Lipoamide dehydrogenase is essential for both bloodstream and procyclic Trypanosoma brucei

Molecular Microbiology (2011) 81(3), 623–639 䊏

doi:10.1111/j.1365-2958.2011.07721.x First published online 4 July 2011

Lipoamide dehydrogenase is essential for both bloodstream and procyclic Trypanosoma brucei mmi_7721 623..639

Angela Roldán,1 Marcelo A. Comini,2 Martina Crispo3 and R. Luise Krauth-Siegel1* 1 Biochemie-Zentrum der Universität Heidelberg (BZH), Im Neuenheimer Feld 328, 69120 Heidelberg, Germany. 2 Group Redox Biology of Trypanosomes and 3 Transgenic and Experimental Animal Unit, Institut Pasteur de Montevideo, Mataojo 2020, CP 11400, Montevideo, Uruguay.

Summary Lipoamide dehydrogenase (LipDH) is a component of four mitochondrial multienzyme complexes. RNA interference or the deletion of both alleles in bloodstream Trypanosoma brucei resulted in an absolute requirement for exogenous thymidine. In the absence of thymidine, lipdh-/- parasites showed a severely altered morphology and cell cycle distribution. Most probably, in bloodstream cells with their only rudimentary mitochondrion, LipDH is required as component of the glycine cleavage complex which generates methylene-tetrahydrofolate for dTMP and thus DNA synthesis. The essential role of LipDH in bloodstream parasites was confirmed by an in vivo model. Lipdh-/- cells were unable to infect mice. Our data further revealed that degradation of branched-chain amino acids takes place but is dispensable. In cultured bloodstream – but not procyclic – African trypanosomes, the total cellular concentration of LipDH increases with increasing cell densities. In procyclic parasites, LipDH mRNA depletion caused an even stronger proliferation defect that was not reversed by thymidine suggesting that in the fully elaborated mitochondrion of these cells the primary effect is not on the glycine cleavage complex. Since the medium used for the cultivation of procyclic cells was not supplemented with glucose, impairment of the 2-ketoglutarate dehydrogenase complex is probably the main effect of LipDH depletion.

Accepted 22 May, 2011. *For correspondence. E-mail [email protected]; Tel. (+49) 6221544187; Fax (+49) 6221545586.

© 2011 Blackwell Publishing Ltd

Introduction Trypanosoma brucei subspecies are the causative agents of African sleeping sickness (T. brucei gambiense, T. brucei rhodesiense) and Nagana cattle disease (T. brucei brucei ). The parasites must differentiate into morphologically and metabolically distinct stages in order to multiply in the bloodstream of the mammalian host and in the digestive tract of the tsetse fly. Long slender bloodstream cells have only a rudimentary mitochondrion that is essentially devoid of cristae (Fenn and Matthews, 2007). Cytochromes and many citric acid cycle enzymes are missing and the parasites are not able to perform oxidative phosphorylation but rely exclusively on glycolysis for energy production (Tielens and Van Hellemond, 1998). In contrast, the procyclic insect form dwelling in the midgut of the tsetse fly has a fully functional mitochondrion and the energy metabolism of these parasites strongly depends on the carbon source (Bochud-Allemann and Schneider, 2002; van Weelden et al., 2003; 2005; Lamour et al., 2005; Coustou et al., 2008). When grown in glucose-rich medium, procyclic T. brucei prefer glucose as nutrient and gain energy by substrate level phosphorylation (Lamour et al., 2005). However, in their natural glucose-deplete habitat, procyclic parasites rely on the catabolism of amino acids, preferably of L-proline (Lamour et al., 2005). Under these conditions, ATP is produced from succinate by oxidative phosphorylation as well as by substrate level phosphorylation catalysed by the citric acid cycle enzyme succinyl-CoA synthetase (Coustou et al., 2008). Lipoamide dehydrogenase (LipDH) is a subunit of four different mitochondrial multienzyme complexes. As part of the pyruvate dehydrogenase (PDH) and 2-ketoglutarate dehydrogenase (KDH) complexes, the flavoenzyme is involved in the energy metabolism of the cell. As a subunit of the branched-chain ketoacid dehydrogenase (BCKDH) it participates in the degradation of the branched-chain amino acids (BCAAs). Finally, as a component of the glycine cleavage complex (GCC), LipDH takes part in the formation of methylene-tetrahydrofolate (Fig. 1). None of these multienzyme complexes has been studied in detail in trypanosomes. For the related trypanosomatid Leishmania major it has been found that parasites that lack glycine decarboxylase grow poorly in the presence of excess glycine or minimal serine (Scott et al., 2008). This

624 A. Roldán, M. A. Comini, M. Crispo and R. L. Krauth-Siegel 䊏

Fig. 1. Methylene-tetrahydrofolate metabolism in T. brucei. H2F, dihydrofolate; H4F, tetrahydrofolate; 5,10-CH2-H4F, methylenetetrahydrofolate; 10-CHO-H4F, 10-formyl-tetrahydrofolate; GCC, glycine cleavage complex containing lipoamide dehydrogenase (Tb11.01.8470) as a subunit; DHFR/TS, bifunctional dihydrofolate reductase/thymidylate synthase (Tb927.7.5480); dUTPase, deoxyuridine triphosphatase (Tb927.7.5160); TK, thymidine kinase (Tb927.10.880); DHCH, bifunctional methylene-tetrahydrofolate dehydrogenase/cyclohydrolase (XP_845784). Genes for TK and DHCH have been annotated but the enzymes have not yet been studied in T. brucei. *In other organisms including Leishmania (Scott et al., 2008) a serine hydroxymethyl transferase can also produce methylene-tetrahydrofolate. The respective gene seems to be absent in T. brucei.

shows that Leishmania can bypass the GCC and synthesize methylene-tetrahydrofolate from serine via the reaction catalysed by serine hydroxymethyl transferase (SHMT). However, the genome of T. brucei lacks a gene for SHMT (Berriman et al., 2005). Methylenetetrahydrofolate is used for thymidylate formation and thus for DNA synthesis. In addition, it is a precursor molecule for the synthesis of formyl-tetrahydrofolate (Scott et al., 2008; Murta et al., 2009). Since trypanosomatids lack the enzymes for purine de novo synthesis, formyltetrahydrofolate seems to be exclusively used for the generation of fMet-tRNA to start mitochondrial protein biosynthesis (Tan et al., 2002; Murta et al., 2009; Vickers et al., 2009). Apart from acting as a dehydrogenase within these multienzyme complexes, LipDH is known for its ability to efficiently catalyse the one-electron reduction of a variety of chemicals such as nitrofuran and quinone derivatives as well as Fe III chelates (Sreider et al., 1992; Blumenstiel et al., 1999; Salmon-Chemin et al., 2001; Petrat et al., 2003). Indeed, yeast LipDH was first discovered as menadione reductase (Misaka et al., 1965). Lipoamide dehydrogenase has been purified from Trypanosoma cruzi, the causative agent of South-American Chagas’ disease (Lohrer and Krauth-Siegel, 1990), the gene has been cloned and overexpressed (Schöneck et al., 1997) and the crystal structure has been solved

(PDB 2QAE) (Werner et al., 2002). For T. brucei, the LipDH gene has also been cloned (Else et al., 1993) and lysates of bloodstream parasites contained about 10% LipDH activity when compared with procyclic cells (Else et al., 1994). The protein has been reported to be closely associated with plasma membrane preparations (Danson et al., 1987) and to be distributed over the whole membrane surface (Jackman et al., 1990). Other authors, however, showed an exclusive mitochondrial localization (Tyler et al., 1997). In pleomorphic T. brucei, the concentration of LipDH strongly increases upon differentiation of the dividing bloodstream long slender to the arrested short stumpy and finally to the insect procyclic parasites (Tasker et al., 2000). Although low KDH activities were found in lysates of bloodstream parasites (Durieux et al., 1991) other authors reported that PDH and KDH activities were undetectable (Else et al., 1994). Thus the physiological – if any – significance of LipDH in the mammalian bloodstream form of T. brucei remained elusive. Here we report on the in vitro and in vivo role of LipDH in African trypanosomes. In both bloodstream and procyclic cells, the enzyme is a mitochondrial protein with comparable local concentrations. In procyclic cells, RNA interference against LipDH resulted in a strong proliferation defect followed by rapid cell death. In contrast, mRNA depletion or the deletion of both alleles in bloodstream cells yielded fully viable parasites if the medium contained thymidine. This strongly suggests that bloodstream T. brucei require LipDH as a component of the GCC. The essential role of LipDH in bloodstream parasites was confirmed by an in vivo model. Lipdh-/- cells were unable to establish an infection in mice.

