Glyoxalase II of African Trypanosomes Is Trypanothione-dependent*

The glyoxalase system is a ubiquitous pathway catalyzing the glutathione-dependent detoxication of ketoaldehydes such as methylglyoxal, which is mainly formed as a by-product of glycolysis. The gene encoding a glyoxalase II has been cloned from Trypanosoma brucei, the causative agent of African sleeping sickness. The deduced protein sequence contains the highly conserved metal binding motif THXHXDH but lacks three basic residues shown to fix the glutathione-thioester substrate in the crystal structure of human glyoxalase II. Recombinant T. brucei glyoxalase II hydrolyzes lactoylglutathione, but does not show saturation kinetics up to 5 mm with the classical substrate of glyoxalases II. Instead, the parasite enzyme strongly prefers thioesters of trypanothione (bis(glutathionyl)spermidine), which were prepared from methylglyoxal and trypanothione and analyzed by high performance liquid chromatography and mass spectrometry. Mono-(lactoyl)trypanothione and bis-(lactoyl)trypanothione are hydrolyzed by T. brucei glyoxalase II with kcat/Km values of 5 × 105 m-1 s-1 and 7 × 105 m-1 s-1, respectively, yielding d-lactate and regenerating trypanothione. Glyoxalase II occurs in the mammalian bloodstream and insect procyclic form of T. brucei and is the first glyoxalase II of the order of Kinetoplastida characterized so far. Our results show that the glyoxalase system is another pathway in which the nearly ubiquitous glutathione is replaced by the unique trypanothione in trypanosomatids.

lesser extent with adenine and cytosine. Glyoxalase I (GLX I) 1 converts the hemithioacetal, which is formed spontaneously from methylglyoxal and glutathione, into S-lactoylglutathione. The thioester is subsequently hydrolyzed by glyoxalase II (GLX II) yielding D-lactate and regenerating glutathione.
Trypanosomatids are the causative agents of severe tropical diseases such as African sleeping sickness (Trypanosoma brucei gambiense and Trypanosoma brucei rhodesiense), Nagana cattle disease (T. brucei brucei and Trypanosoma congolense), Chagas' disease (Trypanosoma cruzi), and the three manifestations of leishmaniasis (Leishmania donovani, Leishmania major, and Leishmania mexicana). All these parasitic protozoa have in common that the nearly ubiquitous glutathione/glutathione reductase system is replaced by trypanothione (N 1 ,N 8bis(glutathionyl)spermidine) and the flavoenzyme trypanothione reductase (4,5). A linkage between glutathione and spermidine metabolism was first discovered in Escherichia coli (6). During stationary phase, all of the cellular spermidine and a large part of glutathione occur as mono(glutathionyl)spermidine whereby the bacterium does not form trypanothione. The trypanothione metabolism is essential for the parasite (7) and is involved in the detoxication of hydroperoxides, the synthesis of deoxyribonucleotides catalyzed by ribonucleotide reductase, as well as the homeostasis of ascorbate (5).
The pathogenic form of T. brucei multiplying in the blood of the mammalian host depends on glycolysis as the sole energy source and has a very high glucose turnover, which is about 200 -300-fold higher than in erythrocytes (8). In E. coli and human red blood cells, the glycolytic rate has been shown to quantitatively correlate with the formation of methylglyoxal (2, 3). These findings together with the fact that trypanothione instead of glutathione is the main low molecular mass thiol prompted us to characterize the glyoxalase system in African trypanosomes.
The T. brucei genome contains two probable glyoxalase II sequences. Here we report on the cloning and overexpression of a glyoxalase II gene encoded on chromosome VI. The recombinant T. brucei protein slowly hydrolyzes lactoylglutathione, the substrate of classical glyoxalases II, but strongly prefers trypanothione thioesters. We provide strong evidence that glutathione is replaced by trypanothione also in the glyoxalase system of trypanosomatid parasites.

Materials
S-Lactoylglutathione, methylglyoxal, yeast glyoxalase I (400 -800 units/mg), bovine liver glyoxalase II (10 units/mg), and Leuconostoc mesenteroides D-lactate dehydrogenase were purchased from Sigma. * This work was supported in part by the Deutsche Forschungsgemeinschaft (Graduiertenkolleg "Pathogene Mikroorganismen: Molekulare Mechanismen und Genome"). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EBI Data Bank with accession number(s) AJ492819.
Hog muscle L-lactate dehydrogenase was from Roche. T. cruzi trypanothione reductase was prepared as described (9). Trypanothione disulfide was obtained from Bachem. Restriction enzymes and PfuTurbo DNA polymerase were from MBI Fermentas. Primer synthesis and DNA sequencing were performed by MWG Biotech. All other chemicals were commercially available reagents of the highest quality.

