Portal de Periódicos da CAPES

Sequence Analysis of the Tryparedoxin Peroxidase Gene fromCrithidia fasciculata and Its Functional Expression in Escherichia coli

1998; Elsevier BV; Volume: 273; Issue: 9 Linguagem: Inglês

10.1074/jbc.273.9.4864

1083-351X

Marisa Montemartini, Everson Nogoceke, Mahavir Singh, Peter M. Steinert, Leopold Flohé, Henryk M. Kalisz,

Enzyme function and inhibition

Tryparedoxin peroxidase from Crithidia fasciculata is an essential component of the trypanothione-dependent hydroperoxide metabolism in the trypanosomatids (Nogoceke, E., Gommel, D. U., Kieβ, M., Kalisz, H. M., and Flohé, L. (1997) Biol. Chem. 378, 827–836). The tryparedoxin peroxidase gene and its flanking regions have been isolated and sequenced from a C. fasciculatagenomic DNA library. It consists of an open reading frame of 564 base pairs encoding a protein of 188 amino acid residues. The gene, modified to encode 6 additional histidine residues, was expressed in Escherichia coli and the recombinant protein was purified to homogeneity by metal chelating chromatography. Recombinant tryparedoxin peroxidase has a subunit molecular mass of 21884 ± 22 and contains two isoforms of pI 6.2 and 6.3. It exhibits a kinetic pattern identical to that of the authentic tryparedoxin peroxidase and has a similar specific activity of 2.51 units mg−1. The enzyme unequivocally belongs to the peroxiredoxin family of proteins, whose members have been found in all phyla. A phylogenetic tree comprising 47 protein and DNA sequences showed tryparedoxin peroxidase and a homologous Trypanosoma brucei sequence to form a distinct molecular clade. The consensus sequence: xnAx5–6Fx9Gx3Vx2Fx1Px2Fx1FVCPTEx21Sx1Dx7Wx16–19Dx15–16Gx3Rx2Fx2Dx27Ax1Qx4–11Cx1–3Wxnwas demonstrated by alignment of the sequences of tryparedoxin peroxidase and 8 other peroxiredoxins with established peroxidase function. Tryparedoxin peroxidase from Crithidia fasciculata is an essential component of the trypanothione-dependent hydroperoxide metabolism in the trypanosomatids (Nogoceke, E., Gommel, D. U., Kieβ, M., Kalisz, H. M., and Flohé, L. (1997) Biol. Chem. 378, 827–836). The tryparedoxin peroxidase gene and its flanking regions have been isolated and sequenced from a C. fasciculatagenomic DNA library. It consists of an open reading frame of 564 base pairs encoding a protein of 188 amino acid residues. The gene, modified to encode 6 additional histidine residues, was expressed in Escherichia coli and the recombinant protein was purified to homogeneity by metal chelating chromatography. Recombinant tryparedoxin peroxidase has a subunit molecular mass of 21884 ± 22 and contains two isoforms of pI 6.2 and 6.3. It exhibits a kinetic pattern identical to that of the authentic tryparedoxin peroxidase and has a similar specific activity of 2.51 units mg−1. The enzyme unequivocally belongs to the peroxiredoxin family of proteins, whose members have been found in all phyla. A phylogenetic tree comprising 47 protein and DNA sequences showed tryparedoxin peroxidase and a homologous Trypanosoma brucei sequence to form a distinct molecular clade. The consensus sequence: xnAx5–6Fx9Gx3Vx2Fx1Px2Fx1FVCPTEx21Sx1Dx7Wx16–19Dx15–16Gx3Rx2Fx2Dx27Ax1Qx4–11Cx1–3Wxnwas demonstrated by alignment of the sequences of tryparedoxin peroxidase and 8 other peroxiredoxins with established peroxidase function. Tryparedoxin peroxidase has recently been identified as a constituent of the complex peroxidase system in the trypanosomatidCrithidia fasiculata (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar). In these parasitic protozoa hydroperoxides are reduced at the expense of NADPH by means of a cascade of three oxidoreductases: the flavoprotein trypanothione reductase, the thioredoxin-related tryparedoxin, and tryparedoxin peroxidase (Fig. 1). The first enzyme of the cascade is homologous to glutathione reductase and thioredoxin reductase (2Krauth-Siegel R.L. Schirmer R.H. Schöllhammer T. Edmonson D.E. McCormick D.B. Flavins and Flavoproteins. Walter de Gruyter & Co., Berlin1987: 69-73Google Scholar), which are involved in NADPH-dependent hydroperoxide reduction in other species (3Tamura T. Gladyshev V. Liu S.-Y. Stadtman T.C. BioFactors. 