Genome mining and biochemical analyses have shown that Leishmania major possesses two pathways for cysteine synthesis – the de novo biosynthesis pathway comprising SAT (serine acetyltransferase) and CS (cysteine synthase) and the RTS (reverse trans-sulfuration) pathway comprising CBS (cystathionine β-synthase) and CGL (cystathionine γ-lyase). The LmjCS (L. major CS) is similar to the type A CSs of bacteria and catalyses the synthesis of cysteine using O-acetylserine and sulfide with Kms of 17.5 and 0.13 mM respectively. LmjCS can use sulfide provided by the action of MST (mercaptopyruvate sulfurtransferase) on 3-MP (3-mercaptopyruvate). LmjCS forms a bi-enzyme complex with Leishmania SAT (and Arabidopsis SAT), with residues Lys222, His226 and Lys227 of LmjCS being involved in the complex formation. LmjCBS (L. major CBS) catalyses the synthesis of cystathionine from homocysteine, but, unlike mammalian CBS, also has high cysteine synthase activity (but with the Km for sulfide being 10.7 mM). In contrast, LmjCS does not have CBS activity. CS was up-regulated when promastigotes were grown in medium with limited availability of sulfur amino acids. Exogenous methionine stimulated growth under these conditions and also the levels of intracellular cysteine, glutathione and trypanothione, whereas cysteine had no effect on growth or the intracellular cysteine levels, correlating with the low rate of transport of cysteine into the cell. These results suggest that cysteine is generated endogenously by promastigotes of Leishmania. The absence of CS from mammals and the clear differences between CBS of mammals and Leishmania suggest that each of the parasite enzymes could be a viable drug target.

INTRODUCTION

Leishmania is a protozoan parasite that is the causative agent of a spectrum of diseases collectively known as the leishmaniases, there being some 60000 new cases annually [1]. Some aspects of the metabolism of Leishmania have been investigated in detail [2], notably that involving the low molecular mass thiol trypanothione, a conjugate of GSH and spermidine, which appears to play a pivotal role in maintaining intracellular redox homoeostasis and providing defence against oxidative stress [3]. The synthesis of glutathione and thus trypanothione depends on the availability of cysteine. This sulfur-containing amino acid is an essential growth factor for the related trypanosomatid Trypanosoma brucei [4], whereas Trypanosoma cruzi can generate it from homocysteine [5]. In contrast, it is synthesized de novo in the microaerophilic parasitic protozoa Entamoeba, Giardia and Trichomonas – in which cysteine itself is the key antioxidant and redox buffer [6,7]. The present study was undertaken to establish how Leishmania obtains cysteine and to characterize the mechanisms involved.

There are two routes known for the generation of cysteine, the de novo biosynthesis pathway, otherwise called the sulfhydrylase pathway [7], and the RTS [reverse TS (trans-sulfuration)] pathway that converts homocysteine to cysteine in two steps, catalysed by CBS (cystathionine β-synthase, EC 4.2.1.22) and CGL (cystathionine γ-lyase, EC 4.4.1.1) (RTS; see Figure 1) [8]. Mammals contain only the latter pathway, but the situation is more complex in other groups. The de novo biosynthetic pathway for cysteine also comprises two steps, catalysed by SAT (serine acetyltransferase, EC 2.3.1.30) and CS (cysteine synthase, EC 4.2.99.8). In plants, distinct isoforms of SAT and CS exist in the mitochondria, plastids and cytosol [9], whereas bacteria contain only one isoform of SAT and CS, although there are two types of the latter, designated CS type A (CS-A) and CS type B (CS-B) [10]. SAT, CS and CBS activities have been identified in T. cruzi, but only the genes for SAT and CBS have been described [5], whereas Entamoeba histolytica contains multiple genes for SAT and CS [6]. Trichomonas vaginalis also has multiple type B CS genes but lacks the SAT gene; instead it uses phosphoserine as the precursor for synthesis of cysteine [11].

Cysteine biosynthesis pathways in L. major

Figure 1
Cysteine biosynthesis pathways in L. major

Enzymes present in L. major (Lmj), and the reactions they catalysed, identified by in silico analysis using BLAST algorithm are given with their geneDB systematic name.

Figure 1
Cysteine biosynthesis pathways in L. major

Enzymes present in L. major (Lmj), and the reactions they catalysed, identified by in silico analysis using BLAST algorithm are given with their geneDB systematic name.

We had identified genes apparently encoding SAT and CS in Leishmania [12], but the genome also contains genes encoding all four enzymes of the trans-sulfuration pathway (both forward and reverse). Thus the in silico evidence suggested a functional redundancy in Leishmania for the generation of cysteine. The aim of the present study was to determine whether or not the predicted enzymes function as postulated and to characterize those that do. Moreover, we wished to determine the interplay between the two routes whereby cysteine may be synthesised, to investigate how they may be regulated to control cysteine homoeostasis, and whether uptake of exogenous cysteine is also important. One notable way in which biosynthesis of cysteine is controlled in some organisms is through complex formation between SAT and CS, thus the potential for such complex formation in Leishmania has been studied. As mammals lack the de novo biosynthesis of cysteine, this pathway, if operational and important in Leishmania, could be exploitable as a drug target.

EXPERIMENTAL

Strains and cultivation of L. major

Promastigotes of L. major (MHOM/IL/80/Friedlin) were used throughout this study. They were grown at 27 °C in either modified Eagle's medium, designated HOMEM (Gibco, UK) supplemented with 10% (v/v) heat-inactivated FCS (fetal calf serum) for routine maintenance [12] or, for the experiments investigating the use of exogenous organic sulfur sources, in SDM (sulfur-depleted medium) comprising MEM (minimum essential medium) (ICN-Biomedical) without an exogenous sulfur source (no cysteine, cystine, methionine or serine) and supplemented with 20% (v/v) dHiFCS (dialysed heat-inactivated FCS) (Sigma-Aldrich), D-glucose (1000 mg/l), glutathione (0.10 mg/l), hypoxanthine (0.70 mg/l), thiamine (0.6 mg/l), uracil (0.6 mg/l), alanine (50 mg/l), aspartic acid (0.06 mg/l), glutamic acid (0.14 mg/l), glutamine (0.02 mg/l), proline (0.08 mg/l), haemin (0.035 mg/l) and folic acid (0.02 mg/l) or in RPMI 1640 without cystine and methionine (Sigma) and supplemented with MgCl2 (0.1 g/l), Hepes, pH 7.2 (40 mM), biopterin (5 μg/ml), haemin (1% v/v) and 20% (v/v) dHiFCS. The sulfur source required for growth of L. major promastigotes was investigated by adding methionine (300 μM), cysteine (300 μM), homocysteine (300 μM), serine (280 μM), OAS (O-acetylserine; 250 μM) or sodium sulfide (80 μM) with and without serine (300 μM). The medium without an exogenous sulfur source acted as the control in these experiments. Cell densities were estimated using an improved Neubauer haemocytometer.

Identification and cloning of SAT, CS and CBS of L. major

The genome database for L. major (http://www.ebi.ac.uk/Tools/blast2/parasites.html) was searched using the TBLASTN algorithm with protein sequences involved in cysteine de novo biosynthesis and RTS pathways from bacteria, yeast, plants and parasitic protozoa as queries to identify putative L. major proteins involved in these pathways. Expectation values (E-value) of the TBLASTN output, amino acid sequence alignments, and identification of putative key domains known to be key components for enzymatic activities were all studied to determine if there was significant orthology. The Leishmania proteins identified were assigned as orthologues potentially involved in cysteine biosynthesis and have been annotated in GeneDB (http://www.geneDB.org) with the systematic names SAT (serine acetyltransferase, LmjF43.2850), CS (cysteine synthase, LmjF36.3590), CGL (cystathionine γ-lyase, LmjF35.3230) and CBS (cystathionine β-synthase, LmjF17.0250). Enzymes potentially comprising a forward trans-sulfuration pathway were also identified and annotated as CBL (cystathionine β-lyase; LmjF32.2640) and CGS (cystathionine γ-synthase; LmjF14.0460).

gDNA (genomic DNA) of L. major was isolated as described in [13]. The ORFs (open reading frames) of the genes putatively involved in the L. major cysteine synthesis pathways were amplified by PCR using the Expand High Fidelity PCR system (Roche Molecular Biochemicals) with gene-specific primers modified with appropriate restriction sites (to facilitate cloning into their respective expression vectors) as detailed in Supplementary Table S1 (at http://www.BiochemJ.org/bj/420/bj4200451add.htm). All PCR assays were carried out in a GeneAmp 9600 PCR system (PerkinElmer Life Sciences) for 30 cycles of denaturation (94 °C, 15 s), annealing (65 °C, 15 s) and extension (72 °C, 2 min). Each ORF was verified by nucleotide sequencing [MBSU (Molecular Biology Support Unit), University of Glasgow, Glasgow, U.K.] and cloned into the pET expression vectors (Novagen) pre-digested with appropriate restriction enzymes to produce the plasmids detailed in Supplementary Table S1. The plasmids were used to transform BL21(DE3) for recombinant protein expression to generate recombinant fusion proteins containing a 6×His-tag, which facilitates purification using affinity chromatography.

Protein expression and purification

L. major SAT, CS and CBS were expressed from a clone of BL21(DE3) transformed with their respective plasmids (Supplementary Table S1). Overnight cultures in LB (Luria–Bertani) broth were used to inoculate fresh LB supplemented with 100 μg/ml ampicillin or 25 μg/ml kanamycin (Supplementary Table S1) and grown at 37 °C until a D600 of 0.6 was reached. Expression of recombinant protein was induced with 1 mM IPTG (isopropyl β-D-thiogalactoside) overnight at 15 °C. Cells were harvested and resuspended in 5 ml of buffer A [50 mM sodium phosphate, 0.3 M NaCl, pH 8.0, and 25 μM PLP (pyridoxal phosphate)] containing 5 mM imidazole and disintegrated by sonication. Soluble fractions were recovered by centrifugation at 16000 g for 30 min at 4 °C. Proteins were affinity purified by applying the soluble fraction to a 13 ml nickel-nitrilotriacetic column (bioCAD 700E workstation) pre-equilibrated with buffer A. The column was washed with 60 ml and 30 ml of buffer A containing 20 mM and 60 mM imidazole respectively. The His-tagged fusion proteins were then eluted with 500 mM imidazole in buffer A. The SAT-CS bi-enzyme complex was purified in a similar manner from BL21(DE3) co-expressing the N-terminal 6×His-tagged SAT and the non-tagged L. major CS from separate plasmids (Supplementary Table S1). The eluants of CS and CBS were dialysed against 50 mM Tris/HCl, pH 7.9, and 25 μM PLP overnight at 4 °C, whereas the eluants containing SAT and the SAT-CS bi-enzyme complex were dialysed against 50 mM Tris/HCl, pH 7.9, at 4 °C overnight. All proteins were stored at 4 °C and their protein concentrations determined with a Bio-Rad protein assay kit (Bradford) using BSA as a standard. SAT and CS of Arabidopsis thaliana were expressed and purified as described previously in [14] using the same expression constructs (gifts from Professor R. Hell) and used as positive controls in the present study. Site-directed mutagenesis of L. major CS and Arabidopsis CS was performed with the primer and plasmid combinations specified in Supplementary Table S1 using the QuickChange™ site-directed mutagenesis kit (Stratagene) according to the manufacturer's instructions.

