein alignment). Like VIT homologues in other organisms, both PfVIT and PbVIT are predicted to have ve transmembrane domains (Supplementary Fig. 1). PfVIT has four phospho-acceptor sites at amino-acid positions 21S, 122S, 140S and 150T, as reported previously15,16. VITs are highly conserved across Plasmodium spp., with open reading frames (ORF) split between two exons. Orthologues in the species infecting humans (P. falciparum,P. vivax and P. knowlesi) form a separate phylogenetic branch from the orthologues of rodent malaria species (P. yoelii,P. chabaudi and P. berghei; Fig. 1a). VIT homologues are also present in other human pathogens of the Apicomplexa phylum, such as Toxoplasma and Cryptosporidium. Furthermore, kinetoplastid parasites Trypanosoma and Leishmania, causing sleeping sickness and leishmaniasis, respectively, also encode homologues of VIT proteins (Fig. 1a).

PfVIT complements a yeast strain lacking CCC1. S. cerevisiae CCC1 most likely transports Fe2 and Mn2 into the vacuole, based on the effect of CCC1 deletion and overexpression on vacuolar accumulation of both of these ions8. Mutants lacking the encoding gene for CCC1 (DCCC1) are susceptible to iron toxicity in the presence of high extracellular iron concentrations but can be rescued by expression of homologous plant VITs, such as AtVIT1 (ref. 11), Tulipa gesneriana VIT1 (ref. 12) and Oryza sativa VIT1 and VIT2 (ref. 13). To assess whether Plasmodium VITs could rescue DCCC1 susceptibility to iron toxicity, a yeast expression vector (pUGpd) containing codon-optimized full-length pfvit (pfvit pUGpd; encoding 273 amino acids) and 50-truncated pfvit (spfvit-pUGpd; encoding 236 amino acids and containing all 5 predicted transmembrane domains, Supplementary Fig. 1) were transformed into S. cerevisiae DCCC1. The latter was used because truncated transporters in heterologous expression systems often demonstrate increased activity, for example, because of autoinhibitory effects17. Parallel transformation with empty vector (pUGpd) and vector containing CCC1 (CCC1-pUGpd) were used as negative and positive controls, respectively. Having conrmed that the appropriate sized gene sequences were present in DCCC1::PfVIT, DCCC1::sPfVIT and DCCC1::pUGpd yeast strains, western blot analysis of whole cell protein extract demonstrated that full-length PfVIT heterologous protein expression was very low compared with that of the truncated version, with the latter being easily detectable in vacuolar membrane fractions (Supplementary Fig. 1). Subsequently, yeast growth assays on solid (qualitative) and in liquid (quantitative) media were performed (Fig. 1b,c). The results show that very low-level expression of PfVIT provided moderate but signicant rescue of the DCCC1 phenotype at 2 and 3 mM

Fe2 (Po0.01, Students t-test; Fig. 1c). However, highly expressed sPfVIT signicantly and strongly rescued the

DCCC1 phenotype at all tested Fe2 concentrations (Po0.01; Fig. 1c), although not to the degree measured in the

CCC1-expressing yeast positive control (which may be the result of relatively low protein expression of heterologous sPfVIT sequence compared with that of CCC1, or else it may represent true differences in transport properties between these two transporters, such as in substrate afnity and transporter capacity). Additionally, Zn2 tolerance was also tested by expression of sPfVIT in a Dzrc1 yeast strain, which is sensitive to increased Zn2 concentrations due to the lack of a vacuolar Zn2 importer18,19. Expression of sPfVIT did not rescue Dzrc1 from sensitivity to high external Zn2 concentration (Supplementary Fig. 2).

2 NATURE COMMUNICATIONS | 7:10403 | DOI: 10.1038/ncomms10403 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10403 ARTICLE

a

Saccharomyces cerevisiae CCC1 Plasmodium yoelii VIT

Plasmodium chabaudi VIT

Plasmodium berghei VIT

Plasmodium vivax VIT

Plasmodium knowlesi VIT

Plasmodium falciparum VIT

Cryptosporidium parvum VIT

Toxoplasma gondii VIT

Trypanosoma brucei VIT

Leishmania braziliensis VIT

Arabidopsis thaliana VIT1

No added Fe2+ 3 mM Fe2+ 5 mM Fe2+

0.0 0.1

b

pUGpd

OD600

PfVIT

sPfVIT

CCC1

0.2 0.04 0.008 0.0016

c

pUGpd PfVIT sPfVIT CCC1

125

100

75

50

25

0 0

**

*** *** *** *** ***

*** **

*** **

OD 600 % of control

1 2 3 5

Fe2+ (mM)

Figure 1 | Iron tolerance of DCCC1 yeast conferred by the functional expression of PfVIT. (a) Phylogenetic analyses of Plasmodium VIT proteins compared with other Apicomplexa and Kinetoplastida parasites and unrelated model organisms. The phylogenetic tree was generated using Phylogeny.fr59 with yeast CCC1 sequence used as an out-group and visualized with TreeGraph 2 (ref. 60). The scale bar for the branch lengths is shown. (b) Complementation of the DCCC1 yeast mutant growth phenotype by expression of PfVIT. The DCCC1 strain (lacking vacuolar iron uptake) was transfected with empty vector pUGpd or vector expressing full-length PfVIT, N-terminally truncated PfVIT (sPfVIT) or CCC1. Transfected strains were diluted (as indicated by OD600 values), spotted onto SD agar plates lacking histidine and supplemented with: no additional Fe2 , 3 mM Fe2 and 5 mM

Fe2 and grown at 30 C for 48 h. Fe2 was provided as ammonium FeSO4 in the presence of 1 mM ascorbic acid. (c) The DCCC1 transfectants described in b were inoculated at a cell density of 0.01 OD600 in SD medium in the presence of indicated Fe2 concentrations (provided as ammonium FeSO4 in the presence of 1 mM ascorbic acid) and grown with shaking at 30 C for 20 h. Yeast cell density was determined by absorbance measurements at 600 nm and for each strain shown the OD600 values are normalized to those obtained in the absence of additional Fe2 in the medium (controls, 0 mM Fe2 ).

Data are shown as the meanss.e.m. of a pool of three independent experiments, each performed in triplicate. The data were analysed using the unpaired, two-tailed Students t-test; **Po0.01, ***Po0.001.

These data show that PfVIT, particularly when truncated, can be expressed in S. cerevisiae and can complement the lack of CCC1, implying that PfVIT acts as an iron transporter.