Cloning and Overexpression of T. brucei Glyoxalase II
T. brucei His 6 -GLX II-The coding region of the glxII gene was amplified from T. brucei genomic DNA (strain TREU 927/4 (10)) by PCR using sequence-specific primers derived from the data base. The 5Ј primer (5Ј-gcgcggatccgaagttgtagtgaagagcatcgg-3Ј) contained a BamHI site (underlined) and gaa encoding glutamate 2 of the protein. The 3Ј primer (5Ј-taggtaccacaccatagttcgcg-3Ј) was placed in the 3Ј-untranslated region directly after the stop codon. The gene was amplified from genomic DNA by PCR (95°C for 2 min; 95°C for 30 s; 55°C for 30 s; 72°C for 2 min; 30 cycles; 72°C for 10 min; Pfu). The PCR product was digested with BamHI and cloned into the pQE-30 vector (Qiagen), digested with BamHI and SmaI. E. coli NovaBlue cells (Novagen) were transformed with the pQE-30/his-glxII plasmid. The plasmid was isolated using the NucleoBond® plasmid purification kit (Macherey-Nagel) and the insert was completely sequenced in both directions. For overexpression of T. brucei His 6 -GLX II, a 1-liter culture of recombinant NovaBlue cells was incubated at 37°C in LB-medium containing 100 g/ml carbenicillin. At an A 600 of about 0.5, expression was induced by adding 200 M isopropyl-␤-D-thiogalactopyranoside and the cells were allowed to grow overnight at 15°C.
T. brucei GLX II-For overexpression of tag-free T. brucei glyoxalase II, the 5Ј primer (5Ј-cgcgccatggaagttgtagtgaagagcatc-3Ј) contained an NcoI site (underlined) and the start ATG (italic). The 3Ј primer (5Јcgcaggatccttacggacacgtattatagaggaag-3Ј) contained the stop codon (tta) and a BamHI site (underlined). The gene was amplified from genomic DNA (95°C for 2 min; 95°C for 30 s; 64°C for 30 s; 72°C for 2 min; 30 cycles; 72°C for 10 min; Pfu) and the PCR product was cloned using the NcoI and BamHI restriction sites into the pQE-60 vector, resulting in the pQE-60/glxII plasmid. The plasmid was isolated and sequenced. NovaBlue cells were transformed with the plasmid and the gene was overexpressed as described above.

Purification of T. brucei His 6 -GLX II
The fusion protein, carrying a 14-residue long N-terminal extension with six histidine residues, was purified at 4°C by chromatography on a TALON® metal affinity resin column (Clontech). Cells from a 1-liter bacterial culture were harvested by centrifugation, resuspended in 50 mM sodium phosphate, 300 mM NaCl, 1 mM imidazol, pH 7.0, 150 nM pepstatin, 4 nM cystatin, and 100 M phenylmethylsulfonyl fluoride and lysed by sonification, and the cell debris was removed by centrifugation at 33,000 ϫ g. The supernatant was applied onto a 5-ml resin preequilibrated in 50 mM sodium phosphate, 300 mM NaCl, 1 mM imidazol, pH 7.0. After washing the column with the equilibration buffer, followed by 50 mM sodium phosphate, 300 mM NaCl, 5 mM imidazol, pH 7.0, the protein was eluted with 250 mM imidazol in 50 mM sodium phosphate, 300 mM NaCl, pH 7.0. The recombinant T. brucei His 6glyoxalase II was Ն95% pure as judged by SDS-polyacrylamide gel electrophoresis. The protein concentration was determined using the bicinchoninic acid kit (BCA, Pierce).

Purification of Tag-free GLX II
The recombinant protein without tag was purified on a Q-Sepharose fast flow cation exchanger (Amersham Biosciences). The column (10 ml) was equilibrated at 4°C in 100 ml of 10 mM MOPS, pH 7.5. The recombinant NovaBlue cells from a 1-liter culture were harvested by centrifugation, the cell pellet was suspended in 5 mM MOPS, pH 7.5, 150 nM pepstatin, 4 nM cystatin, and 100 M phenylmethylsulfonyl fluoride and lysed as described above. The supernatant was diluted with H 2 O to a conductivity lower than that of the equilibration buffer and applied onto the column. After washing the Q-Sepharose with 120 ml of equilibration buffer, the protein was eluted with 90 ml of 20 mM NaCl in 10 mM MOPS, pH 7.5, collecting 5-ml fractions. The purity of the protein was Ն95% as judged by SDS-polyacrylamide gel electrophoresis. For homogeneous His 6 -GLX II and tag-free GLX II a protein concentration of 0.58 mg/ml corresponds to a ⌬A 280 of 1. For long-term storage at Ϫ20°C, 50% glycerol was added to the protein preparations.

Gel Filtration
The subunit composition of T. brucei glyoxalase II was determined by fast protein liquid chromatography on a High Load 26/60 Superdex 75 column (Amersham Biosciences). Purified, tag-free protein (3.6 mg) was loaded onto the column equilibrated in 100 mM sodium phosphate, pH 7.5, and run at room temperature at a flow rate of 2 ml/min, collecting 3-ml fractions. The low molecular weight Gel Filtration Kit (bovine serum albumin, ovalbumin, chymotrypsinogen A, and ribonuclease A) (Amersham Biosciences) served as molecular mass standard.