Enzymic activities of SAT, CS and CBS

CS activity was determined at 37 °C in a 500 μl reaction containing 200 mM potassium phosphate, 1 mg/ml BSA, 1 mM EDTA, 0.2 mM PLP, 30 mM OAS, 3 mM Na2S and 15 ng CS or 0.1 μg CBS as appropriate. The enzyme and substrate were pre-incubated for 5 min at 37 °C before starting the reaction by addition of sodium sulfide. Samples were taken after 5, 10 and 15 min at 37 °C and the cysteine concentration was quantified using the method of Kredich and Tomkins [15]. The first step of this procedure, in which the sample is diluted 5-fold in 0.4 M nitrous acid, effectively stops the reaction. The serine sulfhydrylase activity of CS, in which serine combines with sulfide to form cysteine, was determined in the same way as described for the CS assay above, except that 50 mM serine was used instead of OAS. To determine whether sulfide produced by the action of MST (mercaptopyruvate sulfurtransferase) on 3-MP (3-mercatopyruvate) could be utilized by CS for cysteine biosynthesis, assays were run for 5 min with various concentrations of 3-MP (ranging from 0–10 mM) and excess 2-mercaptoethanol (5 mM) and 0.5 μg MST to generate sulfide that could subsequently condense with OAS. 30 mM OAS and 0.5 μg CS were added to the reaction after the preincubation and incubated for a further 5 min; the amount of cysteine produced was then quantified as described above.

Cystathionine production from homocysteine and serine (CBS activity) was determined in a 100 μl reaction containing 50 mM Tris/HCl, pH 7.3, 10 mM serine, 10 mM homocysteine, 0.25 mM PLP and 0.1 μg enzyme. All components were equilibrated to 37 °C and the reaction was initiated with enzyme. After 2–5 min incubation at 37 °C, the reaction was stopped with 50 μl 20% (w/v) trichloroacetic acid and incubated on ice for 10 min. The mixture was centrifuged for 5 min at 16000 g and the supernatant was used for cystathionine analysis as described in [5]. The amount of cystathionine formed was determined by adding 1 ml of ninhydrin reagent (10 g/l in glacial phosphoric acid) to 130 μl assay supernatant, boiling (5 min), cooling on ice (2 min) and incubating for 20 min at room temperature (25 °C) for colour development. The absorbance was measured at 455 nm. A standard curve was prepared with cystathionine (0–500 nmol) dissolved in ninhydrin reagent and treated as described above to quantify the amount of cystathionine formed. Cystathionine production from homocysteine and cysteine (homocysteine sulfhydrylase activity) was determined in a 100 μl reaction containing 50 mM Tris/HCl, pH 7.3, 0.3 mM homocysteine, 5 mM cysteine, 0.25 mM PLP and 0.3 μg enzyme. Again, all components were equilibrated at 37 °C and the reaction initiated with enzyme. The amount of cystathionine formed was quantified as described above.

Sulfide production in the cysteine desulfhydration reaction, in which cysteine combines with 2-mercaptoethanol to form S-2-hydroxyethylcysteine and hydrogen sulfide [16], was monitored by trapping the sulfide with lead acetate [12]. The 1 ml reaction mixture contained 50 mM Tris/HCl, pH 8.0, 10 mM cysteine, 20 mM 2-mercaptoethanol, 0.1 mM PLP, 0.33 mM lead acetate and 300 ng CBS or CS (to start the reaction). The hydrogen sulfide released during incubation at 37 °C was determined by monitoring the production of lead sulfide through its absorbance at 360 nm. Homocysteine desulfurase activity was measured similarly, except that 5 mM of homocysteine was used instead of cysteine.

SAT catalyses the formation of OAS and CoA from serine and acetyl-CoA. A DTNB [5,5′-dithiobis-(2-nitrobenzoic acid)] assay was used to quantify CoA production as described in [17]. In this assay, DTNB reacts with the sulfhydryl compound CoA to form TNB, which has an extinction coefficient at 412 nm of 13600. The assay were carried out at 37 °C in 100 mM Tris/HCl, pH 8.0, with 0.1 mM acetyl-CoA, 1 mM serine, 0.4 mg/ml DTNB and 5–50 μg of enzyme. The reaction was initiated by the addition of serine and the absorbance at 412 nm was monitored continuously at 37 °C for 5–10 min. Kinetic parameters were calculated using Grafit 5 (Erithacus Software Ltd) software.

Western blot analysis and SDS/PAGE

Rabbit polyclonal α-CS and α-CBS antisera were raised against recombinant proteins of SAT, CS and CBS of L. major by the Scottish Antibody Production Unit (Carluke, U.K.), using standard protocols. Parasite lysates were produced by resuspension of parasite pellets in lysis buffer [0.25 M sucrose, 0.25% (v/v) Triton X-100, 10 mM EDTA, 10 μM E-64, 2 mM 1,10-phenanthroline, 4 μM pepstatin A and 1 mM PMSF]. Lysates were centrifuged at 13000 g for 30 min at 4 °C, and an aliquot of the resulting supernatant (8 μg of protein) was subjected to Western blot analysis as described previously [12]. The polyclonal immune rabbit antisera against SAT, CS, CBS, MST and TDR1 (thiol-dependent reductase 1) were diluted 1:2000 (v/v), 1:5000 (v/v), 1:2000 (v/v), 1:2000 (v/v) and 1:5000 (v/v) respectively, in Tris-buffered saline (137 mM NaCl and 10 mM Tris/HCl, pH 7.4) containing 1% (w/v) non-fat dried skimmed milk and 0.1% (v/v) Tween 20. Monoclonal immune mouse serum recognizing the His-tag of the proteins was used at 1:5000 (v/v) in the same buffer. Bound antibody was detected using horseradish peroxidase-coupled secondary α-mouse (1:5000) and α-rabbit (1:5000) antibodies (Scottish Antibody Production Unit) and enhanced chemiluminescence Western blotting detection reagents (Pierce Ltd) according to the manufacturer's recommendations. The antisera against MST and TDR1, used as loading controls, have been described previously in [12,18]. ImageJ (http://rsb.info.nih.gov/ij/index.html) was used to quantify the Western blot signals in order to make comparisons.

Cysteine transport assay

Transport of [35S]-L-cysteine (American Radiochemicals) into L. major promastigotes was performed using the oil stop technique as described previously in [19]. Briefly, promastigotes were harvested and washed with CBSS (Carter basal salt solution), pH 7.4 (33 mM Hepes, 98 mM NaCl, 4.6 mM KCl, 0.3 mM CaCl2, 0.07 mM MgSO4, 5.8 mM NaH2PO4, 0.3 mM MgCl2 and 14 mM D-glucose), and resuspended in CBSS at 2×108 cells/ml. Cells (100 μl) were then incubated with 25 or 50 nM [35S]-L-cysteine together with non-radioactive cysteine in a microcentrifuge tube containing 200 μl oil [7:1 (v/v) dibutyl-phthalate/mineral oil; d=1.018 g/ml] for a period at 25 °C, as indicated in Results. Incubations were terminated by centrifugation of cells through the oil at 16000 g for 1 min. This separates the radiolabelled cells from extracellular radioactivity. The tubes were frozen in liquid nitrogen and the tips containing the cell pellets cut off, whereupon the contents were solubilized in 2% (w/v) SDS for 30 min. Scintillation fluid (3 ml; Optiphase HiSafe III; Perkin-Elmer) was added and samples were left overnight at room temperature. Radioactivity in the cell pellet was then determined using a 1450 MicroBeta Trilax liquid scintillation counter. The zero uptake background was determined with cells prepared as described above, but with all operations at 0 °C and with immediate termination by centrifugation. The assays were carried out using triplicates and transport values were determined by subtraction of the zero uptake background values. Kinetic parameters were calculated using Grafit 5 (Erithacus Software Ltd) software.

Analysis of intracellular thiol levels using HPLC

L. major promastigotes were harvested from late log phase cultures by centrifuging at 2500 g for 10 min, and resuspended at a density of 2.5×107 in 50 μl of 40 mM N-[2-hydroxyethyl]-piperazine-N′[3-propanesulfonic acid], 4 mM diethylenetriamine penta-acetic acid, pH 8.0, containing 0.7 mM tris(2-carboxyethyl)phosphine and incubated at room temperature for 45 min. Monobromobimane was added to give a final concentration of 1 mM and the mixture was heated for 3 min at 70 °C. Extracts were deproteinized by addition of an equal volume of 4 M methanesulfonic acid, pH 1.6, and incubated on ice for 30 min. Proteins were removed by centrifugation at 16000 g for 7 min at 4 °C and the resultant supernatants were analysed by HPLC [using an UltiMate HPLC system (Dionex) and a GEMINI C18 column (Phenomenex)]. The mobile phase consisted of two solvents: solvent A, 0.25% acetic acid, and solvent B, 100% acetonitrile. Metabolites were separated at a flow rate of 0.55 ml/min by application of the following gradient (% of solvent B): 0 min, 0%; 10 min, 0%; 40 min, 8%; 100 min, 15%; 110 min, 50%; 111 min, 0%; 121 min, 0%. Thiols were detected using a fluorescence detector (excitation, 365 nm; emission, 480 nm). Glutathionine, cysteine and trypanothionine were identified by comparison of their retention times with those of their respective standards. A standard curve for quantification was achieved by integration of peak areas of known quantities of thiols, and experimental peak areas were compared with this curve to determine nmol of thiol/(108 cells). Total thiols were calculated as the sum of cysteine, glutathione and 2×trypanothione concentrations. Differences were analysed using parametric Student's 2-tailed t tests. The results presented are from three independent experiments.

RESULTS

Identification of genes encoding enzymes of cysteine biosynthesis in L. major

We identified Leishmania genes likely to encode enzymes involved in cysteine biosynthesis by searching the Leishmania genome databases (http://www.ebi.ac.uk/Tools/blast2/parasites.html) using protein sequences from plants, bacteria, yeast and parasitic protozoa orthologues as the queries. The L. major sequences identified were then back-searched against the SWISSPROT database to confirm gene identity. These analyses suggest that L. major, Leishmania infantum and Leishmania braziliensis all have enzymes comprising two cysteine biosynthetic routes, the de novo biosynthesis and the RTS pathways (Table 1, Figure 1). The L. major genome contains single copy genes, validated by Southern blot analysis (results not shown), for CBS (LmjCBS; LmjF17.0250) and CGL (LmjCGL; LmjF35.3230) of the RTS pathway, and CS (LmjCS; LmjF36.3590) and SAT (LmjSAT; LmjF34.2850) of the de novo biosynthesis pathway. Each of the four predicted enzymes has the key residues and motifs reported to be required for activity, as detailed below. In addition, L. major encodes CGS (LmjCGS; LmjF14.0460) and CBL (LmjCBL; LmjF32.2640), forming the forward TS pathway, implying that the parasites also can convert cysteine into methionine. A similar search of the T. cruzi genome (http://www.geneDB.org), a related parasite, also identified orthologues of both biosynthetic pathways, although there appear to be multiple copies of all the genes (Table 1). The T. brucei genome (www.geneDB.org), on the other hand, contains single copies of genes encoding enzymes of the RTS pathway but lacks genes encoding the proteins of the cysteine de novo biosynthetic and TS pathways.

