The superoxide reductase from the early diverging eukaryote Giardia intestinalis

Free Radical Biology & Medicine 51 (2011) 1567–1574

Contents lists available at ScienceDirect

Free Radical Biology & Medicine j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / f r e e r a d b i o m e d

Original Contribution

The superoxide reductase from the early diverging eukaryote Giardia intestinalis Fabrizio Testa a, Daniela Mastronicola a, Diane E. Cabelli b, Eugenio Bordi c, Leopoldo P. Pucillo c, Paolo Sarti a, Lígia M. Saraiva d, Alessandro Giuffrè a,⁎, Miguel Teixeira d a

Department of Biochemical Sciences, CNR Institute of Molecular Biology and Pathology, and Istituto Pasteur–Fondazione Cenci Bolognetti, Sapienza Università di Roma, I-00185 Rome, Italy Chemistry Department, Brookhaven National Laboratory, Upton, NY 11973–5000, USA Clinical Chemistry and Microbiology Laboratory, Istituto Nazionale per le Malattie Infettive IRCCS “Lazzaro Spallanzani,” Rome, Italy d Instituto de Tecnologia Química e Biológica, Universidade Nova de Lisboa, 2780–157 Oeiras, Portugal b c

a r t i c l e

i n f o

Article history: Received 22 March 2011 Revised 2 June 2011 Accepted 20 July 2011 Available online 29 July 2011 Keywords: Oxidative stress Anaerobic protozoa Nonheme iron protein Superoxide detoxification Time-resolved spectroscopy Free radicals

a b s t r a c t Unlike superoxide dismutases (SODs), superoxide reductases (SORs) eliminate superoxide anion (O2•−) not through its dismutation, but via reduction to hydrogen peroxide (H2O2) in the presence of an electron donor. The microaerobic protist Giardia intestinalis, responsible for a common intestinal disease in humans, though lacking SOD and other canonical reactive oxygen species-detoxifying systems, is among the very few eukaryotes encoding a SOR yet identified. In this study, the recombinant SOR from Giardia (SORGi) was purified and characterized by pulse radiolysis and stopped-flow spectrophotometry. The protein, isolated in the reduced state, after oxidation by superoxide or hexachloroiridate(IV), yields a resting species (Tfinal) with Fe 3+ ligated to glutamate or hydroxide depending on pH (apparent pKa = 8.7). Although showing negligible SOD activity, reduced SORGi reacts with O2•− with a pH-independent second-order rate constant k1 = 1.0 × 10 9 M− 1 s− 1 and yields the ferric-(hydro)peroxo intermediate T1; this in turn rapidly decays to the Tfinal state with pH-dependent rates, without populating other detectable intermediates. Immunoblotting assays show that SORGi is expressed in the disease-causing trophozoite of Giardia. We propose that the superoxide-scavenging activity of SOR in Giardia may promote the survival of this air-sensitive parasite in the fairly aerobic proximal human small intestine during infection. © 2011 Elsevier Inc. All rights reserved.

The adventitious partial reduction of molecular O2 in a living cell gives rise to formation of a plethora of cytotoxic species, collectively termed “reactive oxygen species” (ROS), which can impair a wide spectrum of biomolecules with key functions. To counteract oxidative damage, all living beings have thus developed enzymatic systems to accomplish ROS detoxification with high selectivity and efficacy. Superoxide anion (O2•−), one of the best studied ROS, is a highly reactive radical species produced by the one-electron reduction of O2. In many (micro)organisms, this radical is detoxified to hydrogen peroxide (H2O2) by superoxide dismutase (SOD), a well-known metalloprotein that efficiently catalyzes the following reactions: •−

−

O2 →O2 þ 1e ;

ð1Þ

Abbreviations: ROS, reactive oxygen species; SOR, superoxide reductase; SOD, superoxide dismutase; SORGi, superoxide reductase from Giardia intestinalis; 1Fe–SOR, one-iron-containing SOR; 2Fe–SOR, two-iron-containing SOR; SDS–PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; MES, 2-(N-morpholino)ethanesulfonic acid; HEPES, N-(2-hydroxyethyl)piperazine-N′-2-ethanesulfonic acid; Tris, tris (hydroxymethyl)aminomethane; SVD, singular value decomposition; EDTA, ethylenediaminetetraacetate. ⁎ Corresponding author. Fax: + 39 06 4440062. E-mail address: [email protected] (A. Giuffrè). 0891-5849/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.freeradbiomed.2011.07.017

•−

þ

−

O2 þ 2H þ 1e →H2 O2 ;

•−

þ

2O2 þ 2H →O2 þ H2 O2 :

ð2Þ

ð1Þ þ ð2Þ

The peroxide produced is then typically eliminated by catalase: 2H2 O2 →O2 þ 2H2 O:

ð3Þ

SOD and catalase promote the removal of O2•− and H2O2, respectively, through their dismutation, which leads to formation of molecular oxygen as a co-product. As O2 production is clearly disadvantageous for air-sensitive microorganisms, it is not surprising that during the past decade many microaerobes or anaerobes have been shown to code for O2•−- and H2O2-detoxifying enzymes that operate without releasing O2. This is the case for superoxide reductase (SOR; see [1–5] for reviews), a nonheme iron-containing enzyme that was shown to eliminate superoxide not through its dismutation (Reaction (1)+(2)), but via reduction to H2O2 (Reaction (2)) [6,7] in the presence of a suitable electron donor.

1568

F. Testa et al. / Free Radical Biology & Medicine 51 (2011) 1567–1574

So far, two classes of SORs have been identified: 1Fe–SORs, also called neelaredoxins (from “neela,” the Sanskrit word for blue [8]) because of their characteristic blue color in the oxidized form, and 2Fe–SORs (or desulfoferrodoxins [9]), which have an extra desulforedoxin-like domain fused to the N-terminus with an additional [FeCys4] center (called “center I”) whose function is still elusive. The reaction of superoxide with 1Fe– and 2Fe–SORs occurs at their common iron center (called “center II” in 2Fe–SORs), which, in the reduced form of the enzyme, consists of a pentacoordinated iron(II) with four equatorial histidines and one axial cysteine in a square pyramidal geometry. In the oxidized state, a monodentate carboxylate from a glutamate residue (E14 in the enzyme from Pyrococcus furiosus) typically binds the Cys(His)4–Fe 3+ center in the sixth axial position, resulting in a hexacoordinated octahedral geometry. The three-dimensional crystallographic structure of a few SORs has been determined [10–14], showing that the active iron site, located in a seven-stranded β-barrel domain, is close to the molecular surface and exposed to the solvent. Site-directed mutagenesis studies revealed that the reaction with superoxide is facilitated electrostatically by the positive charge of a solvent-exposed lysine residue (K15 in the P. furiosus enzyme). The aforementioned glutamate and lysine residues both lie in an EKHVP motif typically (but not strictly) conserved in SORs. The physiological redox partner of SOR is still a matter of debate. Rubredoxins, small proteins (molecular mass of a few thousand kilodaltons) with a [FeCys4] center known to be involved in electron transfer processes, have been generally assumed to be the direct electron donors to SOR, based on the fact that the genes encoding rubredoxin and SOR lie in the same operon in some bacteria. Consistently, reduced rubredoxins were shown to reduce both 1Fe– and 2Fe–SORs with second-order rate constants on the order of 10 6–10 7 M − 1 s − 1 [15–17]. However, physiological electron donors other than rubredoxins must exist because rubredoxins are missing in a large number of organisms that encode SORs. Moreover, a certain degree of promiscuity in the reaction of SOR with electron donor proteins is seemingly allowed. The 2Fe–SOR gene from Desulfoarculus baarsi was indeed shown to complement deletion of the sodAB genes in Escherichia coli [18], in which a rubredoxin is missing.