PfVIT mediates Fe2 uptake. We next decided to test the ability of sPfVIT to transport Fe2 . We measured uptake of 55Fe2 into vacuolar vesicles isolated from transfected DCCC1::sPfVIT yeast. 55Fe2 uptake was over twofold higher in these vacuoles, when compared with vector-only control vacuole preparations (DCCC1::pUGpd) that were made at the same time, reaching 80%

of maximum uptake after 5 min of incubation at 20 C (Fig. 2a). The same fold increase in 55Fe2 uptake was also observed between paired sPfVIT and control vacuolar vesicles made on different days, even though uptake relative to total protein levels varied considerably (compare uptake presented in Fig. 2a with that presented in Supplementary Fig. 3a). Using the same paired

vacuolar preparations used to generate the data presented in Fig. 2a, linear-phase 55Fe2 inux in DCCC1::sPfVIT isolated vacuoles measured over 1 min was 4.20.3 pmol min 1 mg 1 protein (means.e.m.; n 9), which was signicantly higher

than that measured in the same vacuoles placed on ice(1.90.3 pmol min 1 mg 1 protein; means.e.m.; n 9;

P 0.0003; two-tailed, Students t-test) and in control vacuoles

(2.10.3 pmol min 1 mg 1 protein; means.e.m.; n 8;

P 0.0002; two-tailed, Students t-test). sPfVIT-mediated 55Fe2

inux (dened as the inux in DCCC1::sPfVIT vacuoles minus that measured in DCCC1::pUGpd vacuoles and normalized to account for the variability noted above when averaging data generated from more than one paired vacuolar preparation) was inhibited by unlabelled Fe2 in a concentration-dependent manner (Fig. 2b) and was found to be pH-sensitive, with an optimum between pH 6.5 and 7.5 (Fig. 2c). In competition assays with divalent metals at 50-fold higher concentrations (that is, at

NATURE COMMUNICATIONS | 7:10403 | DOI: 10.1038/ncomms10403 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 3

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10403

a b

sPfVIT

pUGpd

55 Fe influx

(pmol g1 of protein)

12 10

8 6 4 2 0

0 5 10 15 20

Time (min)

55 Fe influx

(% of uninhibited control)

c

d

55 Fe influx

(% of controls at pH 7)

125

100

75

50

25

0

**

55 Fe influx

(% of uninhibited control)

100

75

50

25

0 6.0 5.5 5.0 4.5 4.0

Log10[FeSO4](M)

6.0 6.5 7.0 7.5 8.0

pH

150 125 100

75 50 25

0

Control

FeSO 4

ZnSO 4

MnCl 2

CaCl 2

CuCl 2

Bafilomycin

CCCP

Figure 2 | Characterization of iron transport by PfVIT. 55Fe inux was measured into vacuole-enriched vesicles isolated from DCCC1 yeast transfected with empty or sPfVIT-expressing pUGpd vectors. (a) 55Fe uptake over time, measured at pH 7. (b) Inhibition of sPfVIT-mediated 55Fe2 inux (dened as the inux in DCCC1::sPfVIT isolated vacuoles minus that measured in DCCC1::pUGpd vacuoles) by cold unlabelled iron, measured over 1 min at pH 7. Shown is sPfVIT-mediated 55Fe2 inux in the presence of increasing concentrations of FeSO4 normalized to the condition without FeSO4, to allow comparison between independent vacuolar preparations. (c) pH dependence of sPfVIT-mediated 55Fe2 inux, measured over 1 min, normalized to that obtained at pH7. (d) Inhibition of sPfVIT-mediated 55Fe2 inux by 100 mM divalent metals, 1 mM balomycin A1 and 20 mM CCCP, measured over 1 min at pH 7, normalized to uninhibited control sPfVIT-mediated 55Fe2 inux (**Po0.01; one-way analysis of variance, Dunnetts Multiple Comparison Test). In all panels, data are shown as meanss.e.m. of 68 inux measurements performed at room temperature.

100 mM), no other metal inhibited sPfVIT-mediated 55Fe2 uptake (Fig. 2d). This included Zn2 , and thus is in agreement with the Dzrc1 rescue assay described above. These data demonstrate that PfVIT is selective for Fe2 . Given the vacuolar localization of the majority of VIT, they have been proposed to act as Fe2 /H exchangers. Thus it is interesting that the observed transport via sPfVIT occurred without the need to acidify the yeast vacuoles by the addition of ATP, which activates endogenous V-type H pumps. To support this nding, we further demonstrated that Fe2 transport via sPfVIT was insensitive to the V-type H pump inhibitor balomycin A1 or to the proton ionophore CCCP (Fig. 2d). Furthermore, to address the possibility that the transport measurements via sPfVIT are an artefact of Fe2 vesicle surface binding, several additional experiments were performed (Supplementary Fig. 3b). Adding the divalent cation chelator EDTA to the wash solution used during transport experiments (see Methods section) had no effect on

55Fe2 uptake, while lysing vacuolar vesicle preparations by either three freeze-thaw cycles before experimentation or addition of the detergent Triton X-100 at the start of a 1-min transport measurement signicantly reduced 55Fe2 uptake. Finally, lysis of vesicles using 0.1 M HCl added directly following a 1-min transport experiment signicantly reduced accumulated 55Fe2 .

These results are consistent with 55Fe2 accumulating within vesicles rather than adsorbing/binding to vesicles.

PbVIT is expressed throughout the parasites life cycle. Published transcriptomic data show that Pfvit is expressed inP. falciparum blood-stage parasites with its abundance increasing as the parasite progresses from early ring to mature trophozoite

stage2022. We next generated a C-terminal green uorescent protein (GFP) or myc fusion of PbVIT in P. berghei rodent parasites, using a single crossover transfection strategy (Supplementary Fig. 4). This allows the study of VIT expression and localization throughout the parasites life cycle. Confocal analysis after immunostaining with anti-GFP or anti-myc antibodies showed PbVIT expression in asexual blood stages, mosquito and liver stages of infection (Fig. 3 and Supplementary Fig. 6). In blood-stage parasites, both tagged versions of PbVIT mainly co-localized with PbBiP to the parasites endoplasmic reticulum (Fig. 3a,b and Supplementary Fig. 5). We did not observe PbVIT-GFP signal in regions around hemozoin crystals in blood-stage parasites. In oocysts and liver-stage parasites PbVIT-GFP also co-localized with PbBiP (Fig. 3c,d). Altogether, these data strongly support the idea that PbVIT is expressed mainly in the parasite ER throughout the parasites life cycle.

Pbvit parasites yield reduced liver and blood infections. Our data have revealed a novel Plasmodium iron transporter whose homologues in yeast and plants act as iron detoxiers. We therefore asked whether Plasmodium VITs play a similar role during the parasites life cycle. To answer this question, we generated a P. berghei parasite line decient in the Pbvit gene (Pbvit ), using a double crossover transfection strategy (Fig. 4a).

Pbvit parasites were cloned to obtain isogenic mutant lines, which were used for all further analyses. Successful knockout of pbvit demonstrates a non-essential role for this transporter during asexual P. berghei blood stages. However, when controlled experimental infections were performed, parasitemias in Pbvit

(two independent clones A2 and D1) infected mice were

4 NATURE COMMUNICATIONS | 7:10403 | DOI: 10.1038/ncomms10403 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10403 ARTICLE

PbVIT PbBiP Overlay Bright field

GFP

GFP

GFP

a

b

myc

c

PbVIT PbBiP Hoechst Overlay

d

Figure 3 | Expression of PbVIT in blood-, mosquito- and liver-stage parasites. Indirect immunouorescence assay of PbVIT-GFP and PbVIT-myc P. berghei. (a) Blood-stage PbVIT-GFP P. berghei (shown is a trophozoite stage parasite). (b) Blood-stage PbVIT-myc P. berghei (shown is a trophozoite stage parasite). (c) A mosquito midgut oocyst (17 days post infection). (d) A liver-stage EEF, 48 h post infection of Huh7 the hepatoma cell line with PbVIT-GFP sporozoites. Stainings were performed using antibodies against GFP, myc (green) and PbBiP (ER marker, red), and Hoechst nuclei stain (blue).

signicantly reduced, when compared with those in mice infected with wild-type (wt) parasites (Po0.05, Students t-test; Fig. 4b).

