Trypanosoma cruzi expresses a plant-like ascorbatedependent hemoperoxidase localized to the endoplasmic reticulum Shane R. Wilkinson*, Samson O. Obado, Isabel L. Mauricio, and John M. Kelly Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London WC1E 7HT, United Kingdom Edited by John H. Law, University of Arizona, Tucson, AZ, and approved August 13, 2002 (received for review July 17, 2002)
he protozoan parasite Trypanosoma cruzi, the causative agent of Chagas’ disease, infects 16–18 million people in Latin America (1). Only two drugs, benznidazole and nifurtimox, are currently available and these drugs are unsatisfactory because of limited efficacy and toxic side effects. With no prospect of a vaccine, the search for parasite-specific traits exploitable in terms of chemotherapy is a priority (1). It has been reported that trypanosomatids have a limited ability to deal with reactive oxygen species (2, 3). In the parasite these are generated by endogenous processes and as a result of external influences such as host immune responses and drug metabolism. Together with observations that some mechanisms of oxidative defense in trypanosomatids are distinct from those of the mammalian host, this finding has encouraged the view that components of this system could provide targets for drug design (4, 5). In contrast to most eukaryotes, T. cruzi lacks catalase and selenium-dependent glutathione peroxidases, enzymes capable of rapidly metabolizing high levels of H2O2 (2, 4, 5). Instead it has an array of enzyme-mediated mechanisms in which the trypanosomatid-specific thiol trypanothione (N1,N8-bisglutathionylspermidine) plays a pivotal role. In a redox cycle analogous to the glutathione兾 glutathione reductase system, trypanothione is maintained in the reduced form (T[SH]2) by the activity of trypanothione reductase (TR) at the expense of NADPH (4, 6). T[SH]2 can then facilitate the flux of reducing equivalents to either tryparedoxins, a family of thioredoxin-like molecules, or glutathione. In their reduced state, tryparedoxins and glutathione can transfer electrons to peroxidases. Four distinct peroxidases have been identified in T. cruzi. Each exhibits differences in subcellular location, substrate range, and electron donor. Two of these enzymes are 2-cys peroxiredoxins. They are localized in the cytosol (TcCPX) or mitochondrion (TcMPX) (7) and efficiently scavenge H2O2 and small-chain organic hydroperoxides. The remaining enzymes show extensive similarity to glutathione-dependent peroxidases (8, 9). One, TcGPXI is present in both the cytosol and the glycosome. The second, TcGPXII is found in the endoplasmic reticulum (ER). The
T
www.pnas.org兾cgi兾doi兾10.1073兾pnas.202422899
peroxidase activity of these enzymes can be linked to the oxidation of T[SH]2 by glutathione, although tryparedoxin can also act as electron donor to TcGPXI (10). The ability of TcGPXI to use electrons from different donors may reflect the absence of one of the intermediaries in the glycosome. TcGPXI and TcGPXII seem to play specific roles in oxidative defense because both can detoxify fatty acid and phospholipid hydroperoxides but do not have activity toward H2O2. In mammals, ascorbic acid is involved in several important physiological reactions including collagen formation and iron absorption (11). However, its major function is as a powerful antioxidant (11, 12). In plants, it has a pronounced role in oxidative defense, acting as an electron donor to the heme-containing ascorbate peroxidase (APX) family of H2O2-detoxifying proteins (13). These enzymes have been located in all oxy-radical-generating compartments in plants, with multiple isoforms identified (14). In T. cruzi, significant levels of ascorbate and its oxidized form, dehydroascorbate (DHA), have been detected and previous studies have identified a microsomal-associated ascorbate-dependent peroxidase activity (2, 15, 16). Here we show that T. cruzi expresses an unusual plant-like ascorbate-dependent hemoperoxidase whose activity is linked to the trypanothione system. Materials and Methods DNA and RNA Extractions. T. cruzi epimastigotes were grown as described (17). Transformed T. cruzi were maintained in the same medium with 200 g ml⫺1 of G418. Genomic DNA was isolated by the proteinase K兾SDS method and total RNA prepared by guanidinum thiocyanate lysis (18). Isolation of an Ascorbate Peroxidase Gene. T. cruzi ascorbate per-
oxidase (TcAPX) was isolated by PCR by using epimastigote cDNA prepared with the primer GAATTCGATATCGGTACC(T)16. A DNA fragment containing the 5⬘ coding sequence was amplified by using gaattcGTGGGCCCCGATCAAGGCAAC and the spliced leader primer ggatccACAGTTTCTGTACTATTG. The 3⬘ coding sequence was amplified by using aagcttCGTGGATGCAAAGGACGGC and gaattcGATATCGGTACC. Restriction sites (lowercase) were incorporated into primers to facilitate cloning of the product into pBluescript KS(⫺) (Stratagene). A derivative of TcAPX was amplified from genomic DNA by using agatctCTCGGGACGAAAGGTCACG and aagcttTCACTTTGCCTATTTTGACTC. The amplified product was cloned into the BamHI and HindIII sites of the vector pTrcHis-C (Invitrogen). DNA was sequenced by using a dye terminator cycle kit (Applied Biosystems) and an Applied Biosystems Prism 377 Sequencer. This paper was submitted directly (Track II) to the PNAS office. Abbreviations: ER, endoplasmic reticulum; APX, ascorbate peroxidase; DHA, dehydroascorbate; TcAPX, T. cruzi ascorbate peroxidase; T[SH]2, reduced form of trypanothione; TR, trypanothione reductase. Data deposition: The TcAPX sequence reported in this paper has been deposited in the GenBank database (accession no. AJ457987). *To whom correspondence should be addressed. E-mail:
[email protected].