Incidentally, this observation, reported in 1996, provided the first experimental evidence for a role for SOR in superoxide detoxification. This complementation was later observed for other 2Fe– or 1Fe–SORs. The mechanism of the reaction of superoxide with SOR has been extensively investigated mostly by pulse radiolysis [17,19–27], though recently an experimental protocol to follow the reaction by stopped-flow spectrophotometry was also reported [28]. There is general consensus that addition of superoxide to the reduced enzyme yields a first intermediate (T1) with a maximal absorption in the 580–630 nm range, assigned to a ferric-(hydro)peroxo species (see Scheme 1). Formation of T1 occurs at k1 = 10 9 M − 1 s − 1 independent of pH, which indicates that if protonation of bound superoxide takes place at this step, it does not limit the reaction rate. The fate of the T1 intermediate varies among the different SORs that have been investigated. In some enzymes (namely the 2Fe–SOR from Desulfovibrio vulgaris [22,23,28] and the 1Fe–SOR from Treponema pallidum [25]), T1 was shown to rapidly decay, with pH-dependent rates, directly to a final species (Tfinal), with Fe 3+ ligated to either glutamate or hydroxide, depending on pH. Glutamate/hydroxide exchange at the ferric ion was shown (i) to induce remarkable differences in the absorption spectra (up to 90-nm shift of the band in the visible region and the appearance/disappearance of a broad band at ~ 330 nm) and (ii) to have apparent pKa values in the range 6.0–9.6, depending on the SOR. Conversely, in other enzymes (the 1Fe– and 2Fe–SORs from Archaeoglobus fulgidus [17,26] and the 2Fe–SOR from Desulfoa. baarsi [21]), T1 was reported to decay to Tfinal via formation of another intermediate, called T2, with Fe 3+ bound to H2O or OH −: in these cases, the formation rates of both T2 and Tfinal were found to be also dependent on pH. SORs were initially thought to be restricted to the Bacteria and Archaea domains and, accordingly, SOR-encoding genes have been identified in a large number of anaerobic prokaryotes. Recently, however, the occurrence of a putative SOR-encoding gene was also suggested for the early divergent protozoan pathogen Giardia intestinalis [5] and a few more eukaryotes [29]. G. intestinalis is the amitochondriate, microaerophilic parasite responsible for giardiasis, a common intestinal infectious disease with 280 million symptomatic human

Scheme 1. Proposed mechanism for the catalytic cycle of superoxide reductases (modified from [5]). The reduced enzyme from Giardia, like those from Desulfov. vulgaris and T. pallidum, upon reaction with superoxide undergoes a direct transition from T1 to Tfinal with no evidence for the formation of the T2 intermediate.

F. Testa et al. / Free Radical Biology & Medicine 51 (2011) 1567–1574

infections per year [30,31]. Interestingly, this O2-sensitive SODlacking parasite probably has to cope with oxidative stress in vivo, as it colonizes a fairly aerobic tract of the human gut, i.e., the upper part of the small intestine. This prompted us to gain insight into the function of the superoxide reductase from G. intestinalis (SORGi) in this study. Materials and methods Cloning, expression, and purification of SORGi By searching for SOR-coding genes in the genome of G. intestinalis, by using the tBLASTn algorithm in the public genome database of this eukaryotic pathogen (http://www.giardiadb.org/giardiadb), an open reading frame, not yet annotated, putatively coding for a 1Fe–SOR, can be identified [5]. This putative SOR gene is present not only in assemblage A of G. intestinalis (in the genomic element CH991785:AACB02000063, nt 11532–11867), but also in its assemblages B and E (in the genomic elements ACGJ01002424, nt 33108– 33443, and GLP15_1799 on contig 55, nt 8510–8845, respectively). The genomes of the three assemblages have been recently completely sequenced [32,33]. Assemblages A and B, the most common in infected humans, were found in livestock as well as in cats, dogs, and rats, whereas assemblage E was reported in cattle, sheep, pigs, goats, and water buffaloes [31]. Oligonucleotide primers were designed based on the sequence of the putative SORGi-encoding gene from assemblage A. With these primers the gene was amplified by PCR, using lysates of 10 5 trophozoites of the parasite as DNA template and Vent polymerase (Stratagene). The amplified coding region (477 bp) was sequenced and cloned into a pET28a vector (pET28-SORGi). Cultures of E. coli BL21-Gold (DE3) were transformed with pET28-SORGi and grown aerobically in 1-L flasks at 37 °C, in M9 minimum medium supplemented with 30 μg/ml kanamycin and 200 μM ferrous ammonium sulfate ((NH4)2Fe(SO4)2·6H2O). When the cell culture reached ~0.6 OD at 600 nm, temperature was lowered to 25 °C and protein expression induced by addition of 400 μM isopropyl β-D-1-thiogalactopyranoside. The next day the cells were harvested by centrifugation. All protein purification steps were performed at pH 7.5 and 4 °C. Typically, 16 g of E. coli cells was resuspended in 70 ml of 50 mM tris (hydroxymethyl)aminomethane (Tris–HCl) + 500 mM NaCl. Cells were lysed by sonication and the insoluble fraction was separated out by ultracentrifugation (30 min at 14,000 g). The soluble extract was loaded onto a His-Trap column (Amersham) and the recombinant His-tagged SORGi eluted with 300 mM imidazole, which was then removed by gel filtration chromatography. After His-tag cleavage with thrombin, the protein was subjected to a second cycle of nickel affinity and gel filtration chromatography and stored until use at −80 °C in 50 mM Tris, pH 7.2, 100 mM NaCl. Protein purity was assayed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS– PAGE). The concentration and the total iron content of the purified protein were determined by the bicinchoninic acid [34] and the ferrozine assays [35], respectively. Stopped-flow spectroscopy Stopped-flow experiments were carried out at 10 °C with a thermostated instrument (DX.17MV; Applied Photophysics, Leatherhead, UK) equipped with a photodiode array detector. Time-resolved absorption spectra were acquired with an acquisition time of 1 ms per spectrum (light path 1 cm). For pH studies, the instrument was used in the asymmetric mixing mode, with the protein diluted into highly buffered solutions mixed in a 10:1 volume ratio with unbuffered solutions. The following buffers were used at a concentration of 100 or 250 mM: MES (pH 5.0–6.0), HEPES (pH 6.5–7.5), Tris (pH 8.0–9.0), glycine (pH 9.5–10.5). The reaction of SORGi with superoxide was

