ATP generation in the Trypanosoma brucei procyclic form: cytosolic substrate level is essential, but not oxidative phosphorylation.

Trypanosoma brucei is a parasitic protist responsible for sleeping sickness in humans. The procyclic form of this parasite, transmitted by tsetse flies, is considered to be dependent on oxidative phosphorylation for ATP production. Indeed, its respiration was 55% inhibited by oligomycin, which is the most specific inhibitor of the mitochondrial F0/F1-ATP synthase. However, a 10-fold excess of this compound did not significantly affect the intracellular ATP concentration and the doubling time of the parasite was only 1.5-fold increased, suggesting that oxidative phosphorylation is not essential for procyclic trypanosomes. To further investigate the sites of ATP production, we studied the role of two ATP producing enzymes, which are involved in the synthesis of pyruvate from phosphoenolpyruvate: the glycosomal pyruvate phosphate dikinase (PPDK) and the cytosolic pyruvate kinase (PYK). The parasite was not affected by PPDK gene knockout. In contrast, inhibition of PYK expression by RNA interference was lethal for these cells. In the absence of PYK activity, the intracellular ATP concentration was reduced by up to 2.3-fold, whereas the intracellular pyruvate concentration was not reduced. Furthermore, we show that this mutant cell line still excreted acetate from d-glucose metabolism, and both the wild type and mutant cell lines consumed pyruvate present in the growth medium with similar high rates, indicating that in the absence of PYK activity pyruvate is still present in the trypanosomes. We conclude that PYK is essential because of its ATP production, which implies that the cytosolic substrate level phosphorylation is essential for the growth of procyclic trypanosomes.

FIG. 1. Schematic representation of the D-glucose, L-proline, L-threonine, L-glutamine, and pyruvate metabolisms in the procyclic form of T. brucei. The carbon sources used by the procyclic cells grown under standard conditions, i.e. SDM-79/FCS medium, are bold and boxed on a gray background. The number above or below each carbon source represents the percentage by which it, among all carbon sources, is consumed However, it was commonly assumed that succinate is a byproduct of the mitochondrial tricarboxylic acid cycle (12,13). Similarly, CO 2 was considered to be primarily produced by the tricarboxylic acid cycle (12,17). L-Alanine seems to be produced from pyruvate by the alanine aminotransferase (12) but the origin of lactate is not clear. Notwithstanding the debate on end products, the view that ATP production in the procyclic form of T. brucei depended on oxidative phosphorylation, with the respiratory chain fed primarily by reducing equivalents produced by the tricarboxylic acid cycle, was widely held (12,13,17,19).
However, during this last year three independent reports have led to an important revision of this model. First, Bochud-Allemann and Schneider (20) showed, by RNAi inactivation of succinyl-CoA synthetase expression, that mitochondrial substrate level phosphorylation is essential for growth of the parasite (20). Second, we showed that the excreted succinate is produced by a glycosomal NADH-dependent fumarate reductase, which together with the glycosomal malate dehydrogenase, is involved in the maintenance of the glycosomal redox balance (21). Furthermore, in the course of this analysis, we detected a significant PYK activity in the cytosol of the T. brucei procyclic form (21) (PYK activity was previously considered as negligible (12)). Therefore, we proposed that pyruvate is produced by the cytosolic PYK (21) and probably the glycosomal pyruvate phosphate dikinase (PPDK) (22), instead of the cytosolic malic enzyme (12) (Fig. 1). Finally, Van Weelden et al. (23) showed that the tricarboxylic acid cycle enzyme aconitase is not essential for growth and that the acetyl-CoA produced by glycolysis is not metabolized through the tricarboxylic acid cycle (23).
Here we add another important piece to the jigsaw puzzle of energy metabolism in procyclic T. brucei by investigating the role of PYK, PPDK, and oxidative phosphorylation in the generation of ATP in these cells. It has been well established that CHME-5 human microglia cells depend on oxidative phosphorylation for ATP production (24), thus we compared some of the metabolic characteristics of these cells to those of procyclic T. brucei. We observed that oxidative phosphorylation is used, but is not essential for parasite growth, suggesting that it is not the main pathway to produce ATP, under the growth conditions used. Inactivation of PYK expression, by RNA interference, was shown to be lethal for the cells and several lines of evidence indicate that PYK is essential for its ATP production. This clearly shows that cytosolic substrate level phosphorylation is essential for growth of the procyclic form of T. brucei.

EXPERIMENTAL PROCEDURES
Cell Cultures-The human microglial cell line CHME-5 was cultured as a monolayer in 100-mm diameter Petri dishes, in minimal essential medium, supplemented with 20 mM glucose, 4 mM glutamine, 45 mM sodium carbonate, penicillin (100 units⅐ml Ϫ1 ), streptomycin (0.1 mg⅐ ml Ϫ1 ), and 10% heat-inactivated fetal calf serum (24). Cultures were carried out for 5 days in a water-saturated incubator with a 9% CO 2 atmosphere. For the respiratory experiments microglia growth was performed on collagen beads (Cultispher-GL, Percell Biolytica AB, Sweden). The cell suspension and the beads were mixed at least 24 h before the test, at a ratio of 3-5 ϫ 10 6 cells per ml of beads. The procyclic form of T. brucei EATRO1125 was cultured at 27°C in SDM-79 medium containing 10% (v/v) heat-inactivated fetal calf serum and 3.5 mg⅐ml Ϫ1 hemin (SDM-79/FCS) (25). Alternatively, cells were grown in a modified version of the SDM-79/FCS medium, containing 1.6 mM L-glutamine, 4 mM pyruvate, 11 mM D-glucose, and 8 mM L-threonine.