Table 1
Genes identified in the genome of L. major orthologous to genes encoding enzymes of cysteine biosynthesis

1GI number is a unique series of digits assigned consecutively to each sequence record processed by NCBI (http://www.ncbi.nlm.nih.gov).

     Orthologues in other trypanosomatids 
Pathway Gene name GI number of query sequence1 Query organism L. major systematic name in GeneDB L. infantum L. braziliensis T. cruzi T. brucei 
De novo SAT 13241626 T. cruzi LmjF34.2850 LinJ34.V2.2710 Absent Tc00.1047053504013.40 Absent 
       Tc00.1047053510879.80  
 CS 2346964 E. histolytica LmjF36.3590 LinJ36.2007.0420 V3 LbrM35.V2.3820 Tc00.1047053507165.50 Absent 
       Tc00.1047053507793.20  
RTS CBS 70886353 T. cruzi LmjF17.0250 LinJ17.2007.0420 V3 LbrM17.V2.0230 Tc00.1047053510381.10 Tb11.02.5400 
       Tc00.1047053506905.50  
       Tc00.1047053508175.60  
       Tc00.1047053508177.110  
       Tc00.1047053508177.120  
       Tc00.1047053508177.129  
       Tc00.1047053508241.140  
       Tc00.1047053509149.9  
       Tc00.1047053511691.10  
       Tc00.1047053511691.20  
 CGL 399331 S. cerevisiae LmjF35.3230 LinJ35.V3.3280 LbrM34.V2.October Tc00.1047053510739.19 Tb09.221.3330 
       Tc00.1047053510661.250  
       Tc00.1047053510741.10  
TS CBL 16078253 B. subtilis LmjF32.2640 LinJ32.V3.2789 LbrM32.V2.October Absent Absent 
 CGS   LmjF14.0460 LinJ14.V3.0470 LbrM14.V2.October Absent Absent 
     Orthologues in other trypanosomatids 
Pathway Gene name GI number of query sequence1 Query organism L. major systematic name in GeneDB L. infantum L. braziliensis T. cruzi T. brucei 
De novo SAT 13241626 T. cruzi LmjF34.2850 LinJ34.V2.2710 Absent Tc00.1047053504013.40 Absent 
       Tc00.1047053510879.80  
 CS 2346964 E. histolytica LmjF36.3590 LinJ36.2007.0420 V3 LbrM35.V2.3820 Tc00.1047053507165.50 Absent 
       Tc00.1047053507793.20  
RTS CBS 70886353 T. cruzi LmjF17.0250 LinJ17.2007.0420 V3 LbrM17.V2.0230 Tc00.1047053510381.10 Tb11.02.5400 
       Tc00.1047053506905.50  
       Tc00.1047053508175.60  
       Tc00.1047053508177.110  
       Tc00.1047053508177.120  
       Tc00.1047053508177.129  
       Tc00.1047053508241.140  
       Tc00.1047053509149.9  
       Tc00.1047053511691.10  
       Tc00.1047053511691.20  
 CGL 399331 S. cerevisiae LmjF35.3230 LinJ35.V3.3280 LbrM34.V2.October Tc00.1047053510739.19 Tb09.221.3330 
       Tc00.1047053510661.250  
       Tc00.1047053510741.10  
TS CBL 16078253 B. subtilis LmjF32.2640 LinJ32.V3.2789 LbrM32.V2.October Absent Absent 
 CGS   LmjF14.0460 LinJ14.V3.0470 LbrM14.V2.October Absent Absent 

LmjSAT encodes a protein of 411 amino acids (45.5 kDa) with significant identities to the multiple SATs of other protozoa, T. cruzi (38.4–39.1%) and E. histolytica (21.7–30.4%) (Supplementary Figure S1A at http://www.BiochemJ.org/bj/420/bj4200451add.htm). The hexapeptide structural domain required for acyltransferase activity (residues 274–380; L. major numbering) and the β-cluster (residues 277–364) that interacts with CS to form the bi-enzyme complex (Supplementary Figure S1A) [20,21] are also conserved. The residues that are likely to be involved in acetyl-CoA binding comprise Cys376, Pro383, Gln391 and Ser393 respectively, in contrast with Arabidopsis, where they are Ser263, Pro270, Gly277 and Pro279 [22,23].

LmjCS encodes a protein of 342 amino acids (35.5 kDa) with high sequence identities with CSs of other protozoa; 72% for T. cruzi, 42% for E. histolytica and 34–38% for T. vaginalis (Supplementary Figure S1B). LmjCS has the four lysine residues (Lys40, Lys52, Lys67 and Lys199) required for binding the PLP cofactor and thought to be essential for sulfhydrylase activity [24]. All CS enzymes have a similar structure and catalytic mechanism but show differences in substrate specificity, determined in part by the structure of the β8–β9 loop that lines the entry to the active site [10]. The short loop found in the type A isoforms of bacterial and plant CSs [25] severely restricts access to the active site and these enzymes have a strict specificity for OAS and sulfide. The type B CS isoforms of bacteria [25], thermophiles [26] and Trichomonas [11] have an extended loop containing a conserved charged residue that allows the use of larger, negatively charged molecules such as O-phosphoserine and thiosulfate as substrates. Sequence comparison shows that LmjCS has the short β8–β9 loop similar to bacterial type A CS isoforms and thus we predicted that it has a substrate preference for sulfide and OAS. LmjCS also contains the residues Lys222, His226 and Lys227 implicated for plant [27] and bacterial type A [28] CSs in binding to SAT which, together with the presence of the β-cluster residues in LmjSAT, suggested that the two are likely to interact to form a LmjSAT–LmjCS bi-enzyme complex; the equivalent in other organisms plays roles in activating SAT and/or regulating CS activity [27,29]. These residues are absent from type B CS isoforms that do not form a complex with SAT.

CBS is also a PLP-dependent enzyme, evolutionarily related to CS and in most cases with some CS activity [14]. LmjCBS encodes a protein of 359 amino acids (39.2 kDa) which has three (Lys42, Lys53 and Lys202) of the four lysine residues reported to be required for activity of other CSs (Supplementary Figure S1C) [30]. The fourth position in LmjCBS is, instead, Glu66. LmjCBS has relatively high amino acid sequence identity with LmjCS (31%) and quite high identities with the region comprising the catalytic domain (residues 1–359; [5]) of CBSs from Leishmania tarentolae (83%), T. cruzi (56–66%), Homo sapiens (49%) and Saccharomyces cerevisiae (42%). LmjCBS, however, differs from HsCBS (H. sapiens CBS) in lacking the haem-binding motif and regulatory domain at the N- and C-termini respectively [31]. The haem-binding motif acts as a redox sensor [32], whereas the regulatory domain controls the tetrameric state of the protein and so CBS activity [32]. Interestingly, the Saccharomyces protein (ScCBS) also lacks the N-terminal haem-binding motif [33] but does possess the C-terminal regulatory domain [8,34], whereas TcCBS (T. cruzi CBS) lacks the C-terminal regulatory domain but has an N-terminal extension (Supplementary Figure S1C), although this does not bind haem [5]. LmjCBS, TcCBS and ScCBS all also lack the oxidoreductase (CysXXCys) motif of HsCBS [35].

LmjCGL encodes a protein of 552 amino acids (60.6 kDa) and has the PLP-binding moiety (between positions 197–202) and the reactive lysine (Lys199; L. major numbering) required for activity (Supplementary Figure S1D) [36]. LmjCGL's identities with other CGLs are: Bacillus subtilus, 38%; S. cerevisiae, 35%; H. sapiens, 34%; T. brucei, 72%; and T. cruzi, 80%.

CS of L. major catalyses the synthesis of cysteine from OAS

Soluble rLmjCS (recombinant LmjCS) was produced in large quantities (∼25 mg/l) using the pET21a+ expression plasmid in BL21(DE3) Escherichia coli and purified to apparent homogeneity for analysis (Supplementary Figure S2 at http://www.BiochemJ.org/bj/420/bj4200451add.htm). rLmjCS activity was optimal at pH 8.0 and 37 °C and was stable for several weeks without any appreciable loss of activity when stored at 4 °C (results not shown). The apparent Kcat values of rLmjCS using OAS and sodium sulfide were 2047 s−1 and 2669 s−1 for OAS and sulfide, respectively (Table 2). The Km for sulfide was low (0.13 mM). No activity was detected between OAS and methanethiol or sodium thiosulfate (the detection limit being 0.6 μmol/min per mg of protein). rLmjCS was unreactive towards O-acetylhomoserine (50 mM), O-succinylhomoserine (50 mM), O-phosphoserine (100 mM) or serine (50 mM) and sodium sulfide (the detection limit being 0.2 μmol/min per mg of protein). rLmjCS had no detectable CBS activity, which forms cystathionine from serine and homocysteine (the detection limit being 0.4 μmol/min per mg of protein−1). rLmjCS also has cysteine desulfurase activity, hydrolysing cysteine to yield sulfide, but only in the presence of 2-mercaptoethanol, with Kcat values of 88.2 s−1 and 99.0 s−1 for cysteine and 2-mercaptoethanol respectively (Table 2). It has been proposed that this desulfurase activity has a reaction mechanism similar to the sulfhydrylase reaction [16]. In this, cysteine binds to the enzyme and sulfide by β-elimination to form the α-aminoacrylate intermediate, 2-mercaptoethanol then acts as the nucleophile in the second half-reaction to produce S-2-hydroethylcysteine. No desulfurase activity was detected with homocysteine (50 mM) and 2-mercaptoethanol or with cysteine or homocysteine and DTT (dithiothreitol) (the detection limit being 0.2 μmol/min per mg of protein). rLmjCS was relatively insensitive to PAG (propargylglycine) and hydroxylamine, inhibitors of various PLP-dependent enzymes, with 1 mM PAG and hydroxylamine inhibiting rLmjCS sulfhydrylase activity only by 1% and 3% respectively. On the other hand, 1 mM phenylhydrazine-HCl inhibited the CS sulfhydrylase activity of rLmjCS by 36±0.2%.