Furthermore, C57Bl/6 J mice infected with Pbvit parasites (104 infected RBC i.v.) survived signicantly longer compared with mice infected with wt parasites (Po0.01, log-rank MantelCox test; Fig. 4c).

Notably, while Pbvit did not show any defect during transmission to or development within Anopheles stephensi mosquitoes (Supplementary Fig. 7), a signicant reduction in parasite liver load was detected after Pbvit sporozoite infection.

Indeed, to investigate if Plasmodium VITs play a role during the rst obligatory stage of Plasmodium infection in the mammalian host, Pbvit or wt P. berghei sporozoites obtained from

A. stephensi mosquitoes were i.v. injected into C57Bl/6 J mice and parasite liver load was measured 6 and 45 h later. The data show a moderate but signicant reduction in infection with Pbvit

(clone A2) sporozoites 6 h after injection (3710% lower, when compared with wt P. berghei infection; Po0.01, Students t-test; Fig. 4d). Furthermore, 45 h after infection the parasite liver load was reduced by 6419% for Pbvit clone A2 (Po0.01;

Fig. 4d) and by 5319% for Pbvit clone D1 (Po0.05; Supplementary Fig. 7), when compared with wt P. berghei infection. Analysis of infected mouse liver sections, by microscopy, show that the observed reduction in parasite liver load was due to a lower number of infected hepatocytes (Fig. 4e), as well as moderate reduction in size of developing Pbvit exoerythrocytic parasite formsEEFs (Fig. 4f). The combined effect of Pbvit deciency on both liver and blood stages was also observed following an entire course of infection initiated with 500 Pbvit and wt sporozoites (Supplementary Fig. 7).

Pbvit provides an iron detoxication mechanism. In yeast and plants, VIT proteins maintain the homeostasis of iron and some other divalent metals by their sequestration from the cytoplasm into organelles. Thus, to investigate the physiological role of PbVIT and test our hypothesis that it detoxies excess iron by sequestration, we compared the LIP in Pbvit and wt P. berghei parasites. To that end, iRBCs containing either Pbvit or wt parasites were stained with the iron-sensitive uorescent probe,

PhenGreen, and analysed by ow cytometry (Supplementary Fig. 8). The results show that the LIP of Pbvit iRBCs was signicantly higher than that of wt P. berghei iRBCs (3711%

increase; Po0.01, Students t-test; Fig. 5a).

Since the percentage of sporozoite-infected hepatoma cells in vitro is extremely low, the PhenGreen ow cytometry approach is not appropriate to study the role of PbVIT during liver-stage infection. We therefore employed an alternative strategy by comparing sensitivities of Pbvit and wt liver-stage parasites to iron depletion in vitro, by adding an iron chelator (deferoxamine,

DFO), and iron complementation, by adding Fe2 (FeSO4 in the presence of 1 mM ascorbic acid). We determined the DFO EC50

for wt parasites developing in HepG2 cells to be 2.7 mM (95% CI2.53 mM; Supplementary Fig. 9). Therefore, DFO concentrations of 1, 2 and 3 mM were used for further experiments comparing Pbvit and wt parasites. For iron supplementation, concentrations of up to 250 mM FeSO4 in the growth medium did not have deleterious effects on the growth of wt parasites and HepG2 cell viability (Supplementary Fig. 9) and, thus, concentrations of 100 and 200 mM FeSO4 were used in further experiments. While under control conditions parasite liver load of Pbvit parasites was not different from that of wt parasites (P 0.16, one sample

t-test), Pbvit parasite load was lower compared with that of wt parasites in conditions of excess iron (Fe2 ; Fig. 5b), consistent with a role for PbVIT in iron detoxication. Conversely,

Pbvit parasites better tolerated chelation of iron by DFO (Fig. 5c) and established higher parasites loads compared with wt parasites in the presence of DFO in the growth medium. Taken together, we show that Pbvit parasites harbour increased LIPs and are more sensitive to environmental uctuations in iron levels, leading to either growth defects under excess iron conditions or growth rescue under iron chelating conditions. Therefore, these data strongly support a key role for PbVIT in cellular iron detoxication in both blood and liver stages of infection.

DiscussionRegulation of iron is essential for cell survival and should also, therefore, be critical during the entire Plasmodium life cycle2326. Iron withdrawal by the use of iron chelators has been explored as an antimalarial approach for decades27 and novel compounds with improved pharmacokinetic properties are being investigated23,24. During infection, growth of malaria parasites is inuenced by the host iron status both in liver25,28 and blood29. P. falciparum growth is reduced in iron-decient erythrocytes while iron supplementation eliminates this growth attenuation30. However, the inuence of iron deciency and iron supplementation on malaria disease progression is complex and remains a controversial topic because of conicting epidemiological and laboratory data31.

How malaria parasites import, export and store iron is not currently understood in detail. Recently a P. berghei metal transporter, ZIPCO, was described as important for the parasites development in the liver26. While no direct transport studies were performed to characterize the transport properties of ZIPCO; iron and zinc supplementation and depletion experiments suggested it plays a role in uptake of iron and zinc across the plasma

NATURE COMMUNICATIONS | 7:10403 | DOI: 10.1038/ncomms10403 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 5

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10403

a

e d

PbVIT

Kpnl Apal EcoRV Notl

TgDHFR/TS

TgDHFR/TS

Knock-out construct

wt Pbvit 1 3

2 1 2 3

5 UTR 3 UTR

Wild-type locus

a b c f d

Recombined locus

1.65 kb 1 kb

b c

wt

Pbvit clone A2

Pbvit clone D1

wt

Pbvit clone A2

Pbvit clone D1

**** ****

Parasitemia

(% iRBC)

3.0

2.5

2.0

1.0

0.5

0.0

** *** **

Survival (%)

100

75

50

25

0 4 6 8 10 12 14

3 4 5

Days p.i. Days p.i.

d e f

** **

150

100

50

0

***

**

Pb18S/mHprt (% of control)

2.5

2.0

1.5

1.0

0.5

0.0 wt wt

Pbvit Pbvit wt Pbvit wt Pbvit 6 h p.i. 45 h p.i.