Inactivation of the ppdk Gene by Gene Knockout-To construct the p⌬1ppdk plasmids, the DNA fragments flanking the hygromycin resistance gene and the T7 RNA polymerase gene of the pHD328 plasmid (26) were replaced by the 5Ј-and 3Ј-untranslated regions flanking the ppdk gene (5Ј-and 3Ј-UTR) (Fig. 2). The 5Ј-UTR fragment (616 bp) was generated by PCR using as primers the ppdk-5Ј5Ј (5Ј-ATGCGGGCGG-CCGCCAAGCCACATGCAGAAAA-3Ј; the NotI restriction site is underlined) and ppdk-3Ј5Ј (5Ј-CACTTTGCGCGctcgagTGTGTATAGTGG-ATGCTG-3Ј; the BssHII and XhoI restriction sites are underlined and in lower cases, respectively) oligonucleotides and as template the Cos8 cosmid, which contains the cloned ppdk gene and its flanking regions (22). The NotI/BssHII-digested PCR fragment was cloned into the NotI/MluI-digested pHD328 plasmid (BssHII and MluI have compatible ends), to generate the pDH5Ј plasmid. The 3Ј-UTR fragment (660 bp) was generated by PCR using as template the Cos8 cosmid and as primers the ppdk-5Ј3Ј (5Ј-GCCAAGAGGCCTAAAGGACGTGGGAGG-GGA-3Ј; the StuI restriction site is underlined) and ppdk-3Ј3Ј (5Ј-CA-CTAAGCTAGCgcggccgcCGTGATTCTTTCACCCCGA-3Ј; the NheI and NotI restriction sites are underlined and in lower cases, respectively) oligonucleotides. The StuI/NheI-digested PCR fragment was cloned into StuI/NheI-digested pHD5Ј plasmid to produce the p⌬1ppdk plasmid. The p⌬2ppdk plasmid was generated by replacing the XhoI/StuI fragment of the p⌬1ppdk plasmid, which encodes the hygromycin resistance gene and the T7 RNA polymerase gene, by the XhoI/StuI fragment of the pLew114 plasmid (27), which encodes the neomycin resistance gene and the tetracycline repressor gene under the control of the T7 RNA polymerase promoter. The p⌬1ppdk and p⌬2ppdk plasmids were linearized by NotI prior to parasite transfection. The procyclic EATRO1125 strain was sequentially transfected with the NotI-digested p⌬1ppdk and p⌬2ppdk plasmids and selection of hygromycin-resistant and hygromycin/neomycin-resistant clones, respectively, was performed as previously reported (4,22).
Expression of an Anti-PYK Double Stranded RNA-To inhibit PYK expression by RNAi (2), we generated in the pLew79 expression vector (kindly provided by E. Wirtz and G. A. M. Cross) (27) a "sense/antisense" cassette that specifically targets the PYK gene (28) (Fig. 2). The 3Ј-end of the PYK gene was targeted, from position 1088 to 1519 bp (the ATG start and stop codons, of the PYK gene, are at positions 1 and 1500 bp, respectively) (28). The plasmid construction was performed as previously described (4,21). Briefly, a PCR-amplified 484 bp containing the antisense sequence (431 bp of targeted sequence plus 42 bp used as a spacer between the sense and antisense sequences) was inserted in the HindIII and BamHI restriction sites of the pLew79 plasmid. Then a PCR-amplified 442-bp fragment containing the sense sequence was inserted, upstream of the antisense sequence, in HindIII and XhoI restriction sites (XhoI was introduced at the 3Ј extremity of the antisense PCR fragment). The resulting plasmid, pLew79-⌬PYK, contained during the midlog phase (data not shown). Excreted products (acetate, L-alanine, succinate, L-glycine, and CO 2 ) are in white characters on a black background. The enzymatic reaction(s) leading to the production of lactate (from pyruvate and/or D-glucose), acetate (from L-threonine), and glutamate (from L-glutamine) are not known and are indicated by a question mark. The metabolic flux at each enzymatic step is tentatively represented by arrows with different thickness. Dashed arrows indicate steps that are supposed to occur at a background level or not at all, under the standard growth conditions. The glycosomal and mitochondrial compartments are indicated and the boxed numbers represent enzymes targeted for RNAi (pyruvate kinase) and gene knockout (pyruvate phosphate dikinase). The ATP molecules produced in the cytosol and the mitochondrion by substrate level phosphorylation and oxidative phosphorylation are bold and boxed. AA a chimeric construct composed of the sense and antisense version of a PYK gene fragment, separated by a 42-bp fragment, under the control of the tetracycline-inducible procyclin (PARP) promoter (27) (Fig. 2). The procyclic EATRO1125.T7T and ⌬1.2ppdk cell lines, which express the tetracycline repressor (4,27), were transfected with the NotI-digested pLew79-⌬PYK plasmid and selection of phleomycin-resistant cells was performed as previously reported (4,22).