Results LipDH is a mitochondrial protein in both bloodstream and procyclic T. brucei The lipdh gene from the T. brucei 427 strain (EMBL Nucleotide Sequence Database Accession No. FR 696375; Fig. S1) was cloned and overexpressed in Escherichia coli. The recombinant protein served to generate polyclonal rabbit antiserum against T. brucei LipDH used for immunofluorescence microscopy. The LipDH signal matched that displayed by the MitoTracker Red dye in accordance with a mitochondrial localization of the protein in both bloodstream and procyclic parasites (Fig. 2). LipDH-deficient cells did not show any signal which confirmed the specificity of the antibodies observed in Western blots (see Fig. 3). A plasma membrane (Danson et al., 1987) or cytosolic localization (Colasante et al., 2006) suggested for bloodstream parasites could not be confirmed. The latter approach was based on cell frac© 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

Trypanosoma brucei lipoamide dehydrogenase 625

Fig. 2. Immunofluorescence microscopy monitoring LipDH in bloodstream and procyclic T. brucei. LipDH was visualized with the antiserum against the recombinant protein, the mitochondrion was stained with the MitoTracker Red dye and nuclear and kinetoplast DNA with DAPI.

tionation and subsequent Western blot analysis with the polyclonal antiserum against T. cruzi LipDH. Probably a cross-reacting band of similar size that is enriched in the cytosolic fractions was detected (T. Irsch and R.L. KrauthSiegel, unpubl. obs.). Our immunofluorescence data are consistent with the presence of an N-terminal targeting sequence (Schöneck et al., 1997). In addition, Western blot analyses of isolated procyclic mitoplasts revealed a strong enrichment of LipDH in the organelle compared with the total-cell lysate and the fractionated digitonin lysis of the parasites traced the protein to the mitochondrial matrix (Ceylan et al., 2010). A previous immunofluorescence study with the antiserum against the T. cruzi protein readily recognized LipDH in the mitochondrion of short stumpy and procyclic T. brucei but not in long slender bloodstream T. b. rhodesiense isolated from infected mice (Tyler et al., 1997). The reason for this discrepancy is not known. However, other authors report on the detection of LipDH in the mitochondrion of both developmental states (Stephens et al., 2007). Differences between cultured parasites and those isolated from infected animals are unlikely because the bloodstream EATRO 427 cells shown to express LipDH were also harvested from infected mice (Danson et al., 1987). In bloodstream T. brucei, the LipDH level increases at high cell densities; the local protein concentration is comparable in bloodstream and procyclic cells The levels of LipDH were studied in different growth phases by Western blot analyses. In bloodstream cells, the protein concentration increased from early to late logarithmic phase becoming highest in parasites grown to station-

ary phase. In contrast, in procyclic trypanosomes the LipDH level remained constant throughout the different proliferation phases (Fig. 3A and B). In pleomorphic T. brucei, a strong increase in the LipDH concentration is characteristic for the differentiation of long slender to short stumpy and finally procyclic cells (Tasker et al., 2000). The slender to stumpy differentiation has been shown to be induced solely by an increase in cell density (Vassella et al., 1997). Monomorphic bloodstream T. brucei, as studied here, lost their ability to differentiate into morphologically distinct short stumpy forms. However, the in vitro induction of long slender to short stumpy differentiation in a culture-adapted monomorphic T. brucei strain also results in an increase in LipDH protein and activity (Breidbach et al., 2002). The observation that the LipDH concentration was significantly higher in parasites grown to high densities compared with those harvested in the early logarithmic phase indicates that upregulation of the enzyme may be a pre-adaptation step in the differentiation process. Comparative Western blot analyses of total-cell lysates and different amounts of recombinant protein yielded a cellular LipDH concentration of about 2–5 mM for logarithmically growing bloodstream [cell volume of 58 fl (Opperdoes et al., 1984)] and 30–40 mM in procyclic (cell volume of 96 fl, M. Engstler, pers. comm.) cells (Fig. 3C). Thus, the overall cellular concentration of LipDH in bloodstream cells is about 10% that in procyclic parasites in accordance with previous activity measurements (Else et al., 1994). In bloodstream cells, the single mitochondrion occupies only 2.5% of the total-cell volume (Böhringer and Hecker, 1975; Nolan and Voorheis, 1992) whereas in procyclic cells the relative volume of the organelle is 25% (Böhringer and Hecker, 1975). From these data it can be

© 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

626 A. Roldán, M. A. Comini, M. Crispo and R. L. Krauth-Siegel 䊏

Fig. 3. Expression profile and quantification of LipDH in T. brucei. A. Continuous growth curves of bloodstream and procyclic parasites. B. Representative Western blots of total-cell lysates of bloodstream (3 ¥ 106 cells per lane) and procyclic (8 ¥ 105 cells per lane) cells harvested at different cell densities. T. brucei 2-Cys-peroxiredoxin (TXNPx) and aldolase served as loading controls. The blots were developed with both antisera simultaneously. C. Quantification of LipDH in logarithmically growing T. brucei. Bloodstream and procyclic parasites were harvested at a density of 1 ¥ 106 and 8 ¥ 105 cells ml-1, respectively, and subjected to comparative Western blot analysis as described under Experimental procedures. © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

Trypanosoma brucei lipoamide dehydrogenase 627

Fig. 4. LipDH mRNA depletion affects the proliferation of procyclic, but not of bloodstream, parasites. A. LipDH hairpin RNAi vector: PARP, Procyclic Acidic Repetitive Protein; PPARP-ti, tetracycline-inducible PARP promoter. B. Proliferation curves of un-induced (RNAi -tet) and induced (RNAi +tet) bloodstream (clones 1, 3 and 4, see C) and procyclic transfectants (clones 1–3) grown in standard HMI-9 medium (containing 160 mM thymidine) and MEM-Pros medium (lacking thymidine) respectively. The data represent the mean ⫾ standard deviation of three independent experiments with the clones depicted in C. C. Western blot analyses of lysates from four different bloodstream (3 ¥ 106 cells per lane) and three procyclic clones (1 ¥ 106 cells per lane) harvested 2 and 4 days, respectively, after RNAi induction. T. brucei aldolase (antiserum dilution 1: 10000) and 2-Cys-peroxiredoxin (TXNPx) served as loading controls. Each blot was simultaneously developed with both antisera.

concluded that LipDH is a rather abundant protein with a similar concentration in the mitochondrion of both developmental forms. Under normal culture conditions, LipDH downregulation affects proliferation of procyclic – but not of bloodstream – parasites The first approach to investigate the role of LipDH in T. brucei was to generate cell lines that allowed the tetracycline-inducible depletion of its mRNA (Fig. 4A). Bloodstream and procyclic cells were grown in HMI-9 and

MEM-Pros medium respectively. The latter medium contains proline but lacks glucose to mimic the natural habitat of the parasites in the tsetse fly (for details see Experimental procedures). When cultured under these normal conditions, bloodstream cells did not show any growth phenotype, whereas procyclic parasites stopped proliferation 72 h after RNAi induction followed by cell death after 7 days (Fig. 4B). In the case of bloodstream cells, efficient downregulation of LipDH was observed 24 h after starting RNAi induction, whereas in procyclic cells LipDH depletion started after 48 h and became maximal only after 4 days. Figure 4C shows the Western blot analyses of

© 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

628 A. Roldán, M. A. Comini, M. Crispo and R. L. Krauth-Siegel 䊏

Fig. 5. Effect of exogenous thymidine on the proliferation of LipDH-depleted parasites. A. Bloodstream wild-type (WT), un-induced (RNAi -tet) and induced (RNAi +tet) LipDH-RNAi cells were grown in standard HMI-9 medium which contains 160 mM thymidine (+dThd) or in medium lacking thymidine (-dThd). In the absence of thymidine, the LipDH-depleted cells died within a week. However, proliferation was restored when on the fourth day an aliquot of the culture was diluted to 103 cells ml-1 in standard HMI-9 medium. The data presented are the mean ⫾ standard deviation of three independent experiments each with three clones. B. Procyclic WT and LipDH-depleted cells (RNAi +tet) were grown in MEM-Pros medium lacking (-dThd) or supplemented with 1.6 mM thymidine (+dThd). The presence of thymidine had no effect on the growth behaviour of both cell lines. After a few days in culture, two out of the four original procyclic RNAi clones re-expressed LipDH in the presence of tet, a phenomenon often observed for essential T. brucei proteins. Therefore the data presented are those of the two remaining regulable RNAi clones giving identical results.