Liver infection load

EEF number/mm2

(% of WT mean)

EEF size (m2 )

1,250

1,000

750

500

250

0

Figure 4 | Pbvit knockout affects blood- and liver- stage growth of P. berghei. (a) Double crossover strategy for pbvit knockout and genotyping of the Pbvit transgenic clonal line by PCR. Lane 1, detection of knockout construct integration at the 50 end (primers a b, 1.38 kb); lane 2, knockout construct

integration detection at the 30 end (primers c d, 1.23 kb); lane 3, wt pbvit locus (primers e d, 1.45 kb). (b) Parasitemia of C57Bl/6J mice following

infection (i.v.) with 104 wt P. berghei or Pbvit iRBCs, determined by counting of iRBC in Giemsa-stained blood smears (N 10 for wt P. berghei-infected

mice and N 5 for mice infected with Pbvit A2 or D1). (c) Survival of C57Bl/6 J mice infected i.v. with 104 wt or Pbvit iRBC (N 20 for wt P. berghei-

infected mice and N 10 for mice infected with Pbvit A2 or D1). Median survival was 7 and 9 days for wt and Pbvit , respectively, Po0.01, log-rank

MantelCox test. (d) Parasite liver load 6 and 45 h after i.v. injection of wt or Pbvit sporozoites, assessed by reverse transcription PCR measurement of parasite 18s RNA expression, normalized to mouse hypoxanthine-guanine phosphoribosyltransferase, shown are fold expressions relative to the average of controlswt P. berghei (shown is the pool of two independent experiments). (e) Number of EEFs per mm2 of livers 45 h after infection with 50,000 wt and Pbvit sporozoites. Each point represents an average number of EEFs per mm2 per mouse liver by counting the number of EEFs in 58 slices per liver (shown is a pool of two independent experiments, in total 795 wt EEFs and 210 Pbvit EEFs were counted). (f) Size of liver EEFs determined by measuring the PbUIS4 surrounded area by ImageJ in confocal images of liver sections. The means.e.m. size of wt and Pbvit EEFs was 72825 (N 58) and

62923 (N 46) mm2, respectively. In b,d,e and f error bars represent s.e.m. and the asterisks denote signicant differences using the two-tailed, unpaired

Students t-test: *Po0.05; **P o0.01 and ***Po0.001.

membrane26. Mechanisms used by the parasite to store and detoxify excess iron remain unknown. Asexual erythrocyte stage parasites face high demands for maintaining iron homeostasis as they digest iron-containing haemoglobin and the LIP increases with their maturation from ring to schizont forms32. VIT family members have been described in plants and yeast8,1113,33,34 and, more recently, in the human pathogen Trypanosoma brucei35 as important iron regulatory mechanisms. Our study is the rst to characterize VIT homologues in Plasmodium and the rst to conrm the iron transport properties of any VIT homologue.

We show here that expression of PfVIT restores transport of Fe2 into the vacuoles of yeast cells lacking CCC1the yeast

VIT homologue, thereby rescuing the growth defective phenotype of this strain in conditions of increased extracellular Fe2 .

Interestingly, the rescue of the DCCC1 growth phenotype was far greater when the N-terminal tail, before the rst predicted transmembrane region, was removed (resulting in sPfVIT). This is similar to the study of Ca2 /H exchangers (CAXs), including the P. falciparum CAX, where the N-terminal tail has autoinhibitory properties17,36. It is worth noting that the N-terminus of PfVIT contains a phospho-acceptor site, suggesting regulation by phosphorylation. However, our data suggest that removal of the N-terminus improves greatly the expression and subsequent delivery of sPfVIT to the yeast vacuole and, thus, improving its function.

6 NATURE COMMUNICATIONS | 7:10403 | DOI: 10.1038/ncomms10403 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10403 ARTICLE

a b c

** * * *

*

175

150

125

100

75

50

25

0 WT Pbvit

300 250 200 150 100

50

(% of untreated control)

MFI

(% of WT mean)

Parasite load

Parasite load

(% of untreated control)

125 100

75 50 25

0

*

0 0 100 200

Fe2+ (M) DFO (M)

0 1 2 3

Figure 5 | PbVIT functions in iron detoxication by reducing the LIP. (a) The LIP of P. berghei wt and Pbvit iRBCs analysed by ow cytometry. DMFI was determined by evaluating the change in mean uorescence intensity of PhenGreen-loaded iRBCs (SYTO 61-positive subset), after incubation with100 mM DFO (DMFI MFIDFO

treated

MFIDFO untreated). For each independent experiment, the MFI of Pbvit iRBCs was normalized to the mean MFI of wt-iRBCs. Shown is a pool of four independent experiments (N 14), **P 0.0028 (unpaired, two-tailed Students t-test; wt means.e.m. 1006,

Pbvit means.e.m. 137.110). (b) Liver-stage parasite load in HepG2 cells 45 h post infection with P. berghei wt and Pbvit sporozoites in the absence

or presence of FeSO4 and ascorbic acid in the growth medium, normalized to internal untreated control. Liver stage parasite load was determined by reverse transcription PCR quantication of parasite 18s expression normalized to human hypoxanthine-guanine phosphoribosyltransferase expression (shown is a pool of ve independent experiments). (c) Parasite load in HepG2 cells 45 h post infection in the absence or presence of DFO added to the growth medium, determined as in b and normalized to internal untreated control (shown is a pool of ve independent experiments). In b and c the asterisks denote signicant differences using the two-tailed, unpaired Students t-test: *Po0.05.

While numerous VITs have been studied, the functional characteristics of Fe2 transport via VITs have yet to be demonstrated. Here we have measured 55Fe2 uptake into isolated yeast vacuoles expressing sPfVIT (conrmed by western blot). sPfVIT-mediated 55Fe2 uptake was consistent with transport via a Fe2 -specic carrier protein, with an estimated Km value of approximately 15 mM, that was relatively insensitive to pH between 6.5 and 7.5 (note the physiological range of the parasite cytosol is pH 7.17.3 ref. 37). The Km value would suggest that PfVIT is a relatively low afnity/high capacity transport pathway, given that LIP measurements in malaria parasite iRBCs are 0.22 mM (ref. 38). Most VITs have been proposed to be Fe2 /H exchangers, given their localization and proposed role (see below), with Fe2 being accumulated into acidic vacuoles. Yet our data were produced without the need to acidify the yeast vacuoles by the addition of ATP, which activates endogenous V-type H -ATPases (and was unaffected by either a specic inhibitor of this ATPase or a H ionophore). Whether,

PfVIT is an exchanger or facilitative transporter and whether it is a typical member of this family of transporters remains to be established.

In plants and yeast, VITs are usually localized to the membrane of their acidic vacuoles transporting excess iron from the cytoplasm into the vacuole. Consistent with this, a T. brucei VIT homologue has recently been shown to localize to acidocalcisomes35. We initially hypothesized that Plasmodium VIT would be expressed in the membrane of the parasites food vacuole, where haemoglobin digestion and subsequent iron-containing haem detoxication occurs. However, while food vacuoles are only observed in asexual blood stages of development, analysis of published transcriptome and proteome studies suggested expression of Plasmodium VITs throughout the entire parasites life cycle15,3941. Generation of a C-terminal GFP or myc fusion of PbVIT in P. berghei allowed us to conrm complete life cycle expression. Additionally, colocalization studies with selected parasite organelles markers imply PbVIT localization in the parasite ER, in both liver and blood stages of infection. While future studies are necessary to conrm that non-tagged versions of VIT are indeed in the ER, two plant transporters with demonstrated functional homology to CCC1 have been shown to localize to the ER, more specically to ER bodies, where they act to maintain transition metal

homeostasis42. Thus, our study paves the way to explore further the role of the ER in iron detoxication.