Nuclear Magnetic Resonance (NMR) Experiments-4 ϫ 10 9 T. brucei procyclic cells were collected by centrifugation at 1,400 ϫ g for 10 min, washed once in PBS buffer, and incubated in 10 ml of incubation buffer (PBS buffer supplemented with 24 mM NaHCO 3 , pH 7.3) containing 110 mol of D-[1-13 C]glucose (11 mM) for 90 -180 min at 27°C. D-Glucose concentration in the medium was determined with the D-glucose Trinder kit (Sigma). The integrity of the cells during the incubation was checked by microscopic observation. After centrifugation for 10 min at 1,400 ϫ g, the supernatant was lyophilized, re-dissolved in 500 l of D 2 O, and 15 l of pure dioxane was added as an external reference. 13 C NMR spectra were collected at 125.77 MHz with a Bruker DPX500 spectrometer equipped with a 5-mm broad-band probe. Measurements were recorded at 25°C under bilevel broad-band gated proton decoupling and D 2 O lock. Acquisition conditions were: 90°flip angle, 22,150 Hz spectral width, 64,000 memory size, and 21.5 s total recycle time. Measurements were performed overnight with 2,048 scans. Spectra were collected after a 1 Hz exponential line broadening. The specific 13 C-enrichment of lactate (C3), acetate (C2), and succinate (C2 and C3) was determined from 1 H-observed/ 13 C-edited NMR ( 1 H/ 13 C NMR) spectra acquired under 13 C-decoupling (32,33). The sequence enabled the successive acquisitions of a first scan corresponding to a standard spin-echo experiment without any 13 C-excitation and a second scan involving a 13 C-inversion pulse. Subtraction of two alternate scans resulted in the editing of 1 H spins coupled to 13 C spins with a scalar coupling constant J CH ϭ 127 Hz. 13 C-Decoupling during the acquisition collapsed the 1 H/ 13 C-coupling to a single 1 H resonance. Flip angles for rectangular pulses were carefully calibrated on both radiofrequency channels before each experiment. The relaxation delay was 8 s for a nearly complete longitudinal relaxation. The fractional 13 C-enrichment at selected metabolite carbon positions was calculated as the ratio of the area of a given resonance in the 1 H/ 13 C NMR spectrum to its area in the standard spin-echo spectrum. The relative errors in the 13 C-enrichment determinations were Ͻ5%. 3 The amount of excreted products was calculated on the basis of the specific 13 C-enrichment values and the carbon-13 content for each metabolite at the position of interest, using the 13 C-enriched C1 glucose as quantitative reference.
Enzymatic Assays-Sonicated (5 s at 4°C) crude extracts of trypanosomes resuspended in cold hypotonic buffer (10 mM potassium phosphate, pH 7.8) were tested for enzymatic activities. PYK (34) and glycerol-3-phosphate dehydrogenase (GPDH) (31) activities were measured at 340 nm via reduction of NAD ϩ or oxidation of NADH, respectively, according to published procedures. The PYK activity was measured in the presence of its activator (fructose 2,6-bisphosphate) to increase the sensitivity of the assay, as described (34).
Determination of Metabolite Concentrations-The intracellular ATP, PEP, and/or pyruvate concentrations were determined on established procyclic cells in mid-log growth phase or CHME-5 human microglial cells grown on microbeads. Cell pellets (1-2 ϫ 10 8 procyclic or 3-5 ϫ 10 6 CHME-5 human microglial cells) were washed in cold PBS and frozen in liquid nitrogen. Lysis and deproteinization of the cellular pellets involved homogenization in 500 l of cold perchloric acid (0.9 M) and neutralization (pH 6.5) by addition of KOH/MOPS (2/0.5 M). For ATP measurements, the firefly luciferase bioluminescence assay ("Quantitative ATP monitoring kit," ThermoLabsystems) was used (35). Pyruvate and PEP concentrations were determined by enzymatic conversion into L-lactate, with the oxidation of stochiometric amounts of NADH, as described before (36).
To determine the concentration of metabolites consumed or excreted by the EATRO1125 procyclic trypanosomes, the inoculum (10 6 cells⅐ ml Ϫ1 ) was grown in the SDM-79/FCS medium containing 1.6 mM Lglutamine, 4 mM pyruvate, 11 mM D-glucose, and 8 mM L-threonine, until the stationary phase was reached. Aliquots of the growth medium were collected twice a day for the measurements. The quantity of D-glucose present in the medium was determined using the "glucose Trinder kit" (Sigma). Pyruvate concentration was determined enzymatically, as previously described (36). The concentration of the 20 amino acids present in the medium was determined by chromatography on an automatic amino acid analyzer coupled to a computing integrator (Beckman), after deproteinization of the samples by perchloric acid treatment.