PNAS 兩 October 15, 2002 兩 vol. 99 兩 no. 21 兩 13453–13458
BIOCHEMISTRY
In most aerobic organisms hemoperoxidases play a major role in H2O2-detoxification, but trypanosomatids have been reported to lack this activity. Here we describe the properties of an ascorbate-dependent hemoperoxidase (TcAPX) from the American trypanosome Trypanosoma cruzi. The activity of this plant-like enzyme can be linked to the reduction of the parasite-specific thiol trypanothione by ascorbate in a process that involves nonenzymatic interaction. The role of heme in peroxidase activity was demonstrated by spectral and inhibition studies. Ascorbate could saturate TcAPX activity indicating that the enzyme obeys Michaelis–Menten kinetics. Parasites that overexpressed TcAPX activity were found to have increased resistance to exogenous H2O2. To determine subcellular location an epitope-tagged form of TcAPX was expressed in T. cruzi, which was observed to colocalize with endoplasmic reticulum resident chaperone protein BiP. These findings identify an arm of the oxidative defense system of this medically important parasite. The absence of this redox pathway in the human host may be therapeutically exploitable.
Heterologous Expression and Purification of His-Tagged TcAPX. Escherichia coli BL-21, transformed with the plasmid pTrcHis-APX, were grown in NZCYM broth containing 50 g䡠ml⫺1 ampicillin, 1 mM ␦-aminolevulinic acid, and protein expression induced by isopropyl -D-thiogalactoside (8). His-tagged TcAPX was affinity purified on a Ni-NTA column (Qiagen, Chatsworth, CA). Each step was performed in the presence of protease inhibitors (Roche Molecular Biochemicals) and 100 M ascorbate. Protein concentrations were determined by bicinchoninic acid assay system (Pierce). Enzyme Activity. TcAPX activity was measured by following the
change in absorbance at 290 nm due to ascorbate oxidation. A standard reaction (1 ml) containing 50 mM sodium phosphate, pH 6.5, 100 M ascorbate, and 0.1 M TcAPX was incubated at room temperature for 5 min. The background rate of ascorbate oxidation was determined, and the reaction initiated by addition of 20 M H2O2. Enzyme activity was calculated by using a value of 2,880 M⫺1䡠cm⫺1. Data were analyzed by fitting to a rectangular hyperbola (KINENORT program, A. G. Clark, Victoria University of Wellington, Wellington, New Zealand). The time-dependent inactivation of TcAPX was followed by incubating 2 M peroxidase with H2O2 or sodium azide in 50 mM sodium phosphate buffer, pH 6.5. At time intervals, the residual TcAPX activity was determined. To measure activity in parasite extracts, epimastigotes in the logarithmic phase of growth were pelleted and washed in 50 mM sodium phosphate buffer, pH 6.5. The cell pellet was resuspended (2–5 ⫻ 109 cells) in sodium phosphate buffer containing protease inhibitors (Roche Molecular Biochemicals) and 100 M ascorbate, and the cells lysed by freeze-thawing three times. The lysate was clarified and the supernatant dialyzed against 50 mM sodium phosphate buffer, pH 6.5, containing 100 M ascorbate at 4°C. TcAPX activity of extracts was measured by following the change in absorbance at 340 nm due to NADPH oxidation. A standard reaction (1 ml) containing 50 mM sodium phosphate, pH 6.5, 100 M ascorbate, 50 M trypanothione disulfide, 200 M NADPH, 0.5 M TR, and 100 g of parasite extract was incubated at room temperature for 5 min. The background rate of NADPH oxidation was determined, and the reaction initiated by addition of 20 M H2O2. Enzyme activity was calculated by using an value of 6,220 M⫺1䡠cm⫺1. Assays were performed in triplicate. Construction of T. cruzi Vectors and Parasite Transformation. The vector used for the localization of the TcAPX was constructed as follows. The primers ggatccATGGCTTTTTGTTTTGGTTCA and gatatcTTTTGACTCTGCTGGGAGAGA were used to amplify the TcAPX coding sequence from genomic DNA. The amplified fragment was purified and ligated into the BamHI兾EcoRV sites of the T. cruzi vector pTEX-9E10 (19). The ligation was performed such that a c-myc-derived epitope (9E10) was inserted in-frame at the 3⬘ end of the peroxidase gene to produce pTEX-APX9E10. For overexpression studies the entire TcAPX ORF was ligated into the expression vector pTEX (20). After electroporation transformed T. cruzi were selected by using the conditions described (21). Localization by Immunofluorescence. Trypanosomes, expressing an
epitope-tagged form of TcAPX, were fixed, then permeabilized (9). Cells were incubated with 10% FBS (Sigma) diluted in PBS (137 mM NaCl兾4 mM Na2HPO4兾1.7 mM KH2PO4/2.7 mM KCl) and colabeled with a mouse anti-c-myc (9E10) antibody (Santa Cruz Biotechnology) (1:200) and rabbit anti-TbBiP serum (1:400) (J. D. Bangs, University of Wisconsin Medical School, Madison). Dilutions were performed as indicated. After 90 min, the slides were extensively washed in PBS then incubated for 60 min with AlexaFluor 488 goat anti-mouse and Alexa-Fluor 546 goat anti-rabbit serum (Molecular Probes) both diluted 1:400. Parasite DNA was then stained with 5 M TOTO-3 (Molecular Probes) in PBS. 13454 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.202422899
Images were obtained by using a Zeiss axioplan LSM 510 confocal microscope. Drug Sensitivity Experiments. Epimastigotes in the logarithmic phase of growth were seeded at 5 ⫻ 105 ml⫺1 into multiwell plates in 2 ml of growth medium supplemented with the agent under investigation. After 5 days growth at 27°C, cell density was determined and the concentration that inhibited parasite growth by 50% (IC50) established (22).