As VIT and CCC1 homologues in plants and yeast, respectively, are transporters that remove excess iron from the cytoplasm and prevent iron toxicity, we therefore hypothesized that Plasmodium VIT performs a similar function during the parasites life cycle. Consistent with this, Pbvit parasites have higher LIPs in blood stages and are more sensitive to increased iron levels in liver stages, when compared with wt parasites. Most importantly, Pbvit parasites show a reduction in parasite load in both liver and blood stages of infection, also consistent with

Plasmodium VIT playing a major role in iron detoxication and highlighting the necessity for iron detoxication if malaria parasites are to remain viable in the mammalian host.

Notably, while our data show that PbVIT is important for both liver and blood stages of infection, it is not essential during the entire P. berghei life cycle. This may reect a redundancy of function with other compensatory mechanisms preventing iron toxicity. In S. cerevisiae for example lack of vacuolar iron uptake by CCC1 knockout is compensated by increased iron import into mitochondria through MRS3 and MRS4 transporters43,44. Additional protection mechanisms against high iron toxicity inS. cerevisiae involve induction of ironsulfur cluster-binding proteins, such as TYW1 (ref. 45), an enzyme that participates in the synthesis of wybutosine. Thus, in addition to sequestration of iron into organelles, yeast cells avoid iron toxicity by consumption of free cytosolic iron through the formation of protein-bound ironsulfur clusters. Plasmodium orthologues of mitochondrial iron importers mrs3 and mrs4 remain to be identied, as does the putative role of the Plasmodium orthologue of TYW1 and other ironsulfur cluster containing proteins in mediating protection against high iron toxicity in malaria parasites.

In conclusion, our data provides new insights into how iron is regulated within Plasmodium parasites. This knowledge is highly relevant to better understand parasite biology and with regard to malaria treatment and drug resistance. Artemisinins interact with iron in vitro and iron chelators antagonize artemisinins in these models46,47, although acute experiments in uncomplicated malaria have not conrmed antagonism in vivo48. However, the effects of altered iron concentrations within the erythrocyte and the parasite itself remain to be systematically studied with regard to the efcacy of artemisinin therapies. As such, of further interest

NATURE COMMUNICATIONS | 7:10403 | DOI: 10.1038/ncomms10403 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 7

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10403

is to investigate altered drug sensitivity especially to quinolones or artemisinins in parasites lacking VIT.

Methods

Yeast expression of PfVIT. The DCCC1 yeast strain and CCC1-expression plasmid8 were kindly provided by Jerry Kaplans laboratory, University of Utah, USA. A codon-optimised version of pfvit ORF (GenScript, USA Inc.) was used for yeast expression (Supplementary Fig. 1). The expression plasmid was constructed by subcloning the BamHI-XbaI codon-optimised pfvit fragment into the expression vector pUGpd, containing the Ura selectable marker, yeast centromere sequence and autonomously replicating sequence, which confers mitotic and meiotic stability49. The N-truncated fragment of codon-optimized pfvit ORF was amplied using primers sPfVITf and sPfVITr (Supplementary Table 1) and also subcloned into the pUGpd vector. PfVIT and sPfVIT expressing pUGpd constructs as well as CCC1-expression plasmid and empty pUGpd plasmid were transformed into the DCCC1 yeast strain (lacking the VIT CCC1)8. sPfVIT-pUGpd and empty pUGpd were also transformed into the Dzrc1 yeast strain (lacking a vacuolar zinc transporter, ZRC1)18. Yeast strains were transformed using a previously described Liacetate method50 and selected on SD medium lacking uracil.

Preparation of vacuolar membrane vesicles from yeast. Yeast vacuolar vesicles for 55Fe transport assays were isolated from DCCC1::PUG and DCCC1::sPVIT-transfected strains according to a previously described protocol51, with the following modications: ultracentrifugation steps for initial pelleting of microsomal membranes was at 120,000g, the subsequent sucrose gradient was at 150,000g and the nal pelleting of vacuoles was at 150,000g and were all performed for 45 min at 4 C. The nal vacuolar fraction was resuspended in 5 mM Tris-MES, pH 7.6, 0.3 M sorbitol, 1 mM dithiothreitol, 1 mM PMSF and 1x protease inhibitor (cOmplete, EDTA-free, Sigma-Aldrich) and frozen in liquid nitrogen until use. Vacuolar preparations were always prepared as paired from DCCC1::PUG and DCCC1::sPVIT on the same day, and paired preparations were used in the transport experiments.

55Fe transport assays. For 55Fe uptake assays, vacuole vesicles at protein concentration of 510 mg per reaction were prepared in reaction solution containing 0.3 M sorbitol, 5 mM 3-(N-Morpholino)propanesulfonic acid (MOPS) (pH 7), 25 mM KCl, 1 mM dithiothreitol, 0.2 mM Na-azide and 1 mM ascorbic acid. Reactions were performed at pH 7, except for experiment to determine the effect of pH. Uptakes were started by the addition of 55Fe at a nal concentration of 6 mCi ml 1, corresponding to 2 mM Fe. Uptake reactions were performed at room temperature (B20 C) and on ice. At the times indicated, aliquots (100 ml) of the reaction mix were removed and ltered through premoistened 0.45 mm pore-size cellulose acetate GS type lters (Millipore) and washed three times with 1 ml ice-cold washing solution containing 0.3 M sorbitol, 5 mM Tris-MES (pH 7.5), 25 mM KCl, 100 mM FeSO4. For EDTA wash condition used to test the unspecic binding of 55Fe, 100 mM EDTA was added to the wash solution. The lters were air-dried and radioactivity was determined by liquid scintillation counting. 55Fe inux was normalized to the protein content of the each vacuolar sample used in the experiment.

Western blot analysis. Microsomal and vacuolar preps from DCCC1::pUGpd, DCCC1::PVIT and DCCC1::sPVIT yeast were subjected to SDSPAGE (7 mg of protein was loaded for each) and electrotransferred to a nitrocellulose membrane. For a loading control, membranes were stained with Ponceau S. After washing, membranes were blocked in 5% skimmed milk and 0.1% Tween 20 in PBS for 3 h at room temperature. Membranes were then probed with 1:500 diluted anti-PfVIT antibody (afnity puried peptide polyclonal goat antibody, antigen sequence CGLIVTNEDKNE, from Genscript) overnight at 4 C and then incubated with 1:5,000 diluted HRP-conjugated secondary antibody. The signal was detected with Luminata Crescendo Western HRP substrate (Millipore) and imaged with the ChemiDoc XRS system.

Animal work. C57BL/6J and BALB/c wt mice were purchased from the Charles River Breeding Laboratories and were housed in the facilities of the Instituto de Medicina Molecular in Lisbon. All in vivo protocols were approved by the internal animal care committee of the Instituto de Medicina Molecular and were performed according to national and European regulations.