Measurement of Oxygen Consumption-For oxygen consumption measurements, all cells were resuspended in their culture medium (including heat-inactivated fetal calf serum) at a density in the range of 1-2 ϫ 10 6 cells⅐ml Ϫ1 (CHME-5 human microglial cells) or 0.5-2 ϫ 10 8 cells⅐ml Ϫ1 (T. brucei procyclic form). Oxygen uptake was measured polarographically with a Clark type electrode (final volumes of 1 or 2 3 M. Biran and P. Canioni, unpublished data.

Oxidative Phosphorylation Is Not Essential for Procyclic
Trypanosomes-To study the mitochondrial activity of procyclic T. brucei cells, we determined how oxygen consumption, ATP production, and cell growth were affected by effectors of mitochondrial metabolism (SHAM, KCN, CCCP, and oligomycin). This analysis was also performed on the CHME-5 human microglial cells, which depend on oxidative phosphorylation for ATP production (24). SHAM, KCN, and oligomycin are specific inhibitors of the alternative oxidase, cytochrome c oxidase, and F 0 /F 1 -ATP synthase, respectively, and CCCP is a protonophoric uncoupler that dissipates transmembrane proton gradients (37,38). In eukaryotic cells the mitochondrial oligomycin-sensitive F 0 /F 1 -ATP synthase uses the proton gradient generated by the respiratory chain to produce ATP (39). This oxidative phosphorylation was illustrated here by the inhibition of cellular oxygen consumption upon the addition of an excess of oligomycin (40 and 78% inhibition in trypanosome and human cell lines, with 5 and 1.5 g of oligomycin⅐mg Ϫ1 protein, respectively) (Fig. 3A). This inhibition of respiration is because of an increase of the mitochondrial membrane proton gradient and confirms that the proton gradient generated by the respiratory chain is used by the F 0 /F 1 -ATP synthase to produce ATP in procyclic trypanosome as well as in human microglial cell lines (17,24,40). It is noteworthy that the oligomycin activity depends on the cell number (Fig. 3B). Indeed, 0.5 and 1.75 g of oligomycin inhibits completely the oligomycin-sensitive respiration of 5 ϫ 10 7 and 2 ϫ 10 8 T. brucei procyclic cells, which corresponds to 2.5 and 2.2 g of oligomycin⅐mg Ϫ1 protein, respectively. Addition of CCCP, after oligomycin treatment, induced a substantial increase in oxygen consumption, 2.8-and 5.7-fold for the trypanosome and human cells, respectively (Fig.  3A), because of the collapse of the proton gradient.
Respiration properties of procyclic trypanosomes and CHME-5 human microglial cells differ mainly by their sensitivity to SHAM (an inhibitor of mitochondrial alternative oxidase). Indeed, the oxygen consumption of the EATRO1125 procyclic cell line was inhibited by 60 Ϯ 15% in the presence of 0.5 mM SHAM, whereas the human microglial cell line is insensitive to SHAM, because of the absence of an alternative oxidase in mammalian cells.
To determine the effect of mitochondrial effectors on energy transduction, the intracellular ATP concentration was measured periodically over 90 or 120 min of incubation (Fig. 4). In the absence of extracellular carbon sources, the intracellular ATP concentration collapsed after an hour of incubation for both procyclic trypanosomes and human microglia cells. Similarly, the intracellular concentration of ATP in the CHME-5 human microglial cells was considerably reduced after 60 min of incubation with 1 mM KCN (80%) or 5 g of oligomycin⅐mg Ϫ1 protein (60%) (Fig. 4A), which indicates that most of the ATP produced by this cell line depends on oxidative phosphorylation, as previously observed (24). In contrast, these mitochondrial effectors have little or no effect on the steady state ATP concentration in the EATRO1125 procyclic cells (Fig. 4B). Indeed, intracellular ATP was reduced by ϳ20% after 20 min of treatment with 1 mM KCN or 0.5 mM SHAM, and reduced by 35% in the presence of both KCN (1 mM) and SHAM (0.5 mM) (data not shown). More interestingly, no effect was observed following the addition of 6, 12, and 24 g of oligomycin⅐mg Ϫ1 protein (Fig. 4B), which corresponds to up to 10-fold the concentration that completely inhibits the oligomycin-sensitive respiration (Fig. 3B). Furthermore, equivalent oligomycin concentrations (1 to 12 g⅐mg Ϫ1 protein) only had a modest effect on the growth of the parasite (doubling time: 18 versus 12 h for the wild type cells) (Fig. 5). In the course of this 8-day experiment, the intracellular ATP concentration was not affected by oligomycin treatment (data not shown). A very large excess of oligomycin (ϳ1 mg⅐mg Ϫ1 protein) kills 100% of the procyclic trypanosomes after 1 day of treatment (data not shown), probably as a consequence of a nonspecific effect (41). It is noteworthy that the 35% reduction of intracellular ATP concentration, in the presence of SHAM and KCN, is probably the indirect consequence of the complete inhibition of all mitochondrial electron transport activity that appears to be essential for the parasite.
These data indicate that inhibition of the F 0 /F 1 -ATP synthase does not significantly affect ATP levels in procyclic trypanosomes. Other routes such as substrate level phosphorylation, may then be used to generate ATP. To test this hypothesis, we studied the role of two glycolytic enzymes that produce ATP, from PEP, in the glycosomes (PPDK (22)) or in the cytosol (PYK (21)) (see Fig. 1).