Results Isolation of a T. cruzi Ascorbate Peroxidase Gene. A T. cruzi EST
(AI057674) (23) was identified as encoding an internal region of a putative APX. DNA primers were designed and used in reverse transcription–PCR with primers against either the T. cruzi spliced leader or poly(A) tail. This design generated two overlapping products: a 690-bp fragment that contained the 5⬘ end of the gene and a 780-bp fragment that contained the 3⬘ end (Materials and Methods). To show that these fragments were contiguous, primers that overlapped the start and stop codons were used to amplify the complete ORF from T. cruzi genomic DNA. This amplification produced a fragment of 987 bp with the potential to encode a 33-kDa protein (designated TcAPX). Examination of the protein sequence (Fig. 1) indicated that TcAPX is related to the class I group of hemoperoxidases having 30–35% identity to plant APXs. When compared with the cytosolic APXs, TcAPX contains an amino-terminal extension of 55 residues. Many other APXs display such a feature, which is generally associated with targeting to an organelle, commonly the chloroplast. The amino extension of TcAPX is distinct in that it contains a positively charged region followed by a stretch of 24 hydrophobic amino acids that have the potential to form a transmembrane domain (Fig. 1). Several APXs have been shown to be membrane-associated, but usually the transmembrane domain is toward the carboxyl terminus (14). This amino acid topology is also reminiscent of the signal sequences that target proteins to the ER or plasma membrane (24). In addition, TcAPX also contains three regions conserved among plant APXs (Fig. 1). These sequences are believed to form an ascorbateinteracting surface adjacent to the distal heme-binding ligand (14, 25). This configuration brings ascorbate into close proximity with the heme moiety. One feature that does distinguish TcAPX from the plant enzymes is the presence, close to the carboxyl terminus, of a positively charged 17-aa insert of unknown function (Fig. 1). The genomic organization and expression of TcAPX was determined by DNA兾RNA hybridization and by CHEFE analysis. TcAPX is a single-copy gene localized on a 1,700-kb chromosome and is expressed in epimastigotes as a 1.9-kb transcript (data not shown). Biochemical Properties of TcAPX. To investigate peroxidase activity,
TcAPX was cloned into pTrcHis-C and expressed in E. coli. In this system, TcAPX is expressed as a recombinant protein tagged at its amino terminal with a histidine-rich sequence and an epitope detectable with the anti-Xpress monoclonal antibody (Invitrogen). The only construct that gave active, soluble protein was a deletion derivative in which the recombinant protein lacked the aminoterminal extension (Fig. 1). After induction with isopropyl -Dthiogalactoside this construct generated a 30-kDa protein that could be readily purified by one round of affinity chromatography on a nickel-nitrilotriacetic acid column (Fig. 2A). Substrate specificity was investigated by using ascorbate as electron donor, which involved monitoring changes in absorbance at 290 nm in response to either H2O2, t-butyl hydroperoxide, or cumene hydroperoxide. Assays were performed by using purified recombinant TcAPX or E. coli extracts that had been affinitypurified in parallel. Such control assays accounted for any background autooxidation of ascorbate and any E. coli peroxidase activity that may have copurified. These experiments confirmed Wilkinson et al.