Synthesis of Mono-and Bis-(lactoyl)trypanothione
The lactoyltrypanothione thioesters were prepared in two steps. For the generation of mono-(lactoyl)trypanothione, 5 mM trypanothione disulfide in 1 ml of 100 mM MOPS, pH 7.2, was reduced with 7 mM NADPH and 2.75 units of T. cruzi trypanothione reductase for 1 h at room temperature. Then 10 l of 100 mM methylglyoxal in 100 mM MOPS, pH 7.2, and 6.4 units of yeast glyoxalase I were added and the reaction mixture was incubated for another 1 h. For the synthesis of bis-(lactoyl)trypanothione, 1.5 ml of 3 mM reduced trypanothione was mixed with 8 l of 5.5 M methylglyoxal and treated as described above. NADPH, methylglyoxal, and the proteins were removed from the thioesters (and excess trypanothione in the case of mono-(lactoyl)trypanothione) on an Oasis® MCX cartridge (Waters). The cartridge was preequilibrated with 1 ml of methanol and 2 ml of H 2 O and the reaction mixture was applied. After washing with 2 ml of H 2 O and 2 ml of methanol, the thioesters were eluted with 2 ml of 1 M N-ethylmorpholino acetate, pH 6.8. The flow-through already containing some thioester was again applied onto the re-equilibrated cartridge, washed, and eluted with 1 ml of 1 M N-ethylmorpholino acetate, pH 6.8. The thioester containing fractions were dried by evaporation and stored in aliquots at Ϫ20°C. Immediately before use, an aliquot was dissolved in 150 l of 100 mM MOPS, pH 7.2. The concentration was determined by measuring the thiol content with DTNB before and after hydrolysis of the thioester by glyoxalase II (11).

Analysis of S-Lactoyltrypanothione by HPLC and Mass Spectrometry
The S-lactoyltrypanothione thioesters eluted from the Oasis MCX cartridge were analyzed by HPLC using a buffer system described in Ref. 12. After injection of the sample, the column (Vydac 218TP52) was washed at a flow rate of 0.3 ml/min for 5 min with solvent A (0.25% (w/v) D-camphor sulfonate Li-salt, pH 2.64) at 40°C. Then the thioesters were eluted by a 19-min isocratic step of 90% solvent A and 10% solvent B (25% 1-propanol in solvent A). Before applying a new sample, the column was washed with 100% solvent B for 10 min and equilibrated for 23 min with 100% solvent A. The S-lactoyltrypanothione conjugates were detected by the thioester absorption at 240 nm. To show the regeneration of trypanothione in the T. brucei glyoxalase II reaction, the samples were analyzed at 220 nm. The MALDI-TOF (matrix-assisted laser desorption/ionization-time of flight) analyses of the thioesters were carried out by Drs. Jens Pfannstiel and Johannes Lechner (Biochemie-Zentrum der Universitä t Heidelberg) using a 2,5-dihydroxybenzoic acid matrix.

Kinetic Studies
The activities of T. brucei and bovine liver glyoxalase II were determined by measuring the hydrolysis of the thioesters (lactoylglutathione, mono-(lactoyl)trypanothione, and bis-(lactoyl)trypanothione) directly at 240 nm or in a coupled assay following the reaction of the liberated thiol with DTNB at 412 nm (13). The assays were performed in a total volume of 1 ml of 100 mM MOPS, pH 7.2, at 25°C in a Hitachi 150-20 spectrophotometer or in 300-l microcuvettes in a Beckman DU®-65 spectrophotometer. The lactoylglutathione concentration was varied between 0.01 and 5 mM in the coupled and between 0.01 and 0.6 mM in the direct assay. Hydrolysis of the lactoyltrypanothione thioesters was followed only in the direct assay at 240 nm at concentrations between 0.02 and 0.15 mM. The thioester stock solutions were freshly prepared from a dried aliquot stored at Ϫ20°C (see above). The extinction coefficients used are DTNB, ⑀ 412 ϭ 13.6 mM Ϫ1 cm Ϫ1 (11), SLG, ⑀ 240 ϭ 3.3 mM Ϫ1 cm Ϫ1 (14,15), MLT, ⑀ 240 ϭ 3.3 mM Ϫ1 cm Ϫ1 , and BLT, ⑀ 240 ϭ 6.5 mM Ϫ1 cm Ϫ1 , respectively. The ⑀ values of the trypanothione thioesters were derived from measuring the thiol concentration with DTNB after complete hydrolysis.

pH and Ionic Strength Dependence of Glyoxalase II
The pH dependence of T. brucei glyoxalase II was determined by following the activity of the enzyme in the presence of 82 M bis-(lactoyl)trypanothione in a total volume of 200 l of 100 mM MOPS between pH 6.0 and 9.2 at 25°C. To elucidate the optimum of ionic strength, assays were performed at 25°C with 88 M bis-(lactoyl) trypanothione in a total volume of 200 l of 10-250 mM MOPS buffers at a constant pH of 7.2.

Determination of D-Lactate
Production of D-lactate was shown by coupling the T. brucei glyoxalase II reaction to the reaction of D-and L-lactate dehydrogenase, respectively, following formation of NADH. After the glyoxalase II reaction with 87.5 M bis-(lactoyl)trypanothione in 800 l of 100 mM Tris-HCl, pH 8.5, had run to completion, 5 mM NAD and 2.75 units of Dor L-lactate dehydrogenase were added and the absorption increase at 340 nm (⑀ 340 ϭ 6.22 mM Ϫ1 cm Ϫ1 ) was followed at 25°C (16).

Metal Analysis of Recombinant T. brucei Glyoxalase II
The metal content of His 6 -and tag-free T. brucei glyoxalase II was analyzed by Dr. Peter Schramel (GSF-Forschungszentrum, Neuherberg) by ICP-OES (inductively coupled-plasma optical emission spectrometry). For comparison, the concentration of different metals in the buffers before and after the columns used for purification were determined.