1569

investigated following the protocol in [28]. Briefly, 10 mg KO2 was dissolved in 5 ml of a degassed aqueous solution of 0.1 M NaOH and 2 mM diethylenetriaminepentaacetic acid and immediately mixed in a 1:10 volume ratio with a 250 mM buffer solution containing SORGi such that concentrations after mixing were 45 μM protein and, nominally, 2.5 mM superoxide. Under all conditions, the reaction was over within 0.5 s. Over the same time scale, no significant absorption changes were observed in control experiments in which SORGi was mixed with the aforementioned alkaline solution in the absence or presence of H2O2 (2 mM after mixing), the major product of the spontaneous degradation of superoxide; a sluggish oxidation of the protein by H2O2 was observed on time scales much longer than 0.5 s (several tens of seconds). This rules out that the fast reactions observed after mixing with superoxide are due to the peroxide contaminating the KO2 solution. Stopped-flow data were analyzed by singular value decomposition (SVD) and nonlinear regression analyses using the software MATLAB (MathWorks, South Natick, MA, USA). SVD analysis was carried out according to [36]. Pulse radiolysis Pulse radiolysis experiments were carried out using the Brookhaven National Laboratory (BNL) 2-MeV Van de Graaff accelerator as described previously [37]. All samples were prepared using Millipore ultrapurified distilled water. Ethylenediaminetetraacetate (EDTA), sodium formate, ethanol, monobasic phosphate, sodium hydroxide, and hydrochloric acid were of the highest purity commercially available and used as purchased. Reduced SORGi was obtained by addition of ascorbate in a slight molar excess (2:1) over the protein. Unless otherwise mentioned, studies were performed at room temperature pulsing 90 μM SORGi with doses of 1.8–6.0 μM superoxide, in airsaturated solutions containing 2 mM Tris–HCl, 5 mM NaCl, and 50 mM formate (light path 2 cm). Alternatively, 0.5 M ethanol was used as a • OH scavenger instead of formate. Kinetic data were analyzed using the BNL pulse radiolysis program PRWIN (H. Schwarz). The extinction coefficients of the protein intermediates detected in the reaction of reduced SORGi with superoxide were calculated assuming a 1:1 stoichiometry of the reaction. The superoxide dismutase activity of SORGi was assayed by generating O2•− in 0.1 M phosphate, pH 7.1 or pH 7.8, 50 mM formate, and 10 μM EDTA and following its degradation at 265 nm in the presence or absence of oxidized SORGi. For these experiments, the “as isolated” protein was oxidized with a 10-fold excess of hexachloroiridate(IV) (K2IrCl6), followed by removal of the oxidant by gel chromatography. Cultures of G. intestinalis trophozoites and immunoblotting Trophozoites of G. intestinalis strain WB clone C6 (ATCC No. 50803TM) were cultured axenically at 37 °C in bile-supplemented Diamond's TYI-S-33 medium [38]. Whole-cell extracts were prepared from cultures of trophozoites. Typically, the equivalent of 5 × 10 5 cells was loaded per lane. Proteins were separated by SDS–PAGE and blotted onto a nitrocellulose membrane. Blots were incubated with rabbit polyclonal antibodies against SORGi (Davids Biotechnologie GmbH), followed by incubation with alkaline peroxidase-conjugated secondary antibodies (Sigma) and then detected with the ECL kit (GE Healthcare No. RPN2132). Results Properties of the recombinant SOR from G. intestinalis The SOR gene from assemblage A of the G. intestinalis genome was cloned in E. coli, and the recombinant protein was purified to homogeneity with a typical yield of 8 mg protein per gram of cells. Protein purity was assessed by SDS–PAGE and metal content analysis

1570

F. Testa et al. / Free Radical Biology & Medicine 51 (2011) 1567–1574

yielded 1 Fe per monomer, consistent with full occupancy of the metal center in the active site of the enzyme. In a gel chromatography assay, most of the protein elutes with a mass compatible with a tetrameric structure. SORGi from assemblage A is very similar to the one identified in assemblage B (94% identity, 99% similarity) and virtually identical to GLP15_1799 from assemblage E. As expected, the protein from Giardia (SORGi) shares a much lower sequence similarity with previously characterized SORs from other microbial sources, including the 1Fe– SORs from P. furiosus (32% identity, 48% similarity), A. fulgidus (30% identity, 44% similarity), Desulfovibrio gigas (26% identity, 42% similarity), and T. pallidum (33% identity, 52% similarity). The multiple sequence alignment in Supplementary Fig. S1 shows that the protein from Giardia retains all the main residues typically conserved among SORs: the four histidines (H19, H30, H46, and H102) and the cysteine (C99) that bind Fe at the catalytic site; the characteristic EKHVP motif that includes the strictly conserved P20, as well as the residues E17 and K13, whose role in the mechanism of superoxide reduction was extensively investigated by site-directed mutagenesis on several SORs in previous studies (see Ref. [5] and references therein); and the isoleucine (I48) located after the third histidine ligand in the active site. Immunodetection of SOR in Giardia cells The expression of SORGi in the trophozoites of Giardia was assayed by immunoblotting analysis using polyclonal antibodies raised against the recombinant protein (Fig. 1). Giardia trophozoites were cultured at 37 °C in bile-supplemented Diamond's TYI-S-33 medium [38] and assayed for viability by both the trypan blue exclusion test and checking cell motility by light microscopy. As shown in Fig. 1, the SORGi band was clearly detected by immunoblotting in whole-cell extracts (lane 1), thereby confirming that the SOR gene is expressed in the trophozoites of Giardia grown under standard conditions. Reaction of the Giardia SOR with superoxide As isolated, SORGi is mainly in the reduced state, hence the small absorbance in the visible region (Fig. 2, dotted spectrum), and is promptly oxidized by either superoxide or hexachloroiridate(IV). The absorption spectrum of the oxidized protein is characteristically dependent on pH: at pH ≤7.0, the spectrum displays typical bands at 330 and 647 nm (Fig. 2, solid trace), whereas at pH 10.5 the 330 nm band is bleached and the 647 nm band is blue-shifted to 560 nm (Fig. 2, dashed trace). Based on previous studies carried out on different SORs [39,40], the former spectrum can be assigned to the glutamate (E17)-bound form (Fe 3+–Glu) and the latter to the Fe 3+–

Fig. 1. Immunodetection of SORGi in G. intestinalis. Lane 1, whole-cell extract (5 × 105 trophozoites); lane 2, purified recombinant SORGi (2.5 ng).