bloodstream and procyclic T. brucei grown in the presence of tetracycline for 48 h and 96 h respectively. Although LipDH was still detectable in procyclic cells harvested 4 days after RNAi induction this degree of depletion resulted already in a complete proliferation arrest. Exogenous thymidine is sufficient for proliferation of bloodstream, but not of procyclic, LipDH-depleted cells Normal HMI-9 medium for the cultivation of bloodstream T. brucei contains 160 mM thymidine. Therefore we studied if the growth of LipDH-depleted cells necessitates the presence of thymidine. Indeed, omission of the nucleoside from the medium resulted in a proliferation defect which could be reverted by subsequent addition of thymidine (Fig. 5A). The fact that thymidine was able to act as nutritional by-pass suggests that in this developmental stage LipDH plays an essential role mainly as a component of the GCC. Depletion of LipDH should interfere with the formation of methylene-tetrahydrofolate needed for the generation of thymidylate and thus for DNA synthesis (Fig. 1). In the absence of an active GCC, exogenous thymidine is supposed to restore the dTMP levels by the action of a thymidine kinase. This cytosolic enzyme has been characterized in Leishmania species (Thiel et al., 2008) and partially purified extracts from T. b. rhodesiense were shown to have thymidine kinase activity (Chello and Jaffe, 1972). A

thymidine kinase gene has been annotated for T. brucei (Thiel et al., 2008), although the derived protein sequence is much longer than that of the Leishmania enzyme. Our results support the genome-based prediction that a SHMT is missing in African trypanosomes (Scott et al., 2008). In other organisms, including the closely related Leishmania, this enzyme can also generate methylene-tetrahydrofolate and thus circumvent the reaction catalysed by the LipDHcontaining GCC. The MEM-Pros medium used for the cultivation of procyclic T. brucei does not contain thymine or thymidine. However, the severe proliferation defect of LipDHdepleted procyclic cells could not be reversed when supplementing the medium even with 1.6 mM thymidine (Fig. 5B). As reported by Castillo-Acosta et al. (2008), exogenous thymidine can revert the proliferation defect of dUTPase-depleted procyclic parasites in accordance with uptake of thymidine in the cells. This suggests that in the insect form of T. brucei thymidylate production is not the primary reaction affected upon the depletion of LipDH. In the presence of thymidine both LipDH alleles can be deleted in bloodstream cells Since the RNAi experiments revealed the principle viability of LipDH-depleted bloodstream parasites, we decided to generate stable LipDH-deficient cell lines. The coding © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

Trypanosoma brucei lipoamide dehydrogenase 629

Fig. 6. Deletion of the lipdh gene in bloodstream T. brucei. A. Scheme of the lipdh locus, replacement by the puromycin N-acetyltransferase (PAC) and blasticidin S-deaminase (BLA) genes and construct for the inducible expression of an ectopic copy of LipDH-Myc2. UTR, untranslated region; tet, tetracycline; Otet, tet operator; 3′ACT, 3′UTR of the actin gene; Myc, Myc2-tag. The arrows indicate the locations of the primers used for PCR to confirm the correct replacement of the lipdh alleles by PAC and BLA respectively. B. PCR analysis of WT cells and three double-knockout clones (lipdh-/- 1, 2 and 3). Primers to the 5′ and 3′UTR (5′UTR-F and 3′UTR-R) outside the recombination area and internal primers to the PAC and BLA genes (P-F, P-R, B-F and B-R) revealed the replacement of the lipdh alleles and correct insertion of the resistance genes. The absence of lipdh in the double-knockout clones was further confirmed by PCR with primers specific for the coding region of the gene (L-F and L-R), which yielded a 1427 bp fragment in the WT but none in the lipdh-/- clones. C. Western blot analysis of two lipdh+/- clones, WT, an induced (+tet) and un-induced (-tet) representative conditional knockout clone (cKO) and three lipdh-/- clones. The extract of 3 ¥ 106 cells was applied per lane. The cKO cells were harvested 2 days after tet induction. The blot was simultaneously developed with the antisera against T. brucei LipDH and aldolase.

regions of the two alleles were sequentially replaced by drug resistance genes (Fig. 6A) while cultivating the parasites in standard HMI-9 medium. After selection with puromycin (PAC) and/or blasticidin (BLA), resistant parasites lacking one (Dlipdh::PAC/LipDH or Dlipdh::BLA/LipDH, referred to as lipdh+/-) or both (Dlipdh::PAC/Dlipdh::BLA, designated as lipdh-/-) lipdh alleles were obtained. The successful replacement of the coding region was confirmed by different PCR analyses of genomic DNA from

wild-type (WT) cells and double-knockout clones (Fig. 6B). Western blot analysis confirmed the absence of the protein in the lipdh-/- cell lines (Fig. 6C). Next we generated a conditional knockout cell line (Dlipdh/LipDH-cmycti, referred to as cKO). Bloodstream lipdh-/- parasites were transfected with a construct that allowed the tetracycline-inducible expression of an ectopic copy of the protein with C-terminal Myc2-tag (Fig. 6A). After induction, LipDH-Myc2 was detectable by

© 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

630 A. Roldán, M. A. Comini, M. Crispo and R. L. Krauth-Siegel 䊏

Fig. 7. Proliferation of LipDH-deficient bloodstream cells in the presence and absence of thymidine. A. Growth curves of WT, induced conditional knockout cells (cKO +tet) and lipdh-/- cells cultivated with (+dThd) and without (-dThd) 160 mM thymidine. Independent of thymidine, the cKO cells attained the stationary phase although at a slower rate when compared with WT cells. The same was observed for lipdh-/- cells when cultured in thymidine-supplemented medium. However, in the absence of thymidine, the lipdh-/cells failed to multiply and were below detectable levels after 3 days. B. Lipdh-/- cells were grown in the presence of different thymidine concentrations and counted after 15 h and 27 h. The asterisks (*) denote significant differences relative to lipdh-/- cells in standard HMI-9 medium containing 160 mM thymidine (P < 0.05, ANOVA, LSD Fischer). All values shown are the mean ⫾ standard deviation of three independent experiments, each with three clones.

Western blot analysis using either the polyclonal rabbit antiserum against the recombinant protein (Fig. 6C) or the monoclonal antibodies towards the Myc-tag (not shown). A single band running at a slightly higher molecular mass compared with the WT protein was recognized in accordance with the presence of LipDH-Myc2. Effect of different thymidine concentrations on the proliferation of LipDH-deficient bloodstream cells As observed for the LipDH-depleted cells (Fig. 5A), the proliferation of lipdh-/- bloodstream parasites was severely impaired when thymidine was omitted from the medium. As expected, in the knockout cell lines the effect was even stronger. Whereas the RNAi cell lines stopped growing 2 days after starting induction but were still alive for up to 5 days, the cell number of the lipdh-/- trypanosomes decreased already during the first day in thymidinefree medium and after 72 h living cells were below detection levels (Fig. 7A). To verify that the effects observed are specifically caused by disruption of the lipdh gene, the proliferation of the cKO cell line was also studied. In the absence of both thymidine and tetracycline, the cKO cells showed the same growth defect as the parental lipdh-/- cells (data not shown). In the presence of tetracycline and thus expression of the ectopic LipDH-Myc2, the cells grew in medium lacking thymidine although at a slightly lower rate than WT cells. In the presence of thymidine, induced cKO cells showed a growth behaviour intermediate between that of lipdh-/- and WT cells. Proliferation of the lipdh-/- cell lines was sensitive to the starting cell density. At a density of 5 ¥ 105 instead of

1 ¥ 105 cells ml-1, the cells grew faster resulting in a doubling time of 6.4 h compared with 5.4 h for WT cells (data not shown). Thus, although exogenous thymidine efficiently rescued the growth defect of lipdh-/- cells, the proliferation rate was slightly lower than that of WT cells. One possible explanation is that in the absence of methylene-tetrahydrofolate the dUMP levels raise with a concomitant increase of dUTP (Blount et al., 1997; Castillo-Acosta et al., 2008). Mis-incorporation of dUTP into DNA would require its repair associated with growth retardation. To estimate the minimal thymidine concentration for survival and proliferation, bloodstream lipdh-/- cells were cultivated in the presence of various thymidine concentrations. At 5 or 20 mM thymidine in the medium, the parasites died within 27 h (Fig. 7B). This result is particularly relevant when considering that in mice and human serum the thymidine concentration is about 1 and 0.1 mM respectively (Clarke et al., 2000). LipDH-deficient bloodstream cells exhibit an altered cell cycle distribution and morphology Wild-type and lipdh-/- parasites were labelled with propidium iodide and their DNA content was analysed by fluorescence-activated cell sorting (FACS). WT cells grown for 24 h with or without thymidine as well as lipdh-/- cells cultivated in the presence of thymidine showed a nearly identical cell cycle pattern (Fig. 8A). As expected, about 70% of the cells were in G1 phase, 10% in S phase and 20% in G2/M phase (Fig. 8B, 160 mM thymidine) in good agreement with published data (Sien© 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