P. berghei transfection and culture. Transfection experiments were performed onP. berghei ANKA strain 2.34 parasites according to a described protocol52. The pbvit knockout vector was constructed for a double crossover homologous recombination, as previously described53. Primer sequences used to amplify 50 and 30 untranslated regions are given in Supplementary Table 1. The nal knockout construct was digested with KpnI and NotI to release the fragment for transfection. The pyrimethamine-resistant parasite population containing the correct genomic integration of the pbvit knockout construct was cloned by injecting one parasite per mouse (BALB/c male mice, 68 weeks of age). To generate the pbvit-gfp transfection construct (single crossover homologous recombination), a 0.8 kb

region of the pbvit was amplied without the stop codon and inserted in frame and upstream of the gfp sequence in the transfection plasmid containing the human dhfr cassette and conveying resistance to pyrimethamine. Similarly, Pbvit-myc transfection construct was generated by inserting the same 0.8 kb region of the pbvit in frame and upstream of the myc sequence in the transfection plasmid containing the human dhfr cassette. Transfected P. berghei parasites were selected with pyrimethamine selection pressure according to a described protocol52.

For liver-stage experiments, A. stephensi mosquitoes (produced by Instituto de Medicina Molecular insectary) were fed on BALB/c male mice (68 weeks of age) infected with wt, Pbvit and PbVIT-GFP P. berghei. For collection of salivary gland sporozoites, infected mosquitoes were dissected on the 21st day post infection.

Genotype analysis of P. berghei transfectants. PCR analysis performed on genomic DNA isolated from transgenic P. berghei was used to inspect if the transfection constructs integrated into the correct loci in pyrimethamine-resistant parasites. Sequences of primers used for genotyping are provided in Supplementary Table 1.

Culturing and infection of hepatoma cells. HepG2 and Huh7 hepatoma cells (ATCC, USA) were cultured in supplemented Dulbeccos modied Eagles medium or RPMI 1640, respectively, and maintained in a 5% CO2 humidied incubator at 37 C. For determination of the parasite infection load in vitro, 50,000 HepG2 cells were plated per well of a 24-well culture plate and 24 h later infected with 30,000 wt or Pbvit sporozoites per well. Post infection, hepatoma cells were cultured in the presence of 0.3% Fungizone (added to the culture medium). After 45 h of incubation under standard culture conditions in the presence or absence of DFO (13 mM) or FeSO4 (100200 mM) added to the culture medium, infected cells were collected for RNA extraction and parasite load was analysed by Real-time PCR (Supplementary Table 1).

Analysis of the PbVIT-GFP localization. BALB/c male mice (68 weeks of age) infected with PbVIT-GFP P. berghei were bled and the suspension of infected blood in RPMI medium was passed through a CF-11 cellulose column to remove leukocytes. After three washes in PBS, blood stages of PbVIT-GFP P. berghei were immunostained according to a previously described protocol54. For liver-stage localization experiments, HepG2 or Huh7 hepatoma cells were seeded on imaging coverslips in a 24-well culture plate and infected with PbVIT-GFP P. berghei sporozoites. Coverslips were xed at indicated time points post infection in 4% paraformaldehyde/PBS solution for 15 min at room temperature. After three washes in PBS, coverslips were incubated for 45 min in a permeabilization/blocking solution containing 0.1% Triton X-100 and 2% bovine serum albumin in PBS. Coverslips were then incubated for a minimum of 2 h at room temperature in the primary antibody solution diluted accordingly in the permeabilization/blocking solution. Following three washes in PBS, coverslips were then stained with a mixture of appropriate secondary antibodies, diluted 1:500 in the permeabilization/ blocking solution. All stainings for indirect immunouorescence assay (IFA) presented in Fig. 3 were performed using a monoclonal mouse anti-GFP antibody (Abcam; 1:500). Additional, control stainings presented in Supplementary Fig. 4 were performed using a rabbit polyclonal anti-GFP antibody (Abcam; 1:500). For colocalization experiments, rabbit anti-PbBiP was used at a 1:600 dilution and goat anti-PbUIS4 diluted 1:1,000. The secondary antibodies used were: donkey anti-mouse conjugated to Alexa Fluor 488, donkey anti-rabbit conjugated to Alexa Fluor 549 and donkey anti-goat conjugated to Alexa Fluor 660. All images of PbVIT-GFP-expressing parasites were captured with a Zeiss LSM 710 confocal point-scanning microscope. Hoechst 33342 was used for nuclear staining.

The PbBiP antibody (rabbit, polyclonal) was designed to recognize a highly conserved C-terminal region of PBANKA_081890 (GANTPPPGDEDVDS) based on previously widely used P. falciparum BiP antibody55, which also cross-reacted with P. berghei56.

Immunohistochemical staining of liver sections. Livers isolated from infected mice were xed with 4% paraformaldehyde at room temperature for 2 h. The xed liver lobes were cut into 50-mm-thick sections using the Vibratome VT 1000S (Leica). Following blocking in 2% bovine serum albumin and 0.3% Triton X-100 at 4 C overnight, liver sections were stained with goat anti-P. berghei UIS4 (1:1,000) (ref. 57) and mouse anti-P. berghei HSP70 (1:1,000) (ref. 58). The secondary antibodies used for detection were: Alexa Fluor 555 donkey anti-goat antibody and donkey anti-mouse conjugated to Alexa Fluor 488 (all 1:500). Cell nuclei were stained with diamidino-2-phenylindole. Stained liver sections were mounted on microscope slides with Fluoromount-G (SouthernBiotech). Images were acquired on a LSM 710 confocal point-scanning microscope (Zeiss).

Determination of the LIP of iRBCs. The LIP of RBCs infected with wt and Pbvit parasites was determined by ow cytometry using the PhenGreen uorescent iron probe. The staining protocol before ow cytometry measurement was based on a recently published method for P. falciparum32 with the following modication: infected BALB/c male mice (68 weeks of age) with 1.52.8% parasitemia were bled and the RBCs incubated in culture overnight to enrich the culture for mature

8 NATURE COMMUNICATIONS | 7:10403 | DOI: 10.1038/ncomms10403 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10403 ARTICLE

parasite stages. RBCs were washed in PBS and stained with PhenGreen for 45 min (10 mM probe in serum-free RPMI 1640 medium). Following two washes in PBS, cells were incubated for 1 h in standard culture conditions with 0.5 mM Syto61 DNA stain, in the presence or absence of 100 mM DFO or 100 mM FeSO4 1 mM

ascorbic acid. After washing, stained cells were analysed on a FACSCalibur. The geometric mean of PhenGreen uorescence for the FL1-H, FL4-H subset (Supplementary Fig. 7) was determined for all samples. The amount of labile iron was estimated for each sample as that relative to the DFO condition (DMFI).

References

1. WHO. World Malaria Report 2014 http://www.who.int/malaria/publications/world_malaria_report_2014/report/en/

Web End =http://www.who.int/malaria/publications/ http://www.who.int/malaria/publications/world_malaria_report_2014/report/en/

Web End =world_malaria_report_2014/report/en/ (2014).

2. Heppner, D. G., Hallaway, P. E., Kontoghiorghes, G. J. & Eaton, J. W. Antimalarial properties of orally active iron chelators. Blood 72, 358361 (1988).

3. Lytton, S. D., Mester, B., Libman, J., Shanzer, A. & Cabantchik, Z. I. Mode of action of iron (III) chelators as antimalarials: II. Evidence for differential effects on parasite iron-dependent nucleic acid synthesis. Blood 84, 910915 (1994).