that TcAPX could use ascorbate as an electron donor and that the substrate range was restricted to H2O2; no activity toward the other hydroperoxides was observed. To investigate the type of kinetics that TcAPX displays toward ascorbate and H2O2, assays were performed by using various concentrations of H2O2 against fixed concentrations of ascorbate (Fig. 2). The interaction of TcAPX with H2O2 was analyzed by plotting [H2O2]兾TcAPX activity against H2O2 (Fig. 2C). This analysis generated a series of linear slopes for all ascorbate concentrations and these slopes converged to allow the true KM for H2O2 of 3.1 ⫾ 0.5 M to be determined (Fig. 2C). Secondary plot analysis of these data gave a Vmax value of 9,940 ⫾ 489 nmol of ascorbate oxidized min⫺1䡠mg⫺1 and a substrate specificity (Kcat兾KM) value of 3.5 ⫻ 106 M⫺1䡠s⫺1. To determine whether TcAPX activity could be saturated by ascorbate, secondary plot analysis of [ascorbate]兾app Vmax against [ascorbate] was performed (Fig. 2B), which produced a linear plot and allowed the true KM for ascorbate of 192 ⫾ 19 M to be determined. Both enzymatic and nonenzymatic mechanisms for the regeneration of ascorbate from its oxidized DHA form have been postulated in T. cruzi (16, 26). To address this issue, the soluble fraction of dialyzed T. cruzi cell extracts was assayed for DHA reductase activity by monitoring the oxidation of NADPH to NADP⫹ (data not shown). These reactions were performed in the presence of recombinant TcAPX and ascorbate and initiated by addition of H2O2. No activity was detected unless trypanothione and excess TR were added to the reaction. Moreover, this activity could be reconstituted in the presence of trypanothione and its associated reductase when cell extracts were omitted from the assay, which suggests that DHA reduction occurs at the expense of T[SH]2 by a nonenzymatic process (Fig. 2), in agreement with earlier studies (26). The spectral properties of TcAPX were examined to determine whether it had the typical absorption spectrum of a heme-binding Wilkinson et al.
APX (Fig. 3A). In its resting state TcAPX has a Soret absorption peak at 413 nm with minor peaks in the visible spectrum at 531 and 559 nm (Fig. 3A). Addition of H2O2 resulted in a shift in the Soret absorption peak to 421 nm and an increase in absorption at the other two wavelengths (Fig. 3A). In other APXs, spectral changes of this type stem from the H2O2-mediated oxidation of the ferric iron within the ‘‘resting enzyme’’ to the oxyferryl (Fe4⫹ ⫽ O) form. Initially, on formation of the oxyferryl form, a free radical is generated on the heme ring (compound I), which is then sequentially reduced by ascorbate, first to compound II (the oxyferryl form without the associated free radical) and then finally back to the resting state. In the experiments conducted here, ascorbate was present within the assay and, therefore, the formation of compound I was not observed. The spectral properties observed in this experiment for the oxidized enzyme are due to the compound II form of TcAPX. Sodium azide and H2O2 have been shown to inactivate the heme moiety in APXs (27). To ascertain whether TcAPX displayed similar properties, the enzyme was incubated with different concentrations of these agents and the residual peroxidase activity was determined at various time intervals (Fig. 3 B and C). In the absence of ascorbate, both sodium azide and H2O2 inhibited TcAPX activity in a concentration- and time-dependent manner. These inhibition profiles, in conjunction with the spectral and sequence data, strongly indicate that TcAPX is a heme-containing enzyme. Consistent with this finding, addition of the heme precursor ␦-aminolevulinic acid significantly increased the yield of active recombinant protein when added to the bacterial growth medium. Localization of TcAPX in T. cruzi. The subcellular location of TcAPX
was examined by expressing a tagged version of the enzyme by using the T. cruzi vector pTEX-9E10 (19). We have used this system for the subcellular location of mitochondrial and cytosolic peroxirePNAS 兩 October 15, 2002 兩 vol. 99 兩 no. 21 兩 13455
BIOCHEMISTRY
Fig. 1. Sequence analysis of T. cruzi APX. The sequence of TcAPX was aligned with other ascorbate-dependent peroxidases; each of the different subcellular isoforms in Arabidopsis thaliana and the pea cytosolic form. The residues that are common with the TcAPX sequence are represented by dots; dashes represent gaps in the sequence made to optimize the alignments. Differences between the sequences when compared with TcAPX are indicated. Amino acids implicated in the redox activity of APXs are noted in capital letters above each aligned block and d represents residues involved in electrostatic interactions between dimers in the pea cytosolic APX. The boxed region in TcAPX represents a hydrophobic domain and the three conserved segments that form the ascorbate-interacting surface are highlighted. The arrow corresponds to the primer used for expression of the recombinant protein (Materials and Methods). The sequences aligned are: 1, T. cruzi TcAPX (AJ457987). 2, A. thaliana stromal APX (CAA67425). 3, A. thaliana thylakoid APX (CAA67426). 4, A. thaliana cytosolic, soluble APX (Q05431). 5, A. thaliana membrane associated APX (CAA66640). 6, P. sativum cytosolic, soluble APX (P48534).