Purification of the Polyclonal Rabbit Antiserum against T. brucei Glyoxalase II
About 12 mg of T. brucei His 6 -GLX II in 50 mM MES, pH 6.5, was coupled to 0.7 ml of Affi-Gel 10 (Bio-Rad) overnight at 4°C as described by the manufacturer. The resin was then washed with 50 mM MES, pH 6.5, and the remaining reactive groups were blocked with 0.7 ml of 1 M ethanolamine acetate, pH 8.0, for 1 h at 4°C. The column was washed 2 times each with 3 ml of PBS, containing 250 mM NaCl, 3 ml of 100 mM glycine, pH 2.5, 3 ml of PBS with 500 mM NaCl, and again with 3 ml of PBS containing 250 mM NaCl. The IgGs from 30 ml of rabbit polyclonal antiserum against T. brucei His 6 -GLX II (Eurogentec) were precipitated by 33% (NH 4 ) 2 SO 4 overnight at 4°C. After centrifugation for 20 min at 33,000 ϫ g, the pellet was dissolved in 10 ml of PBS containing 250 mM NaCl, and the solution was applied onto the Affi-Gel 10-His 6 -GLX II column at room temperature. The flow-through was again applied, and the column was washed with 7 ml of PBS containing 250 mM NaCl. After additional washing with 5 ml of PBS with 500 mM NaCl, the antibodies were eluted with 100 mM glycine, pH 2.5, collecting 1-ml fractions in Eppendorf cups containing 60 l of 1 M Tris, pH 9.5. The protein concentration was determined at 280 nm. Antibody containing fractions were pooled (3 ml), glycerol was added at 50% saturation, and aliquots were stored at Ϫ20°C.

Western Blot Analysis
Recombinant T. brucei GLX II and cell lysates of bloodstream and procyclic T. brucei were mixed with 4ϫ SDS loading buffer (250 mM Tris-HCl, pH 6.8, 8% SDS, 40% glycerol, 0.004% bromphenol blue, containing 100 mM Tris(2-carboxyethyl)phosphine solution, neutral pH (Bond-Breaker®, Pierce)) and boiled for 5 min. The proteins were separated by polyacrylamide gel electrophoresis on a 10% SDS gel and transferred onto a Hybond-P polyvinylidene difluoride membrane (Amersham Biosciences) by a Bio-Rad Mini Trans Blot, at 4°C for 2 h at 150 mA. The membrane was treated with blocking solution (5% milk powder, 1ϫ Tris-buffered saline, 0.05% Tween) overnight at 4°C. The blot was washed at room temperature three times for 5 min with 1ϫ Trisbuffered saline, 0.05% Tween and then incubated for 1 h with the purified T. brucei GLX II antibodies (see above, dilution 1:500). The membrane was again washed as described above and treated with the secondary antibody (horseradish peroxidase-conjugated anti-rabbit IgG (Santa Cruz, dilution 1:20,000)) for 1 h. After washing with Tris-buffered saline/Tween, the blots were developed using the SuperSignal® West Pico chemiluminescent substrate (Pierce) with exposition times from 30 s to 60 min.