1995/1996; 5: 99-102Google Scholar). The other components of the trypanosomatid system also belong to protein families occasionally constituting peroxidase systems. Preliminary amino acid sequencing data indicated that tryparedoxin is phylogenetically related to thioredoxin (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar), whereas the tryparedoxin peroxidase belongs to the peroxiredoxins (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar) comprising the thioredoxin peroxidases of yeast and mammals (4Cha M.-K. Kim I.-H. Biochem. Biophys. Res. Commun. 1995; 217: 900-907Google Scholar) and the alkyl hydroperoxide reductases of bacteria (5Tartaglia L.A. Storz G. Brodsky M.H. Lai A. Ames B.N. J. Biol. Chem. 1990; 265: 10535-10540Google Scholar).The unique feature of the trypanosomatidal peroxidase system is its dependence on the peculiar redox mediator trypanothione which so far has not been discovered in any species outside the Trypanosomatidae. Its biosynthesis from spermidine and glutathione requires two distinct enzymes, glutathionylspermidine synthetase (6Koenig K. Menge U. Kieβ M. Wray V. Flohé L. J. Biol. Chem. 1997; 272: 11908-11915Google Scholar) and trypanothione synthetase (7Henderson G.B. Yamaguchi M. Novoa L. Fairlamb A.H. Cerami A. Biochemistry. 1990; 29: 3924-3929Google Scholar). With a cascade of oxidoreductases plus the redox mediator trypanothione and the two auxiliary enzymes for its synthesis, the trypanosomatids have developed the most complicated system for the removal of hydroperoxides so far discovered in nature. This is not to imply a particular efficiency or robustness of the system. On the contrary, trypanosomatids are reported to be highly susceptible to oxidative stress (8Docampo R. Chem. Biol. Interactions. 1990; 73: 1-27Google Scholar). Correspondingly, their extraordinary metabolism is being discussed as a potential target area of specific trypanocidal agents (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar, 9Fairlamb A.H. The Biochemist. 1996; 18: 11-16Google Scholar, 10Jacoby E.M. Schlichting I. Lantwin C.B. Kabsch W. Krauth-Siegel R.L. Proteins. 1996; 24: 73-80Google Scholar).The present possibilities available for the treatment of trypanosomal diseases, such as Chagas disease, African sleeping sickness, and the various forms of leishmaniasis, necessitate improvement. We (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar, 6Koenig K. Menge U. Kieβ M. Wray V. Flohé L. J. Biol. Chem. 1997; 272: 11908-11915Google Scholar), like many others (11Lohrer H. Krauth-Siegel R.L. Eur. J. Biochem. 1990; 194: 863-869Google Scholar, 12Calonge M. Cubrı́a J.C. Balaña-Fouce R. Ordóñez D. Biol. Chem. 1996; 377: 233-238Google Scholar, 13Cazzulo J.J. Stoka V. Turk V. Biol. Chem. 1997; 378: 1-10Google Scholar, 14Navas I.M. Garcı́a-Fernández A.J. Johnson R.A. Reguera R.M. Balaña-Fouce R. Ordóñez D. Biol. Chem. 1996; 377: 833-836Google Scholar), have therefore embarked on the identification and characterization of potential molecular targets typical of the trypanosomatids. Here we report for the first time the full-length DNA, deduced amino acid sequence, and expression of a tryparedoxin peroxidase and its relatedness to peroxiredoxins with established or unknown functions.DISCUSSIONWe have cloned a genomic DNA fragment from C. fasciculata that encodes multiple copies of the entire sequence of tryparedoxin peroxidase, which is an essential constituent of the trypanosomatidal peroxide metabolism (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar). The reading frame is very similar to a DNA sequence of T. brucei rhodesiense encoding a protein of unknown function (26El-Sayed N.M.A. Alarcon C.M. Beck J.C. Sheffield V.C. Donelson J.E. Mol. Biochem. Parasitol. 