Trypanosoma brucei lipoamide dehydrogenase 631

kiewicz et al., 2008). However, the distribution of lipdh-/cells grown for 24 h in medium lacking thymidine (corresponding to ⱕ 20 mM) differed significantly. A clear decrease of cells in G1 phase (to 34%) and a concomitant increase of cells in S phase (to 24%) were observed. Additionally, we detected many cells with either sub-G1 or > G2/M DNA content. In the presence of 80–460 mM thymidine, the lipdh-/- cells displayed a normal cell cycle composition, although 80 mM thymidine resulted already in a significantly reduced proliferation rate (Fig. 7B). The atypical DNA content was accompanied by dramatic changes in the cell morphology. The general appearance of WT parasites grown with or without thymidine and of lipdh-/- cells cultivated in the presence of thymidine was normal (Fig. 8C). In contrast, when we imaged lipdh-/cells after 24 h cultivation without exogenous thymidine most of the parasites appeared bloated. Some cells presented more than one flagellum and nucleus which may be an indication for impaired cytokinesis (Hammarton, 2007; Castillo-Acosta et al., 2008). The severely affected overall cell morphology rendered it difficult to distinguish individual parasites. Taken together, our data clearly evidenced that the lack of LipDH in bloodstream T. brucei, under thymidine starvation, strongly disturbs cell cycle progression and cellular architecture. An impaired thymidylate synthesis and thus defects in DNA replication are probably the main effects upon lipdh deletion. Effect of exogenous BCAAs on LipDH-depleted bloodstream parasites The catabolism of Ile, Val and Leu begins with two common steps. Transamination by the branched-chain amino acid aminotransferase (BCAT) (Tb927.2.4610, Tb927.2.4590, putative) results in the corresponding 2-keto acids which are then oxidatively decarboxylated to the respective acyl-CoAs by the LipDH-containing BCKDH complex (Tb10.v4.0043, BCKDH e1-beta subunit precursor, putative; Tb927.5.4330, dihydrolipoamide branched-chain transacylase, putative). The standard HMI-9 medium contains 0.72 mM of each Ile, Val and Leu, and thus a total BCAA concentration of 2.16 mM. A 20-fold excess of each individual BCAA did not affect the growth of bloodstream lipdh-/- cells over the course of 72 h. However, the simultaneous 20-fold excess of all three BCAA reduced the proliferation of lipdh-/- cells to about 30% compared with growth in standard medium (Fig. 9A). The proliferation of WT cells was unaffected in either case. This was the first indication that BCAA catabolism takes place in bloodstream T. brucei. Since the proliferation defect became evident at an overall BCAA concentration of 43 mM, we supposed that very high concentrations of Ile, Val and Leu would be required to envisage a putative effect of an individual BCAA. Therefore the proliferation of WT and

lipdh-/- parasites was followed in medium containing 72 mM of the different BCAAs. In the presence of Leu and Val, but not of Ile, lipdh-/- cells died. Again WT cells grew under these conditions (Fig. 9B). Because a blockage of the BCKDH complex results in the accumulation of both BCAA and branched-chain ketoacid (BCKA) (Chuang et al., 2006) we do not know whether the lipdh-/- cells die as a consequence of the accumulation of Leu or Val or their respective 2-keto acid, and if Ile did not show an effect because of differences in the uptake of the individual BCAAs. Nonetheless, our data revealed that the catabolism of BCAAs is active in bloodstream T. brucei but the pathway is dispensable. In human serum, the concentration of Ile, Leu and Val is about 60, 130 and 220 mM respectively (Aussel et al., 1986). Thus, the unphysiologically high concentrations of Leu and Val needed to show an effect on lipdh-/- cells suggests that also in the mammalian host, an active BCKDH is not required for the parasites to proliferate and/or sustain an infection. Finally, we can exclude that the slightly lower proliferation rate of lipdh-/parasites (in thymidine-supplemented medium) when compared with WT cells is due to the lack of an active BCKDH. LipDH is essential for in vivo growth and infectivity of T. brucei BALB/cJ mice were injected with WT, lipdh-/- and cKO cells and animal survival as well as the blood parasite levels were followed. In animals infected with WT T. brucei, the parasite density started to rise at day 7 post infection (Fig. S2) coinciding with the appearance of sickness symptoms such as rough hair coat, fatigue, hunched posture and abnormal respiration. Between weeks 2 and 4, all mice died or had to be euthanized for ethical reasons (Fig. 10). In contrast, no parasites could be detected in blood samples from mice injected with lipdh-/- trypanosomes throughout the observation period. This clearly demonstrated that LipDH is indispensable for the viability of T. brucei in vivo. In a third approach, mice were infected with cKO cells and kept either without or with oxytet (to induce expression of LipDH-Myc2) in the drinking water. The animals did not show any disease symptoms and no parasites were detectable in the blood samples. To prove the expression of the ectopic copy of LipDH-Myc2, cKO parasites were pre-cultured for 72 h in the presence of oxytet prior to animal infection. A Western blot and subsequent densitometric analysis (program ImageJ; Abramoff et al., 2004) revealed that un-induced and induced cKO cells expressed LipDH-Myc2 at a level of about 8% and 33%, respectively, compared with the authentic protein in WT cells (inset of Fig. 10). Since no parasites were detectable in the blood samples taken at different time points, it was

© 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

632 A. Roldán, M. A. Comini, M. Crispo and R. L. Krauth-Siegel 䊏

not possible to confirm expression of the ectopic copy of LipDH-Myc2 in vivo. Assuming that upon inoculation of the mice the parasite still expressed LipDH-Myc2, our results suggest that this LipDH level was not sufficient for parasite survival and infectivity in the mammalian host.

Discussion Lipoamide dehydrogenase is the common component of four mitochondrial multienzyme complexes, namely the three 2-ketoacid dehydrogenase complexes as well as © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

Trypanosoma brucei lipoamide dehydrogenase 633

Fig. 8. Phenotypic analysis of lipdh-/- bloodstream parasites. A. Flow cytometry of propidium iodide (PI) stained WT and lipdh-/- cells grown for 24 h in the presence (+dThd) or absence (-dThd) of 160 mM thymidine. B. Percentage of cells in different cell cycle phases after 24 h cultivation in HMI-9 medium containing variable thymidine concentrations. No differences were observed for WT cells (not shown). However, at ⱕ 20 mM thymidine, lipdh-/- cells showed a significantly altered distribution. The values represent the mean ⫾ standard deviation of three independent experiments, each with three clones. The asterisks (*) indicate significant differences compared with WT cells (P < 0.05, LSD Fischer). C. Microscopy of WT and lipdh-/- parasites cultured for 24 h with (+dThd) or without (-dThd) 160 mM thymidine. In the absence of thymidine, lipdh-/- cells showed a highly abnormal overall morphology.

the GCC. Here we showed that the flavoenzyme is not only essential in the procyclic form of T. brucei which has a fully functional mitochondrion but also in the mammalian bloodstream form with its rudimentary organelle. The viability and almost full proliferation of bloodstream lipdh-/- cells when cultured in the presence of thymidine renders thymidylate production by the GCC the main process in which the enzyme is involved and suggests the dispensability of PDH and KDH in the mammalian infective form of the parasite. However, lipdh-/- bloodstream parasites rapidly developed multiple cell cycle abnormalities when they were transferred into medium deprived of thymidine. Notably, we recognized a significant decrease of cells in G1 phase and increase of parasites in S phase, together with the emergence of < G1 and > G2/M populations. The rise of a > G2/M peak correlated with the appearance of trypanosomes with enlarged nucleus or

aggregates with several nuclei, as observed upon immunofluorescence of DAPI stained parasites (not shown). Very similar phenotypes have been reported for bloodstream T. brucei cells that lack the DHFR/TS gene (Sienkiewicz et al., 2008) and for procyclic parasites where the dUTPase activity has been depleted by RNAi (CastilloAcosta et al., 2008). As suggested for DHFR/TS-deficient trypanosomes (Sienkiewicz et al., 2008), cells arrested in S phase may evidence their inability to complete DNA synthesis. The appearance of a sub-G1 population, which is also observed in procyclic cells depleted of dUTPase (Castillo-Acosta et al., 2008) or subjected to ER stress (Goldshmidt et al., 2010), suggests that many cells undergo DNA fragmentation and loss of DNA fragments. In cells with any kind of impaired thymidylate synthesis, DNA strand breaks probably accumulate during the required repair of mis-incorporated dUTP caused by alter-

Fig. 9. Effect of exogenous branched-chain amino acids on the proliferation of lipdh-/- bloodstream parasites. A. Lipdh-/- cells were cultured in standard HMI-9 medium containing 0.72 mM of each branched-chain amino acid (n Ile, Val, Leu) and in medium supplemented with a 20-fold excess (14.4 mM) of either Ile, Val or Leu (20 ¥ Ile, 20 ¥ Val, 20 ¥ Leu). After different time points, the cells were counted. No significant differences compared with cells grown in standard medium (set to 100%) were observed. However, a 20-fold excess of Ile, Val and Leu at the same time (20 ¥ Ile, Val, Leu) resulted in impaired proliferation of the lipdh-/- cells. The growth of WT cells was unaffected by either treatment (not shown). B. Lipdh-/- and WT cells were seeded at an initial cell density of 2 ¥ 105 cells ml-1 in standard HMI-9 medium (n Ile, Val, Leu) and in medium containing a 100-fold excess (72 mM) of one of the amino acids, with the other two present at their normal concentration (100 ¥ Ile, 100 ¥ Val or 100 ¥ Leu). All WT cultures as well as the lipdh-/- cells in standard HMI-9 and in 100 ¥ Ile media, proliferated and every 24 h were diluted back to a density of 2 ¥ 105 cells ml-1 in the corresponding fresh medium. Since Leu and Val caused a decrease of the lipdh-/- populations, these cultures were not diluted. Instead, every day the cells were harvested and completely resuspended in the same volume of fresh medium. The values presented are the mean ⫾ standard deviation of data obtained for three different lipdh-/- clones. © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