PPDK Is Not Essential for Procyclic Cells-PPDK is a glycosomal enzyme present in all trypanosomatids analyzed so far, which produces pyruvate, ATP, and inorganic phosphate from PEP, AMP, and pyrophosphate (22). To study the role of this enzyme, we replaced both PPDK alleles with two DNA fragments (p⌬1ppdk and p⌬2ppdk) through homologous recombination using the upstream and downstream PPDK UTRs flanking these fragments (Fig. 2). The EATRO1125 procyclic form of T. brucei was sequentially transfected with p⌬1ppdk and p⌬2ppdk to generate the ⌬1ppdk (⌬ppdk::HYG-T7RNApol) and ⌬1.2ppdk (⌬ppdk::HYG-T7RNApol/⌬ppdk::NEO-TetR) cell lines. The ⌬1ppdk cell line, which contains a single PPDK copy, showed a decrease of the PPDK signal by Western blot analysis, which is correlated with a 47% decrease of the PPDK activity, as compared with the wild type EATRO1125.T7T cell line (Fig. 6). For the ⌬1.2ppdk cell line, in which both PPDK alleles are deleted, no PPDK is detectable by activity assays nor by Western blot (Fig. 6) and immunofluorescence analyses (data not shown). The doubling time of the ⌬1.2ppdk cell line is not affected (Fig. 7), indicating that PPDK is not essential for cell viability under the standard growth conditions. However, the rate of D-glucose consumption is slightly reduced in the ⌬1.2ppdk cell line, as compared with the wild type cells (1.0 versus 1.3 mol⅐h Ϫ1 ⅐mg Ϫ1 protein), which may be an adaptation to the PPDK gene knockout.
To further study the role of PPDK in glucose metabolism of procyclic trypanosomes, we used carbon-13 NMR ( 13 C NMR) spectroscopy to compare the metabolic end products excreted by the ⌬1.2ppdk and untransfected cell lines fed with D-[1-13 C]glucose. The parasites were incubated in PBS/NaHCO 3 medium containing 110 mol of D-[1-13 C]glucose, as the only carbon source, until 30 -60 mol of D-glucose were consumed by each cell line. The incubation medium was then analyzed by NMR spectroscopy. To determine the [ 13 C] amount in each excreted end product, the analysis was performed on fully relaxed spectra. For each of the T. brucei procyclic cell lines (wild type and mutants) analyzed by this approach, we calculated that approximately two molecules are excreted (out of the products succinate, acetate, lactate, malate, and fumarate) per molecule of glucose consumed (21). For convenience, we have expressed the quantity of individual excreted molecules as the percentage of all of the excreted end products (Table I). The wild type procyclic form excretes mainly succinate (69.4% of all the excreted 13 C-enriched molecules), acetate (19.4%), and lactate (9.5%), with traces of malate (1.3%) and fumarate (0.4%) ( Fig. 8A and Table I). The 13 C-enriched end products of glucose metabolism excreted by the wild type and ⌬1.2ppdk cell lines are in the same range, except the mutant cell line ⌬1.2ppdk, which shows a 3.8-fold reduction of lactate (9.5% versus 2.5%) ( Table I). In conclusion, ppdk gene knockout does not significantly affect D-glucose metabolism nor the doubling time of the procyclic trypanosomes. This indicates that the involvement of PPDK in ATP production is not essential. However, in the absence of PPDK, another ATP producing activity (yet to be determined) may compensate for this loss in the ⌬1.2ppdk cell line.