1995; 73: 75-90Google Scholar). The high degree of similarity suggests that this trypanosomal protein is also a tryparedoxin peroxidase. However, the multiplication of genes in the trypanosomatids, which may lead to pseudogenes, precludes a definite functional interpretation of DNA sequences without experimental evidence. In the case of the C. fasciculata DNA sequence shown in Fig. 3, its functional relevance was unambiguously established by heterologous expression and comparison of the gene product with the authentic tryparedoxin peroxidase.Tryparedoxin peroxidase unequivocally belongs to the family of peroxiredoxins. This protein family is obviously widespread in nature. The phylogenetic tree showing the molecular evolution of the peroxiredoxins practically covers all phyla from bacteria to vertebrates but it does not simply reflect the phylogenetic divergence of the species. The sequence of the parasitic plathelmint Brugia malayi (T), for example, debranches close to protozoal and vertebrate sub-trees, whereas another plathelmint protein (U) belongs to a plant sub-tree. There are two distinct molecular clades found in chlorophyta plus a separate type in the diatomean algaOdonthella, each diverging at distant points from bacterial branches. Similarly, one of the two yeast peroxiredoxins debranches from a plant branch, which in turn diverges from bacterial ancestors. This puzzling situation suggests multiple gene acquisition by means of endosymbiosis. Such gene transfer has also been implicated for variousEuglenazoa, including the Trypanosomatidae (31Henze K. Badr A. Wettern M. Cerff R. Martin W. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 9122-9126Google Scholar). Even the cytosolic GapC of T. brucei and Leishmania mexicana, for example, is believed to be acquired from endosymbiotic γ-purple bacteria and to be secondarily integrated into the nuclear genome of the trypanosomatids. The debranching point of the two trypanosomatid peroxiredoxins, however, does not lend any support to speculations about secondary gene acquisition. Debranching between the metazoa and yeast, bacteria, and plants, the molecular evolution of the trypanosomal peroxiredoxins appears congruent with taxonomical development.Molecular evolution has given ample room for functional diversification within the peroxiredoxin family and it can by no means be uncritically presumed that all peroxiredoxins are peroxidases or “antioxidant proteins.” Conceivably, many of the peroxiredoxins will turn out to reduce hydroperoxides, but the reducing substrates differ substantially in the examples investigated to date. TSA of yeast, like human NKEF-α and -β, utilizes thioredoxin as a reductant; the crithidial homolog is reduced by a remote relative of thioredoxin, tryparedoxin, which is 50% larger than the typical thioredoxins and is characterized by a WCPPC motif in its active site; the alkyl hydroperoxide reductases of bacteria are directly reduced by flavoproteins also containing vicinal thiols (AhpF); and from their history of discovery we know that the activity of TSAs depends on dithiols such as dithiothreitol. The common denominator of the peroxidase activity of peroxiredoxin thus appears to be the regeneration by dithiols of a residue reacting with hydroperoxides. In the absence of any reasonable alternative such a redox active residue can only be a cysteine. In the case of thioredoxin peroxidase of yeast, the active site was identified by site-directed mutagenesis as the N-proximal VCP motif (32Chae H.Z. Uhm T.B. Rhee S.G Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 7022-7026Google Scholar). Interestingly, replacement of the conserved C-terminal cysteine residue by a serine resulted in an active antioxidant protein when tested with the non-physiological substrate dithiothreitol, but the activity with thioredoxin was lost (21Chae H.Z. Chung S.J. Rhee S.G. J. Biol. Chem. 1994; 269: 27670-27678Google Scholar, 32Chae H.Z. Uhm T.B. Rhee S.G Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 7022-7026Google Scholar). With tryparedoxin peroxidase, the participation of two cysteine residues in the catalysis was demonstrated by the substrate-dependent inactivation withN-ethylmaleimide and confirmed by matrix-assisted laser desorption/ionization-time of flight-mass spectrometry (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar). The consensus sequence of the peroxiredoxins with experimentally demonstrated peroxidase activity reveals two conserved cysteines as the only potentially redox-active functional groups. Only the first cysteine appears to be obligatorily integrated into a VCP motif within a highly conserved region. Although, the second cysteine also usually forms a VCP motif, it appears to tolerate substantial changes in its intimate sequence context. However, based on the yeast TSA example, it is tempting to speculate that the latter is crucial for the reaction with the specific