Fig. 2. Investigating the ascorbate-dependent activity of TcAPX. (A) Fractions obtained during the purification of TcAPX were resolved on a 10% SDS兾PAGE gel and visualized by Coomassie staining. A clarified fraction of an E. coli (pTrcHisAPX) cell lysate after 3 h isopropyl -D-thiogalactoside induction (lane 1) was loaded onto a Ni-NTA column and the flowthrough collected (lane 2). The column was washed extensively with 10 mM imidazole (lanes 3 and 4). TcAPX eluted at 200 mM imidazole (lane 5). Markers show molecular mass in kilodaltons. The band of 30 kDa (indicated) corresponds to recombinant TcAPX. (B) TcAPX activity was assayed by following the oxidation of ascorbate (25–300 M) in the presence TcAPX (0.1 M) and H2O2 (0 –10 M). At each ascorbate concentration, the app Vmax was calculated and then examined further by secondary plot analysis. (C) TcAPX activity was assayed by following the oxidation of ascorbate [25 (F), 50 (E), 100 (), and 300 (ƒ) M] in the presence TcAPX (0.1 M) and H2O2 (0 –10 M). All assays were initiated by the addition of the H2O2. TcAPX activities are expressed as nanomoles of ascorbate oxidised per minute per milligram, while [ascorbate] and [H2O2] are expressed in micromolar. The postulated scheme for the metabolism of H2O2 in T. cruzi by an ascorbate peroxidase pathway is shown. Trypanothione disulfide (TS2) is reduced to dihydrotrypanothione (T[SH]2) by the NADPHdependent TR. TcAPX can reduce H2O2 to H2O by using T[SH]2 as electron donor in the presence of ascorbate.
doxins (10). The TcAPX ORF was ligated upstream of a 10-aa epitope from the human c-myc protein. The vector was introduced into T. cruzi and G418-resistant cells selected. These were verified by hybridization (data not shown) and by Western blot analysis with a monoclonal antibody against the epitope (Fig. 4A), which recognized a 33-kDa band corresponding to the tagged protein in transformed cell lines and did not crossreact with wild-type T. cruzi cell extracts. Transformed T. cruzi cells were fixed, permeablized, and costained with the monoclonal antibody and the DNA dye, TOTO-3, which resulted in a punctate pattern within the cell body, with a high degree of labeling around but not within the nucleus. This type of pattern has been observed in trypanosomes for proteins found within the ER (9, 28). To confirm this observation, parasites were costained additionally with antiserum raised against the T. brucei ER-resident chaperone, BiP (28) (Fig. 4B, 2 and 3). When the images were superimposed, the pattern of colocalization (Fig. 4B, 4 and 6) indicated that TcAPX is predominately located in the same compartment as BiP. Overexpression of TcAPX in T. cruzi Confers Resistance Toward Exogenous H2O2. The ORF of TcAPX was amplified from genomic DNA
and cloned into the vector pTEX (20). The resultant construct,
13456 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.202422899
Fig. 3. TcAPX is a hemoperoxidase. (A) The absorption spectra from 400 to 620 nm are shown for 10 ⌴ TcAPX in 50 mM sodium phosphate buffer (pH 6.5), 30 M ascorbate before (a) and after (b) the addition of 100 M H2O2. The shift in the Soret peak from 413 to 421 nm is indicated. The inhibition of TcAPX (2 M) activity was examined in the presence of 0 (F), 10 (E), and 100 () M H2O2 (B) or 0 (F), 100 (E), and 1000 () M sodium azide (C) in 50 mM sodium phosphate buffer (pH 6.5). At time intervals 50-l samples were assayed for APX activity in the presence of 100 M ascorbate and 20 M H2O2. The data are expressed as a percentage of activity compared with the initial activity.
pTEX-APX, was used to transform T. cruzi. The presence of multiple copies of the episome within the parasite was verified by Southern hybridization and elevated expression confirmed by RNA hybridization (data not shown). The higher level of expression was also detected as an approximately 5-fold increase in trypanothionedependent, ascorbate-mediated peroxidase activity in transformed cell extracts compared with that of the wild-type (512 ⫾ 31 vs. 108 ⫾ 15 nmol NADPH oxidized min⫺1䡠mg⫺1). The transformed and wild-type cell lines were grown in the presence of H2O2 and the IC50 was determined (Fig. 5). When exposed to exogenous H2O2, the TcAPX overexpressing cell line showed an approximately 2-fold increase in resistance compared with controls. Cells expressing the c-myc-tagged version of the protein also displayed increased APX activity and exhibited a similar H2O2-resistance phenotype. The generation of reactive oxygen species has been implicated in the mechanism of action of the trypanocidal agents nifurtimox, benznidazole, and gentian violet (29, 30). By using the approach described in Fig. 5, we investigated whether overexpression of TcAPX conferred resistance against these agents. No significant differences between overexpressing and control cell lines were observed. Discussion This article describes the first instance, to our knowledge, of an ascorbate-dependent hemoperoxidase in a nonphotosynthetic organism. On the basis of amino acid sequence and subcellular location, hemoperoxidases can be divided into three major classes (31). Class I includes the intracellular peroxidases, which can be further subdivided into the bacterial catalase-peroxidases and plant APXs, with TcAPX falling into the latter group. The secreted peroxidases of fungal and plant origin form the remainWilkinson et al.