Table 2
Kinetic parameters of enzymes of cysteine synthesis

For kinetic analyses, ≥6 different substrate concentrations were used with at least two replicate assays. 1O-acetylserine sulfhydrylase activity of LmjCS: the Km for OAS was determined using 3 mM Na2S with 1–40 mM OAS and the Km for Na2S was determined using 30 mM OAS with 0.1–1 mM Na2S. 2Cysteine desulfurase activity of LmjCS: the Km for cysteine was determined using 50 mM 2-mercaptoethanol with 0.2–20 mM cysteine and the Km for 2-mercaptoethanol was determined using 20 mM cysteine and 0.5–50 mM 2-mercaptoethanol. 3O-acetylserine sulfhydrylase activity of LmjCBS: the Km for OAS was determined using 3 mM Na2S with 0.1–100 mM OAS and the Km for Na2S was determined using 100 mM OAS with 1–3 mM Na2S. 4Cysteine desulfurase activity of LmjCBS: the Km for cysteine was determined using 50 mM 2-mercaptoethanol with 0.1–20 mM cysteine and the Km for 2-mercaptoethanol was determined using 20 mM cysteine and 1–50 mM 2-mercaptoethanol. 5Serine sulfhydrylase activity of LmjCBS: the Km for serine was determined using 50 mM Na2S with 1–50 mM serine and the Km for Na2S was determined using 50 mM serine with 1–50 mM Na2S. 6Homocysteine sulfhydrylase activity of LmjCBS: the Km for cysteine was determined using 0.3 mM homocysteine with 1–5 mM cysteine and the Km for homocysteine was determined using 5 mM cysteine with 0.01–0.3 mM homocysteine. 7 LmjCBS: the Km for serine was determined using 50 mM homocysteine with 1–50 mM serine and the Km for homocysteine was determined using 50 mM serine with 1–50 mM homocysteine.

Enzyme Activity Substrates Km (mM) Kcat (s−1Kcat/Km (M−1·s−1
LmjCS O-acetylserine sulfhydrylase1 O-acetylserine 17.5±4.8 2047 1.2×105 
  Sulfide 0.13±0.04 2669 2.0×107 
 Cysteine desulfurase2 Cysteine 3.1±0.7 88.2 2.9×104 
  2-Mercaptoethanol 13.8±0.6 99.0 7.2×103 
LmjCBS O-acetylserine sulfhydrylase3 O-acetylserine 1.6±0.4 2269 1.4×106 
  Sodium sulfide 10.7±0.9 1516 1.4×105 
 Cysteine desulfurase4 Cysteine 0.8±0.5 38.3 4.8×104 
  2-Mercaptoethanol 4.4±1.9 20.7 4.7×103 
 Serine sulfhydrylase5 Serine 3.0±1.3 14.2 4.7×103 
  Sodium sulfide 6.6±4.2 14.1 2.1×103 
 Homocysteine sulfhydrylase6 Cysteine 7.1±2.0 11.1 1.6×103 
  Homocysteine 0.54±0.10 26.2 4.9×104 
 CBS7 Homocysteine 6.9±1.8 97.7 1.4×104 
  Serine 1.1±0.8 51.9 4.7×104 
Enzyme Activity Substrates Km (mM) Kcat (s−1Kcat/Km (M−1·s−1
LmjCS O-acetylserine sulfhydrylase1 O-acetylserine 17.5±4.8 2047 1.2×105 
  Sulfide 0.13±0.04 2669 2.0×107 
 Cysteine desulfurase2 Cysteine 3.1±0.7 88.2 2.9×104 
  2-Mercaptoethanol 13.8±0.6 99.0 7.2×103 
LmjCBS O-acetylserine sulfhydrylase3 O-acetylserine 1.6±0.4 2269 1.4×106 
  Sodium sulfide 10.7±0.9 1516 1.4×105 
 Cysteine desulfurase4 Cysteine 0.8±0.5 38.3 4.8×104 
  2-Mercaptoethanol 4.4±1.9 20.7 4.7×103 
 Serine sulfhydrylase5 Serine 3.0±1.3 14.2 4.7×103 
  Sodium sulfide 6.6±4.2 14.1 2.1×103 
 Homocysteine sulfhydrylase6 Cysteine 7.1±2.0 11.1 1.6×103 
  Homocysteine 0.54±0.10 26.2 4.9×104 
 CBS7 Homocysteine 6.9±1.8 97.7 1.4×104 
  Serine 1.1±0.8 51.9 4.7×104 

The source of the sulfide for the CS activity in Leishmania is unclear. Prokaryotes, plants and enteric protozoa assimilate inorganic sulfur into organic sulfur via the sulfate reduction pathway [6]. Our genome search suggests that this pathway is absent from L. major. Sulfide produced by the action of MST on 3-MP in a linked assay with rLmjCS gave a specific activity for CS of 1114±208 μmol/min per mg of protein, with the amount of cysteine produced roughly proportional to the amount of 3-MP added when between 0.5 and 3 mM (Figure 2).

Activity of enzymes potentially involved in the de novo biosynthesis of cysteine

Figure 2
Activity of enzymes potentially involved in the de novo biosynthesis of cysteine

Relationship between the cysteine synthase activity and the sulfide produced from the action of MST on different concentrations of 3-MP. The reaction mixture containing 3-MP, 2-mercaptoethanol (5 mM) and 0.5 μg rLmjMST was incubated for 5 min prior to addition of OAS (30 mM) and 0.5 μg LmjCS and further incubation for 5 min. The amount of cysteine produced was quantified as described in [15].

Figure 2
Activity of enzymes potentially involved in the de novo biosynthesis of cysteine

Relationship between the cysteine synthase activity and the sulfide produced from the action of MST on different concentrations of 3-MP. The reaction mixture containing 3-MP, 2-mercaptoethanol (5 mM) and 0.5 μg rLmjMST was incubated for 5 min prior to addition of OAS (30 mM) and 0.5 μg LmjCS and further incubation for 5 min. The amount of cysteine produced was quantified as described in [15].

LmjCBS can synthesize both cystathionine and cysteine

rLmjCBS (recombinant LmjCBS) was produced at high quantities (∼32 mg/l) in E. coli and the enzyme, when purified to apparent homogeneity (Supplementary Figure S2), showed multiple enzymic activities. It was optimally active at pH 8.0 and 37 °C for all of its activities and was stable as purified for several weeks at 4 °C without any appreciable loss of activity (results not shown). The CBS activity of rLmjCBS, utilizing homocysteine and serine to generate cystathionine (Figure 3, reaction I), had Kcat values of 97.7 s−1 and 51.9 s−1 for homocysteine and serine respectively (Table 2). rLmjCBS was also capable of the β-replacement reaction, ‘homocysteine sulfhydrylase’ (Figure 3, reaction II), in which cysteine and homocysteine form cystathionine and hydrogen sulfide. The homocysteine sulfhydrylase reaction was only apparent at concentrations of >0.1 mM cysteine, whereas homocysteine at concentrations greater than 0.3 mM were inhibitory (Supplementary Figure S3 at http://www.BiochemJ.org/bj/420/bj4200451add.htm). Consequently, the kinetic parameters of LmjCBS for this reaction were carried out at homocysteine concentrations (up to 0.3 mM) that were not inhibitory and obeyed Michealis–Menten kinetics (Supplementary Figure S3). The resultant apparent Kcat values were 11.1 s−1 and 26.2 s−1 for homocysteine and cysteine respectively (Table 2). rLmjCBS also showed remarkably high CS activity with OAS and sodium sulfide (Figure 3, reaction III), the apparent Kcat values being 2269 s−1 and 1516 s−1 respectively. However, the Km for sulfide was relatively high (10.7 mM). rLmjCBS also had serine sulfhydrylase activity (cysteine being formed from serine and sodium sulfide; Figure 3, reaction IV) with Kcat values for serine and sodium sulfide being 14.2 s−1 and 14.1 s−1 respectively (Table 2). LmjCBS was also capable of the ‘activated serine sulfydrase’ activity, otherwise known as cysteine desulfurase (which generates S-hydroxyethylcysteine and hydrogen sulfide from cysteine and 2-mercaptoethanol; Figure 3, reaction V), with Kcat values of 38.3 s−1 and 20.7 s−1 for cysteine and 2-mercaptoethanol respectively (Table 2). In the absence of 2-mercaptoethanol, there was a small activity, approx. 10% of that in the presence of β-mercaptoethanol; this is called reverse serine sulfydrase and forms serine and hydrogen sulfide from cysteine.

A general scheme of reactions catalysed by LmjCBS

Figure 3
A general scheme of reactions catalysed by LmjCBS

Substrates used by the enzyme to synthesize cystathionine and cysteine as well as degrade cysteine are given. Key to reactions: I, CBS; II, homocysteine desulfurase; III, CS; IV, serine sulfydrase; V, cysteine desulfurase.

Figure 3
A general scheme of reactions catalysed by LmjCBS

Substrates used by the enzyme to synthesize cystathionine and cysteine as well as degrade cysteine are given. Key to reactions: I, CBS; II, homocysteine desulfurase; III, CS; IV, serine sulfydrase; V, cysteine desulfurase.

LmjCS enhances the activity of LmjSAT in vitro

LmjSAT was generated as recombinant enzyme, rLmjSAT, in soluble form using an E. coli system, but expression was low (∼0.5 mg/l). Using standard assay conditions described for other SATs, we failed to detect SAT activity towards serine and acetyl-CoA with the purified rLmjSAT (the detection limit being 0.2 μmol/min per mg of protein), whereas recombinant SAT of Arabidopsis, used as a positive control, showed activity of 2.8 μmol/min per mg of protein. This finding for rLmjSAT was surprising and contrasted with that reported for TcSAT (T. cruzi SAT) [5], although LmjSAT has only 17% identity with TcSAT. We hypothesized that LmjSAT needs to be activated by LmjCS. When LmjSAT, tagged at the N-terminus with a 6×His-tag, was co-expressed with LmjCS (without a His-tag) in E. coli and the resultant LmjSAT was then affinity-purified (see below) and assayed, SAT activity was detected (specific activity of 90 nmol/min per mg of protein; a rate 4-fold higher than background under the experimental conditions used) at a rate comparable with that reported for TcSAT [5]. No kinetic analysis was possible due to low availability of rLmjSAT. These data suggested that LmjCS stabilizes and/or enhances LmjSAT activity, perhaps through formation of a complex.

LmjSAT interacts with LmjCS to form a complex in vitro

The formation of a SAT–CS complex has been described previously for plant proteins [14,37]. We investigated whether the Leishmania proteins do likewise by His-tagging rLmjSAT at its N-terminus (to give His-LmjSAT) and co-expressing in E. coli with LmjCS (devoid of any tag). Purification was then conducted using affinity chromatography targeting the His-tag. The expectation was that, if LmjCS interacts with His-LmjSAT, the complex should be retained on the column; elution could then be achieved with either OAS (10 mM), which would remove CS from the complex, or imidazole (500 mM), which would remove the whole complex by interfering with the His-binding. This procedure has been validated for plant enzymes [14,29,37]. SDS/PAGE analysis of the eluates in our experiments revealed that rLmjCS and rLmjSAT were indeed retained on the column and eluted by 500 mM imidazole (Figures 4A and 4Bi, lane 1) similarly to a positive control involving the Arabidopsis SAT–LmjCS complex (Figure 4Bii, lane 1). Initial elution of the column with wash buffer (50 mM Tris/HCl, pH 8.8, 300 mM NaCl, 80 mM imidazole) containing 10 mM OAS resulted in just rLmjCS being recovered in the eluate (Figure 4Bi, lane 3). rLmjSAT was subsequently eluted by washing the column with elution buffer containing 500 mM imidazole (Figure 4Bi, lane 4). rLmjSAT expressed alone in E. coli and purified under similar conditions is shown in Figure 4(Bi, lane 2).