Cloning and Structural Comparison of T. brucei Glyoxalase
II with Glyoxalases II from Other Species-Blast searches with human glyoxalase II as template revealed two sets of putative glyoxalase II sequences in the T. brucei genome. The deduced protein sequences encoded on chromosomes IV and VI were only 25% identical to each other but showed 30 and 36% identity with human glyoxalase II, respectively. Based on this information, the gene on chromosome VI was cloned from genomic DNA of strain TREU 927/4 (10). The complete coding region was amplified with two gene-specific primers and se-quenced in both directions. PCR on cDNA from bloodstream T. brucei with a sequence-specific primer and a poly(dT) primer amplified a fragment containing the 3Ј end of the coding sequence followed by a 390-bp long 3Ј-untranslated region (data not shown). The deduced protein sequence consists of 296 amino acid residues and clearly classifies the T. brucei protein as glyoxalase II. The highest degree of similarity is found with putative proteins from T. cruzi and L. major where 66 and 51%, respectively, of all residues are identical. In the functionally characterized human and Arabidopsis thaliana glyoxalases II 36 and 31%, respectively, of all residues are conserved. The sequence of a probable hydroxyacylglutathione hydrolase from E. coli shows 30% identical residues (Fig. 1).
The isoelectric point (pI) of T. brucei glyoxalase II calculated from the protein sequence is 6.0 -6.5, which is comparable with those of the proteins from A. thaliana (6.2) (17) and Candida albicans (6.0) (18) determined by isoelectric focusing. The putative other kinetoplastid glyoxalases II (Fig. 1) also have theoretical acidic pI values. In general, plant glyoxalases II are acidic proteins with pI values ranging from 4.7 to 6.2, whereas animal enzymes have basic pI values. Isoelectric focusing of recombinant human glyoxalase II and the enzyme isolated from erythrocytes yielded isoelectric points of 8.5 (13) and 8.3 (19), respectively. Several glyoxalases II such as the enzyme from Aloe vera (20), spinach leaves (21), and bovine liver mitochondria (22) show multiple protein bands when subjected to isoelectric focusing. At least some of the bands may be explained by a varying content of their metal cofactors that obviously exchange or get lost easily (see below). Because the molecular and kinetic properties of the known glyoxalases II are very similar, the diversities of their isoelectric points probably reflect the evolutionary distance rather than functional differences of the enzymes.
Glyoxalases II contain the highly conserved metal binding motif THXHXDH (23). In total, five His and two Asp residues interact directly with two metal ions as shown in the threedimensional structure of human glyoxalase II (24). All these residues are conserved in the trypanosomatid proteins, suggesting that the parasite proteins also possess metal cofactors (Fig. 1). The structure of human glyoxalase II in complex with a substrate analogue revealed three conserved basic residues that are involved in the fixation of the thioester in the active site. Arg-249, Lys-252, and Lys-143 (numbering of human glyoxalase II) are in close proximity to the glycine carboxylate of the glutathione moiety of the substrate analogue (24). These residues, present in all glyoxalases II studied so far, are not conserved in the proteins from T. brucei and the other kinetoplastid organisms. This was the first indication that glutathione thioesters are probably not the physiological substrates of the parasite glyoxalases II.
Overexpression and Purification of T. brucei Glyoxalase II-The glyoxalase II gene was overexpressed from pQE vectors with and without the N-terminal His 6 tag. Purification of Histagged T. brucei glyoxalase II by metal affinity chromatography yielded 10 mg of pure protein from a 1-liter bacterial culture. The tag-free recombinant protein was purified also in a single step using Q-Sepharose as cation exchanger. About 16 mg of pure protein were obtained from a 1-liter culture. Gel filtration of the recombinant protein on Superdex 75 yielded a relative molecular mass of about 25,000 in accordance with a monomeric structure. This corresponds to all glyoxalases II studied so far, e.g. from A. vera (20), human red blood cells (19), human liver (13), and A. thaliana (17). The molecular mass calculated from the protein sequence is 32,507 in good agreement with 32,000 estimated by SDS-polyacrylamide gel electrophoresis.
Synthesis and Structural Analysis of S-Lactoyltrypanothione-Mono-and bis-(lactoyl)trypanothione thioesters were obtained by reacting reduced trypanothione with methyl-glyoxal. The spontaneous reaction resulted in a hemithioacetal (14,25), which is isomerized into the lactoyltrypanothione thioester by yeast glyoxalase I. The method takes advantage of the FIG. 1. Comparison of T. brucei glyoxalase II with glyoxalases II from other organisms. Residues that are conserved in at least five of the seven sequences are depicted in bold. ϩ, residues responsible for metal ion fixation, and *, residues that bind the glutathione moiety of a substrate analogue in the crystal structure of human glyoxalase II (24). T. brucei, T. brucei glyoxalase II, characterized here (AJ492819); T. brucei 2, T. brucei putative second glyoxalase II (AC091702); T. cruzi, T. cruzi putative glyoxalase II (AAL96759); L. major, L. major putative glyoxalase II (LMFLCHR12_92 possible protein); human, Homo sapiens glyoxalase II (Q16775); A. thaliana, A. thaliana cytoplasmic glyoxalase II (O24496); E. coli, E. coli probable hydroxyacyl-glutathione hydrolase (Q47677). The sequences were aligned using CLUSTALW (55). fact that glyoxalase I, in contrast to glyoxalase II, is not highly specific for glutathione but also accepts derivatives of glutathione modified at the glycine carboxylate (26,27). The thioesters were freed from the other reaction components by chromatography on an Oasis MCX cation exchanger cartridge that retains the positively charged trypanothione derivatives. HPLC analysis of bis-(lactoyl)trypanothione showed two peaks with retention times of 17.5 and 19.5 min. The first, minor peak represents mono-(lactoyl)trypanothione and the second, major peak is bis-(lactoyl)trypanothione ( Fig. 2A). Mass spectrometry confirmed bis-(lactoyl)trypanothione as the main reaction product and the presence of a small amount of mono-(lactoyl)trypanothione (Fig. 2B). Mono-(lactoyl)trypanothione (Ն90% pure) was obtained by applying a 10-fold excess of reduced trypanothione over methylglyoxal (data not shown). (23,28). To reveal if hydrolysis of the lactoyltrypanothione thioester by T. brucei glyoxalase II yields also the D-isomer of lactate, the reaction was coupled to that of D-and L-lactate dehydrogenase, respectively. As shown in Fig. 3, addition of L-lactate dehydrogenase to the products of the glyoxalase II reaction did not lead to any NAD reduction. In contrast, in the presence of D-lactate dehydrogenase NADH was readily formed. Because the equilibrium of the lactate dehydrogenase reaction strongly favors pyruvate reduction, the reaction does not run to completion under these assay conditions. Only at alkaline pH and addition of hydrazine to trap pyruvate, can the reaction be completely driven in the reverse reaction (28).