Fig. 2. Redox absorption spectra of SORGi. Absorption spectra of 45 μM “as isolated” SORGi (dotted) or 100 ms after mixing with a slight excess of K2IrCl6 at pH ≤7.0 (Fe3+– Glu, solid) or pH 10.5 (Fe3+–OH, dashed). Inset: pH dependence of the transition between the Fe3+–Glu and the Fe3+–OH forms of the oxidized protein.

OH form. As shown in the inset to Fig. 2, in the Giardia enzyme, interconversion between these two species (Tfinal in Scheme 1) occurs with an apparent pKa of 8.7 ± 0.4 (Scheme 2). The kinetics of the reaction of SORGi with superoxide was investigated by both stopped-flow spectrophotometry and pulse radiolysis. Stopped-flow experiments were carried out at 10 °C in the pH range 5.0–10.5, following the protocol described in [28] (see Materials and methods for experimental details). Briefly, a highly buffered solution containing the protein “as isolated” was rapidly mixed in a 10:1 ratio with a freshly prepared alkaline solution of KO2, and the reaction was followed spectrophotometrically in the UV/ visible range. As expected, largely pH-dependent kinetics were observed, with higher rates measured at lower pH. At acidic pH values, the reaction is too fast and the final Fe 3+–Glu protein form (Tfinal in Scheme 1) was fully populated within the dead-time of the instrument (not shown). At pH ≥7.0, part of the reaction could be time-resolved. Fig. 3 shows that at pH 7.5 the protein, initially almost completely reduced (dashed trace), within the instrumental deadtime converts into an intermediate with an absorption maximum at ~595 nm. This species resembles the ferric-(hydro)peroxo T1 intermediate (see Scheme 1) previously detected on a submillisecond time scale following the reaction of superoxide with reduced SORs from other microbial sources [17,19–27]. At longer times (≤100 ms), this intermediate decays into the final, fully oxidized glutamate-ligated Tfinal species, with its characteristic absorption bands at 647 and 330 nm (compare Fig. 3, top, with Fig. 2). As shown in Fig. 3, the absorption increase at ~330 nm and the 50-nm red shift of the 595nm band occur synchronously. Consistently, global fitting of the spectral data, as obtained by SVD analysis, shows that the reaction occurs with a single optical transition following an exponential time course (k′ = 98 s − 1, Fig. 3, bottom). The analysis, thus, indicates that

Scheme 2. pH dependent ligand exchange in the Tfinal intermediate.

F. Testa et al. / Free Radical Biology & Medicine 51 (2011) 1567–1574

1571

within 1 μs and the reaction was monitored at selected wavelengths in the 500–750 nm range. As shown in Supplementary Fig. S2, after radiolytic generation of superoxide, two kinetic phases are detected. A fast phase, corresponding to the formation of the T1 intermediate with a maximum absorption at ~600 nm (Fig. 4, open symbols), is followed by the decay of T1 to the fully oxidized Tfinal species, with Fe 3+–Glu or Fe 3+– OH depending on pH (Fig. 4, closed symbols). Consistent with the stopped-flow data, no other intermediates are detected during the reaction. The formation of the T1 intermediate proceeds at a secondorder rate constant k1 ~ 1.0 × 10 9 M − 1 s − 1, which is independent of pH as well as the absorption spectrum of the intermediate. In contrast, the observed rate constant for the T1 → Tfinal decay is largely dependent on pH in the range 5.0–8.0 and almost invariant at higher pH values (Fig. 5). According to [23], the pH/rate profile observed with SORGi was fitted using the equation k2 = k2′[H3O+] + k2″, which allowed us to estimate a second-order rate constant k2′ ~ 1.0 × 1010 M− 1 s − 1 and a value of ~320 s− 1 for k2″ (Fig. 5). Summing up, pulse radiolysis and stopped-flow data consistently show that in the Giardia enzyme, as previously observed with the SORs from Desulfov. vulgaris and T. pallidum [22,23,25,28], T1 is the only intermediate detected in the

Fig. 3. Reaction with superoxide followed by stopped-flow spectrophotometry. (Top) Time-resolved absorption spectra collected within the first 100 ms after mixing, in a 10:1 ratio, “as isolated” SORGi in 250 mM HEPES, pH 7.5, with superoxide (see Materials and methods for details). Concentrations after mixing: [SORGi] = 47 μM; [superoxide]= 2.5 mM (nominal concentration). Temperature= 10 °C. Arrows indicate the direction of the spectral changes. The spectrum of the “as isolated” protein, i.e., the initial species, is also shown (dotted). (Bottom) Reaction time course (normalized to 100%), as obtained by SVD analysis of the spectra set depicted at the top after subtraction of the final, 647 nm species. Fitted rate constant: k′ = 98 s− 1. (Inset) Single optical component obtained by SVD analysis, referenced to the 647 nm species. Arrows indicate the direction of the absorption changes.

the “dead-time intermediate” (presumably T1) decays into the Tfinal species directly with no other detectable intermediates. As expected, lower decay rates (~50 s − 1) were measured at higher pH (≥8.0) and formation of the Fe3+–OH Tfinal species was eventually observed in experiments at the highest pH values (consistent with the pKa = 8.7 ± 0.4 measured for the hydroxide/glutamate exchange at Fe 3+, inset in Fig. 2). At high pH, some kinetic heterogeneity was noted, possibly arising from protein instability. The stopped-flow data clearly show that SORGi in the reduced state is highly reactive toward superoxide, but these experiments raise the question as to whether the dead-time intermediate actually represents the T1 intermediate. To address this question, the reaction of ascorbate-reduced SORGi with superoxide was investigated at higher time resolution by pulse radiolysis. Reduced SORGi was allowed to react at room temperature and various pH values (from 5.0 to 9.5) with largely substoichiometric amounts of superoxide generated

Fig. 4. Reaction intermediates detected by pulse radiolysis. Kinetic absorption spectra of the T1 (open symbols) and Tfinal (closed symbols) species, as detected by pulse radiolysis following the reaction of ascorbate-reduced SORGi with superoxide, at (A) pH 6.5 or (B) pH 9.0.