The LmjSAT–LmjCS complex from L. major

Figure 4
The LmjSAT–LmjCS complex from L. major

(A) His-LmjSAT co-purified with untagged LmjCS and analysed by SDS/PAGE analysis including staining with Coomassie Blue. Lane 1 shows the proteins eluted with 1 M imidazole under denaturing conditions; they are consistent with there being a LmjSAT–LmjCS complex. (Bi) Western blot analysis using α-His and α-LmjCS on eluants from affinity chromatography of LmjSAT in complex with LmjCS and eluted with 500 mM imidazole (lane 1) or first treated with wash buffer containing 10 mM OAS (lane 3) and subsequently with 500 mM imidazole (lane 4). Lane 2 shows the eluant from lysates of E. coli expressing rLmjSAT alone. (Bii) SDS/PAGE analysis, stained with Coomassie Blue, of eluants from affinity chromatography of AtSAT in complex with LmjCS (lane 1), LmjCS(K222A) (lane 2) and LmjCS(H226A/K227A) (lane 3). (Ci). Sequence alignment of LmjCS (the present study), AtOASTL [9] and TvCS [11] showing positively charged residues involved in SAT binding marked with asterisks; they were subsequently replaced with an alanine by site-directed mutagenesis to generate the mutants LmjCS(K222A) and LmjCS(H226A/K227A). (Cii) Western blot analysis using α-LmjSAT and α-LmjCS on eluants from affinity chromatography of LmjSAT (shown in top panel) in complex with various CS proteins (bottom panel) and eluted with 1 M imidazole. α-LmjCS recognizes AtOASTL and TvCS. Key: LmjCS (lane 1), LmjCS(K222A) (lane 2), LmjCS(H226A/K227A) (lane 3), AtOASTL (lane 4), AtOASTL(K217A) (lane 5) and TvCS (lane 6). Molecular masses (in kDa) are shown on the left of all gels.

Figure 4
The LmjSAT–LmjCS complex from L. major

(A) His-LmjSAT co-purified with untagged LmjCS and analysed by SDS/PAGE analysis including staining with Coomassie Blue. Lane 1 shows the proteins eluted with 1 M imidazole under denaturing conditions; they are consistent with there being a LmjSAT–LmjCS complex. (Bi) Western blot analysis using α-His and α-LmjCS on eluants from affinity chromatography of LmjSAT in complex with LmjCS and eluted with 500 mM imidazole (lane 1) or first treated with wash buffer containing 10 mM OAS (lane 3) and subsequently with 500 mM imidazole (lane 4). Lane 2 shows the eluant from lysates of E. coli expressing rLmjSAT alone. (Bii) SDS/PAGE analysis, stained with Coomassie Blue, of eluants from affinity chromatography of AtSAT in complex with LmjCS (lane 1), LmjCS(K222A) (lane 2) and LmjCS(H226A/K227A) (lane 3). (Ci). Sequence alignment of LmjCS (the present study), AtOASTL [9] and TvCS [11] showing positively charged residues involved in SAT binding marked with asterisks; they were subsequently replaced with an alanine by site-directed mutagenesis to generate the mutants LmjCS(K222A) and LmjCS(H226A/K227A). (Cii) Western blot analysis using α-LmjSAT and α-LmjCS on eluants from affinity chromatography of LmjSAT (shown in top panel) in complex with various CS proteins (bottom panel) and eluted with 1 M imidazole. α-LmjCS recognizes AtOASTL and TvCS. Key: LmjCS (lane 1), LmjCS(K222A) (lane 2), LmjCS(H226A/K227A) (lane 3), AtOASTL (lane 4), AtOASTL(K217A) (lane 5) and TvCS (lane 6). Molecular masses (in kDa) are shown on the left of all gels.

With A. thaliana CS [designated AtOASTL (OAS thiol lyase) or AtCS], the positively charged residues Lys217, His221 and Lys222 located in loops β8–β9 govern its interaction with SAT, and removal of the residues through site-directed mutagenesis disrupts complex formation [27]. Alignment of LmjCS with the AtOASTL revealed that the residues are conserved in LmjCS (Lys222, His226 and Lys227) (Figure 4Ci). Thus we generated two mutants [designated LmjCS(K222A) and LmjCS(H226A/K227A)] and analysed them for interactions with LmjSAT. As predicted, LmjCS(K222A) and LmjCS(H226A/K227A) failed to interact with the native LmjSAT (Figure 4Cii, lanes 2 and 3), whereas the native LmjCS bound well (Figure 4Cii, lane 1). Furthermore, our interaction assay also showed that rLmjSAT can interact with CS from Arabidopsis (AtOASTL) to form an rLmjSAT/AtOASTL complex (Figure 4Cii, lane 4). As expected, an AtOASTL mutant designated AtOASTL(K217A) showed no interaction with rLmjSAT (Figure 4Cii, lane 5). The CS of T. vaginalis, TvCS, also did not interact with rLmjSAT (Figure 4Cii, lane 6), as expected as it is a type B CS and lacks the key residues (Figure 4Ci and [11]). We also showed that SAT from A. thaliana interacts with rLmjCS in a similar manner, although LmjCS(K222A) and LmjCS(H226A/K227A) showed a much reduced binding to the Arabidopsis SAT (Figure 4Bii).

Methionine and not cysteine is essential for the growth of L. major promastigotes

It has been reported that cysteine is an essential growth factor for T. brucei [4]. We now know that this protozoon encodes genes for only the RTS pathway for cysteine generation, which may be inadequate to satisfy its cysteine requirement. It seemed likely, however, that the existence of the two cysteine biosynthetic pathways in Leishmania would make it insensitive to the absence of an exogenous source of cysteine. This hypothesis was tested using L. major wild-type promastigotes cultured in a semi-defined medium depleted of serine, cysteine and methionine (SDM) and supplemented with 20% (v/v) dHiFCS. Cultures were initiated with promastigotes at 2.5×105 cells/ml and growth was monitored over 5 days. Promastigotes in the control experiments were maintained in normal medium supplemented with 10% (v/v) normal FCS. As expected, promastigotes maintained in normal medium grew well but growth was retarded in promastigotes maintained in SDM, although they remained motile and viable (Figure 5). The addition of exogenous cysteine or methionine at 300 μM had no significant effect on the growth of L. major promastigotes (Figure 5A), nor did addition of thiosulfate. Exogenous serine (at 300 μM), however, resulted in enhanced growth (Figure 5A). These results are consistent with the absence of genes encoding proteins involved in the de novo synthesis of serine in this protozoon [6] and thus with exogenous serine being essential [38]. Thus serine was added to SDM (SSDM) and a reanalysis of the importance of the other additions was carried out (Figure 5B). Of all the substances added, only methionine restored promastigote growth to that of the control cells maintained in normal medium (Figure 5B). Addition of cysteine failed to significantly stimulate growth of promastigotes maintained in SSDM (P>0.5). Higher concentrations of cysteine (>300 μM) retarded promastigote growth; indeed death ensued within 24 h at concentrations of 10 mM (results not shown). In total, these results suggest that L. major differs from T. brucei in that methionine and not cysteine is markedly beneficial for growth.

Cell density and thiol levels of L. major promastigotes in sulfur depleted media

Figure 5
Cell density and thiol levels of L. major promastigotes in sulfur depleted media

(A) L. major promastigotes at 2.5×105 cells/ml were incubated at 27 °C for 5 days in SDM (lacking an exogenous sulfur source) and supplemented with cysteine (300 μM), methionine (300 μM), thiosulfate (150 μM) or serine (300 μM). The cell densities were determined initially (represented by grey dashed lines) and at day 5 (black bars). Results are means±S.E.M. from 3 independent experiments. (B) Experiments were carried out as detailed in (A) but all treatments were supplemented with serine at 300 μM to give SSDM. Additional supplements were OAS (250 μM) or 3-MP (180 μM). Results are means±S.E.M. from 2 independent experiments. (Ci) Total thiol content of L. major grown under different conditions: RPMI alone (1), RPMI supplemented with 300 μM methionine (2) or cysteine (3); HOMEM medium (4). Results are means±S.D. from 3 extracts. (Cii) Intracellular levels of cysteine, trypanathione and glutathione of L. major grown under conditions as in (Ci). Results are means±S.D. from 3 extracts. (Di) Western blot analysis of lysates of promastigotes from (B). SSDM (lane 1), HOMEM nutrient-rich medium (lane 2), SSDM supplemented with cysteine (lane 3) and SSDM supplemented with methionine (lane 4). Molecular masses (in kDa) are shown on the left. (ii) Densitometric analysis of signals from LmjCBS compared with TDR1 (top panel) and LmjCS compared with TDRI (bottom panel) using ImageJ software (http://rsb.info.nih.gov/ij/index.html).

Figure 5
Cell density and thiol levels of L. major promastigotes in sulfur depleted media

(A) L. major promastigotes at 2.5×105 cells/ml were incubated at 27 °C for 5 days in SDM (lacking an exogenous sulfur source) and supplemented with cysteine (300 μM), methionine (300 μM), thiosulfate (150 μM) or serine (300 μM). The cell densities were determined initially (represented by grey dashed lines) and at day 5 (black bars). Results are means±S.E.M. from 3 independent experiments. (B) Experiments were carried out as detailed in (A) but all treatments were supplemented with serine at 300 μM to give SSDM. Additional supplements were OAS (250 μM) or 3-MP (180 μM). Results are means±S.E.M. from 2 independent experiments. (Ci) Total thiol content of L. major grown under different conditions: RPMI alone (1), RPMI supplemented with 300 μM methionine (2) or cysteine (3); HOMEM medium (4). Results are means±S.D. from 3 extracts. (Cii) Intracellular levels of cysteine, trypanathione and glutathione of L. major grown under conditions as in (Ci). Results are means±S.D. from 3 extracts. (Di) Western blot analysis of lysates of promastigotes from (B). SSDM (lane 1), HOMEM nutrient-rich medium (lane 2), SSDM supplemented with cysteine (lane 3) and SSDM supplemented with methionine (lane 4). Molecular masses (in kDa) are shown on the left. (ii) Densitometric analysis of signals from LmjCBS compared with TDR1 (top panel) and LmjCS compared with TDRI (bottom panel) using ImageJ software (http://rsb.info.nih.gov/ij/index.html).