Fig. 4. Immunolocalization of TcAPX to the ER in T. cruzi epimastigotes. (A) Specificity of the c-myc (9E10) antiserum was examined by probing a blot containing 10 g of cell lysates from T. cruzi wild-type (lane 1) and T. cruzi expressing epitope-tagged TcAPX (lane 2). Protein loading was judged by amido black staining. An arrow indicates the c-myc (9E10)-tagged TcAPX (33 kDa). (B) T. cruzi epimastigote cells expressing epitope-tagged TcAPX were stained with TOTO-3 (blue; 1), anti-c-myc (9E10; green; 2), and anti-BiP (red; 3) and the pattern of TcAPX and BiP colocalization observed (yellow; 4). A phase image of the cell (5) was obtained, on which all images were superimposed (6).
ing classes. Extensive structural analysis of class I hemoperoxidases has revealed that, although they share low sequence identity, their three-dimensional structures are conserved, suggesting that similar mechanisms are used to mediate H2O2 reduction (25, 32). This conservation is pronounced in the two regions that together allow binding of the heme cofactor and thus constitute part of the redox active center of the enzyme. The distal heme-binding ligand involves hydrogen bond formation between tryptophan (W41), histidine (H42), and arginine (R38) (Fig. 1; numbering in accordance with the Pisum sativum cytosolic APX crystal structure; ref. 25) with R38 binding through a water molecule to the heme ring. The proximal heme-binding ligand involves hydrogen bonding between histidine (H163), tryptophan (W179), and aspartate (D208), with H163 binding directly to heme. In TcAPX, all six amino acids are present at equivalent positions (Fig. 1; distal binding ligand residues at R89, W92, and H93; proximal binding ligand residues at H217, W233, and D278), suggesting that this enzyme functions in a manner mechanistically similar to plant APXs. Analysis of the absorbance spectra (Fig. 3) confirms that TcAPX is a hemoperoxidase, an activity previously thought absent from trypanosomatids. The profiles show the characteristic shift in the Wilkinson et al.
Soret peak after H2O2 treatment that corresponds to a change in the valance state of the iron atom within the heme moiety. Furthermore, TcAPX activity could be inhibited by H2O2 and sodium azide (Fig. 3) and we observed increased production of active recombinant protein in the presence of the heme precursor ␦-aminolevulinic acid. APXs display different types of enzymatic properties toward ascorbate. For those that exist predominantly as homodimers, such as the cytosolic APXs from peas and soybean, pseudo first-order kinetics have been observed (33, 34). In contrast, monomeric forms such as the peroxisomal APX from spinach and APXs from various unicellular algae obey Michaelis–Menten kinetics (35, 36), which has led to the suggestion that the kinetics displayed by the homodimeric enzymes may be due to variation in the oligomeric form of the protein, especially when ascorbate concentrations are low (33). For TcAPX, a KM value for ascorbate could be determined when using saturating concentrations of H2O2, indicating that this enzyme displays kinetics similar to the monomeric APXs (Fig. 2) (35, 35, 46). Further evidence that TcAPX is monomeric can be inferred from the sequence; TcAPX lacks many of the charged residues present in P. sativum cytosolic APX that have been implicated in homodimer formation (Fig. 1). When the effect of ascorbate on TcAPX activity was examined further at a range of H2O2 concentrations, a group of converging plots was obtained (Fig. 2), which allowed the KM value for H2O2 to be calculated (3.1 M). In general, APXs, including TcAPX, have high affinity for H2O2, in contrast to other H2O2-metabolizing enzymes such as catalase. These differences may reflect biological function with enzymes such as APX being more effective when H2O2 levels are low (37). Two peroxide-metabolizing enzymes have now been localized to the ER of T. cruzi, TcAPX and the glutathione-dependent peroxidase, TcGPXII (9). These enzymes have distinct substrate ranges with TcGPXII restricted to fatty acid and phospholipid hydroperoxides. Therefore, it seems that TcAPX complements TcGPXII giving the parasite the ability to deal with a range of oxidant challenges in this organelle. In eukaryotes many proteins are synthesized and undergo posttranslational modification in the ER. These processes require an oxidized environment, which is rePNAS 兩 October 15, 2002 兩 vol. 99 兩 no. 21 兩 13457
BIOCHEMISTRY
Fig. 5. Susceptibility of pTEX-APX-transformed and wild-type T. cruzi epimastigotes to H2O2. Wild-type (shaded symbols) and TcAPX-overexpressing (open symbols) parasites were grown in the presence of different concentrations of H2O2 and the cell numbers determined (Materials and Methods). The concentration of H2O2 that inhibited parasite growth by 50% (IC50) was established. The differences observed in sensitivity toward H2O2 in the wild-type and transformed cell lines were statistically significant (P ⬍ 0.01), as assessed by Student’s t test.