PYK Is Essential in Procyclic Trypanosomes-According to the model recently proposed (Fig. 1), PYK may play a role in glucose metabolism in procyclic cells (21). Indeed, we showed that a significant PYK activity is present in the cytosol of the procyclic form, suggesting that PYK can produce pyruvate from cytosolic PEP. To study the role of PYK, we inactivated the expression of its gene by RNAi. To inhibit PYK expression by RNAi in the procyclic form of T. brucei, the pLew79 vector (27) was used to express double stranded RNA molecules containing linked sense and antisense copies of the targeted PYK sequence, under control of a tetracycline-inducible promoter (Fig.  2). The recombinant plasmid (pLew79-⌬PYK) was inserted into the rDNA spacer of the EATRO1125.T7T cell line expressing the tetracycline repressor (4,27), to generate the ⌬PYK cell line. The PYK activity and the doubling time of the tetracycline-induced (⌬PYK.i) or non-induced (⌬PYK.ni) ⌬PYK cell line were determined. The PYK activity of the ⌬PYK.ni cell line is reduced as compared with the wild type trypanosomes (Fig.  7), whereas the double stranded RNA should not be expressed in the absence of tetracycline. This is probably because of a leakage of the tetracycline-repressor control, as previously described (4,27). Interestingly, the reduction of the PYK activity (2.3-fold) is directly correlated with the increase of the doubling time (2.4-fold) observed for the ⌬PYK.ni cell line (Fig. 7). The importance of PYK for cell growth was confirmed by inducing A) and procyclic T. brucei (panel B). Monolayer CHME-5 human microglia cells in Petri dishes (ϳ4 ϫ 10 6 cells) were incubated at 37°C in 10 ml of minimal essential medium in the absence or presence of oligomycin (1.5 g⅐ml Ϫ1 ) or KCN (1 mM). The EATRO1125 procyclic cells (4 ϫ 10 7 cells⅐ml Ϫ1 ) were incubated at 27°C in 10 ml of SDM-79/FCS medium in the absence or presence of oligomycin (1, 2, or 4 g⅐ml Ϫ1 ), SHAM (0.5 mM), or KCN (1 mM). The intracellular amount of ATP was determined periodically in the cells from one Petri dish (panel A) or on 2-ml samples (panel B), as described under "Experimental Procedures." The cells incubated in PBS/dialyzed FCS (10%) showed an amount of intracellular ATP reduced by 90% after 90 min of incubation in the absence of carbon source, indicating that the intracellular ATP pool cannot be maintained, in the absence of extracellular carbon sources. However, for the human microglia cells (panel A), the intracellular ATP concentration did not change significantly during 40 min of incubation, suggesting the contribution of their high intracellular phosphocreatine content to maintain the ATP steady state, as described for this cell type in response to heat stress situation (24). It is noteworthy that the steady state amount of ATP is stable (as observed for the curve, Ⅺ) when the same experiment is conducted on trypanosomes incubated in PBS/dialyzed FCS (10%) containing all the carbon sources (data not shown). The values represent the means of two to four independent experiments. the expression of double stranded RNA with tetracycline (⌬PYK.i). The complete inhibition of PYK expression after 6 days of tetracycline induction leads to cell death (Fig. 7), indicating that PYK is essential for cell viability, as previously observed (42).

FIG. 4. Effect of metabolic inhibitors on the steady state amount of intracellular ATP in human microglia cells (panel
Because, PYK and PPDK may compete for the same substrate (PEP) or be complementary because they have the same products (pyruvate and ATP), we determined the effect of the simultaneous loss of both enzymes. The ⌬1.2ppdk cell line, which expresses the tetracycline repressor, was transfected by the pLew79-⌬PYK plasmid to generate the ⌬1.2ppdk/⌬PYK cell line. The death of the ⌬1.2ppdk/⌬PYK.i trypanosomes, 6 days after tetracycline induction, correlated with the loss of the PYK activity, as observed for the ⌬PYK.i cell line (Fig. 7). We also observed a reduction of the PYK activity in the non-induced ⌬1.2ppdk/⌬PYK cell line (⌬1.2ppdk/⌬PYK.ni) as compared with the wild type cells. However, this substantial reduction of the PYK activity (3-fold) was associated with only a small increase in the doubling time (1.3-fold), as opposed to the ⌬PYK.i cell line (Fig. 7). Comparison of the ⌬1.2ppdk/⌬PYK.ni and ⌬PYK.ni cell lines, shows that the procyclic cells are less affected by the reduction of the PYK activity, when PPDK is not expressed. This indicates that the absence of PPDK compensates for the decrease of the PYK activity in the ⌬1.2ppdk/ ⌬PYK.ni mutant. Because PYK and PPDK may compete for the same PEP substrate, these data could suggest that the metabolic flux through the PYK pathway is increased in the absence of PPDK.
ATP Produced by PYK Is Essential for Procyclic Cells-PYK could be essential either for its role in ATP production or in pyruvate production. To distinguish which of these roles were essential, we conducted three different analyses. First, the metabolic end products excreted by the tetracycline-induced (⌬PYK.i and ⌬1.2ppdk/⌬PYK.i) or non-induced (⌬PYK.ni and ⌬1.2ppdk/⌬PYK.ni) ⌬PYK and ⌬1.2ppdk/⌬PYK cell lines, fed with D-[1-13 C]glucose as the only carbon source, were monitored by NMR spectroscopy, as described above (Fig. 8B and Table I). For the tetracycline-induced cells, this NMR analysis was conducted 5 days after addition of tetracycline, when the PYK activity reached a value Ͻ1 milliunit⅐mg Ϫ1 protein. At this stage, the rate of D-glucose consumption is not affected by the inhibition of PYK expression. Interestingly, the amount of excreted 13 C-enriched acetate is only reduced by a small amount in the absence of PYK activity (13.9% Ϯ 0.1 and 12.2% Ϯ 0.05 for the ⌬PYK.i and ⌬1.2ppdk/⌬PYK.i cell lines, respectively, versus 19.4% Ϯ 1.2 for the wild type cells) (Table  I). This implies that [ 13 C]pyruvate is produced, from D-[1-13 C]glucose, in the mutant cell lines. Indeed, under our experimental conditions (where D-[1-13 C]glucose is the only carbon source), excreted [ 13 C]acetate is produced from [ 13 C]pyruvate by the pyruvate dehydrogenase complex and acetate:succinyl-CoA transferase (see Fig. 1). In the absence of detectable PYK activity, succinate excretion is only slightly increased, whereas the amounts of excreted malate and fumarate are increased ϳ10and 4-fold, respectively (Table I), in both the ⌬PYK.i and ⌬1.2ppdk/⌬PYK.i cell lines. This NMR analysis suggests that, in the absence of PYK, the glycolytic flux is probably reoriented through PEPCK, which leads to an increase in intracellular malate concentration. Approximately half of the malate is converted to pyruvate by the malic enzyme. Pyruvate is then converted to acetate that is excreted along with the remaining half of the malate (13.9% Ϯ 0.1 excreted acetate versus 13.9% Ϯ 0.8 excreted malate in the ⌬PYK.i cell line and 12.2% Ϯ 0.05 versus 12.6% Ϯ 0.4 in the ⌬1.2ppdk/⌬PYK.i cell lines) (Table I).