634 A. Roldán, M. A. Comini, M. Crispo and R. L. Krauth-Siegel 䊏

Fig. 10. Kaplan-Meier survival analysis of mice infected with WT, lipdh-/- or cKO T. brucei. Female BALB/cJ mice (six animals per group) were infected intraperitoneally with 104 parasites. In the case of the cKO cells, parasites cultured in the presence (pre-induced, cKO +pi) or absence (cKO -pi) of oxytetracycline (oxytet) were used. The animals were kept with (+oxytet) or without (-oxytet) 1 mg ml-1 oxytet in the drinking water. Mice infected with WT T. brucei died within 28 days independent of oxytet. In contrast, all animals infected with lipdh-/parasites survived for the whole experimental time span. The same was observed for the cKO cells. Also parasites that prior to inoculation were treated with oxytet and shown to express LipDH-Myc2 at a level of about one-third of WT cells (see inset) were unable to cause an in vivo infection. Inset: Western blot analysis of total extracts from cKO -pi, cKO +pi, lipdh-/- and WT parasites prior to animal infection. The blot was simultaneously developed with the antisera against T. brucei LipDH and 2-Cysperoxiredoxin (TXNPx) as loading control.

ations in the dTTP/dUTP ratio (Castillo-Acosta et al., 2008). Taken together, our results strongly suggest that LipDH is essential in bloodstream T. brucei because its lack would abolish the synthesis of dTTP necessary for DNA synthesis. In contrast, the severe growth defect of procyclic cells upon depletion of LipDH was not abolished by addition of thymidine to the culture medium indicating that thymidylate synthesis is not the primary reaction affected. This is corroborated by studies on the parasite dUTPase, another enzyme involved in thymidylate synthesis (Castillo-Acosta et al., 2008). RNAi against dUTPase results also in a severe growth defect of procyclic T. brucei but in this case proliferation of the cells is efficiently restored by exogenous thymidine. The involvement of LipDH in other metabolic processes may be even more important. Depletion of LipDH in procyclic cells is supposed to affect both the PDH complex and the KDH complex. For procyclic parasites that are cultured in the presence of glucose, pyruvate and proline, it has been shown that the cells are viable if only one – but not both – E1 subunits of these complexes are depleted (BochudAllemann and Schneider, 2002). In the present study the cells were grown in MEM-Pros medium which contains 5.2 mM proline but lacks glucose, glycerol and pyruvate. As shown by Coustou et al. (2008), the proline catabolism of procyclic parasites strongly depends on the glucose level in the medium. In glucose-deplete (0.15 mM) medium – but not in glucose-rich (6 mM) medium – proline is metabolized to alanine and glutamate and ATP is produced from succinate by oxidative phosphorylation as well as in the reaction of succinyl-CoA synthetase. The

final MEM-Pros medium used here contains 10% fetal calf serum and therefore about 0.5 mM glucose. Under the assumption that this resembles more the glucose-deplete than glucose-rich medium, the (activity of the) PDH complex should be dispensable. The detrimental effect of LipDH depletion in procyclic cells may be mainly caused by the impairment of the KDH complex. Being among the most hydrophobic amino acids, Ile, Val and Leu play important roles in the stability of folded proteins (Brosnan and Brosnan, 2006). Leishmania species, and to a minor extent T. cruzi, can synthesize sterols starting from Leu (Ginger et al., 2000). Intact promastigote Leishmania donovani have been shown to decarboxylate Leu and ketocapronate, its deamination product, suggesting that BCAA catabolism takes place in the parasites (Blum, 1991). The second step in the degradation of Ile, Val and Leu is the oxidative decarboxylation of the corresponding 2-keto acid to the respective acyl-CoA. This reaction is catalysed by the LipDHcontaining BCKDH complex. In humans, an impaired BCAA catabolism causes maple syrup urine disease and a neuronal cell model with inhibited BCKDH identified Leu as the most toxic BCAA (Kasinski et al., 2004). Unphysiologically high concentrations of Val and Leu, but not Ile, were lethal for bloodstream lipdh-/- cells. This indicates that, although being active in bloodstream T. brucei, the pathway is not essential under culture conditions and most probably not in vivo either. Our data revealed that (i) methylene-tetrahydrofolate synthesis is absolutely essential for bloodstream trypanosomes when thymidine is absent, and (ii) thymidylate production is the main metabolic route of methylene© 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

Trypanosoma brucei lipoamide dehydrogenase 635

tetrahydrofolate. Although this metabolite may be used for supplementary reactions, the proliferation rescue of lipdh-/- cells solely by thymidine addition demonstrates that such processes are not essential in bloodstream cells. In T. brucei a formate-tetrahydrofolate lyase gene is missing; the only pathway to synthesize 10-formyltetrahydrofolate appears to be the conversion of methylene-tetrahydrofolate by the bifunctional methylenetetrahydrofolate dehydrogenase/cyclohydrolase (Vickers et al., 2009) (Fig. 1). 10-Formyl-tetrahydrofolate serves to generate formyl-Met-tRNA for the initiation of mitochondrial protein biosynthesis. The fact that exogenous thymidine efficiently recovered the proliferation of lipdh-/bloodstream cells indicates either that mitochondrial protein biosynthesis is dispensable or that formylation of Met-tRNA is not essential. In yeast and apicomplexan organisms (Li et al., 2000; Pino et al., 2010), mitochondrial translation can start with an unmodified Met-tRNA and this may also be the case in T. brucei. The adequate rescue of DHFR/TS-depleted bloodstream cells by thymidine (Sienkiewicz et al., 2008) supports the conclusion that the synthesis of formyl-Met-tRNA is not essential. The inactivation of DHFR/TS, although at an earlier step, has the same consequence in terms of impairing methylenetetrahydrofolate synthesis. Protein biosynthesis is supposed to take place also in the rudimentary mitochondrion of bloodstream T. brucei because different components of the translation machinery are essential (Cristodero et al., 2010). In addition, transcription and editing of some respiratory chain genes are not downregulated in comparison with procyclic cells (Ochsenreiter et al., 2008). However, a direct proof for the synthesis of a mitochondrially encoded protein in bloodstream T. brucei is still awaiting (Schnaufer et al., 2005; Richterová et al., 2011). The work presented here revealed that LipDH is essential for the infectious bloodstream form of African trypanosomes, both in vitro and in vivo in a mouse model. Recently it has been reported for the enzymes from T. cruzi (Ramos et al., 2009; Gutiérrez-Correa, 2010) and mycobacteria (Bryk et al., 2010) that LipDH may act as a drug target molecule. This is supported by the fact that LipDH mRNA is upregulated in T. cruzi parasites resistant to the drug benznidazole (Murta et al., 2008). However, the development of a selective inhibitor may be difficult. Mammalian LipDHs share about 95% of all residues. Although the similarity between human and T. brucei LipDH is lower (50% identity), the active sites of the enzymes are highly conserved; and the crystal structure of T. cruzi LipDH did not reveal significant differences to the proteins from other sources (PDB 2QAE; Werner et al., 2002). Nevertheless there are several examples that the selective uptake/enrichment of a compound can result in considerable selectivity for the parasite. To validate LipDH as a putative drug target molecule in African

trypanosomes, future work will include studies on the cellular effect of known inhibitors or subversive substrates on lipdh-/- cells in comparison with LipDH-overproducing and WT parasites.

Experimental procedures Materials All T. brucei cells used in this work are derived from the culture-adapted cell line 449, which is a descendant of the strain Lister 427 (Cunningham and Vickerman, 1962) carrying an integrated gene for the tetracycline repressor (Biebinger et al., 1997). Genomic DNA was prepared from procyclic cells using the DNeasy Blood & Tissue Kit (Qiagen) following the manufacturer’s instructions. All primers were designed based on the Tb11.01.8470 GeneDB LipDH sequence of the TREU 927/4 strain. They were synthesized by MWG Biotech and are listed in Table S1. The pET9d expression vector and His-tagged TEV protease were kind gifts of Gunther Stier (Heidelberg). The vectors pHD678, pHD1747, pHD1748, pHD1700 and pHD615 as well as the T. brucei aldolase antibody were provided by Dr Christine Clayton (Heidelberg). The correctness of each construct was confirmed by restriction analysis and sequencing (MWG Biotech). Table S2 shows all plasmids obtained. Polyclonal rabbit antibodies against recombinant T. brucei LipDH (see supplementary experimental procedures in Supporting information) were generated by Eurogentec.