flected in a relatively low GSH:GSSG ratio (approximately 2:1 compared with approximately 50:1 in the cytosol) and the presence of enzymes that produce oxidizing compounds (38–40). Paradoxically, in the ER and associated vesicular structures, millimolar levels of ascorbate have been reported (41, 42). To maintain these levels the oxidized form of the vitamin, DHA, is preferentially transported into the lumen of the ER where it undergoes reduction to ascorbate (43, 44). In most organisms ascorbate regeneration can occur nonenzymatically by, for example, interaction with glutathione, or by means of the activity of enzymes including protein disulfide isomerase and DHA reductase (44, 45). In trypanosomatids both enzymatic and nonenzymatic mechanisms have been postulated for this recycling process (16, 26). We were able to confirm the findings of Krauth–Siegel and Ludemann (26) regarding the nonenzymatic mechanism of this reaction in T. cruzi (Fig. 2). Overexpression of components of the T. cruzi oxidative defense system can lead to altered sensitivity toward trypanocidal agents and exogenous hydroperoxides (7, 10, 46). Here we found that elevated levels of TcAPX confer increased resistance toward exogenous H2O2 (Fig. 6). Because H2O2 is an uncharged molecule, it can cross biological membranes and is able to access organelles. Thus, overexpression of TcAPX can prevent potentially lethal damage that results from the interaction of H2O2 with components of the ER. Although the precise mechanisms by which the trypanocidal agents nifurtimox, benznidazole, and gentian violet function are unknown, the generation of reactive oxygen species has been
implicated (29, 30). Cell lines expressing elevated levels of TcAPX did not show increased resistance to these agents. Therefore, it can be inferred that their trypanocidal activity does not result from H2O2-mediated damage within the ER. With the characterization of TcAPX, we have now identified five peroxidases in T. cruzi, each of which has a distinctive substrate specificity and subcellular location. Our finding that peroxide metabolism in T. cruzi is mediated by a series of linked and complementary pathways contrasts with earlier reports that the parasite is deficient in this aspect of oxidative defense. The plantlike nature of this hemoperoxidase suggests that it may have an origin early in the evolution of the trypanosomatid lineage. Some other trypanosome enzymes, including the T. cruzi peroxidase, TcGPXI (9, 10), share surprising similarity with their plant counterparts and it has been suggested that this similarity may reflect an engulfment and gene transfer event in the evolution of these organisms (9, 47). No mammalian equivalent exists to the ascorbate-dependent peroxidase and, therefore, the redox pathway together with the mechanisms by which trypanosomes scavenge and兾or synthesize ascorbate should be considered as potential drug targets.
1. World Health Organization (1997) Fourteenth Programme Report of the United Nations Development Program兾World Bank兾World Health Organization, Research and Training in Tropical Diseases (TDR) (W.H.O., Geneva). 2. Boveris, A., Sies. H., Martino, E. E., Docampo, R., Turrens, J. F. & Stoppani, A. O. (1980) Biochem. J. 188, 643–648. 3. Carnieri, E. V. S., Moreno, S. N. J. & Docampo, R. (1993) Mol. Biochem. Parasitol. 61, 79–86. 4. Fairlamb, A. H. & Cerami, A. (1992) Annu. Rev. Microbiol. 46, 695–729. 5. Flohe, L., Hecht, H. J. & Steinert, P. (1999) Free Radical Biol. Med. 27, 966–984. 6. Shames, S. L., Fairlamb, A. H., Cerami, A. & Walsh, C. T. (1986) Biochemistry 25, 3519–3526. 7. Wilkinson, S. R., Temperton, N. J., Mondragon, A. & Kelly, J. M. (2000) J. Biol. Chem. 275, 8220–8225. 8. Wilkinson, S. R., Meyer, D. J. & Kelly, J. M. (2000) Biochem. J. 352, 755–761. 9. Wilkinson, S. R., Taylor, M. C., Touitha, S., Mauricio, I. L., Meyer, D. J. & Kelly, J. M. (2002) Biochem. J. 364, 787–794. 10. Wilkinson, S. R., Meyer, D. J., Taylor, M. C., Bromley, E. V., Miles, M. A. & Kelly, J. M. (2002) J. Biol. Chem. 277, 17062–17071. 11. Halliwell, B. (2001) Mutat. Res. 475, 29–35. 12. Smirnoff, N. (2001) Curr. Opin. Plant Biol. 3, 229–235. 13. Asada, K. (1992) Physiol. Plant. 85, 235–241. 14. Jespersen, H. M., Kjaersgard, I. V., Ostergaard, L. & Welinder, K. G. (1997) Biochem. J. 326, 305–310. 15. Docampo, R., de Boiso, J. F., Boveris, A. & Stoppani, A. O. (1976) Experientia 32, 972–975. 16. Clark, D., Albrecht, M. & Arevalo, J. (1994) Mol. Biochem. Parasitol. 66, 143–145. 17. Kendall, G., Wilderspin, A. F., Ashall, F., Miles, M. A. & Kelly, J. M. (1992) EMBO J. 9, 2751–2758. 18. Kelly, J. M. (1993) Protocols in Molecular Parasitology (Humana, Totowa, NJ). 19. Tibbetts, R. S., Klein, K. G. & Engman, D. M. (1995) Exp. Parasitol. 80, 572–574. 20. Kelly, J. M., Ward, H. M., Miles, M. A. & Kendall, G. (1992) Nucleic Acids Res. 20, 3963–3969. 21. Kelly, J. M., Taylor, M. C., Rudenko, G. & Blundell, P. A. (1995) Methods in Molecular Biology (Humana, Totowa, NJ), Vol. 47. 22. Kelly, J. M., Taylor, M. C., Smith, K., Hunter, K. J. & Fairlamb, A. H. (1993) Eur. J. Biochem. 218, 29–37. 23. Porcel, B. M., Tran, A. N., Tammi, M., Nyarady, Z., Rydaker, M., Urmenyi, T. P., Rondinelli, E., Pettersson, U., Andersson, B. & Aslund, L. (2000) Genome Res. 10, 1103–1107.