1572

F. Testa et al. / Free Radical Biology & Medicine 51 (2011) 1567–1574

Fig. 5. pH dependence of the T1 intermediate decay. Rate constant of the T1 → Tfinal transition (see Scheme 1), measured by pulse radiolysis as a function of pH (n = 5). Data were fitted using the equation k2 = k2′[H3O+] + k2″ with k2′ = 1.0 × 1010 M− 1 s− 1 and k2″ = 320 s− 1.

reaction with superoxide, with no evidence for the formation of the so-called T2 intermediate. Because very low (if not negligible) superoxide dismutase activities were reported for other SORs [41–43], we tested whether the enzyme from Giardia was also endowed with such an activity. In these assays, 2.4–24 μM superoxide was generated by pulse radiolysis and its degradation was followed spectrophotometrically in the presence or absence of 4–8 μM oxidized SORGi. Under these conditions, however, the protein did not enhance significantly the spontaneous degradation of superoxide (Supplementary Fig. S3). On this basis, we conclude that SORGi, while being highly reactive toward superoxide in the reduced state, does not display a measurable superoxide dismutase activity. Discussion The amitochondriate protozoan pathogen G. intestinalis has a relatively simple life cycle. After oro-fecal transmission of the cyst, the parasite develops into the vegetative trophozoite, which attaches to the epithelial mucosa of the proximal small intestine causing the disease. The trophozoite is highly sensitive to O2, as it expresses both O2-labile key metabolic enzymes (such as pyruvate:ferredoxin oxidoreductase [44]) and ROS-producing proteins (such as the NAD(P)H:menadione oxidoreductase [45]). Moreover, the recent completion of the genome sequencing project [33] confirmed that Giardia is devoid of some of the conventional ROS-scavenging enzymes, namely SOD, catalase, and glutathione peroxidase [46], which may also account for the high vulnerability of this parasite to oxidative damage. Despite its O2 susceptibility, Giardia colonizes a fairly aerobic tract of the human gut (duodenum and jejunum) [47], which implies that survival of the trophozoite in the host should rely on its ability to cope with oxidative/nitrosative stress. In this regard, it is worth mentioning that two H2O-forming O2-reducing flavoenzymes (NADH oxidase [48] and flavodiiron protein [49–51]) and a flavohemoglobin [52,53], all probably acquired from Prokarya by lateral gene transfer [54,55], were identified in Giardia and suggested to play a role in O2 and NO detoxification, respectively. No information is currently available about Giardia and its defense mechanisms against O2•−. Only a modest killing effect (~30% dead cells) was reported on trophozoites after 2 h incubation with 1 mM xanthine and 20 mU/ml xanthine oxidase [56], a well-known superoxide donor system; this effect may not be entirely due to O2•−, because Giardia is reportedly sensitive to H2O2 [57], which is a side product of the xanthine/xanthine oxidase system [58]. The relatively low sensitivity of

Giardia toward extracellularly administered O2•− points to the presence of a superoxide-detoxification system in this SOD-lacking parasite. In this regard, it is interesting to notice that in 2006 Li and Wang [45], working on DT-diaphorase overexpressing trophozoites, had predicted “some yet unidentified, although limited, means of converting superoxide to hydrogen peroxide” in Giardia. In this study, we show that the Giardia trophozoite expresses a SOR (Fig. 1), probably acquired from a prokaryote by lateral gene transfer, possibly involved in superoxide detoxification. To gain insight into the biochemical and functional properties of SORGi, the recombinant protein was purified and characterized spectroscopically and kinetically. As shown in Fig. 2, the enzyme displays the spectral features of a bona fide 1Fe–SOR with the two characteristic spectrally distinct forms of the oxidized protein (Tfinal): the glutamate-ligated (Fe 3+–Glu) species, prevailing at lower pH, and the hydroxide-ligated (Fe 3+–OH) SOR appearing at higher pH values with an apparent pKa of 8.7. Based on site-directed mutagenesis studies carried out on prokaryotic SORs [23,24,26,40,59,60], E17 in the conserved EKHVP motif is likely to be the residue involved in iron coordination in the Giardia enzyme. Glutamate ligation to Fe 3+ has been invariantly observed in SORs whenever this residue was present. However, the function of glutamate binding remains elusive, because the initially proposed role for this residue in product release or in proton delivery to the catalytic site was ruled out by site-directed mutagenesis and in studies on the glutamate-lacking Nanoarchaeum equitans 1Fe–SOR [27]. To acquire information on the reaction mechanism of SORGi, we have investigated the reaction of this protein with superoxide, by both pulse radiolysis and stopped-flow spectrophotometry. Pulse radiolysis experiments were carried out at room temperature, whereas stopped-flow measurements were performed at 10 °C (to slow down the reaction). In analogy with its prokaryotic homologues, the enzyme from Giardia proved to be highly reactive toward superoxide in the reduced form, but had only a marginal SOD activity. The negligible SOD activity is consistent with the information that no SOD activity was detected in extracts of the parasitic trophozoites, when assayed by bulk spectrophotometry or in gel activity tests after native polyacrylamide gel electrophoresis [46]. The reaction mechanism of reduced SOR with superoxide is still a matter of debate. The main controversy concerns the actual number of intermediates populated during the reaction. Although there is general consensus on the finding that the first detectable intermediate is a ferric-(hydro)peroxo T1 species and the last one is the final (Tfinal) species (whether glutamate- or hydroxide-ligated depending on pH), there are some SORs in which the T1 → Tfinal decay occurs directly without other intermediates [22,23,25,28] and other protein homologues in which the additional T2 intermediate is detected [17,21,26]. Our pulse-radiolysis data show that in the Giardia protein the T1 intermediate forms at diffusion-controlled rates (~1.0× 109 M− 1 s− 1), independent of pH, as observed for all other SORs investigated. Afterward, this intermediate decays directly to Tfinal without populating additional intermediates, as reported for the SORs from Desulfov. vulgaris and T. pallidum [22,23,25,28]. Such a result was confirmed also by multiwavelength stopped-flow spectroscopy combined with global fitting analysis of the spectral changes. The reason why the T2 intermediate has not been detected in some prokaryotic SORs, as well as in the Giardia enzyme, remains to be clarified. A possibility that we cannot rule out is that the intermediate is transiently formed, but at undetectable levels, i.e., that in the reaction with superoxide T2 forms at a rate significantly lower than its intrinsic decay rate to the Tfinal species (k3 N N k2 in Scheme 1). Analysis of the pulse-radiolysis data shows that in the Giardia enzyme the rate constant of the T1 → Tfinal decay follows a characteristic pH dependence (Fig. 5): in the pH range 5.0–8.0, it displays a steep dependence (with lower values measured at higher pH), whereas at pH N8.0 it tends to a plateau value. A similar pH profile