Cultivation of trypanosomes Bloodstream and procyclic T. brucei were cultivated in HMI-9 and MEM-Pros medium, respectively, as previously described (Schlecker et al., 2005). Thymidine-free HMI-9 medium for the experiments with the RNAi cell lines was prepared by omitting thymidine and dialysing the heatinactivated fetal calf serum (GibcoBRL-Invitrogen, Karlsruhe) which should reduce its thymidine concentration at least by a factor of 10 (Schaer et al., 1978). In the case of the experiments with the knockout cell lines, thymidine free HMI-9 medium was prepared by omitting thymidine without dialysis of the fetal calf serum. HMI-9 medium with excess of BCAAs was prepared by adding each individual amino acid or all three amino acids to the desired concentrations. L-Leu, L-Val and L-Ile were purchased from Sigma-Aldrich. The MEM-Pros medium for procyclic cells is a customer-defined medium produced by Biochrom (C. Clayton, pers. comm.). It corresponds essentially to the DTM medium described in http:// tryps.rockefeller.edu/trypsru2_culture_media_compositions except that it lacks glycerol, mercaptoethanol, bathocupronic acid as well as most of the non-essential amino acids. Importantly, the medium does not contain glucose, pyruvate, thymidine and thymine.

Cloning, overexpression and purification of recombinant T. brucei LipDH The procedure is described in Supporting information.

© 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

636 A. Roldán, M. A. Comini, M. Crispo and R. L. Krauth-Siegel 䊏

Western blot analyses Bloodstream and procyclic T. brucei cells were subjected to Western blot analyses as described previously (Ceylan et al., 2010) with the T. brucei LipDH antiserum (1:5000 to 7000). Antibodies against T. brucei 2-Cys-peroxiredoxin (TXNPx, 1:4000) and aldolase (1:40000) served as loading controls, and goat anti-rabbit immunoglobulin G (IgG) (Santa Cruz Biotechnology) as secondary antibody (1:20 000). The immune complexes were visualized with the SuperSignal® West Pico chemiluminescent substrate (Pierce). For the cellular quantification of LipDH, logarithmic phase bloodstream (harvested at 1 ¥ 106 cells ml-1) and procyclic T. brucei were harvested, and different amounts of cell pellets and recombinant T. brucei LipDH were subjected to Western blot analyses as described above. The intensities of the protein bands were quantified as the absolute integrated optical density using the program Gel Pro Analyser 3.1.

To generate lipdh-/- parasites with a tetracycline-inducible (ti) ectopic copy of the LipDH gene with C-terminal Myc2-tag (Dlipdh/LipDH-cmycti), a BamHI site in the coding region had to be removed. The complete coding region was amplified from genomic DNA by PCR using lipdh-cko.cmyc-HindIII-11F and lipdh-cko.cmyc-BamHI-12R as primers, and cloned into a pGEM-T vector (Promega). The purified plasmid-DNA served as template for the replacement of adenine 618 by a thymine using lipdh-mut-BamHI-13F and lipdh-mutBamHI-14R as primers and the QuikChange Site-Directed Mutagenesis Kit (Stratagene). After treatment with DnpI, the pGEM-T-mut-lipdh plasmid was amplified in XL1-Blue cells (Stratagene) and digested with BamHI and HindIII. The resulting fragment was purified on a 1% agarose gel, extracted using the Zymoclean™ Gel DNA Recovery Kit (Zymo Research) and ligated into pHD1700. After amplification in E. coli SURE 2 cells, the NotI-linearized cKO-lipdhmyc construct was used for transfection of bloodstream double-knockout trypanosomes.

Immunofluorescence microscopy A total of 1–2 ¥ 106 parasites from logarithmic phase (1–2 ¥ 106 cells ml-1) cultures were collected by 10 min centrifugation at 2000 g, 27°C, and subjected to immunofluorescence microscopy as previously described (Comini et al., 2008). The T. brucei LipDH and Alexa Fluor® 488 goat antirabbit IgG (H+L) antibodies were diluted 1:2000 and 1:250 respectively. The slides were analysed under a Carl Zeiss Axiovert 200 M light and fluorescence microscope equipped with an AxioCam MRm digital camera using the program AxioVision (Zeiss, Jena).

Plasmid constructs for transfection of T. brucei For RNA interference, a 409 bp stretch of the coding region of LipDH was amplified from genomic DNA by PCR using lipdhrnai-HpaI-3F and lipdh-rnai-EcoRI-4R as primers. The same sequence plus additional 65 bp was amplified with the primers lipdh-rnai-HindIII-5F and lipdh-rnai-EcoRI-6R. Both fragments were digested and simultaneously ligated with the HindIII- and HpaI-digested, tetracycline-inducible pHD678 vector. The hairpin construct pHD678-lipdh was amplified in E. coli SURE 2 cells (Stratagene) at 30°C in LB medium. Lowering the temperature proved to be critical for the successful amplification of the plasmid. After isolation, the NotI-linearized, ethanol precipitated plasmid was used for transfection of WT trypanosomes. For the replacement of the lipdh alleles, segments of the lipdh 5′ and 3′ untranslated regions (UTR) were amplified by PCR using lipdh-5′UTR-XhoI-7F, lipdh-5′UTR-HindIII-8R and lipdh-3′UTR-PstI-9F, lipdh-3′UTR-NotI-10R as primers. The XhoI/HindIII digested 5′UTR and PstI/NotI digested 3′UTR fragments were then placed at either side of the puromycin (PAC; pHD1747) or blasticidin (BLA; pHD1748) resistance cassettes. The resulting KO-lipdh-pur and KO-lipdh-bla plasmids were amplified in SURE 2 cells at 37°C, digested with KpnI and SacI and used for the transfection of bloodstream parasites. To confirm the correct replacement of the lipdh alleles by PAC and BLA, PCR analyses were perform using Pfu DNA polymerase (recombinant) (Fermentas) or PfuUltra Hotstart DNA polymerase AD (Stratagene).

Transfection Stable RNAi and knockout cell lines were generated by transfecting 1–2 ¥ 107 parasites from logarithmic phase cultures using the Amaxa nucleofector system (Amaxa AG, Köln, Germany) as described previously (Ceylan et al., 2010). Single clones were obtained by serial dilutions in medium containing the appropriate selective drug (for bloodstream RNAi and cKO cells, 10 mg ml-1 hygromycin; for procyclic RNAi cells, 50 mg ml-1 hygromycin; for single and doubleknockout parasites 0.2 mg ml-1 puromycin and/or 5 mg ml-1 blasticidin). To induce RNA interference or the expression of LipDH-Myc2, 0.5 mg ml-1 and 1 mg ml-1 tetracycline was added to bloodstream and procyclic cultures respectively.

Growth phenotypes Bloodstream and procyclic parasites were seeded at an initial density of 0.1 to 5 ¥ 105 cells ml-1 and 1 to 8 ¥ 105 cells ml-1, respectively, allowed to grow for 4–7 days and counted every 24 h in a Neubauer chamber. Long-term proliferation was observed by starting at a cell density of 1 ¥ 105 to 1 ¥ 106 cells ml-1 and diluting the cultures daily to the initial density. Tetracycline was added every 24 h when induction of RNAi or ectopic expression of LipDH-Myc2 was studied. LipDHdeficient cells (single- and double-knockout) were cultivated without selecting antibiotics. In experiments where bloodstream cells were grown in HMI-9 medium lacking thymidine or containing an excess of BCAAs, the cells were harvested, washed twice in the respective medium and then seeded in this and/or in normal HMI-9 medium for control.

FACS analysis At least 5 ¥ 105 T. brucei cells were harvested by 10 min centrifugation at 2000 g, 4°C and resuspended in 1 ml of 70% ethanol in PBS while slowly vortexing. After overnight incubation at 4°C, the cells were washed with PBS and resuspended for 30 min in 500 ml of PBS containing 10 mg ml-1 RNase A (Sigma) and 30 mg ml-1 propidium iodide (Sigma). © 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

Trypanosoma brucei lipoamide dehydrogenase 637

The samples were analysed using a FACSCanto™ II flow cytometer and the software FACSDiva (Becton Dickinson, San Jose, CA).

Statistical analyses The data normality (Kolmogorov–Smirnov normality test) and homogeneity of variances (Levene’s test) was analysed. If both assumptions were verified we performed a single-factor ANOVA (P ⱕ 0.05). When significant differences were found, a Fisher’s least significant difference test (LSD, P ⱕ 0.05) was conducted to determine which treatment differed from the control group. The software Statistica v9 (StatSoft, Inc. 2010, Tulsa, USA) was used for all analyses.

In vivo experiments Six- to 8-week-old female BALB/cJ mice, bred at the SPF animal facility of the Transgenic and Experimental Animal Unit of Institut Pasteur de Montevideo (IPMon), were housed in individual ventilated cages with negative pressure (Sealsafe rack, Tecniplast, Milano, Italy), at 20 ⫾ 1°C, a relative humidity of 40–60% and in a 14/10 light–dark cycle. Food and water were administered ad libitum. All animal procedures were approved by IPMon Animal Care Committee and were in accordance with national law and international guidelines of FELASA regarding laboratory animal’s protocols. Six mice per group were infected by a single intraperitoneal injection of 104 parasites suspended in 0.3 ml of fresh HMI-9 medium. Where appropriate, 1 mg ml-1 oxytetracycline (oxytet) was added to the drinking water starting at day 3 prior to infection and freshly every 48 h. Six experimental groups were studied: animals infected with WT parasites without (group 1) and with (group 2) oxytet in the drinking water, animals infected with cKO parasites without (group 3) or with (group 4) oxytet, animals infected with cKO parasites that were cultured for 72 h in the presence of oxytet to pre-induce expression of LipDH-Myc2 and then kept with oxytet in the drinking water (group 5) and animals infected with lipdh-/parasites and without oxytet (group 6). The health status of the animals was monitored on a daily basis. Parasitaemia levels were determined regularly as described in Supporting information (Fig. S2). Mice showing an impaired health status and/or with a parasite load ⱖ 108 cells per ml of blood were euthanized.