Next, we investigated the effect of the exogenous sources of organic sulfur on the thiol levels in L. major promastigotes cultured in RPMI (lacking cysteine and methionine) supplemented with cysteine or methionine using HPLC (Figure 5C). Growth of Leishmania promastigotes in RPMI and SSDM was equivalent under all conditions tested (results not shown). Promastigotes cultured in normal medium were used as controls (Figure 5C, lane 4). Compared with the control promastigotes, the total thiol levels were reduced 3.8-fold (P<0.005) in promastigotes cultured in RPMI and to 59% (P<0.05) in RPMI supplemented with cysteine. In contrast, the total thiol levels in promastigotes cultured in RPMI supplemented with methionine were slightly higher (1.4-fold, P<0.05) than those in the control promastigotes in normal medium (Figure 5Ci). The decreased thiol levels in L. major promastigotes cultured in RPMI or RPMI supplemented with cysteine correlated well with the reduced growth observed (Figure 5B). Analysis of the levels of cysteine, glutathione and trypanothione revealed detailed differences (Figure 5Cii). The thiol concentrations in control cells are similar to those previously reported [39]. Compared with promastigotes grown in normal medium, cells grown in RPMI showed reduced levels of all thiols. Cells grown in RPMI with methionine had similar levels of cysteine and glutathione to the control but trypanothione levels were increased 2.4-fold (P<0.005). Cells grown in RPMI supplemented with cysteine showed increased levels of glutathione (2.3-fold, P<0.05) and trypanothione (2.8-fold, P<0.05) compared with promastigotes cultured in just RPMI, which apparently accounted for the increase in total thiol content (2.2-fold higher, P<0.05). The levels of glutathione and trypanothione in RPMI supplemented with cysteine were relatively similar to the levels in the control cells grown in normal medium. Importantly, however, addition of exogenous cysteine did not result in an increase in the intracellular level of cysteine, which was only 12% of the level seen in the control cells.

Promastigotes maintained in SSDM had increased levels of LmjCS, as judged by Western blot analysis (Figure 5Di, lane 1). This increase was validated by densitometric analysis using ImageJ software (Figure 5Dii). LmjCS and LmjCBS in promastigotes growing in SSDM with cysteine (Figure 5C, lane 3) or methionine (Figure 5C, lane 4) were relatively similar to the level in wild-type cells in normal nutrient-rich medium (Figure 5C, lane 2). These data suggest that enzymes from cysteine biosynthetic pathways are present and presumably functional in promastigotes, and that the level of CS can be adjusted in attempts to accommodate variations in the availability of exogenous amino acids.

Transport of [35S]-L-cysteine into L. major promastigotes

Having established that an exogenous source of cysteine does not stimulate growth of L. major promastigotes and has little effect on the intracellular cysteine levels, whereas exogenous methionine does, we examined the ability of promastigotes to utilize an exogenouse source of cysteine. Leishmania is known to take up exogenous methionine [40]. A transport assay with [35S]-L-cysteine showed that L. major promastigotes can take up cysteine and that the transport was linear for at least 5 min (Figure 6A, insert) but declined markedly by 25 min (Figure 6A). [35S]-L-cysteine transport measured as a function of different concentrations of extracellular cysteine showed a typical Michaelis–Menten hyperbolic curve and displayed an apparent Km of 229±39 μM (n=3) and a Vmax of 16.9±1.3 pmol/min per (107 cells) (Figure 6B). The transport of cysteine was investigated using a range of non-radioactive cysteine, which showed that, as expected, cysteine inhibited [35S]-L-cysteine transport into promastigotes (Supplementary Figure S4 at http://www.BiochemJ.org/bj/420/bj4200451add.htm). These results show that cysteine can be transported into L. major promastigotes, but only at a low rate.

Cysteine transport in L. major promastigotes

Figure 6
Cysteine transport in L. major promastigotes

(A) Time dependence of cysteine incorporation using 50 nM [35S]-L-cysteine and 25 μM non-radioactive cysteine with 2×107 promastigotes. Cysteine incorporation was linear for 5 min (insert). (B) Transport of cysteine during 5 min as a function of cysteine concentration. The experiment involved 2×107 promastigotes and various concentrations of cysteine, in all cases provided from a stock solution of 50 nM [35S]-L-cysteine and 500 μM non-radioactive cysteine. The reciprocal plot is shown in the inset.

Figure 6
Cysteine transport in L. major promastigotes

(A) Time dependence of cysteine incorporation using 50 nM [35S]-L-cysteine and 25 μM non-radioactive cysteine with 2×107 promastigotes. Cysteine incorporation was linear for 5 min (insert). (B) Transport of cysteine during 5 min as a function of cysteine concentration. The experiment involved 2×107 promastigotes and various concentrations of cysteine, in all cases provided from a stock solution of 50 nM [35S]-L-cysteine and 500 μM non-radioactive cysteine. The reciprocal plot is shown in the inset.

DISCUSSION

This study has shown that L. major possesses enzymes of each of the two pathways for cysteine synthesis – the de novo biosynthesis pathway comprising SAT and CS, and the RTS pathway comprising CBS and CGL. Biochemical analysis of the recombinant CS showed that the enzyme has high activity with OAS and sulfide as substrates (Table 2), but is unable to utilize thiosulfate and phosphoserine. This substrate preference suggests that the Leishmania enzyme is similar to type A CSs of bacteria and has a β8–β9 surface loop that restricts access to the active site pocket. The protein structure predicted from the gene sequence is in agreement with this (Supplementary Figure S1B). The type B CSs of bacteria [25] and the CSs of Aeropyrum pernix [26] and T. vaginalis [11] all have a different structure in this region, centred around a conserved positively charged residue that confers greater flexibility and allows the use of larger, negatively charged substrates (Supplementary Figure S1B).

It has been shown that type A CSs of bacteria and Arabidopsis CS form bi-enzyme complexes with SAT, that are dissociated by OAS in the absence of sulfide [14,29,41]. The C-terminal residues of SAT bind to sequence motifs in the β8–β9 surface loop and the substrate-binding loop in the active-site cleft; this completely fills the active site of CS and results in inhibition of its activity [27,29,42]. The substrate-binding loop is conserved in all CS enzymes, whereas the SAT-binding motif in the β8–β9 loop is not found in the type B CSs of bacteria, which do not form a complex with SAT. Consistent with this, the CS enzymes of organisms that do not contain genes for SAT, notably A. pernix [42] and T. vaginalis [11], also lack this binding motif in the β8–β9 loop. The gene sequences of the CSs of Leishmania show that the proteins do contain this SAT-binding motif and the consequent prediction that the enzyme will form a complex with SAT was confirmed in this study (Figure 4). Mutagenesis of LmjCS also indicated that residues equivalent to those of AtOASTL are involved in the binding to SAT (Figure 4).

The function of the CS-SAT bi-enzyme complex in Leishmania remains to be proven, but based on the findings for bacteria and plants [14,17,37,44] it is likely to have a regulatory role. The two enzymes show opposite responses to being part of the complex: SAT is inactivated or less active when released from the complex, whereas CS is completely inactive when in complex with SAT but highly active as a free enzyme [14,29,45]. The data for Leishmania SAT is consistent with this, as SAT activity could only be detected when co-expressed with CS, but how this plays a part in cysteine homoeostasis requires additional study.

The source of sulfide for cysteine synthesis in L. major is unclear. Plants, bacteria and Entamoeba assimilate inorganic sulfur via the sulfide reduction pathway, but genes encoding enzymes of this pathway appear to be absent from the L. major genome. We postulated that the sulfide required by CS may be provided by the action of MST on 3-MP. This enzyme has previously been implicated in sulfide production and antioxidant defence [12] and we have now shown that LmjCS can indeed use sulfide emanating from the activity of MST to form cysteine at a rate similar to that when sulfide itself is provided (Figure 2). Such activity in vitro, however, does not mean that this is the In vivo source; this suggestion needs to be validated by genetic studies. An alternative is that sulfide could potentially be provided exogenously for the parasite in the parasitophorous vacuole, and/or the insect gut, although evidence that this is so is currently lacking.

This study has also established that L. major CBS is clearly different from mammalian CBSs; it lacks the N-terminal haem-binding motif and the C-terminal regulatory domain present in mammalian CBSs, moreover it is capable of using a wide variety of substrates (Table 2). LmjCBS does, unsurprisingly, possess CBS activity to produce cystathionine (Figure 3; Table 1), but can also form cystathionine and sulfide with homocysteine and cysteine via β-replacement reaction, and was capable of producing sulfide via the cysteine desulfurase reaction by hydrolysing cysteine (Figure 3 and Table 1). LmjCBS, however, can also form cysteine from OAS and sulfide. The Kcat/Km of LmjCBS towards OAS is ∼7-fold higher than that of LmjCS, however its Kcat/Km for sulfide is some 140-fold lower (Table 2). This suggests that CS and CBS are adapted to different physiological conditions, in which the relative concentrations of the two substrates are changed. Unfortunately, there is little or no information on the concentration of the two substrates in the various developmental stages of the parasite. However, there is some evidence that amastigotes have increased capacities for β-oxidation, the tricarboxylic acid cycle, mitochondrial electron transport and oxidative phosphorylation [2,46]. This could result in increased availability of acetyl-CoA for OAS synthesis and a decreased availability of sulfide due to increased utilization for the synthesis of iron sulfur clusters and lipoic acid. Under these circumstances, one would predict that CS rather than CBS would be functional in de novo cysteine biosynthesis. However, these speculative suggestions need to be rigorously tested experimentally.

Other CBSs also have CS activity. Human CBS, previously thought to lack CS activity [47], has recently been reported to have such activity but with a very low catalytic efficiency of Kcat/Km=135 s−1·M−1 [48]. The difference between HsCBS and CBS from lower eukaryotes (which have significant CS activity, detailed below) is thought to be due to the active-site pocket of HsCBS being constricted by the N-terminal haem-binding and C-terminal regulatory domains (Supplementary Figure S1C), such that it cannot accommodate the large acetyl group of OAS at its serine-binding site, whereas CBSs lacking one or other of the domains can [48]. The CBSs from T. cruzi (TcCBS) and S. cerevisiae (ScCBS) have significantly greater CS activities than HsCBS, with the Kcat values for OAS being 12.5 s−1 and 19.2 s−1 respectively, and Km values of 4.9 mM and 1.3 mM for OAS and 4.1 mM and 16.6 mM for sodium sulfide respectively [5,8]. These Kcat values are similar to those of the CBS activities of the same enzymes [43]. This is a very different situation from the L. major CBS, which has a CS activity 23-fold higher than its CBS activity (Table 2). We suggest that the explanation for the much higher CS activity of LmjCBS is that it lacks both the N- and C-terminal extensions that in the human enzyme constrict entry to the active-site cleft, whereas both ScCBS and TcCBS have one of these extensions (Supplementary Figure S1C). Interestingly, only three of the eight T. cruzi CBS genes have an N-terminal extension (http://www.geneDB.org), and the isoform with the longest extension is the one characterized [5]. The other T. cruzi CBS isoforms encoded in the genome may have more similar biochemical properties to the L. major CBS.

The functional significance of the multiple reactions carried out by L. major CBS is unknown but the kinetic parameters of this protein do not rule out that these reactions potentially occur within the cell and may confer selective advantage and flexibility to the parasite within its hosts. The possibility that CBS functionally acts In vivo as a CS cannot yet be ruled out. Indeed, the finding that the T. cruzi CBS gene can overcome a growth defect of both S. cerevisiae (WB63yCBSΔ lines; deficient in CBS) and E. coli [NK3 lines; a double mutant deficient in the bacterial type A (cysK) and type B (cysM) CSs] suggests that this trypanosomatid CBS can functionally act as a CBS and a CS In vivo [5]. Thus L. major CBS has both CS and CBS activities, and this clearly distinguishes it from LmjCS in that the latter has no detectable CBS activity.