4. Shvartsman, M. & Ioav Cabantchik, Z. Intracellular iron trafcking: role of cytosolic ligands. Biometals 25, 711723 (2012).

5. Philpott, C. C. & Ryu, M. S. Special delivery: distributing iron in the cytosol of mammalian cells. Front. Pharmacol. 5, 173 (2014).

6. Dixon, S. J. & Stockwell, B. R. The role of iron and reactive oxygen species in cell death. Nat. Chem. Biol. 10, 917 (2014).

7. Arosio, P., Ingrassia, R. & Cavadini, P. Ferritins: a family of molecules for iron storage, antioxidation and more. Biochim. Biophys. Acta 1790, 589599 (2009).

8. Li, L., Chen, O. S., McVey Ward, D. & Kaplan, J. CCC1 is a transporter that mediates vacuolar iron storage in yeast. J. Biol. Chem. 276, 2951529519 (2001).

9. Portnoy, M. E., Liu, X. F. & Culotta, V. C. Saccharomyces cerevisiae expresses three functionally distinct homologues of the nramp family of metal transporters. Mol. Cell. Biol. 20, 78937902 (2000).

10. Singh, A., Kaur, N. & Kosman, D. J. The metalloreductase Fre6p in Fe-efux from the yeast vacuole. J. Biol. Chem. 282, 2861928626 (2007).

11. Kim, S. A. et al. Localization of iron in Arabidopsis seed requires the vacuolar membrane transporter VIT1. Science 314, 12951298 (2006).

12. Momonoi, K. et al. A vacuolar iron transporter in tulip, TgVit1, is responsible for blue coloration in petal cells through iron accumulation. Plant J. 59, 437447 (2009).

13. Zhang, Y., Xu, Y. H., Yi, H. Y. & Gong, J. M. Vacuolar membrane transporters OsVIT1 and OsVIT2 modulate iron translocation between ag leaves and seeds in rice. Plant J. 72, 400410 (2012).

14. Scholl, P. F., Tripathi, A. K. & Sullivan, D. J. in Malaria: Drugs, Disease and Post-genomic Biology Vol. 295 Current Topics in Microbiology and Immunology (eds Sullivan, D. J. & Krishna, S.) 293324 (Springer, 2005).

15. Solyakov, L. et al. Global kinomic and phospho-proteomic analyses of the human malaria parasite Plasmodium falciparum. Nat. Commun. 2, 565 (2011).

16. Treeck, M., Sanders, J. L., Elias, J. E. & Boothroyd, J. C. The phosphoproteomes of Plasmodium falciparum and Toxoplasma gondii reveal unusual adaptations within and beyond the parasites boundaries. Cell Host Microbe 10, 410419 (2011).

17. Pittman, J. K., Shigaki, T., Cheng, N. H. & Hirschi, K. D. Mechanism of N-terminal autoinhibition in the Arabidopsis Ca(2 )/H( ) antiporter CAX1.

J. Biol. Chem. 277, 2645226459 (2002).18. Kamizono, A., Nishizawa, M., Teranishi, Y., Murata, K. & Kimura, A. Identication of a gene conferring resistance to zinc and cadmium ions in the yeast Saccharomyces cerevisiae. Mol. Gen. Genet. 219, 161167 (1989).

19. Simm, C. et al. Saccharomyces cerevisiae vacuole in zinc storage and intracellular zinc distribution. Eukaryot. Cell 6, 11661177 (2007).

20. Bozdech, Z. et al. The transcriptome of the intraerythrocytic developmental cycle of Plasmodium falciparum. PLoS Biol. 1, E5 (2003).

21. Llinas, M., Bozdech, Z., Wong, E. D., Adai, A. T. & DeRisi, J. L. Comparative whole genome transcriptome analysis of three Plasmodium falciparum strains. Nucleic Acids Res. 34, 11661173 (2006).

22. Otto, T. D. et al. New insights into the blood-stage transcriptome of Plasmodium falciparum using RNA-Seq. Mol. Microbiol. 76, 1224 (2010).

23. Ferrer, P. et al. Antimalarial iron chelator, FBS0701, shows asexual and gametocyte Plasmodium falciparum activity and single oral dose cure in a murine malaria model. PLoS ONE 7, e37171 (2012).

24. Ferrer, P., Vega-Rodriguez, J., Tripathi, A. K., Jacobs-Lorena, M. & Sullivan, Jr.D. J. Antimalarial iron chelator FBS0701 blocks transmission by Plasmodium falciparum gametocyte activation inhibition. Antimicrob. Agents Chemother. 59, 14181426 (2015).25. Portugal, S. et al. Host-mediated regulation of superinfection in malaria. Nat. Med. 17, 732737 (2011).

26. Sahu, T. et al. ZIPCO, a putative metal ion transporter, is crucial for Plasmodium liver-stage development. EMBO Mol. Med. 6, 13871397 (2014).

27. Mabeza, G. F., Loyevsky, M., Gordeuk, V. R. & Weiss, G. Iron chelation therapy for malaria: a review. Pharmacol. Ther. 81, 5375 (1999).

28. Ferrer, P., Castillo-Neyra, R., Roy, C. N. & Sullivan, Jr. D. J. Dynamic control of hepatic Plasmodium numbers by hepcidin despite elevated liver iron during iron supplementation. Microbes Infect. S1286-4579, 0019200196 (2015).

29. Clark, M. A., Goheen, M. M. & Cerami, C. Inuence of host iron status on Plasmodium falciparum infection. Front. Pharmacol. 5, 84 (2014).

30. Clark, M. A. et al. Host iron status and iron supplementation mediate susceptibility to erythrocytic stage Plasmodium falciparum. Nat. Commun. 5, 4446 (2014).31. Spottiswoode, N., Duffy, P. E. & Drakesmith, H. Iron, anemia and hepcidin in malaria. Front. Pharmacol. 5, 125 (2014).

32. Clark, M., Fisher, N. C., Kasthuri, R. & Cerami Hand, C. Parasite maturation and host serum iron inuence the labile iron pool of erythrocyte stage Plasmodium falciparum. Br. J. Haematol. 161, 262269 (2013).

33. Fu, D., Beeler, T. & Dunn, T. Sequence, mapping and disruption of CCC1, a gene that cross-complements the Ca(2 )-sensitive phenotype of csg1 mutants.

Yeast 10, 515521 (1994).34. Lapinskas, P. J., Lin, S. J. & Culotta, V. C. The role of the Saccharomyces cerevisiae CCC1 gene in the homeostasis of manganese ions. Mol. Microbiol. 21, 519528 (1996).

35. Huang, G. et al. Proteomic analysis of the acidocalcisome, an organelle conserved from bacteria to human cells. PLoS Pathog. 10, e1004555 (2014).

36. Guttery, D. S. et al. The Plasmodium berghei Ca(2 )/H( ) exchanger,

PbCAX, is essential for tolerance to environmental Ca(2 ) during sexual

development. PLoS Pathog. 9, e1003191 (2013).37. Saliba, K. J. & Kirk, K. pH regulation in the intracellular malaria parasite, Plasmodium falciparum. H( ) extrusion via a V-type H( )-ATPase. J. Biol.