Acknowledgements We would like to thank Natalie Dirdjaja and Kathrin Diederich for their help with the cloning and purification of the recombinant LipDH. The technical assistance of BSc Gabriel Fernandez (Transgenic and Experimental Animal Unit, Institut Pasteur de Montevideo) during the infection experiments is gratefully acknowledged. Dr. André Schneider, Bern is kindly acknowledged for helpful discussions on mitochondrial protein biosynthesis. This work was supported by the Deutsche Forschungsgemeinschaft (SFB 544, project B3 and Kr 1242/1-2 to L.K.-S.). A.R. is a fellow of the Hartmut HoffmannBerling International Graduate School of Molecular and Cellular Biology (HBIGS) of Heidelberg University.

References Abramoff, M.D., Magelhaes, P.J., and Ram, S.J. (2004) Image processing with ImageJ. Biophotonics Int 11: 36–42. Aussel, C., Cynober, L., Lioret, N., Coudray-Lucas, C., Vaubourdolle, M., Saizy, R., and Giboudeau, J. (1986) Plasma branched-chain keto acids in burn patients. Am J Clin Nutr 44: 825–831. Berriman, M., Ghedin, E., Hertz-Fowler, C., Blandin, G., Renauld, H., Bartholomeu, D.C., et al. (2005) The genome of the African trypanosome Trypanosoma brucei. Science 309: 416–422. Biebinger, S., Wirtz, L.E., Lorenz, P., and Clayton, C. (1997) Vectors for inducible expression of toxic gene products in bloodstream and procyclic Trypanosoma brucei. Mol Biochem Parasitol 85: 99–112. Blount, B.C., Mack, M.M., Wehr, C.M., MacGregor, J.T., Hiatt, R.A., Wang, G., et al. (1997) Folate deficiency causes uracil misincorporation into human DNA and chromosome breakage: implications for cancer and neuronal damage. Proc Natl Acad Sci USA 94: 3290–3295. Blum, J.J. (1991) Oxidation of leucine by Leishmania donovani. J Protozool 38: 527–531. Blumenstiel, K., Schöneck, R., Yardley, V., Croft, S.L., and Krauth-Siegel, R.L. (1999) Nitrofuran drugs as common subversive substrates of Trypanosoma cruzi lipoamide dehydrogenase and trypanothione reductase. Biochem Pharmacol 58: 1791–1799. Bochud-Allemann, N., and Schneider, A. (2002) Mitochondrial substrate level phosphorylation is essential for growth of procyclic Trypanosoma brucei. J Biol Chem 277: 32849– 32854. Böhringer, S., and Hecker, H. (1975) Quantitative ultrastructural investigations of the life cycle of Trypanosoma brucei: a morphometric analysis. J Protozool 22: 463–467. Breidbach, T., Ngazoa, E., and Steverding, D. (2002) Trypanosoma brucei: in vitro slender-to-stumpy differentiation of culture-adapted, monomorphic bloodstream forms. Exp Parasitol 101: 223–230. Brosnan, J.T., and Brosnan, M.E. (2006) Branched-chain amino acids: enzyme and substrate regulation. J Nutr 136: 207S–211S. Bryk, R., Arango, N., Venugopal, A., Warren, J.D., Park, Y.H., Patel, M.S., et al. (2010) Triazaspirodimethoxybenzoyls as selective inhibitors of mycobacterial lipoamide dehydrogenase. Biochemistry 49: 1616–1627. Castillo-Acosta, V.M., Estévez, A.M., Vidal, A.E., Ruiz-Perez, L.M., and González-Pacanowska, D. (2008) Depletion of dimeric all-alpha dUTPase induces DNA strand breaks and impairs cell cycle progression in Trypanosoma brucei. Int J Biochem Cell Biol 40: 2901–2913. Ceylan, S., Seidel, V., Ziebart, N., Berndt, C., Dirdjaja, N., and Krauth-Siegel, R.L. (2010) The dithiol glutaredoxins of African trypanosomes have distinct roles and are closely linked to the unique trypanothione metabolism. J Biol Chem 45: 35224–35237. Chello, P.L., and Jaffe, J.J. (1972) Comparative properties of trypanosomal and mammalian thymidine kinases. Comp Biochem Physiol B 43: 543–562. Chuang, D.T., Chuang, J.L., and Wynn, R.M. (2006) Lessons from genetic disorders of branched-chain amino acid metabolism. J Nutr 136: 243S–249S.

© 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

638 A. Roldán, M. A. Comini, M. Crispo and R. L. Krauth-Siegel 䊏

Clarke, S.J., Farrugia, D.C., Aherne, G.W., Pritchard, D.M., Benstead, J., and Jackman, A.L. (2000) Balb/c mice as a preclinical model for raltitrexed-induced gastrointestinal toxicity. Clin Cancer Res 6: 285–296. Colasante, C., Alibu, V.P., Kirchberger, S., Tjaden, J., Clayton, C., and Voncken, F. (2006) Characterization and developmentally regulated localization of the mitochondrial carrier protein homologue MCP6 from Trypanosoma brucei. Eukaryot Cell 5: 1194–1205. Comini, M.A., Rettig, J., Dirdjaja, N., Hanschmann, E.M., Berndt, C., and Krauth-Siegel, R.L. (2008) Monothiol glutaredoxin-1 is an essential iron-sulfur protein in the mitochondrion of African trypanosomes. J Biol Chem 283: 27785–27798. Coustou, V., Biran, M., Breton, M., Guegan, F., Rivière, L., Plazolles, N., et al. (2008) Glucose-induced remodeling of intermediary and energy metabolism in procyclic Trypanosoma brucei. J Biol Chem 283: 16342–16354. Cristodero, M., Seebeck, T., and Schneider, A. (2010) Mitochondrial translation is essential in bloodstream forms of Trypanosoma brucei. Mol Microbiol 78: 757–769. Cunningham, M.P., and Vickerman, K. (1962) Antigenic analysis in the Trypanosoma brucei group, using the agglutination reaction. Trans R Soc Trop Med Hyg 56: 48– 59. Danson, M.J., Conroy, K., McQuattie, A., and Stevenson, K.J. (1987) Dihydrolipoamide dehydrogenase from Trypanosoma brucei. Characterization and cellular location. Biochem J 243: 661–665. Durieux, P.O., Schütz, P., Brun, R., and Köhler, P. (1991) Alterations in Krebs cycle enzyme activities and carbohydrate catabolism in two strains of Trypanosoma brucei during in vitro differentiation of their bloodstream to procyclic stages. Mol Biochem Parasitol 45: 19–27. Else, A.J., Hough, D.W., and Danson, M.J. (1993) Cloning, sequencing, and expression of Trypanosoma brucei dihydrolipoamide dehydrogenase. Eur J Biochem 212: 423– 429. Else, A.J., Clarke, J.F., Willis, A., Jackman, S.A., Hough, D.W., and Danson, M.J. (1994) Dihydrolipoamide dehydrogenase in the Trypanosoma subgenus, trypanozoon. Mol Biochem Parasitol 64: 233–239. Fenn, K., and Matthews, K.R. (2007) The cell biology of Trypanosoma brucei differentiation. Curr Opin Microbiol 10: 539–546. Ginger, M.L., Prescott, M.C., Reynolds, D.G., Chance, M.L., and Goad, L.J. (2000) Utilization of leucine and acetate as carbon sources for sterol and fatty acid biosynthesis by Old and New World Leishmania species, Endotrypanum monterogeii and Trypanosoma cruzi. Eur J Biochem 267: 2555–2566. Goldshmidt, H., Matas, D., Kabi, A., Carmi, S., Hope, R., and Michaeli, S. (2010) Persistent ER stress induces the spliced leader RNA silencing pathway (SLS), leading to programmed cell death in Trypanosoma brucei. PLoS Pathog 6: e1000731. Gutiérrez-Correa, J. (2010) Trypanosoma cruzi dihydrolipoamide dehydrogenase as target of reactive metabolites generated by cytochrome c/hydrogen peroxide (or linoleic acid hydroperoxide)/phenol systems. Free Radic Res 44: 1345–1358.