Western blot analyses showed that CS and CBS are both present in promastigotes of L. major (Figure 5), results confirmed for other Leishmania species by the global proteomic analyses reported to date, with there being some evidence of protein levels changing between different developmental stages [46]. In an attempt to unravel which pathways operate in promastigotes, we carried out the experiments monitoring parasite growth when different sulfur amino acids are available exogenously (Figure 5). These experiments showed that the parasite does not benefit significantly from an exogenous source of cysteine, in contrast with that reported for T. brucei [4]. Interestingly, however, methionine promoted growth (Figure 5), which correlates with a previous report that is is transported into promastigotes and augments the amino acid pool; importantly it is used predominatly for cystathionine biosynthesis [49]. Our results on intracellular thiol levels confirm that addition of exogenous methionine results in thiol levels, including cysteine, relatively similar to those in normal promastigotes grown in full medium, presumably from conversion of methionine via the trans-sulfuration pathway, whereas addition of exogenous cysteine had little effect on intracellular cysteine levels, although glutathione and trypanothione levels were relatively similar to those in cells grown in normal medium (Figure 5Ci). These results suggested that, unlike methionine, cysteine may not be transported at a sufficient rate to maintain the intracellular pool of cysteine at the level required for normal growth.

Our transport data (Figure 6) confirm this, for although transport occurred, its rate was some 200-fold lower than cysteine transport into T. brucei [50] and the Km for cysteine is some 10-fold higher than that for transport of methionine into Leishmania [40]. The low Vmax for transport of cysteine coupled with the Km of >200 μM, when the concentration of cysteine in the medium is approx. 70 μM, is likely to mean that this is not a mechanism used by the parasite for obtaining cysteine. The importance of the cysteine de novo biosynthetic pathway was reiterated by the observation that promastigotes in medium lacking an exogenous organic source of sulfur had an up-regulated level of CS (Figure 5D), presumably in an attempt to counter the lack of intracellular cysteine. Thus our results show that Leishmania differs considerably not only from T. brucei, which relies upon exogenous cysteine [4,50], but also T. cruzi, which transports cysteine via a highly specific transporter with a relatively low Km (49 μM) to maintain its intracellular cysteine pool and iron–sulfur formation [51]. Our growth data (Figure 5A) also confirmed that serine was an essential amino acid for Leishmania that is salvaged by promastigotes and used for growth [38]. This is consistent with the absence from its genome of genes involved in serine biosynthesis [52]. Together these data are consistent with both the CS and CBS pathways operating in promastigotes and that methionine can be converted to cysteine, thus suggesting that CBS certainly functions in cystathionine synthesis In vivo.

A key question is why Leishmania has two pathways for generating cysteine when many cells cope very well with just a single source. One likely explanation is that the availability of exogenous nutrients differs considerably between the parasite's environments in its sandfly and mammalian hosts and that the potential for two functional synthetic routes in the parasite reflects this. For instance, maybe de novo synthesis via CS occurs in one stage and synthesis from methionine occurs in another. The generation of genetic mutants lacking key genes could be an informative approach to answer these questions. A second possibility is that the two routes occur in different sub-cellular compartments within Leishmania. However, neither CS nor SAT contains clear targeting signals for location to organelles and preliminary evidence suggests both are cytosolic. It would be interesting to investigate whether intentionally targeting them to an organelle would have any impact upon cysteine homoeostasis in the parasite.

As CS is absent from humans, and L. major CBS is divergent from the mammalian homologue, both proteins potentially could represent good drug targets. The discovery that the ten C-terminal residues of SAT are inhibitory to CS [53] could form a framework for which specific inhibitors of CS can be designed, whereas the ability of L. major CBS to use more bulky substrates than can the human enzyme suggests a way to specifically target the parasite CBS.

We thank Professor Rudiger Hell, Heidelberg Institute of Plant Sciences, University of Heidelberg, Germany, for the pET3dAtOASTL and pET28a+SAT plasmids used in the present study. We thank Professor Sylke Müller (University of Glasgow) for assistance with the thiol analytical methodology and very helpful comments on the manuscript, Professor Mike Barrett and Ms Isabel Vincent (University of Glasgow) and Ms Kirstyn Brunker (University of Strathclyde) for their help with the cysteine transport assays and HPLC analyses respectively.

Abbreviations

     
  • AtOASTL

    Arabidopsis thaliana O-acetylserine (thiol) lyase

  •  
  • CBL

    cystathionine β-lyase

  •  
  • CBS

    cystathionine β-synthase

  •  
  • CBSS

    Carter basal salt solution

  •  
  • CGL

    cystathionine γ-lyase

  •  
  • CGS

    cystathionine γ-synthase

  •  
  • CS

    cysteine synthase

  •  
  • dHiFCS

    dialysed heat-inactivated FCS

  •  
  • DTNB

    5,5′-dithiobis-(2-nitrobenzoic acid)

  •  
  • FCS

    fetal calf serum

  •  
  • HsCBS

    Homo sapiens CBS

  •  
  • LmjCS

    Leishmania major CS

  •  
  • LmjCBS

    L. major CBS

  •  
  • LmCGL

    L. major CGL

  •  
  • LmjSAT

    L. major SAT

  •  
  • 3-MP

    3-mercaptopyruvate

  •  
  • MST

    mercaptopyruvate sulfurtransferase

  •  
  • OAS

    O-acetylserine

  •  
  • ORF

    open reading frame

  •  
  • PAG

    propargylglycine

  •  
  • PLP

    pyridoxal phosphate

  •  
  • rLmjCBS

    recombinant LmjCBS

  •  
  • rLmjCS

    recombinant LmjCS

  •  
  • RTS

    reverse trans-sulfuration

  •  
  • SAT

    serine acetyltransferase

  •  
  • ScCBS

    Saccharomyces cerevisiae CBS

  •  
  • SDM

    sulfur-depleted medium

  •  
  • SSDM

    serine-supplemented SDM

  •  
  • TcCBS

    Trypanosoma cruzi CBS

  •  
  • TcSAT

    T. cruzi SAT

  •  
  • TDR1

    thiol-dependent reductase 1

  •  
  • TvCS

    Trypanosoma vaginalis CS

  •  
  • TS

    trans-sulfuration

AUTHOR CONTRIBUTION

Roderick Williams carried out the genome mining and gene analyses, many of the enzyme characterization components, and the growth and transport studies. Gareth Westrop carried out the protein–protein interaction studies and thiol analyses. Graham Coombs conceived and co-ordinated the project. All participated in data interpretation and writing of the manuscript.

FUNDING

This study was partially funded by the Medical Research Council [grant number G0700127].