24. 25. 26. 27. 28.
13458 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.202422899
We thank M. Taylor and D. Meyer for critical reading of this manuscript, D. Engman (Northwestern University, Chicago) for pTEX-9E10, J. D. Bangs (University of Wisconsin Medical School, Madison) for anti-BiP serum, and members of the T. cruzi genome project. This study was supported by The Wellcome Trust.
29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47.
Martoglio, B. & Dobberstein, B. (1998) Trends Cell Biol. 8, 410–415. Patterson, W. R. & Poulos, T. L. (1995) Biochemistry 34, 4331–4341. Krauth-Siegel, R. L. & Ludemann, H. (1996) Mol. Biochem. Parasitol. 80, 203–208. Mittler, R. & Zilinskas, B. A. (1993) Anal. Biochem. 212, 540–546. Bangs, J. D., Uyetake, L., Brickman, M. J., Balber, A. E. & Boothroyd, J. C. (1993) J. Cell Sci. 105, 1101–1113. Docampo, R. & Stoppani, A. O. M. (1979) Arch. Biochem. Biophys. 197, 317–321. Docampo, R., Moreno, S. N. J. & Cruz, F. S. (1988) Mol. Biochem. Parasitol. 27, 241–248. Welinder, K. G. (1992) Curr. Opin. Struct. Biol. 2, 388–393. Goodin, D. B. & McRee, D. E. (1993) Biochemistry 32, 3313–3324. Patterson, W. R. & Poulos, T. L. (1994) J. Biol. Chem. 269, 17020–17024. Marquez, L. A., Quitoriano, M., Zilinskas, B. A. & Dunford, H. B. (1996) FEBS Lett. 389, 153–156. Ishikawa, T., Yoshimura, K., Sakai, K., Tamoi, M., Takeda, T. & Shigeoka, S. (1998) Plant Cell Physiol. 39, 23–34. Sano, S., Ueda, M., Kitajima, S., Takeda, T., Shigeoka, S., Kurano, N., Miyachi, S., Miyake, C. & Yokota, A. (2001) Plant Cell Physiol. 42, 433–440. Willekens, H., Chamnongpol, S., Davey, M., Schraudner, M., Langebartels, C., Van Montagu, M., Inze, D. & Van Camp, W. (1997) EMBO J. 16, 4806–4816. Hwang, C., Sinskey, A. J. & Lodish, H. F. (1992) Science 257, 1496–1502. Suh, J.-K., Poulsen, L. L., Ziegler, D. M. & Robertus, J. D. (1999) Proc. Natl. Acad. Sci. USA 96, 2687–2691. Banhegyi, G., Lusini, L., Puskas, F., Rossi, R., Fulceri, R., Braun, L., Mile, V., di Simplicio, P., Mandl, J. & Benedetti, A. (1999) J. Biol. Chem. 274, 12213–12216. Daniels, A. J., Dean, G., Viveros, O. H. & Diliberto, E. J., Jr. (1982) Science 216, 737–739. von Zastrow, M., Tritton, T. R. & Castle, J. D. (1986) Proc. Natl. Acad. Sci. USA 83, 3297–3301. Banhegyi, G., Marcolongo, P., Puskas, F., Fulceri, R., Mandl, J. & Benedetti, A. (1998) J. Biol. Chem. 273, 2758–2762. Nardai, G., Braun, L., Csala, M., Mile, V., Csermely, P., Benedetti, A., Mandl, J. & Banhegyi, G. (2001) J. Biol. Chem. 276, 8825–8828. Wells, W. W., Xu, D. P., Yang, Y. F. & Rocque, P. A. (1990) J. Biol. Chem. 265, 15361–15364. Temperton, N. J., Wilkinson, S. R., Meyer, D. J. & Kelly, J. M. (1998) Mol. Biochem. Parasitol. 96, 167–176. Cavalier-Smith, T. (1993) Microbiol. Rev. 57, 953–994.
Wilkinson et al.