Hydrolysis of S-Lactoyltrypanothione by T. brucei Glyoxalase II Yields D-Lactate and Trypanothione-Hydrolysis of lactoylglutathione by classical glyoxalases II generates D-lactate
Hydrolysis of bis-(lactoyl)trypanothione by T. brucei glyoxalase II regenerates trypanothione as shown by following the time course of the reaction by HPLC. Before starting the reaction by adding the enzyme, the sample contains bis-(lactoyl) trypanothione and a small amount of mono-(lactoyl)trypanothione (Fig. 4A). Incubation of the thioester with glyoxalase II for 8 min leads to the decrease of bis-(lactoyl)trypanothione with the concomitant increase of mono-(lactoyl)trypanothione. Two new peaks appear that were identified as trypanothione and trypanothione disulfide (Fig. 4B). After 72 min, bis-(lactoyl)trypanothione is completely hydrolyzed. The reaction mixture still contains a small amount of mono-(lactoyl)trypano-thione but the main product is trypanothione with a minor fraction being oxidized to trypanothione disulfide (Fig. 4C). Thus hydrolysis of bis-(lactoyl)trypanothione occurs via formation of the monothioester to finally generate free trypanothione. Taken together, T. brucei glyoxalase II cleaves lactoyltrypanothione thioesters into D-lactate and trypanothione (Fig. 5).
Kinetic Studies of the Glyoxalase II-The kinetic parameters of recombinant T. brucei glyoxalase II were determined for lactoylglutathione and the newly synthesized mono-and bis-(lactoyl)trypanothione as substrates and compared with those of the bovine liver enzyme ( Table I). Hydrolysis of lactoylglutathione was measured in the coupled assay where glutathione freed in the reaction reacts with DTNB and formation of the thio-nitrobenzoate is followed at 412 nm (11). The mammalian enzyme yielded a V max of 12 units/mg in accordance with literature data (14). The K m of 191 M corresponded to the values reported for the mitochondrial bovine liver (22) and human liver glyoxalases II (1). T. brucei glyoxalase II hydrolyzed lactoylglutathione but did not show saturation kinetics up to 5 mM. The activities with the trypanothione thioesters were measured in the direct assay following hydrolysis of the thioester at 240 nm. Under the conditions used, bovine liver glyoxalase II did not have any activity with the trypanothione thioesters. In contrast, T. brucei glyoxalase II hydrolyzed bis-(lactoyl)trypanothione and mono-(lactoyl)trypanothione with V max values of about 100 units/mg and K m values of 86 and 108 M, respectively (Table I). The catalytic efficiencies (k cat /K m ) are in the order of 6 ϫ 10 5 M Ϫ1 s Ϫ1 . The preference of T. brucei glyoxalase II for trypanothione instead of glutathione-based thioesters is probably because of the replacement of basic residues shown to locate the glycine carboxylate of glutathione in the mammalian enzyme (24). The structural differences between the trypanosomatid enzymes and all other glyoxalases II reflect the opposed charge distribution of their substrates. Under physiological conditions, glutathionylspermidine-based thioesters carry an overall positive charge, whereas glutathione thioesters are negatively charged. The inability of bovine liver glyoxalase II to hydrolyze the thioesters of trypanothione may be attributed to the fact that the bulky positively charged spermidine moiety of the substrate cannot be accommodated in the active site of the mammalian enzyme (24,29).
T. brucei glyoxalase II shows a rather broad pH-optimum at pH 7.0 -8.0 (Fig. 6A). A corresponding behavior has been reported for human liver glyoxalase II (pH optimum 6.8 -7.5) (30), whereby the optimum of the T. brucei enzyme is slightly shifted to more alkaline pH values. Because the conductivity of the 100 mM MOPS buffer increased from 0.3 to 3.4 mS when changing the pH between 6.1 and 9.2 we measured the activity of the enzyme in 10 -250 mM MOPS buffer at a constant pH of 7.2 (corresponding to 0.22-4 mS) (Fig. 6B). T. brucei glyoxalase II did not show any dependence on the ionic strength within this range. These findings contrast with glyoxalase II from rat erythrocytes, which has been reported to be highly sensitive to the ionic strength of the buffer, yielding the highest activity at the lowest ionic strength (31).