References

References
1
Croft
S. L.
Yardley
V.
Chemotherapy of leishmaniasis
Curr. Pharm. Des.
2002
, vol. 
8
 (pg. 
319
-
342
)
2
Opperdoes
F. R.
Coombs
G. H.
Metabolism of Leishmania: proven and predicted
Trends Parasitol.
2007
, vol. 
23
 (pg. 
49
-
158
)
3
Krauth-Siegel
R. L.
Comini
M. A.
Redox control in trypanosomatids, parasitic protozoa with trypanothione-based thiol metabolism
Biochim. Biophys. Acta
2008
, vol. 
1780
 (pg. 
1236
-
1248
)
4
Duszenko
M.
Muhlstadt
K.
Broder
A.
Cysteine is an essential growth factor for Trypanosoma brucei bloodstream forms
Mol. Biochem. Parasitol.
1992
, vol. 
50
 (pg. 
269
-
273
)
5
Nozaki
T.
Shigeta
Y.
Saito-Nakano
Y.
Imada
M.
Kruger
W. D.
Characterization of transsulfuration and cysteine biosynthetic pathways in the protozoan hemoflagellate, Trypanosoma cruzi. Isolation and molecular characterization of cystathionine beta-synthase and serine acetyltransferase from Trypanosoma
J. Biol. Chem.
2001
, vol. 
276
 (pg. 
6516
-
6523
)
6
Ali
V.
Nozaki
T.
Current therapeutics, their problems, and sulfur-containing-amino-acid metabolism as a novel target against infections by “amitochondriate” protozoan parasites
Clin. Microbiol. Rev.
2007
, vol. 
20
 (pg. 
164
-
187
)
7
Muller
S.
Liebau
E.
Walter
R. D.
Krauth-Siegel
R. L.
Thiol-based redox metabolism of protozoan parasites
Trends Parasitol.
2003
, vol. 
19
 (pg. 
320
-
328
)
8
Ono
B.
Kijima
K.
Inoue
T.
Miyoshi
S.
Matsuda
A.
Shinoda
S.
Purification and properties of Saccharomyces cerevisiae cystathionine β-synthase
Yeast
1994
, vol. 
10
 (pg. 
333
-
339
)
9
Hell
R.
Jost
R.
Berkowitz
V.
Wirtz
M.
Molecular and biochemical analysis of the enzymes of cysteine biosynthesis in the plant Arabidopsis thaliana
Amino Acids
2002
, vol. 
22
 (pg. 
245
-
257
)
10
Chattopadhyay
A.
Meier
M.
Ivaninskii
S.
Burkhard
P.
Speroni
F.
Campanini
B.
Bettati
S.
Mozzarelli
A.
Rabeh
W. M.
Li
L.
Cook
P. F.
Structure, mechanism, and conformational dynamics of O-acetylserine sulfhydrylase from Salmonella typhimurium: comparison of A and B isozymes
Biochemistry
2007
, vol. 
46
 (pg. 
8315
-
8330
)
11
Westrop
G. D.
Goodall
G.
Mottram
J. C.
Coombs
G. H.
Cysteine biosynthesis in Trichomonas vaginalis involves cysteine synthase utilizing O-phosphoserine
J. Biol. Chem.
2006
, vol. 
281
 (pg. 
25062
-
25075
)
12
Williams
R. A.
Kelly
S. M.
Mottram
J. C.
Coombs
G. H.
3-Mercaptopyruvate sulfurtransferase of Leishmania contains an unusual C-terminal extension and is involved in thioredoxin and antioxidant metabolism
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
1480
-
1486
)
13
Medina-Acosta
E.
Cross
G. A.
Rapid isolation of DNA from trypanosomatid protozoa using a simple ‘mini-prep’ procedure
Mol. Biochem. Parasitol.
1993
, vol. 
59
 (pg. 
327
-
329
)
14
Wirtz
M.
Berkowitz
O.
Droux
M.
Hell
R.
The cysteine synthase complex from plants. Mitochondrial serine acetyltransferase from Arabidopsis thaliana carries a bifunctional domain for catalysis and protein-protein interaction
Eur. J. Biochem.
2001
, vol. 
268
 (pg. 
686
-
693
)
15
Kredich
N. M.
Tomkins
G. M.
The enzymic synthesis of L-cysteine in Escherichia coli and Salmonella typhimurium
J. Biol. Chem.
1966
, vol. 
241
 (pg. 
4955
-
4965
)
16
Flint
D. H.
Tuminello
J. F.
Miller
T. J.
Studies on the synthesis of the Fe-S cluster of dihydroxy-acid dehydratase in Escherichia coli crude extract. Isolation of O-acetylserine sulfhydrylases A and B and β-cystathionase based on their ability to mobilize sulfur from cysteine and to participate in Fe-S cluster synthesis
J. Biol. Chem.
1996
, vol. 
271
 (pg. 
16053
-
16067
)
17
Mino
K.
Imamura
K.
Sakiyama
T.
Eisaki
N.
Matsuyama
A.
Nakanishi
K.
Increase in the stability of serine acetyltransferase from Escherichia coli against cold inactivation and proteolysis by forming a bienzyme complex
Biosci. Biotechnol. Biochem.
2001
, vol. 
65
 (pg. 
865
-
874
)
18
Denton
H.
McGregor
J. C.
Coombs
G. H.
Reduction of anti-leishmanial pentavalent antimonial drugs by a parasite-specific thiol-dependent reductase, TDR1
Biochem. J.
2004
, vol. 
381
 (pg. 
405
-
412
)
19
Al Salabi
M. I.
de Koning
H. P.
Purine nucleobase transport in amastigotes of Leishmania mexicana: involvement in allopurinol uptake
Antimicrob. Agents Chemother.
2005
, vol. 
49
 (pg. 
3682
-
3689
)
20
Hindson
V. J.
Moody
P. C.
Rowe
A. J.
Shaw
W. V.
Serine acetyltransferase from Escherichia coli is a dimer of trimers
J. Biol. Chem.
2000
, vol. 
275
 (pg. 
461
-
466
)
21
Zhu
X.
Yamaguchi
T.
Masada
M.
Complexes of serine acetyltransferase and isozymes of cysteine synthase in spinach leaves
Biosci. Biotechnol. Biochem.
1998
, vol. 
62
 (pg. 
947
-
952
)
22
Ruffet
M. L.
Lebrun
M.
Droux
M.
Douce
R.
Subcellular distribution of serine acetyltransferase from Pisum sativum and characterization of an Arabidopsis thaliana putative cytosolic isoform
Eur. J. Biochem.
1995
, vol. 
227
 (pg. 
500
-
509
)
23
Vaara
M.
Eight bacterial proteins, including UDP-N-acetylglucosamine acyltransferase (LpxA) and three other transferases of Escherichia coli, consist of a six-residue periodicity theme
FEMS Microbiol. Lett.
1992
, vol. 
76
 (pg. 
249
-
254
)
24
Byrne
C. R.
Monroe
R. S.
Ward
K. A.
Kredich
N. M.
DNA sequences of the cysK regions of Salmonella typhimurium and Escherichia coli and linkage of the cysK regions to ptsH
J. Bacteriol.
1988
, vol. 
170
 (pg. 
3150
-
3157
)
25
Claus
M. T.
Zocher
G. E.
Maier
T. H.
Schulz
G. E.
Structure of the O-acetylserine sulfhydrylase isoenzyme CysM from Escherichia coli
Biochemistry
2005
, vol. 
44
 (pg. 
8620
-
8626
)
26
Oda
Y.
Mino
K.
Ishikawa
K.
Ataka
M.
Three-dimensional structure of a new enzyme, O-phosphoserine sulfhydrylase, involved in l-cysteine biosynthesis by a hyperthermophilic archaeon, Aeropyrum pernix K1, at 2.0Å resolution
J. Mol. Biol.
2005
, vol. 
351
 (pg. 
334
-
344
)
27
Bonner
E. R.
Cahoon
R. E.
Knapke
S. M.
Jez
J. M.
Molecular basis of cysteine biosynthesis in plants: structural and functional analysis of O-acetylserine sulfhydrylase from Arabidopsis thaliana
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
38803
-
38813
)
28
Liszewska
F.
Lewandowska
M.
Plochocka
D.
Sirko
A.
Mutational analysis of O-acetylserine (thiol) lyase conducted in yeast two-hybrid system
Biochim. Biophys. Acta
2007
, vol. 
1774
 (pg. 
450
-
455
)
29
Droux
M.
Ruffet
M. L.
Douce
R.
Job
D.
Interactions between serine acetyltransferase and O-acetylserine (thiol) lyase in higher plants – structural and kinetic properties of the free and bound enzymes
Eur. J. Biochem.
1998
, vol. 
255
 (pg. 
235
-
245
)
30
Alexander
F. W.
Sandmeier
E.
Mehta
P. K.
Christen
P.
Evolutionary relationships among pyridoxal-5′-phosphate-dependent enzymes. Regio-specific α, β and γ families
Eur. J. Biochem.
1994
, vol. 
219
 (pg. 
953
-
960
)
31
Bateman
A.
The structure of a domain common to archaebacteria and the homocystinuria disease protein
Trends Biochem. Sci.
1997
, vol. 
22
 (pg. 
12
-
13
)
32
Banerjee
R.
Zou
C. G.
Redox regulation and reaction mechanism of human cystathionine-β-synthase: a PLP-dependent hemesensor protein
Arch. Biochem. Biophys.
2005
, vol. 
433
 (pg. 
144
-
156
)
33
Jhee
K. H.
McPhie
P.
Miles
E. W.
Domain architecture of the heme-independent yeast cystathionine β-synthase provides insights into mechanisms of catalysis and regulation
Biochemistry
2000
, vol. 
39
 (pg. 
10548
-
10556
)
34
Maclean
K. N.
Janosik
M.
Oliveriusova
J.
Kery
V.
Kraus
J. P.
Transsulfuration in Saccharomyces cerevisiae is not dependent on heme: purification and characterization of recombinant yeast cystathionine β-synthase
J. Inorg. Biochem.
2000
, vol. 
81
 (pg. 
161
-
171
)
35
Meier
M.
Janosik
M.
Kery
V.
Kraus
J. P.
Burkhard
P.
Structure of human cystathionine β-synthase: a unique pyridoxal 5′-phosphate-dependent heme protein
EMBO J.
2001
, vol. 
20
 (pg. 
3910
-
3916
)
36
Stipanuk
M. H.
Sulfur amino acid metabolism: pathways for production and removal of homocysteine and cysteine
Annu. Rev. Nutr.
2004
, vol. 
24
 (pg. 
539
-
577
)
37
Wirtz
M.
Hell
R.
Functional analysis of the cysteine synthase protein complex from plants: structural, biochemical and regulatory properties
J. Plant Physiol.
2006
, vol. 
163
 (pg. 
273
-
286
)
38
Scott
D. A.
Hickerson
S. M.
Vickers
T. J.
Beverley
S. M.
The role of the mitochondrial glycine cleavage complex in the metabolism and virulence of the protozoan parasite Leishmania major
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
155
-
165
)
39
Ariyanayagam
M. R.
Fairlamb
A. H.
Ovothiol and trypanothione as antioxidants in trypanosomatids
Mol. Biochem. Parasitol.
2001
, vol. 
115
 (pg. 
189
-
198
)
40
Mukkada
A. J.
Simon
M. W.
Leishmania tropica: uptake of methionine by promastigotes
Exp. Parasitol.
1977
, vol. 
42
 (pg. 
87
-
96
)
41
Feldman-Salit
A.
Wirtz
M.
Hell
R.
Wade
R. C.
A mechanistic model of the cysteine synthase complex
J. Mol. Biol.
2008
, vol. 
386
 (pg. 
37
-
59
)
42
Huang
B.
Vetting
M. W.
Roderick
S. L.
The active site of O-acetylserine sulfhydrylase is the anchor point for bienzyme complex formation with serine acetyltransferase
J. Bacteriol.
2005
, vol. 
187
 (pg. 
3201
-
3205
)
43
Mino
K.
Ishikawa
K.
Characterization of a novel thermostable O-acetylserine sulfhydrylase from Aeropyrum pernix K1
J. Bacteriol.
2003
, vol. 
185
 (pg. 
2277
-
2284
)
44
Bogdanova
N.
Hell
R.
Cysteine synthesis in plants: protein-protein interactions of serine acetyltransferase from Arabidopsis thaliana
Plant J.
1997
, vol. 
11
 (pg. 
251
-
262
)
45
Mino
K.
Yamanoue
T.
Sakiyama
T.
Eisaki
N.
Matsuyama
A.
Nakanishi
K.
Effects of bienzyme complex formation of cysteine synthetase from Escherichia coli on some properties and kinetics
Biosci. Biotechnol. Biochem.
2000
, vol. 
64
 (pg. 
1628
-
1640
)
46
Rosenzweig
D.
Smith
D.
Opperdoes
F.
Stern
S.
Olafson
R. W.
Zilberstein
D.
Retooling Leishmania metabolism: from sand fly gut to human macrophage
FASEB J.
2008
, vol. 
22
 (pg. 
590
-
602
)
47
Braunstein
A. E.
Goryachenkova
E. V.
Tolosa
E. A.
Willhardt
I. H.
Yefremova
L. L.
Specificity and some other properties of liver serine sulphhydrase: evidence for its identity with cystathionine-synthase
Biochim. Biophys. Acta
1971
, vol. 
242
 (pg. 
247
-
260
)
48
Frank
N.
Kent
J. O.
Meier
M.
Kraus
J. P.
Purification and characterization of the wild type and truncated human cystathionine β-synthase enzymes expressed in E. coli
Arch. Biochem. Biophys.
2008
, vol. 
470
 (pg. 
64
-
72
)
49
Mukkada
A. J.
Simon
N. W.
Leishmania tropica: uptake of methionine by promastigotes
Exp. Parasitol.
1977
, vol. 
42
 (pg. 
87
-
96
)
50
Duszenko
M.
Ferguson
M. A.
Lamont
G. S.
Rifkin
M. R.
Cross
G. A.
Cysteine eliminates the feeder cell requirement for cultivation of Trypanosoma brucei bloodstream forms in vitro
J. Exp. Med.
1985
, vol. 
162
 (pg. 
1256
-
1263
)
51
Canepa
G. E.
Bouvier
L. A.
Miranda
M. R.
Uttaro
A. D.
Pereira
C. A.
Characterization of Trypanosoma cruzi L-cysteine transport mechanisms and their adaptive regulation
FEMS Microbiol. Lett.
2009
, vol. 
292
 (pg. 
27
-
32
)
52
Dos Santos
M. G.
Paes
L. S.
Zampieri
R. A.
da Silva
M. F.
Silber
A. M.
Floeter-Winter
L. M.
Biochemical characterization of serine transport in Leishmania (Leishmania) amazonensis
Mol. Biochem. Parasitol.
2009
, vol. 
163
 (pg. 
107
-
113
)
53
Kumaran
S.
Jez
J. M.
Thermodynamics of the interaction between O-acetylserine sulfhydrylase and the C-terminus of serine acetyltransferase
Biochemistry
2007
, vol. 
46
 (pg. 
5586
-
5594
)

Supplementary data