Second, we observed that pyruvate (1 mM), as well as Lthreonine (2.5 mM) and L-glutamine (0.46 mM) present in the SDM-79/FCS medium commonly used for cultivation of the parasites, are rapidly consumed. The SDM-79/FCS medium was modified by increasing the pyruvate (4 mM final), L-glutamine (1.6 mM final), and L-threonine (8 mM final) concentrations (D-glucose concentration was also increased up to 11 mM). In the mid-log phase, the EATRO1125.T7T cell line consumes 0.8 mol of pyruvate⅐h Ϫ1 ⅐mg Ϫ1 protein. Interestingly, pyruvate is one of the major carbon sources present in the growth medium. Indeed, only D-glucose (1.4 mol⅐h Ϫ1 ⅐mg Ϫ1 protein) and L-threonine (1.25 mol⅐h Ϫ1 ⅐mg Ϫ1 protein) showed a higher consumption rate, whereas L-proline (0.45 mol⅐h Ϫ1 ⅐mg Ϫ1 protein) and L-glutamine (0.4 mol⅐h Ϫ1 ⅐mg Ϫ1 protein) are consumed with a lower rate (data not shown). Interestingly, the rate of pyruvate consumption is not significantly affected by arrest of PYK expression in the tetracycline-induced and non-induced ⌬1.2ppdk/⌬PYK (0.6 and 1.1 mol⅐h Ϫ1 ⅐mg Ϫ1 protein, before and 3 days after induction, respectively) and ⌬PYK (0.65 and 1.25 mol⅐h Ϫ1 ⅐mg Ϫ1 protein, before and 3 days after induction, respectively) cell lines. Similarly, the rate of L-proline consumption is not affected by inhibition of PYK expression (data  Table I). In the case of cell lines grown in the presence of tetracycline (⌬1.2ppdk/⌬PYK.i and ⌬PYK.i), the PYK activity decreases over time to become undetectable (ND), as indicated in the table of the lower panel (the PYK activities are normalized with the GPDH activity measured in the same samples). not shown). This indicates that, in the absence of PYK activity, the procyclic cells still take up extracellular pyruvate and the proline metabolism is not affected.
Third, the steady state concentration of intracellular PEP, pyruvate, and ATP in the mutant and wild type cell lines was determined (Table II). As expected, the intracellular concentration of PEP (the PYK substrate) was increased up to 3.4-fold (⌬PYK.i) and 6.4-fold (⌬1.2ppdk/⌬PYK.i) in the tetracyclineinduced mutant cell lines, as compared with the wild type trypanosomes. Similarly, the intracellular pyruvate concentration was moderately increased in the mutant cell lines (Table  II), indicating that PYK depletion does not result in a reduction of the intracellular pyruvate concentration, under the growth conditions used. In contrast, the intracellular ATP concentration was reduced 1.5-fold (⌬PYK.i) and 2.3-fold (⌬1.2ppdk/ ⌬PYK.i) 4 days after tetracycline induction, when the PYK activity has dropped below the detection threshold but cells are still alive. This shows that there is a direct correlation between reduction of the PYK activity and the steady state concentration of intracellular ATP, whereas the steady state concentration of intracellular pyruvate is not significantly affected (Table  II). Taken together, these data demonstrate that in procyclic T. brucei PYK serves primarily for ATP synthesis and pyruvate production may be considered as a secondary aspect of the reaction.

DISCUSSION
In 1979, Brun and Schönenberger (25) developed a semidefined medium, called SDM-79, which became the most popular medium for growing the procyclic form of T. brucei (25). When grown in SDM-79, or equivalent media, this insect stage parasite primarily consumes D-glucose and L-threonine as carbon sources, and to a lower extent L-proline and L-glutamine (Ref. 10, see Fig. 1). D-Glucose metabolism has received more attention than the metabolism of the other carbon sources. Here, we confirm the previous data and we show for the first time, that pyruvate is also consumed with a rate higher than observed for L-proline (0.8 versus 0.45 mol⅐h Ϫ1 ⅐mg Ϫ1 protein).