Hammarton, T.C. (2007) Cell cycle regulation in Trypanosoma brucei. Mol Biochem Parasitol 153: 1–8. Jackman, S.A., Hough, D.W., Danson, M.J., Stevenson, K.J., and Opperdoes, F.R. (1990) Subcellular localisation of dihydrolipoamide dehydrogenase and detection of lipoic acid in bloodstream forms of Trypanosoma brucei. Eur J Biochem 193: 91–95. Kasinski, A., Doering, C.B., and Danner, D.J. (2004) Leucine toxicity in a neuronal cell model with inhibited branched chain amino acid catabolism. Brain Res Mol Brain Res 122: 180–187. Lamour, N., Rivière, L., Coustou, V., Coombs, G.H., Barrett, M.P., and Bringaud, F. (2005) Proline metabolism in procyclic Trypanosoma brucei is down-regulated in the presence of glucose. J Biol Chem 280: 11902–11910. Li, Y., Holmes, W.B., Appling, D.R., and RajBhandary, U.L. (2000) Initiation of protein synthesis in Saccharomyces cerevisiae mitochondria without formylation of the initiator tRNA. J Bacteriol 182: 2886–2892. Lohrer, H., and Krauth-Siegel, R.L. (1990) Purification and characterization of lipoamide dehydrogenase from Trypanosoma cruzi. Eur J Biochem 194: 863–869. Misaka, E., Kawahara, Y., and Nakanishi, K. (1965) Studies on menadione reductase of bakers’ yeast. III. The identity of menadione reductase with lipoamide dehydrogenase. J Biochem 58: 436–443. Murta, S.M., Nogueira, F.B., Dos Santos, P.F., Campos, F.M., Volpe, C., Liarte, D.B., et al. (2008) Differential gene expression in Trypanosoma cruzi populations susceptible and resistant to benznidazole. Acta Trop 107: 59–65. Murta, S.M., Vickers, T.J., Scott, D.A., and Beverley, S.M. (2009) Methylene tetrahydrofolate dehydrogenase/ cyclohydrolase and the synthesis of 10-CHO-THF are essential in Leishmania major. Mol Microbiol 71: 1386– 1401. Nolan, D.P., and Voorheis, H.P. (1992) The mitochondrion in bloodstream forms of Trypanosoma brucei is energized by the electrogenic pumping of protons catalysed by the F1Fo-ATPase. Eur J Biochem 209: 207–216. Ochsenreiter, T., Cipriano, M., and Hajduk, S.L. (2008) Alternative mRNA editing in trypanosomes is extensive and may contribute to mitochondrial protein diversity. PLoS ONE 3: e1566. Opperdoes, F.R., Baudhuin, P., Coppens, I., De Roe, C., Edwards, S.W., Weijers, P.J., and Misset, O. (1984) Purification, morphometric analysis, and characterization of the glycosomes (microbodies) of the protozoan hemoflagellate Trypanosoma brucei. J Cell Biol 98: 1178–1184. Petrat, F., Paluch, S., Dogruöz, E., Dörfler, P., Kirsch, M., Korth, H.G., et al. (2003) Reduction of Fe(III) ions complexed to physiological ligands by lipoyl dehydrogenase and other flavoenzymes in vitro: implications for an enzymatic reduction of Fe(III) ions of the labile iron pool. J Biol Chem 278: 46403–46413. Pino, P., Aeby, E., Foth, B.J., Sheiner, L., Soldati, T., Schneider, A., and Soldati-Favre, D. (2010) Mitochondrial translation in absence of local tRNA aminoacylation and methionyl tRNA Met formylation in Apicomplexa. Mol Microbiol 76: 706–718. Ramos, E.I., Garza, K.M., Krauth-Siegel, R.L., Bader, J., Martinez, L.E., and Maldonado, R.A. (2009) 2,3-Diphenyl© 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639

Trypanosoma brucei lipoamide dehydrogenase 639

1,4-naphthoquinone: a potential chemotherapeutic agent against Trypanosoma cruzi. J Parasitol 95: 461–466. Richterová, L., Vávrová, Z., and Lukeš, J. (2011) DEAD-box RNA helicase is dispensable for mitochondrial translation in Trypanosoma brucei. Exp Parasitol 127: 300–303. Salmon-Chemin, L., Buisine, E., Yardley, V., Kohler, S., Debreu, M.A., Landry, V., et al. (2001) 2- and 3-substituted 1,4-naphthoquinone derivatives as subversive substrates of trypanothione reductase and lipoamide dehydrogenase from Trypanosoma cruzi: synthesis and correlation between redox cycling activities and in vitro cytotoxicity. J Med Chem 44: 548–565. Schaer, J.C., Maurer, U., and Schindler, R. (1978) Determination of thymidine in serum used for cell culture media. Exp Cell Biol 46: 1–10. Schlecker, T., Schmidt, A., Dirdjaja, N., Voncken, F., Clayton, C., and Krauth-Siegel, R.L. (2005) Substrate specificity, localization, and essential role of the glutathione peroxidase-type tryparedoxin peroxidases in Trypanosoma brucei. J Biol Chem 280: 14385–14394. Schnaufer, A., Clark-Walker, G.D., Steinberg, A.G., and Stuart, K. (2005) The F1–ATP synthase complex in bloodstream stage trypanosomes has an unusual and essential function. EMBO J 24: 4029–4040. Schöneck, R., Billaut-Mulot, O., Numrich, P., Ouaissi, M.A., and Krauth-Siegel, R.L. (1997) Cloning, sequencing and functional expression of dihydrolipoamide dehydrogenase from the human pathogen Trypanosoma cruzi. Eur J Biochem 243: 739–747. Scott, D.A., Hickerson, S.M., Vickers, T.J., and Beverley, S.M. (2008) The role of the mitochondrial glycine cleavage complex in the metabolism and virulence of the protozoan parasite Leishmania major. J Biol Chem 283: 155–165. Sienkiewicz, N., Jaroslawski, S., Wyllie, S., and Fairlamb, A.H. (2008) Chemical and genetic validation of dihydrofolate reductase-thymidylate synthase as a drug target in African trypanosomes. Mol Microbiol 69: 520–533. Sreider, C.M., Grinblat, L., and Stoppani, A.O. (1992) Reduction of nitrofuran compounds by heart lipoamide dehydrogenase: role of flavin and the reactive disulfide groups. Biochem Int 28: 323–334. Stephens, J.L., Lee, S.H., Paul, K.S., and Englund, P.T. (2007) Mitochondrial fatty acid synthesis in Trypanosoma brucei. J Biol Chem 282: 4427–4436. Tan, T.H., Bochud-Allemann, N., Horn, E.K., and Schneider, A. (2002) Eukaryotic-type elongator tRNAMet of Trypanosoma brucei becomes formylated after import into mitochondria. Proc Natl Acad Sci USA 99: 1152–1157. Tasker, M., Wilson, J., Sarkar, M., Hendriks, E., and Matthews, K. (2000) A novel selection regime for differentiation

defects demonstrates an essential role for the stumpy form in the life cycle of the African trypanosome. Mol Biol Cell 11: 1905–1917. Thiel, M., Harder, S., Wiese, M., Kroemer, M., and Bruchhaus, I. (2008) Involvement of a Leishmania thymidine kinase in flagellum formation, promastigote shape and growth as well as virulence. Mol Biochem Parasitol 158: 152–162. Tielens, A.G., and Van Hellemond, J.J. (1998) Differences in energy metabolism between trypanosomatidae. Parasitol Today 14: 265–272. Tyler, K.M., Matthews, K.R., and Gull, K. (1997) The bloodstream differentiation-division of Trypanosoma brucei studied using mitochondrial markers. Proc Biol Sci 264: 1481–1490. Vassella, E., Reuner, B., Yutzy, B., and Boshart, M. (1997) Differentiation of African trypanosomes is controlled by a density sensing mechanism which signals cell cycle arrest via the cAMP pathway. J Cell Sci 110: 2661–2671. Vickers, T.J., Murta, S.M., Mandell, M.A., and Beverley, S.M. (2009) The enzymes of the 10-formyl-tetrahydrofolate synthetic pathway are found exclusively in the cytosol of the trypanosomatid parasite Leishmania major. Mol Biochem Parasitol 166: 142–152. van Weelden, S.W., Fast, B., Vogt, A., van der Meer, P., Saas, J., van Hellemond, J.J., et al. (2003) Procyclic Trypanosoma brucei do not use Krebs cycle activity for energy generation. J Biol Chem 278: 12854–12863. van Weelden, S.W., van Hellemond, J.J., Opperdoes, F.R., and Tielens, A.G. (2005) New functions for parts of the Krebs cycle in procyclic Trypanosoma brucei, a cycle not operating as a cycle. J Biol Chem 280: 12451–12460. Werner, C., Stubbs, M.T., Klebe, G., and Krauth-Siegel, R.L. (2002) The crystal structure of lipoamide dehydrogenase from Trypanosoma cruzi: a target for the design of new drugs against Chagas’ disease. In Flavins and Flavoproteins. Chapman, S., Perham, R., and Scrutton, N. (eds). Berlin: Dr. Rudolf Weber Agency for Scientific Publications, pp. 917–922.

Supporting information Additional supporting information may be found in the online version of this article. Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

© 2011 Blackwell Publishing Ltd, Molecular Microbiology, 81, 623–639