reductant, whereas the former is indispensible for the reaction with the oxidant.The kinetics of tryparedoxin peroxidase have been elucidated in detail (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar). The apparent net forward rate constants for the reaction of the reduced enzyme (calculated per subunit) with a variety of hydroperoxides were found to be around 105m−1 s−1. To achieve such rate constants the active site cysteine residue must be highly activated. These rate constants are virtually identical to those of sulfur homologs of the selenocysteine containing glutathione peroxidases (33Rocher C. Lalanne J.L. Chaudiere J. Eur. J. Biochem. 1992; 205: 955-960Google Scholar,34Maiorino M. Aumann K.-D. Brigelius-Flohé R. Doria D. van den Heuvel J. McCarthy J. Roveri A. Ursini F. Flohé L. Biol. Chem. Hoppe-Seyler. 1995; 376: 651-660Google Scholar), where the (seleno)cysteine residue, together with a tryptophan and a glutamine residue, forms a unique catalytic triad and thereby is forced into full dissociation and is correspondingly reactive (34Maiorino M. Aumann K.-D. Brigelius-Flohé R. Doria D. van den Heuvel J. McCarthy J. Roveri A. Ursini F. Flohé L. Biol. Chem. Hoppe-Seyler. 1995; 376: 651-660Google Scholar). Interestingly, two tryptophans and one glutamine are also strictly conserved in the peroxiredoxin-type peroxidases and it may therefore be speculated that their cysteines might be activated in an analogous way to the structurally unrelated glutathione peroxidases. In fact, with the exception of a single arginine, none of the other residues of the consensus sequence would be able to force a cysteine into dissociation. The three aspartate residues, if situated near the active site cysteine, would be contraproductive, the lipophilic residues would be largely indifferent, and the conserved prolines and glycines are most probably involved in the preservation of the three-dimensional structure. Unfortunately, not a single three-dimensional structure of a peroxiredoxin is presently available to confirm the hypothesis of the convergent evolution of an analogous active site by unrelated proteins. Tryparedoxin peroxidase has recently been identified as a constituent of the complex peroxidase system in the trypanosomatidCrithidia fasiculata (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar). In these parasitic protozoa hydroperoxides are reduced at the expense of NADPH by means of a cascade of three oxidoreductases: the flavoprotein trypanothione reductase, the thioredoxin-related tryparedoxin, and tryparedoxin peroxidase (Fig. 1). The first enzyme of the cascade is homologous to glutathione reductase and thioredoxin reductase (2Krauth-Siegel R.L. Schirmer R.H. Schöllhammer T. Edmonson D.E. McCormick D.B. Flavins and Flavoproteins. Walter de Gruyter & Co., Berlin1987: 69-73Google Scholar), which are involved in NADPH-dependent hydroperoxide reduction in other species (3Tamura T. Gladyshev V. Liu S.-Y. Stadtman T.C. BioFactors. 1995/1996; 5: 99-102Google Scholar). The other components of the trypanosomatid system also belong to protein families occasionally constituting peroxidase systems. Preliminary amino acid sequencing data indicated that tryparedoxin is phylogenetically related to thioredoxin (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar), whereas the tryparedoxin peroxidase belongs to the peroxiredoxins (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar) comprising the thioredoxin peroxidases of yeast and mammals (4Cha M.-K. Kim I.-H. Biochem. Biophys. Res. Commun. 1995; 217: 900-907Google Scholar) and the alkyl hydroperoxide reductases of bacteria (5Tartaglia L.A. Storz G. Brodsky M.H. Lai A. Ames B.N. J. Biol. Chem. 1990; 265: 10535-10540Google Scholar). The unique feature of the trypanosomatidal peroxidase system is its dependence on the peculiar redox mediator trypanothione which so far has not been discovered in any species outside the Trypanosomatidae. Its biosynthesis from spermidine and glutathione requires two distinct enzymes, glutathionylspermidine synthetase (6Koenig K. Menge U. Kieβ M. Wray V. Flohé L. J. Biol. Chem. 1997; 272: 11908-11915Google Scholar) and trypanothione synthetase (7Henderson G.B. Yamaguchi M. Novoa L. Fairlamb A.H. Cerami A. Biochemistry. 