Three cytosolic or mitochondrial enzymatic activities of Dglucose catabolism produce ATP by phosphorylation at the substrate level, i.e. phosphoglycerate kinase, PYK, and succinyl-CoA synthetase (steps 7, 15, and 21 in Fig. 1). The recent development in T. brucei of an RNAi approach permitting conditional gene expression (2-7), provides an excellent tool to study genes potentially essential for cell viability, such as phosphoglycerate kinase, PYK, and succinyl-CoA synthetase. Inactivation of phosphoglycerate kinase gene expression would probably not be informative in determining whether it plays an essential role in ATP generation, because phosphoglycerate kinase catalyzes an essential step of glycolysis. Recently, Bochud-Allemann et al. (20,23) showed that succinyl-CoA synthetase, which catalyzes the last step of both D-glucose and L-proline metabolism, is essential for the procyclic cells (20, 23) (see Fig. 1). Interestingly, succinyl-CoA synthetase is also involved in the metabolism of other carbon sources, i.e. pyruvate, L-glutamine, and probably L-threonine (if part of acetyl-CoA is metabolized into acetate by acetate:succinate-CoA transferase) (see Fig. 1). Consequently, one cannot rule out the possibility that inhibition of succinyl-CoA synthetase expression affects the rate by which all of these carbon sources are metabolized and thus may also affect the rate of ATP production by oxidative phosphorylation. Indeed, the NADH produced in the mitochondrion by the metabolism of these different carbon sources (steps 19, 25, 28, 30, and 31 in Fig. 1) is probably reoxidized by the mitochondrial electron transport chain, which generates a proton gradient in part used by the F 0 /F 1 -ATP synthase to produce ATP. Here, we showed that the cytosolic PYK is also essential for procyclic cells. It is noteworthy that, under the growth conditions used, the formation of PEP (PYK substrate) from sources other than glucose (including amino acids), is unlikely. PYK is absolutely required for ATP production, which significantly drops in the mutant cell lines, whereas the amount of intracellular pyruvate is not affected. In addition, the production of ATP by oxidative phosphorylation is not affected by the inactivation of PYK expression, because (i) the mutant still produces acetate from D-glucose and (ii) the rate of amino acids consumption, which is metabolized in the mitochondrion, is not affected. This clearly demonstrates that the cytosolic production of ATP, through PYK, at the substrate level phosphorylation is essential for the cell viability.
In contrast, we observed that the glycosomal ATP producing enzymes, i.e. PPDK (Fig. 7) and PEPCK (step 10 in Fig. 1), 4 are not essential for the viability of the procyclic trypanosomes. The glycosomes show a reduced exchange of glycolytic intermediary products with the other subcellular compartments (43) and the same has been proposed for nucleotides (AMP, ADP, and ATP) and NAD ϩ /NADH (12)(13)(14)(15)17). Our data suggest that one or more alternative routes are used to regenerate the ATP consumed by the first glycolytic steps (steps 1 and 3 in Fig. 1), or that the glycosomal nucleotide pool is exchanged with other subcellular compartments. Detailed analyses of mutants, including the PEPCK mutant, will be helpful in understanding how the ATP/ADP, as well as the NAD ϩ /NADH, balances are conserved in the glycosomes of these parasites.
We showed that the production of ATP by oxidative phosphorylation is not essential for the procyclic cells. Indeed, inhibition of the F 0 /F 1 -ATP synthase (step 36 in Fig. 1), by a very specific inhibitor (oligomycin), does not affect the intracellular steady state concentration of ATP and only moderately affects the doubling time of the parasite (18 versus 12 h). However, inhibition of cytochrome oxidase by 1 mM cyanide (IV in Fig. 1) considerably affects the doubling time of the procyclic cells (23), and addition of both 1 mM cyanide and 0.5 mM SHAM (a specific inhibitor of the alternative oxidase, step 35 in Fig. 1), kills all parasites within 2 days of incubation (data not shown), indicating that the mitochondrial electron transport chains (cyanide-sensitive and SHAM-sensitive) are essential for this parasite. An important question, which has not yet been completely answered, is what is(are) the role(s) of these mitochondrial electron transport chains in procyclic T. brucei? As in most eukaryotes, a primary role is probably reoxidizing reduced equivalents, such as NADH produced in the mitochondrion during the metabolism of D-glucose/pyruvate (step 19 in Fig. 1), L-threonine (step 31), and L-proline/L-glutamine (steps 25, 28, and 30), as well as that of other substrates. The other essential role of the cyanide-sensitive respiratory chain is to generate a transmembrane proton gradient, which is used for essential mitochondrial functions, such as ion (Ca 2ϩ ) regulation and the mitochondrial protein and metabolite import. The proton gradient can also be used by the F 0 /F 1 -ATP synthase to produce ATP by oxidative phosphorylation, however, our data strongly suggest that this is not essential for the procyclic T. brucei. Thus, it appears that, in procyclic trypanosomes, production of ATP is not the primary role of the mitochondrial electron transport chains. It is noteworthy that the bloodstream forms of this parasite lack a full respiratory chain and use the F 0 /F 1 -ATP synthase to generate the transmembrane proton gradient from ATP hydrolysis (41). As discussed above, procyclic trypanosomes use the electron chain transport to generate the transmembrane proton gradient. The role of the 4 V. Coustou and F. Bringaud, unpublished data. F 0 /F 1 -ATP synthase in mitochondrial ATP metabolism of the procyclic form of T. brucei therefore remains uncertain. The generation and analysis of RNAi mutants affecting expression of F 0 /F 1 -ATP synthase subunits, will be helpful to answer this question.