F. Testa et al. / Free Radical Biology & Medicine 51 (2011) 1567–1574

was observed for the reaction of superoxide with Desulfov. vulgaris SOR and was interpreted as arising from a second-order ratedetermining protonation of the T1 intermediate at low pH, outcompeted by a first-order pH-independent decay of the intermediate at high pH (pH N8.0) [23]. In the case of the Giardia enzyme, based on our data, the former protonation process proceeds at an estimated second-order rate constant k2′ ~ 1.0 × 10 10 M − 1 s − 1, whereas the latter decay proceeds at k2″ ~ 320 s − 1. It is worth noting that these rates are somewhat greater than those previously reported for other SORs. For instance, at neutral pH, in the Giardia enzyme T1 decays with a rate constant k2 ~ 1700 s − 1, which is approximately fourfold higher than observed in the A. fulgidus 1Fe–SOR and 30–40 times higher than measured for the 2Fe–SORs from A. fulgidus and Desulfov. vulgaris (see Table 3 in [5] and references therein). A higher rate constant (4800 s − 1) was instead reported under similar experimental conditions for the enzyme from T. pallidum [25]. The faster kinetics observed with T. pallidum SOR, however, could be attributed to the use of phosphate in these measurements. Phosphate was indeed shown to act as a proton donor for the T1 intermediate, thereby increasing its decay rate in a concentration-dependent manner [26], and for this reason we avoided using such a buffer in this study on Giardia SOR. 5-Nitroimidazole metronidazole is the drug of choice against giardiasis [61]. Introduced more than 50 years ago, it is still debated how metronidazole kills Giardia and the other sensitive anaerobic microorganisms. To exert microbicidal effects, metronidazole needs to be activated by reduction to the cytotoxic nitroradical [61]. Relevant to the present study, under aerobic conditions such a nitroradical is reportedly reoxidized by O2 to the nontoxic parent compound (“futile cycle” [61]), thereby generating O2•−. In this regard, the herein documented discovery of an efficient superoxide-scavenging enzyme in Giardia opens new perspectives; in particular, future studies should aim to establish if SORGi is implicated in the degradation of the superoxide anion intracellularly produced by metronidazole upon redox cycling with O2. In conclusion, in this work we have shown that the SOD-lacking microaerobic pathogen G. intestinalis is endowed with a bona fide SOR. The enzyme is extremely reactive toward superoxide and is expressed in the disease-causing parasitic trophozoite. Future work should aim at testing if this superoxide-scavenging enzyme promotes the survival of Giardia in the proximal human small intestine, thereby enhancing its pathogenicity, and/or if it contributes to tuning metronidazole toxicity against the parasite.

Acknowledgments This work was partially supported by the Ministero dell'Istruzione, dell'Università e della Ricerca, Italy (FIRB RBFR08F41U_001 to A.G. and PRIN 2008FJJHKM_002 to P.S.), by the European Society of Clinical Microbiology and Infectious Diseases (Research Grant 2009 to A.G.), and by the Fundação para a Ciência e Tecnologia, Portugal, Research Grant PTDC/BIA-Pro/6726372006 (to M.T.). We also acknowledge a bilateral grant award (to A.G. and M.T.) by the Consiglio Nazionale delle Ricerche of Italy and Fundação para a Ciência e Tecnologia of Portugal. The Brookhaven National Laboratory pulse radiolysis facility is funded by the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences of the U.S. Department of Energy through Grant DE-AC02-98-CH10886.

Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10. 1016/j.freeradbiomed.2011.07.017.