1990; 29: 3924-3929Google Scholar). With a cascade of oxidoreductases plus the redox mediator trypanothione and the two auxiliary enzymes for its synthesis, the trypanosomatids have developed the most complicated system for the removal of hydroperoxides so far discovered in nature. This is not to imply a particular efficiency or robustness of the system. On the contrary, trypanosomatids are reported to be highly susceptible to oxidative stress (8Docampo R. Chem. Biol. Interactions. 1990; 73: 1-27Google Scholar). Correspondingly, their extraordinary metabolism is being discussed as a potential target area of specific trypanocidal agents (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar, 9Fairlamb A.H. The Biochemist. 1996; 18: 11-16Google Scholar, 10Jacoby E.M. Schlichting I. Lantwin C.B. Kabsch W. Krauth-Siegel R.L. Proteins. 1996; 24: 73-80Google Scholar). The present possibilities available for the treatment of trypanosomal diseases, such as Chagas disease, African sleeping sickness, and the various forms of leishmaniasis, necessitate improvement. We (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar, 6Koenig K. Menge U. Kieβ M. Wray V. Flohé L. J. Biol. Chem. 1997; 272: 11908-11915Google Scholar), like many others (11Lohrer H. Krauth-Siegel R.L. Eur. J. Biochem. 1990; 194: 863-869Google Scholar, 12Calonge M. Cubrı́a J.C. Balaña-Fouce R. Ordóñez D. Biol. Chem. 1996; 377: 233-238Google Scholar, 13Cazzulo J.J. Stoka V. Turk V. Biol. Chem. 1997; 378: 1-10Google Scholar, 14Navas I.M. Garcı́a-Fernández A.J. Johnson R.A. Reguera R.M. Balaña-Fouce R. Ordóñez D. Biol. Chem. 1996; 377: 833-836Google Scholar), have therefore embarked on the identification and characterization of potential molecular targets typical of the trypanosomatids. Here we report for the first time the full-length DNA, deduced amino acid sequence, and expression of a tryparedoxin peroxidase and its relatedness to peroxiredoxins with established or unknown functions. DISCUSSIONWe have cloned a genomic DNA fragment from C. fasciculata that encodes multiple copies of the entire sequence of tryparedoxin peroxidase, which is an essential constituent of the trypanosomatidal peroxide metabolism (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar). The reading frame is very similar to a DNA sequence of T. brucei rhodesiense encoding a protein of unknown function (26El-Sayed N.M.A. Alarcon C.M. Beck J.C. Sheffield V.C. Donelson J.E. Mol. Biochem. Parasitol. 1995; 73: 75-90Google Scholar). The high degree of similarity suggests that this trypanosomal protein is also a tryparedoxin peroxidase. However, the multiplication of genes in the trypanosomatids, which may lead to pseudogenes, precludes a definite functional interpretation of DNA sequences without experimental evidence. In the case of the C. fasciculata DNA sequence shown in Fig. 3, its functional relevance was unambiguously established by heterologous expression and comparison of the gene product with the authentic tryparedoxin peroxidase.Tryparedoxin peroxidase unequivocally belongs to the family of peroxiredoxins. This protein family is obviously widespread in nature. The phylogenetic tree showing the molecular evolution of the peroxiredoxins practically covers all phyla from bacteria to vertebrates but it does not simply reflect the phylogenetic divergence of the species. The sequence of the parasitic plathelmint Brugia malayi (T), for example, debranches close to protozoal and vertebrate sub-trees, whereas another plathelmint protein (U) belongs to a plant sub-tree. There are two distinct molecular clades found in chlorophyta plus a separate type in the diatomean algaOdonthella, each diverging at distant points from bacterial branches. Similarly, one of the two yeast peroxiredoxins debranches from a plant branch, which in turn diverges from bacterial ancestors. This puzzling situation suggests multiple gene acquisition by means of endosymbiosis. Such gene transfer has also been implicated for variousEuglenazoa, including the Trypanosomatidae (31Henze K. Badr A. Wettern M. Cerff R. Martin W. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 9122-9126Google Scholar). Even the cytosolic GapC of T. brucei and Leishmania mexicana, for example, is believed to be acquired from endosymbiotic γ-purple bacteria and to be secondarily integrated into the nuclear genome of the trypanosomatids. The debranching point of the two trypanosomatid peroxiredoxins, however, does not lend any support to speculations about secondary gene acquisition. Debranching between the metazoa and yeast, bacteria, and plants, the molecular evolution of the trypanosomal peroxiredoxins appears congruent with taxonomical development.Molecular evolution has given ample room for functional diversification within the peroxiredoxin family and it can by no means be uncritically presumed that all peroxiredoxins are peroxidases or “antioxidant proteins.” Conceivably, many of the peroxiredoxins will turn out to reduce hydroperoxides, but the reducing substrates differ substantially in the examples investigated to date. TSA of yeast, like human NKEF-α and -β, utilizes thioredoxin as a reductant; the crithidial homolog is reduced by a remote relative of thioredoxin, tryparedoxin, which is 50% larger than the typical thioredoxins and is characterized by a WCPPC motif in its active site; the alkyl hydroperoxide reductases of bacteria are directly reduced by flavoproteins also containing vicinal thiols (AhpF); and from their history of discovery we know that the activity of TSAs depends on dithiols such as dithiothreitol. The common denominator of the peroxidase activity of peroxiredoxin thus appears to be the regeneration by dithiols of a residue reacting with hydroperoxides. In the absence of any reasonable alternative such a redox active residue can only be a cysteine. In the case of thioredoxin peroxidase of yeast, the active site was identified by site-directed mutagenesis as the N-proximal VCP motif (32Chae H.Z. Uhm T.B. Rhee S.G Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 7022-7026Google Scholar). Interestingly, replacement of the conserved C-terminal cysteine residue by a serine resulted in an active antioxidant protein when tested with the non-physiological substrate dithiothreitol, but the activity with thioredoxin was lost (21Chae H.Z. Chung S.J. Rhee S.G. J. Biol. Chem. 1994; 269: 27670-27678Google Scholar, 32Chae H.Z. Uhm T.B. Rhee S.G Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 7022-7026Google Scholar). With tryparedoxin peroxidase, the participation of two cysteine residues in the catalysis was demonstrated by the substrate-dependent inactivation withN-ethylmaleimide and confirmed by matrix-assisted laser desorption/ionization-time of flight-mass spectrometry (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar). The consensus sequence of the peroxiredoxins with experimentally demonstrated peroxidase activity reveals two conserved cysteines as the only potentially redox-active functional groups. Only the first cysteine appears to be obligatorily integrated into a VCP motif within a highly conserved region. Although, the second cysteine also usually forms a VCP motif, it appears to tolerate substantial changes in its intimate sequence context. However, based on the yeast TSA example, it is tempting to speculate that the latter is crucial for the reaction with the specific reductant, whereas the former is indispensible for the reaction with the oxidant.The kinetics of tryparedoxin peroxidase have been elucidated in detail (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar). The apparent net forward rate constants for the reaction of the reduced enzyme (calculated per subunit) with a variety of hydroperoxides were found to be around 105m−1 s−1. To achieve such rate constants the active site cysteine residue must be highly activated. These rate constants are virtually identical to those of sulfur homologs of the selenocysteine containing glutathione peroxidases (33Rocher C. Lalanne J.L. Chaudiere J. Eur. J. Biochem. 1992; 205: 955-960Google Scholar,34Maiorino M. Aumann K.-D. Brigelius-Flohé R. Doria D. van den Heuvel J. McCarthy J. Roveri A. Ursini F. Flohé L. Biol. Chem. Hoppe-Seyler. 1995; 376: 651-660Google Scholar), where the (seleno)cysteine residue, together with a tryptophan and a glutamine residue, forms a unique catalytic triad and thereby is forced into full dissociation and is correspondingly reactive (34Maiorino M. Aumann K.-D. Brigelius-Flohé R. Doria D. van den Heuvel J. McCarthy J. Roveri A. Ursini F. Flohé L. Biol. Chem. Hoppe-Seyler. 