Recombinant T. brucei Glyoxalase II Contains Metal Ions-
The metal content of recombinant His 6 and tag-free glyoxalase II was determined in several enzyme preparations by inductively coupled-plasma optical emission spectrometry (Dr. P. Schramel). The analyses resulted in quite varying compositions with about 1.5 and 0.7-0.9 mol of total metal ions per mole of His 6 and tag-free glyoxalase II, respectively. A typical analysis of the fusion protein yielded 0.45 mol of zinc, 0.42 mol of iron, and 0.6 mol of cobalt ions per mol of protein. Some preparations also contained small amounts of manganese. In the tag-free protein, zinc was the main metal with some iron. The high concentration of cobalt in the fusion protein purified by TALON metal affinity chromatography but not in the tag-free protein obtained by ion exchange chromatography indicates that T. brucei glyoxalase II easily exchanges its metal cofactors. Despite the varying metal compositions, fresh enzyme preparations had very similar specific activities suggesting that T. brucei glyoxalase II is not specific for its metal ligands.
Glyoxalase II Is Expressed in Bloodstream and Procyclic T. brucei-Western blot analyses revealed the presence of glyoxalase II in bloodstream as well as procyclic T. brucei (Fig.  7A). The cellular concentration of the enzyme was estimated from a standard line derived from blots with different amounts of recombinant glyoxalase II and cell extracts (Fig. 7B). With a cell volume of 58 femtoliters (32,33), a concentration of glyoxalase II of about 7 M is obtained for bloodstream T. brucei. When identical cell numbers are applied, procyclic parasites yield slightly weaker signals for glyoxalase II (Fig. 7A). Together with the larger cell volume of procyclic parasites, one can estimate that the glyoxalase II concentration is about 50% of that of bloodstream parasites. The mammalian forms of T. brucei have a very high glucose turnover (8), which probably correlates with formation of large amounts of methylglyoxal and the need for an efficient glyoxalase system (2, 3). On the other hand, cultured procyclic T. brucei contain also significant amounts of glyoxalase II. Because these cells are grown in a proline-rich medium and glycolysis should not play a major role for their energy supply the physiological role(s) of the glyoxalase system in T. brucei remains to be elucidated. DISCUSSION Kinetoplastid organisms possess a unique thiol metabolism based on the dithiol trypanothione and trypanothione reductase instead of the ubiquitous glutathione system. The trypanothione system is the electron donor for the synthesis of DNA precursors and is involved in the antioxidative defense in these parasites (5,34,35). As shown here, trypanothione replaces glutathione also in the glyoxalase system of African trypanosomes. T. brucei glyoxalase II strongly prefers thioesters of trypanothione instead of glutathione as substrates. The glutathione-dependent glyoxalase system is found constitutively in a wide variety of organisms. The pathway has been analyzed in bacteria (36), yeast (37), plants (17,20), and mammals (13,19,26,38). The T. brucei protein is the first glyoxalase II characterized so far that does not use glutathione as cofactor.
The T. brucei glyoxalase II analyzed here, the gene of which is located on chromosome VI, shows 25% overall identity with a protein encoded on chromosome IV that has also been annotated as glyoxalase II. When comparing the four putative glyoxalase II sequences available in trypanosomatid data bases, 66 and 51% of all residues are conserved between the T. brucei glyoxalase II described here and a T. cruzi and L. major protein, respectively, whereas only 26 and 22% are identical in the T. brucei protein encoded in chromosome IV. In addition, the FIG. 5. Proposed reaction mechanism for the detoxication of methylglyoxal by the glyoxalase system in African trypanosomes. Trypanothione reacts spontaneously with methylglyoxal (bold letters) to a mono-or bis-hemithioacetal (shown here), which is then isomerized into (lactoyl)trypanothione by GLX I. The thioester formed is subsequently hydrolyzed into D-lactate and trypanothione by GLX II as characterized here. A glyoxalase I has not yet been detected in T. brucei but putative genes are present in the genomes of T. cruzi and L. major.
latter protein contains a long insertion (following residue 190 of the human protein, Fig. 1) that does not occur in any glyoxalase II characterized so far. Only the functional characterization will clarify if the protein indeed represents a second glyoxalase II in T. brucei.
All (putative) glyoxalases II of trypanosomatid organisms lack three conserved basic residues that are responsible for binding the glutathione moiety of a substrate analogue in the three-dimensional structure of the human enzyme. The glutathione part is bound in the active site mainly by direct interactions between the glycine carboxylate and the side chains of Arg-249 and Lys-252 (24). Thioesters of trypanothione, the preferred substrates of T. brucei glyoxalase II, do not carry a negative charge at the glycine but are positively charged because of the spermidine bridge linking the two glutathione molecules (Fig. 5). The replacement of the basic active site residues in the parasite proteins supports the conclusion that lactoyltrypanothione is a physiological substrate of T. brucei glyoxalase II and the other kinetoplastid enzymes.
The four trypanosomatid proteins contain the metal binding motif THXHXDH, which together with another two His and an Asp residue are present in all glyoxalases II studied so far (23,24). The metal analysis of recombinant T. brucei glyoxalase II yielded different metals and between 0.7 and 1.5 mol of total metal ions per mol of enzyme. In the tag-free protein zinc was the main metal. The His 6 -tagged protein contained zinc, cobalt, and iron (and sometimes manganese). The metal content and composition of glyoxalases II is a continuous matter of discussion. Whereas the old literature reported that the enzymes do not need any metal cofactor, more recent studies demonstrated the presence of two metal ions bound to the protein (23,24,39). The crystal structure of human glyoxalase II revealed two metals that were assigned as zinc ions although the authors mention that the analysis could not discriminate between zinc and iron (24). Cytosolic A. thaliana glyoxalase II contains an iron-zinc binuclear metal center that is essential for substrate binding and catalysis (23). A most recent report shed some light on the variable metal composition and content of these enzymes. Recombinant A. thaliana glyoxalase II is able to incorporate zinc, iron, and manganese depending on the metal added to the bacterial culture medium resulting in enzyme species with similar catalytic efficiency. The high degree of  6. Optima of pH and ionic strength of T. brucei glyoxalase II. The activity (q) of His-tagged T. brucei glyoxalase II was followed (A) in 100 mM MOPS buffers between pH 6.0 and 9.2 and (B) at pH 7.2 in MOPS buffers from 10 to 250 mM as described under "Experimental Procedures." The data are the mean of duplicate measurements, which differed by less than 15%. These series were conducted twice. OE, conductivity of the buffer.