1573

References [1] Adams, M. W.; Jenney Jr., F. E.; Clay, M. D.; Johnson, M. K. Superoxide reductase: fact or fiction? J. Biol. Inorg. Chem. 7:647–652; 2002. [2] Imlay, J. A. What biological purpose is served by superoxide reductase? J. Biol. Inorg. Chem. 7:659–663; 2002. [3] Kurtz Jr., D. M. Avoiding high-valent iron intermediates: superoxide reductase and rubrerythrin. J. Inorg. Biochem. 100:679–693; 2006. [4] Nivière, V.; Fontecave, M. Discovery of superoxide reductase: an historical perspective. J. Biol. Inorg. Chem. 9:119–123; 2004. [5] Pinto, A. F.; Rodrigues, J. V.; Teixeira, V. M. Reductive elimination of superoxide: structure and mechanism of superoxide reductases. Biochim. Biophys. Acta 1804: 285–297; 2010. [6] Jenney Jr., F. E.; Verhagen, M. F.; Cui, X.; Adams, M. W. Anaerobic microbes: oxygen detoxification without superoxide dismutase. Science 286:306–309; 1999. [7] Liochev, S. I.; Fridovich, I. A mechanism for complementation of the sodA sodB defect in Escherichia coli by overproduction of the rbo gene product (desulfoferrodoxin) from Desulfoarculus baarsii. J. Biol. Chem. 272:25573–25575; 1997. [8] Chen, L.; Sharma, P.; Le Gall, J.; Mariano, A. M.; Teixeira, M.; Xavier, A. V. A blue nonheme iron protein from Desulfovibrio gigas. Eur. J. Biochem. 226:613–618; 1994. [9] Moura, I.; Tavares, P.; Moura, J. J.; Ravi, N.; Huynh, B. H.; Liu, M. Y.; LeGall, J. Purification and characterization of desulfoferrodoxin: a novel protein from Desulfovibrio desulfuricans (ATCC 27774) and from Desulfovibrio vulgaris (strain Hildenborough) that contains a distorted rubredoxin center and a mononuclear ferrous center. J. Biol. Chem. 265:21596–21602; 1990. [10] Adam, V.; Royant, A.; Nivière, V.; Molina-Heredia, F. P.; Bourgeois, D. Structure of superoxide reductase bound to ferrocyanide and active site expansion upon Xray-induced photo-reduction. Structure 12:1729–1740; 2004. [11] Coelho, A. V.; Matias, P.; Fulop, V.; Thompson, A.; Gonzalez, A.; Carrondo, M. A. Desulfoferrodoxin structure determined by MAD phasing and refinement to 1.9-Å resolution reveals a unique combination of a tetrahedral FeS4 centre with a square pyramidal FeSN4 centre. J. Biol. Inorg. Chem. 2:680–689; 1997. [12] Katona, G.; Carpentier, P.; Nivière, V.; Amara, P.; Adam, V.; Ohana, J.; Tsanov, N.; Bourgeois, D. Raman-assisted crystallography reveals end-on peroxide intermediates in a nonheme iron enzyme. Science 316:449–453; 2007. [13] Santos-Silva, T.; Trincao, J.; Carvalho, A. L.; Bonifacio, C.; Auchere, F.; Raleiras, P.; Moura, I.; Moura, J. J.; Romão, M. J. The first crystal structure of class III superoxide reductase from Treponema pallidum. J. Biol. Inorg. Chem. 11:548–558; 2006. [14] Yeh, A. P.; Hu, Y.; Jenney Jr., F. E.; Adams, M. W.; Rees, D. C. Structures of the superoxide reductase from Pyrococcus furiosus in the oxidized and reduced states. Biochemistry 39:2499–2508; 2000. [15] Emerson, J. P.; Coulter, E. D.; Phillips, R. S.; Kurtz Jr., D. M. Kinetics of the superoxide reductase catalytic cycle. J. Biol. Chem. 278:39662–39668; 2003. [16] Rodrigues, J. V.; Abreu, I. A.; Saraiva, L. M.; Teixeira, M. Rubredoxin acts as an electron donor for neelaredoxin in Archaeoglobus fulgidus. Biochem. Biophys. Res. Commun. 329:1300–1305; 2005. [17] Rodrigues, J. V.; Saraiva, L. M.; Abreu, I. A.; Teixeira, M.; Cabelli, D. E. Superoxide reduction by Archaeoglobus fulgidus desulfoferrodoxin: comparison with neelaredoxin. J. Biol. Inorg. Chem. 12:248–256; 2007. [18] Pianzzola, M. J.; Soubes, M.; Touati, D. Overproduction of the rbo gene product from Desulfovibrio species suppresses all deleterious effects of lack of superoxide dismutase in Escherichia coli. J. Bacteriol. 178:6736–6742; 1996. [19] Abreu, I. A.; Saraiva, L. M.; Carita, J.; Huber, H.; Stetter, K. O.; Cabelli, D.; Teixeira, M. Oxygen detoxification in the strict anaerobic archaeon Archaeoglobus fulgidus: superoxide scavenging by neelaredoxin. Mol. Microbiol. 38:322–334; 2000. [20] Abreu, I. A.; Saraiva, L. M.; Soares, C. M.; Teixeira, M.; Cabelli, D. E. The mechanism of superoxide scavenging by Archaeoglobus fulgidus neelaredoxin. J. Biol. Chem. 276:38995–39001; 2001. [21] Bonnot, F.; Houee-Levin, C.; Favaudon, V.; Nivière, V. Photochemical processes observed during the reaction of superoxide reductase from Desulfoarculus baarsii with superoxide: re-evaluation of the reaction mechanism. Biochim. Biophys. Acta 1804:762–767; 2010. [22] Coulter, E. D.; Emerson, J. P.; Kurtz Jr., D. M.; Cabelli, D. Superoxide reactivity of rubredoxin oxidoreductase (Desulfoferrodoxin) from Desulfovibrio vulgaris: a pulse radiolysis study. J. Am. Chem. Soc. 122:11555–11556; 2000. [23] Emerson, J. P.; Coulter, E. D.; Cabelli, D. E.; Phillips, R. S.; Kurtz Jr., D. M. Kinetics and mechanism of superoxide reduction by two-iron superoxide reductase from Desulfovibrio vulgaris. Biochemistry 41:4348–4357; 2002. [24] Lombard, M.; Houee-Levin, C.; Touati, D.; Fontecave, M.; Nivière, V. Superoxide reductase from Desulfoarculus baarsii: reaction mechanism and role of glutamate 47 and lysine 48 in catalysis. Biochemistry 40:5032–5040; 2001. [25] Nivière, V.; Lombard, M.; Fontecave, M.; Houee-Levin, C. Pulse radiolysis studies on superoxide reductase from Treponema pallidum. FEBS Lett. 497:171–173; 2001. [26] Rodrigues, J. V.; Abreu, I. A.; Cabelli, D.; Teixeira, M. Superoxide reduction mechanism of Archaeoglobus fulgidus one-iron superoxide reductase. Biochemistry 45:9266–9278; 2006. [27] Rodrigues, J. V.; Victor, B. L.; Huber, H.; Saraiva, L. M.; Soares, C. M.; Cabelli, D. E.; Teixeira, M. Superoxide reduction by Nanoarchaeum equitans neelaredoxin, an enzyme lacking the highly conserved glutamate iron ligand. J. Biol. Inorg. Chem. 13: 219–228; 2008. [28] Huang, V. W.; Emerson, J. P.; Kurtz Jr., D. M. Reaction of Desulfovibrio vulgaris twoiron superoxide reductase with superoxide: insights from stopped-flow spectrophotometry. Biochemistry 46:11342–11351; 2007. [29] Lucchetti-Miganeh, C.; Goudenege, D.; Thybert, D.; Salbert, G.; Barloy-Hubler, F. SORGOdb: superoxide reductase gene ontology curated database. BMC Microbiol. 11:105; 2011.

1574

F. Testa et al. / Free Radical Biology & Medicine 51 (2011) 1567–1574

[30] Adam, R. D. Biology of Giardia lamblia. Clin. Microbiol. Rev. 14:447–475; 2001. [31] Ankarklev, J.; Jerlstrom-Hultqvist, J.; Ringqvist, E.; Troell, K.; Svärd, S. G. Behind the smile: cell biology and disease mechanisms of Giardia species. Nat. Rev. Microbiol. 8:413–422; 2010. [32] Franzen, O.; Jerlstrom-Hultqvist, J.; Castro, E.; Sherwood, E.; Ankarklev, J.; Reiner, D. S.; Palm, D.; Andersson, J. O.; Andersson, B.; Svärd, S. G. Draft genome sequencing of Giardia intestinalis assemblage B isolate GS: is human giardiasis caused by two different species? PLoS Pathog. 5:e1000560; 2009. [33] Morrison, H. G.; McArthur, A. G.; Gillin, F. D.; Aley, S. B.; Adam, R. D.; Olsen, G. J.; Best, A. A.; Cande, W. Z.; Chen, F.; Cipriano, M. J.; Davids, B. J.; Dawson, S. C.; Elmendorf, H. G.; Hehl, A. B.; Holder, M. E.; Huse, S. M.; Kim, U. U.; LasekNesselquist, E.; Manning, G.; Nigam, A.; Nixon, J. E.; Palm, D.; Passamaneck, N. E.; Prabhu, A.; Reich, C. I.; Reiner, D. S.; Samuelson, J.; Svärd, S. G.; Sogin, M. L. Genomic minimalism in the early diverging intestinal parasite Giardia lamblia. Science 317:1921–1926; 2007. [34] Smith, P. K.; Krohn, R. I.; Hermanson, G. T.; Mallia, A. K.; Gartner, F. H.; Provenzano, M. D.; Fujimoto, E. K.; Goeke, N. M.; Olson, B. J.; Klenk, D. C. Measurement of protein using bicinchoninic acid. Anal. Biochem. 150:76–85; 1985. [35] Stookey, L. L. Ferrozine—a new spectrophotometric reagent for iron. Anal. Chem. 42:779–781; 1970. [36] Henry, E. R.; Hofrichter, J. Singular value decomposition: application to analysis of experimental data. Methods Enzymol. 210:129–192; 1992. [37] Goto, J. J.; Gralla, E. B.; Valentine, J. S.; Cabelli, D. E. Reactions of hydrogen peroxide with familial amyotrophic lateral sclerosis mutant human copper–zinc superoxide dismutases studied by pulse radiolysis. J. Biol. Chem. 273:30104–30109; 1998. [38] Keister, D. B. Axenic culture of Giardia lamblia in TYI-S-33 medium supplemented with bile. Trans. R. Soc. Trop. Med. Hyg. 77:487–488; 1983. [39] Mathe, C.; Nivière, V.; Mattioli, T. A. Fe3+-hydroxide ligation in the superoxide reductase from Desulfoarculus baarsii is associated with pH dependent spectral changes. J. Am. Chem. Soc. 127:16436–16441; 2005. [40] Todorovic, S.; Rodrigues, J. V.; Pinto, A. F.; Thomsen, C.; Hildebrandt, P.; Teixeira, M.; Murgida, D. H. Resonance Raman study of the superoxide reductase from Archaeoglobus fulgidus, E12 mutants and a 'natural variant'. Phys. Chem. Chem. Phys. 11:1809–1815; 2009. [41] Lumppio, H. L.; Shenvi, N. V.; Summers, A. O.; Voordouw, G.; Kurtz Jr., D. M. Rubrerythrin and rubredoxin oxidoreductase in Desulfovibrio vulgaris: a novel oxidative stress protection system. J. Bacteriol. 183:101–108; 2001. [42] Romão, C. V.; Liu, M. Y.; Le Gall, J.; Gomes, C. M.; Braga, V.; Pacheco, I.; Xavier, A. V.; Teixeira, M. The superoxide dismutase activity of desulfoferrodoxin from Desulfovibrio desulfuricans ATCC 27774. Eur. J. Biochem. 261:438–443; 1999. [43] Silva, G.; Oliveira, S.; Gomes, C. M.; Pacheco, I.; Liu, M. Y.; Xavier, A. V.; Teixeira, M.; Legall, J.; Rodrigues-pousada, C. Desulfovibrio gigas neelaredoxin: a novel superoxide dismutase integrated in a putative oxygen sensory operon of an anaerobe. Eur. J. Biochem. 259:235–243; 1999. [44] Townson, S. M.; Upcroft, J. A.; Upcroft, P. Characterisation and purification of pyruvate:ferredoxin oxidoreductase from Giardia duodenalis. Mol. Biochem. Parasitol. 79:183–193; 1996.