1995; 376: 651-660Google Scholar). Interestingly, two tryptophans and one glutamine are also strictly conserved in the peroxiredoxin-type peroxidases and it may therefore be speculated that their cysteines might be activated in an analogous way to the structurally unrelated glutathione peroxidases. In fact, with the exception of a single arginine, none of the other residues of the consensus sequence would be able to force a cysteine into dissociation. The three aspartate residues, if situated near the active site cysteine, would be contraproductive, the lipophilic residues would be largely indifferent, and the conserved prolines and glycines are most probably involved in the preservation of the three-dimensional structure. Unfortunately, not a single three-dimensional structure of a peroxiredoxin is presently available to confirm the hypothesis of the convergent evolution of an analogous active site by unrelated proteins. We have cloned a genomic DNA fragment from C. fasciculata that encodes multiple copies of the entire sequence of tryparedoxin peroxidase, which is an essential constituent of the trypanosomatidal peroxide metabolism (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar). The reading frame is very similar to a DNA sequence of T. brucei rhodesiense encoding a protein of unknown function (26El-Sayed N.M.A. Alarcon C.M. Beck J.C. Sheffield V.C. Donelson J.E. Mol. Biochem. Parasitol. 1995; 73: 75-90Google Scholar). The high degree of similarity suggests that this trypanosomal protein is also a tryparedoxin peroxidase. However, the multiplication of genes in the trypanosomatids, which may lead to pseudogenes, precludes a definite functional interpretation of DNA sequences without experimental evidence. In the case of the C. fasciculata DNA sequence shown in Fig. 3, its functional relevance was unambiguously established by heterologous expression and comparison of the gene product with the authentic tryparedoxin peroxidase. Tryparedoxin peroxidase unequivocally belongs to the family of peroxiredoxins. This protein family is obviously widespread in nature. The phylogenetic tree showing the molecular evolution of the peroxiredoxins practically covers all phyla from bacteria to vertebrates but it does not simply reflect the phylogenetic divergence of the species. The sequence of the parasitic plathelmint Brugia malayi (T), for example, debranches close to protozoal and vertebrate sub-trees, whereas another plathelmint protein (U) belongs to a plant sub-tree. There are two distinct molecular clades found in chlorophyta plus a separate type in the diatomean algaOdonthella, each diverging at distant points from bacterial branches. Similarly, one of the two yeast peroxiredoxins debranches from a plant branch, which in turn diverges from bacterial ancestors. This puzzling situation suggests multiple gene acquisition by means of endosymbiosis. Such gene transfer has also been implicated for variousEuglenazoa, including the Trypanosomatidae (31Henze K. Badr A. Wettern M. Cerff R. Martin W. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 9122-9126Google Scholar). Even the cytosolic GapC of T. brucei and Leishmania mexicana, for example, is believed to be acquired from endosymbiotic γ-purple bacteria and to be secondarily integrated into the nuclear genome of the trypanosomatids. The debranching point of the two trypanosomatid peroxiredoxins, however, does not lend any support to speculations about secondary gene acquisition. Debranching between the metazoa and yeast, bacteria, and plants, the molecular evolution of the trypanosomal peroxiredoxins appears congruent with taxonomical development. Molecular evolution has given ample room for functional diversification within the peroxiredoxin family and it can by no means be uncritically presumed that all peroxiredoxins are peroxidases or “antioxidant proteins.” Conceivably, many of the peroxiredoxins will turn out to reduce hydroperoxides, but the reducing substrates differ substantially in the examples investigated to date. TSA of yeast, like human NKEF-α and -β, utilizes thioredoxin as a reductant; the crithidial homolog is reduced by a remote relative of thioredoxin, tryparedoxin, which is 50% larger than the typical thioredoxins and is characterized by a WCPPC motif in its active site; the