FIG. 7. Quantification of glyoxalase II in bloodstream and procyclic T. brucei.
A, Western blot of total cell lysates of 1 ϫ 10 6 and 2 ϫ 10 6 bloodstream (BF) and procyclic (PC) cultured T. brucei and 10, 20, and 30 ng of recombinant tag-free glyoxalase II as well as 5 ng of tagged protein (His 6 -GLX II). Purified polyclonal rabbit antibodies against the recombinant His 6 -GLX II together with the SuperSignal West Pico chemiluminescent substrate were used for visualization as described under "Experimental Procedures." B, standard diagram based on 5 amounts of tag-free recombinant glyoxalase II (q) using the "Quantity One 1-D Analysis Software" (Bio-Rad). *, 1 ϫ 10 6 cells of bloodstream, and ؉, procyclic parasites. INT, intensity. structural flexibility within the binuclear active site observed in the plant enzyme is probably the basis for the broad metal selectivity. It appears that the catalytic and metal binding properties of the enzyme allow full functionality with various types and ratios of bound metal ions (39).
The substoichiometric metal content in T. brucei glyoxalase II may be attributed to a loss during purification or to a limited metal availability during overexpression. On the other hand, recombinant A. thaliana glyoxalase II isolated from E. coli grown in the presence of excess metals also contained only 1.4 to 1.76 metals per protein (39). Interestingly, storage of T. brucei glyoxalase II causes a simultaneous decrease of k cat and K m . The reason for this may be the loss of metal as described for the A. thaliana enzyme.
Hydrolysis of lactoyltrypanothione by T. brucei glyoxalase II generates D-lactate. Promastigotes of Leishmania braziliensis and L. donovani produce D-lactate from methylglyoxal (40,41), which is obviously not metabolized but excreted into the culture medium (42). In contrast, in isolated rat liver mitochondria, specific translocators mediate the import of D-lactate where a putative D-lactate dehydrogenase metabolizes D-lactate to pyruvate (43). The fate of D-lactate produced in T. brucei still needs to be elucidated.
The two thioester substrates, mono-and bis-(lactoyl) trypanothione, were obtained by reacting different ratios of methylglyoxal and reduced trypanothione. Formation of bis-(lactoyl)trypanothione is favored in the presence of a 3-4-fold excess of methylglyoxal over trypanothione. In contrast, a 10fold excess of reduced trypanothione over methylglyoxal leads to formation of mono-(lactoyl)trypanothione. Considering a cellular concentration of trypanothione of Ն350 M in bloodstream T. brucei (44) and a methylglyoxal concentration of 1-2 M reported for human blood (45) and Saccharomyces cerevisiae (46), the monothioester of trypanothione is probably the main physiological substrate of the parasite enzyme.
The catalytic efficiency of T. brucei glyoxalase II with the lactoyltrypanothione thioesters is comparable with that of the C. albicans enzyme with lactoylglutathione (18). The recombinant glyoxalases II from man (13), A. thaliana (17), and yeast (47) have k cat /K m values for lactoylglutathione that are about an order of magnitude higher. This may at least partially be attributed to the higher assay temperature (37°C instead of 25°C) and differing buffer conditions. The commercially available bovine liver glyoxalase II used here showed under identical assay conditions with lactoylglutathione a 20-fold lower catalytic efficiency than the T. brucei enzyme with the lactoyltrypanothione thioesters.
The physiological function(s) of the trypanothione-dependent glyoxalase system in T. brucei remains to be elucidated. The enzyme occurs in both bloodstream and procyclic parasites. In yeast (47), spinach (21), as well as bovine (22) and rat liver (38), glyoxalase II is present in the cytoplasma and mitochondria. In yeast, and A. thaliana the cytosolic and mitochondrial forms of the enzyme are the products of two different genes (47,48). The protein sequence of T. brucei glyoxalase II does not show a sorting signal and preliminary Western blots with cell fractions of bloodstream T. brucei indicate a cytosolic localization (data not shown). If methylglyoxal is indeed the main physiological substrate the question arises where the ketoaldehyde is formed and detoxified. In trypanosomes, the first seven enzymes of glycolysis reside in glycosomes, peroxisome-related organelles. Thus methylglyoxal may be generated in the glycosomes and subsequently exported to the cytosol for detoxication. Alternatively and more likely, the ketoaldehyde is formed in the cytosol. In bloodstream parasites, the NAD/NADH balance of glycolysis is maintained by reduction of dihydroxyacetone phosphate to glycerol 3-phosphate within the glycosome. Glycerol 3-phosphate is then reconverted into dihydroxyacetone phosphate by a mitochondrial glycerolphosphate oxidase (49). This shuttle implies the presence of cytosolic dihydroxyacetone phosphate. Triose-phosphate isomerase deficiency is associated with accumulation of dihydroxyacetone phosphate and concomitantly methylglyoxal (50). As shown by Helfert et al. (51) triose-phosphate isomerase is essential for bloodstream T. brucei. At enzyme levels of about 15% of wild-type parasites the growth rate is halved and the total cellular dihydroxyacetone phosphate concentration is increased. Because dihydroxyacetone phosphate spontaneously degrades to methylglyoxal the lethality of the triose-phosphate isomerase knock-out may at least partially be because of increased formation of toxic methylglyoxal.
Inhibitors of glyoxalase I and glyoxalase II are investigated as potential anti-tumor and anti-malarial agents (52)(53)(54). The high glucose turnover of T. brucei together with the unique trypanothione dependence renders the glyoxalase system an attractive new target for anti-trypanosomal chemotherapy. Future work will include kinetic studies on other physiological ketoaldehydes such as glyoxal. RNA-interference experiments are in progress to elucidate if glyoxalase II is essential for T. brucei.