[45] Li, L.; Wang, C. C. A likely molecular basis of the susceptibility of Giardia lamblia towards oxygen. Mol. Microbiol. 59:202–211; 2006. [46] Brown, D. M.; Upcroft, J. A.; Upcroft, P. Free radical detoxification in Giardia duodenalis. Mol. Biochem. Parasitol. 72:47–56; 1995. [47] Sheridan, W. G.; Lowndes, R. H.; Young, H. L. Intraoperative tissue oximetry in the human gastrointestinal tract. Am. J. Surg. 159:314–319; 1990. [48] Brown, D. M.; Upcroft, J. A.; Upcroft, P. A H2O-producing NADH oxidase from the protozoan parasite Giardia duodenalis. Eur. J. Biochem. 241:155–161; 1996. [49] Di Matteo, A.; Scandurra, F. M.; Testa, F.; Forte, E.; Sarti, P.; Brunori, M.; Giuffrè, A. The O2-scavenging flavodiiron protein in the human parasite Giardia intestinalis. J. Biol. Chem. 283:4061–4068; 2008. [50] Mastronicola, D.; Giuffrè, A.; Testa, F.; Mura, A.; Forte, E.; Bordi, E.; Pucillo, L. P.; Fiori, P. L.; Sarti, P. Giardia intestinalis escapes oxidative stress by colonizing the small intestine: a molecular hypothesis. IUBMB Life 63:21–25; 2011. [51] Vicente, J. B.; Testa, F.; Mastronicola, D.; Forte, E.; Sarti, P.; Teixeira, M.; Giuffrè, A. Redox properties of the oxygen-detoxifying flavodiiron protein from the human parasite Giardia intestinalis. Arch. Biochem. Biophys. 488:9–13; 2009. [52] Mastronicola, D.; Testa, F.; Forte, E.; Bordi, E.; Pucillo, L. P.; Sarti, P.; Giuffrè, A. Flavohemoglobin and nitric oxide detoxification in the human protozoan parasite Giardia intestinalis. Biochem. Biophys. Res. Commun. 399:654–658; 2010. [53] Rafferty, S.; Luu, B.; March, R. E.; Yee, J. Giardia lamblia encodes a functional flavohemoglobin. Biochem. Biophys. Res. Commun. 399:347–351; 2010. [54] Andersson, J. O.; Sjogren, A. M.; Davis, L. A.; Embley, T. M.; Roger, A. J. Phylogenetic analyses of diplomonad genes reveal frequent lateral gene transfers affecting eukaryotes. Curr. Biol. 13:94–104; 2003. [55] Nixon, J. E.; Wang, A.; Field, J.; Morrison, H. G.; McArthur, A. G.; Sogin, M. L.; Loftus, B. J.; Samuelson, J. Evidence for lateral transfer of genes encoding ferredoxins, nitroreductases, NADH oxidase, and alcohol dehydrogenase 3 from anaerobic prokaryotes to Giardia lamblia and Entamoeba histolytica. Eukaryotic Cell 1: 181–190; 2002. [56] Fernandes, P. D.; Assreuy, J. Role of nitric oxide and superoxide in Giardia lamblia killing. Braz. J. Med. Biol. Res. 30:93–99; 1997. [57] Lloyd, D.; Harris, J. C.; Maroulis, S.; Biagini, G. A.; Wadley, R. B.; Turner, M. P.; Edwards, M. R. The microaerophilic flagellate Giardia intestinalis: oxygen and its reaction products collapse membrane potential and cause cytotoxicity. Microbiology 146 (Pt 12):3109–3118; 2000. [58] Kelley, E. E.; Khoo, N. K.; Hundley, N. J.; Malik, U. Z.; Freeman, B. A.; Tarpey, M. M. Hydrogen peroxide is the major oxidant product of xanthine oxidase. Free Radic. Biol. Med. 48:493–498; 2010. [59] Berthomieu, C.; Dupeyrat, F.; Fontecave, M.; Vermeglio, A.; Nivière, V. Redoxdependent structural changes in the superoxide reductase from Desulfoarculus baarsii and Treponema pallidum: a FTIR study. Biochemistry 41:10360–10368; 2002. [60] Mathe, C.; Weill, C. O.; Mattioli, T. A.; Berthomieu, C.; Houee-Levin, C.; Tremey, E.; Nivière, V. Assessing the role of the active-site cysteine ligand in the superoxide reductase from Desulfoarculus baarsii. J. Biol. Chem. 282:22207–22216; 2007. [61] Edwards, D. I. Nitroimidazole drugs—action and resistance mechanisms. I. Mechanisms of action. J. Antimicrob. Chemother. 31:9–20; 1993.