alkyl hydroperoxide reductases of bacteria are directly reduced by flavoproteins also containing vicinal thiols (AhpF); and from their history of discovery we know that the activity of TSAs depends on dithiols such as dithiothreitol. The common denominator of the peroxidase activity of peroxiredoxin thus appears to be the regeneration by dithiols of a residue reacting with hydroperoxides. In the absence of any reasonable alternative such a redox active residue can only be a cysteine. In the case of thioredoxin peroxidase of yeast, the active site was identified by site-directed mutagenesis as the N-proximal VCP motif (32Chae H.Z. Uhm T.B. Rhee S.G Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 7022-7026Google Scholar). Interestingly, replacement of the conserved C-terminal cysteine residue by a serine resulted in an active antioxidant protein when tested with the non-physiological substrate dithiothreitol, but the activity with thioredoxin was lost (21Chae H.Z. Chung S.J. Rhee S.G. J. Biol. Chem. 1994; 269: 27670-27678Google Scholar, 32Chae H.Z. Uhm T.B. Rhee S.G Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 7022-7026Google Scholar). With tryparedoxin peroxidase, the participation of two cysteine residues in the catalysis was demonstrated by the substrate-dependent inactivation withN-ethylmaleimide and confirmed by matrix-assisted laser desorption/ionization-time of flight-mass spectrometry (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar). The consensus sequence of the peroxiredoxins with experimentally demonstrated peroxidase activity reveals two conserved cysteines as the only potentially redox-active functional groups. Only the first cysteine appears to be obligatorily integrated into a VCP motif within a highly conserved region. Although, the second cysteine also usually forms a VCP motif, it appears to tolerate substantial changes in its intimate sequence context. However, based on the yeast TSA example, it is tempting to speculate that the latter is crucial for the reaction with the specific reductant, whereas the former is indispensible for the reaction with the oxidant. The kinetics of tryparedoxin peroxidase have been elucidated in detail (1Nogoceke E. Gommel D.U. Kieβ M. Kalisz H.M. Flohé L. Biol. Chem. 1997; 378: 827-836Google Scholar). The apparent net forward rate constants for the reaction of the reduced enzyme (calculated per subunit) with a variety of hydroperoxides were found to be around 105m−1 s−1. To achieve such rate constants the active site cysteine residue must be highly activated. These rate constants are virtually identical to those of sulfur homologs of the selenocysteine containing glutathione peroxidases (33Rocher C. Lalanne J.L. Chaudiere J. Eur. J. Biochem. 1992; 205: 955-960Google Scholar,34Maiorino M. Aumann K.-D. Brigelius-Flohé R. Doria D. van den Heuvel J. McCarthy J. Roveri A. Ursini F. Flohé L. Biol. Chem. Hoppe-Seyler. 1995; 376: 651-660Google Scholar), where the (seleno)cysteine residue, together with a tryptophan and a glutamine residue, forms a unique catalytic triad and thereby is forced into full dissociation and is correspondingly reactive (34Maiorino M. Aumann K.-D. Brigelius-Flohé R. Doria D. van den Heuvel J. McCarthy J. Roveri A. Ursini F. Flohé L. Biol. Chem. Hoppe-Seyler. 1995; 376: 651-660Google Scholar). Interestingly, two tryptophans and one glutamine are also strictly conserved in the peroxiredoxin-type peroxidases and it may therefore be speculated that their cysteines might be activated in an analogous way to the structurally unrelated glutathione peroxidases. In fact, with the exception of a single arginine, none of the other residues of the consensus sequence would be able to force a cysteine into dissociation. The three aspartate residues, if situated near the active site cysteine, would be contraproductive, the lipophilic residues would be largely indifferent, and the conserved prolines and glycines are most probably involved in the preservation of the three-dimensional structure. Unfortunately, not a single three-dimensional structure of a peroxiredoxin is presently available to confirm the hypothesis of the convergent evolution of an analogous active site by unrelated proteins. We thank Dr. Michael Kieβ and Rita Getzlaff for amino acid sequencing.