Chem. 274, 3321333219 (1999).38. Loyevsky, M. et al. Chelation of iron within the erythrocytic Plasmodium falciparum parasite by iron chelators. Mol. Biochem. Parasitol. 101, 4359 (1999).

39. Lasonder, E. et al. Proteomic proling of Plasmodium sporozoite maturation identies new proteins essential for parasite development and infectivity. PLoS Pathog. 4, e1000195 (2008).

40. Lindner, S. E. et al. Total and putative surface proteomics of malaria parasite salivary gland sporozoites. Mol. Cell. Proteomics 12, 11271143 (2013).

41. Silvestrini, F. et al. Protein export marks the early phase of gametocytogenesis of the human malaria parasite Plasmodium falciparum. Mol. Cell. Proteomics 9, 14371448 (2010).

42. Yamada, K., Nagano, A. J., Nishina, M., Hara-Nishimura, I. & Nishimura, M. Identication of two novel endoplasmic reticulum body-specic integral membrane proteins. Plant Physiol. 161, 108120 (2013).

43. Li, L. & Kaplan, J. A mitochondrial-vacuolar signaling pathway in yeast that affects iron and copper metabolism. J. Biol. Chem. 279, 3365333661 (2004).

44. Yu, D. et al. High-resolution genome-wide scan of genes, gene-networks and cellular systems impacting the yeast ionome. BMC Genomics 13, 623 (2012).

45. Li, L., Jia, X., Ward, D. M. & Kaplan, J. Yap5 protein-regulated transcription of the TYW1 gene protects yeast from high iron toxicity. J. Biol. Chem. 286, 3848838497 (2011).

46. Haynes, R. K. et al. The Fe2 -mediated decomposition, PfATP6 binding, and

antimalarial activities of artemisone and other artemisinins: the unlikelihood of C-centered radicals as bioactive intermediates. ChemMedChem 2, 14801497 (2007).47. Meshnick, S. R. et al. Iron-dependent free radical generation from the antimalarial agent artemisinin (qinghaosu). Antimicrob. Agents Chemother. 37, 11081114 (1993).

48. Looareesuwan, S. et al. Co-administration of desferrioxamine B with artesunate in malaria: an assessment of safety and tolerance. Ann. Trop. Med. Parasitol. 90, 551554 (1996).

49. Pulcini, S. et al. Expression in yeast links eld polymorphisms in PfATP6 to in vitro artemisinin resistance and identies new inhibitor classes. J. Infect. Dis. 208, 468478 (2013).

50. Gietz, R. D. & Schiestl, R. H. Quick and easy yeast transformation using the LiAc/SS carrier DNA/PEG method. Nat. Protoc. 2, 3537 (2007).

51. Nakanishi, Y., Saijo, T., Wada, Y. & Maeshima, M. Mutagenic analysis of functional residues in putative substrate-binding site and acidic domains of vacuolar H -pyrophosphatase. J. Biol. Chem. 276, 76547660 (2001).

52. Janse, C. J., Ramesar, J. & Waters, A. P. High-efciency transfection and drug selection of genetically transformed blood stages of the rodent malaria parasite Plasmodium berghei. Nat. Protoc. 1, 346356 (2006).

53. Slavic, K. et al. Life cycle studies of the hexose transporter of Plasmodium species and genetic validation of their essentiality. Mol. Microbiol. 75, 14021413 (2010).

54. Tonkin, C. J. et al. Localization of organellar proteins in Plasmodium falciparum using a novel set of transfection vectors and a new immunouorescence xation method. Mol. Biochem. Parasitol. 137, 1321 (2004).

55. Kumar, N., Koski, G., Harada, M., Aikawa, M. & Zheng, H. Induction and localization of Plasmodium falciparum stress proteins related to the heat shock protein 70 family. Mol. Biochem. Parasitol. 48, 4758 (1991).

NATURE COMMUNICATIONS | 7:10403 | DOI: 10.1038/ncomms10403 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 9

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10403

56. Wiser, M. F., Lanners, H. N., Bafford, R. A. & Favaloro, J. M. A novel alternate secretory pathway for the export of Plasmodium proteins into the host erythrocyte. Proc. Natl Acad. Sci. USA 94, 91089113 (1997).

57. Lopes da Silva, M. et al. The host endocytic pathway is essential for Plasmodium berghei late liver stage development. Trafc 13, 13511363 (2012).

58. Tsuji, M., Mattei, D., Nussenzweig, R. S., Eichinger, D. & Zavala, F. Demonstration of heat-shock protein 70 in the sporozoite stage of malaria parasites. Parasitol. Res. 80, 1621 (1994).

59. Dereeper, A. et al. Phylogeny.fr: robust phylogenetic analysis for the non-specialist. Nucleic Acids Res. 36, W465W469 (2008).

60. Stover, B. C. & Muller, K. F. TreeGraph 2: combining and visualizing evidence from different phylogenetic analyses. BMC Bioinform. 11, 7 (2010).

Acknowledgements

We are thankful to Bryan Mackenzie for helpful advice regarding 55Fe transport

experiments and to Jerry Kaplan for kindly providing the DCCC1 yeast strain. We thank

Ana Parreira for producing the P. berghei-infected Anopheles mosquitoes and Soa Guia

Marques for helping with ow cytometry procedures. This work was supported by grants

from the European Research Council ERC-2012-StG_311502 to M.M.M., Fundaao para

a Cincia e Tecnologia EXPL/BIM-MET/0753/2013 and The European Union

Seventh Framework Programme (FP7/2007-2013) under grant agreement n.

304948NANOMAL to S.K. and H.M.S. K.S. was supported by an EMBO long-term

fellowship (EMBO ALTF 1584-2011).

Author contributions

K.S., S.K., J.K.P., H.M.S. and M.M.M. conceived and designed the study. K.S., A.L., G.B.,

K.K.H. and I.V. performed the experiments. K.S., S.K., G.B., J.K.P., H.M.S. and M.M.M.

analysed the data. K.S., S.K., H.M.S. and M.M.M. wrote the manuscript.

Additional information

Supplementary Information accompanies this paper at http://www.nature.com/naturecommunications

Web End =http://www.nature.com/

http://www.nature.com/naturecommunications

Web End =naturecommunications Competing nancial interests: The authors declare no competing nancial interests. Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/

Web End =http://npg.nature.com/

http://npg.nature.com/reprintsandpermissions/

Web End =reprintsandpermissions/

How to cite this article: Slavic, K. et al. A vacuolar iron-transporter homologue acts as a

detoxier in Plasmodium. Nat. Commun. 7:10403 doi: 10.1038/ncomms10403 (2016).

This work is licensed under a Creative Commons Attribution 4.0

International License. The images or other third party material in this

article are included in the articles Creative Commons license, unless indicated otherwise

in the credit line; if the material is not included under the Creative Commons license,

users will need to obtain permission from the license holder to reproduce the material.

To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Web End =http://creativecommons.org/licenses/by/4.0/

10 NATURE COMMUNICATIONS | 7:10403 | DOI: 10.1038/ncomms10403 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications