A Vacuolar-type H+-Pyrophosphatase Governs Maintenance of Functional Acidocalcisomes and Growth of the Insect and Mammalian Forms of Trypanosoma brucei*

Vacuolar proton pyrophosphatases (V-H+-PPases) are electrogenic proton pumps found in many organisms of considerable industrial, environmental, and clinical importance. V-H+-PPases of several parasites were shown to be associated with acidic vacuoles named acidocalcisomes, which contain polyphosphate and calcium. In this work we functionally characterized a Trypanosoma brucei V-H+-PPase gene by using double-stranded RNA interference methodology to produce inducible V-H+-PPase-deficient strains of procyclic and bloodstream forms (PFiVP1 and BFiVP1). Acidocalcisomes of these mutated parasites lost acidity and contained 90% less polyphosphate. PFiVP1 did not release calcium after the addition of nigericin, and its total acidity was reduced by 70%. This mutant also failed to stabilize its intracellular pH on exposure to external basic pH >7.4 and recovered from intracellular acidification at a slower rate and to a more acidic final intracellular pH. In the absence of T. bruceiV-H+-PPase expression, PFiVP1 and BFiVP1 grew at a slower rate with doubling times of 27 h instead of 15 h, and 10 h instead of 7.5 h, respectively. Moreover, BFiVP1 could not grow over 5 × 105 cells/ml corresponding to a cell density reduction of five times for bloodstream form stationary phase growth.

Intracellular acidic vacuoles containing polyphosphate (polyP), 1 initially called volutin or polyP bodies, have been described in bacteria, algae, yeast, and protozoa (1). In trypano-somatids, these polyP vacuoles were called acidocalcisomes (2) and shown to be electron dense and contain large concentrations of PP i , calcium, magnesium, and other elements (3). Similar organelles have been identified in apicomplexan parasites (4,5) as well as in the green algae, Chlamydomonas reinhardtii (6) and the slime mold, Dictyostelium discoideum (7). Acidocalcisomes were postulated to play an important role in the regulation of both cytosolic Ca 2ϩ concentration and intracellular pH (pH i ). For example, polyP hydrolysis (8) and activation of the Na ϩ /H ϩ antiporter (9,10) were postulated to protect the cells against alkaline pH stress and increase the intracellular Ca 2ϩ concentration ([Ca 2ϩ ] i ). In addition, these organelles possess a vacuolar-type H ϩ -translocating pyrophosphatase (V-H ϩ -PPase) (11), which is an electrogenic proton pump initially discovered in photosynthetic bacteria and plants (12). It has been shown to be associated with the plasma membrane or vacuoles in plants and with the chromatophore membranes of Rhodospirullum rubrum. The biochemical function of this enzyme in plants and unicellular eukaryotes is to couple hydrolysis of the high energy phosphate bond of PP i with H ϩ translocation from the cytosol to acidify the plant vacuole (tonoplast) or the acidocalcisome, respectively (10,13). Working on isolated acidocalcisomes, Rodrigues et al. (14) demonstrated that the Trypanosoma brucei H ϩ -PPase was able to generate a membrane potential by PP i -dependent proton uptake. Moreover, Scott et al. (11,15) showed that PP i could drive Ca 2ϩ uptake into the compartments of permeabilized Trypanosoma cruzi epimastigotes (11) or isolated acidocalcisomes (15). However, the physiological importance of the V-H ϩ -PPase was investigated less thoroughly because genetic approaches were not used. Studies of V-H ϩ -PPase expression in plants have revealed higher expression levels in young growing tissue and during growth under stress (12). Recently, it has been shown that transgenic plants overexpressing V-H ϩ -PPase become more resistant to high concentrations of NaCl and water deprivation than the isogenic wild-type strains (16).
Using RNA interference methodology (17), we have now shown that T. brucei V-H ϩ -PPase not only plays roles in the regulation of Ca 2ϩ , polyP, and H ϩ content of acidocalcisomes and pH i regulation of procyclics but also is required for normal in vitro growth of bloodstream and procyclic forms.

EXPERIMENTAL PROCEDURES
Strains Used-T. brucei brucei procyclic form host cell line 29 -13 and bloodstream form host cell line 90 -13 co-expressing the T7 RNA polymerase and the Tet repressor were gifts from G. A.M. Cross (18). Procyclic forms were cultivated to late log-phase in SDM-79 medium (19) supplemented with 10% fetal calf serum. Bloodstream forms were cultured in vitro in modified essential medium supplemented with 10% fetal calf serum (20). * This work was supported by the CNRS, the Conseil Régional d'Aquitaine, the GDR CNRS-Parasitologie and the Ministère de l'Education Nationale de la Recherche et de la Technologie (Action Microbiologie), and the United Nations Development Project/World Bank/World Health Organization Special Programme for Research and Training in Tropical Diseases (to R. D.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EBI Data Bank with accession number(s) AF159881 and AY043295.
Sequence Analysis-DNA sequence data were generated at the High Throughput Sequencing and Genotyping Unit of the Keck Center for Comparative and Functional Genomics at the University of Illinois at Urbana-Champaign. Sequence analysis was done using the Biology Workbench 3.0 utility (workbench.sdsc.uiuc.edu), the Wisconsin Package (version 10.0-UNIX, Genetics Computer Group, Madison, WI). The PSORT program was from Kenta Nakai (University of Tokyo, Tokyo, Japan). The complete gene sequence for TbVP1 was submitted to the GenBank TM data base (GenBank TM accession number AY 043295).
Cloning, Expression, and Purification of the TbVP1 Loop III (TbVP1L3) in Escherichia coli-A 357-bp fragment comprising the loop III and adjacent regions of TbVP1 gene was generated by PCR. The 5Ј primer (5Ј-GCGGCGCATATGGCTCTCTTCTGCACGGTG-3Ј) contained a 12-nucleotide linker with an NdeI restriction site to facilitate subcloning and 5Ј-adjacent N-terminal residues. The 3Ј primer (5Ј-CAGGACCTCGAGGTCTGCGCTGAGCTCGGC-3Ј) included a 12-nucleotide linker with a XhoI site for cloning. The PCR product was inserted into the NdeI/XhoI sites of the pET23a plasmid (Novagen). The resulting recombinant protein was expressed in E. coli BL21 (DE3) from Novagen according to the manufacturer's instructions. Cells were lysed, and the recombinant protein was purified as previously described after solubilization in 6 M urea (23).
Production and Western Blot Analysis of Immune Serum against the Recombinant TbVP1L3 Protein-Mice were first injected with 10 g of TbVP1L3 in complete Freund's adjuvant and with another 10 g in incomplete Freund's adjuvant 15 and 30 days later. Blood was collected before the first injection (pre-immune serum) and 10, 13, and 15 days following the last injection. Western blotting was performed as described previously (24) with serum diluted 1/200 and incubated overnight at 4°C. The second antibody, a 1/5000 dilution of goat anti-mouse IgG conjugated to horseradish peroxidase (Sigma), was incubated for 2 h in phosphate-buffered saline-Tween 20 milk. Immunoreactive bands were revealed by washing in 50 mM Tris-HCl, pH 7.5, 20 mM NaCl, and a ECL revelation kit (Amersham Biosciences).
For observation of Acridine Orange accumulation, cells were washed twice in a buffer containing 116 mM NaCl, 5 mM KCl, 0.8 mM MgSO 4 , 5.5 mM glucose, 50 mM K-Hepes, pH 7.4, and incubated for 15 min at 30°C in the same buffer with 6 M Acridine Orange. Parasites were then partially immobilized by centrifugation at 130 ϫ g for 10 min on poly-L-lysine glass coverslips as described previously (25). Coverslides with immobilized living parasites were observed under a fluorescent microscope fitted with a fluorescein filter set. Cells were examined on a Zeiss UV microscope, and images taken with a camera (Photometrics) with Metaview and Adobe Photoshop 6 (Adobe Systems) software.
For observation of DAPI accumulation, cells were washed twice in a buffer containing 116 mM NaCl, 5 mM KCl, 0.8 mM MgSO 4 , 5.5 mM glucose, 50 mM K-Hepes, pH 7.4, and incubated for 10 min at 30°C in the same buffer with 10 g/ml DAPI (8). Cells were then treated and observed for Acridine Orange accumulation.
Double-stranded sRNA Expression and Trypanosome Transformation-The inducible T7 RNA polymerase-based protein expression system developed by Wirtz et al. (18) was used in this study. The pLew 100 vector (18) was kindly provided by G.A.M. Cross. For double-stranded RNA expression, the following construct was produced as described previously (26). DNA fragments corresponding to the coding regions of TbVP1 from nucleotides 99 to 416 and from nucleotides 99 to 464 were PCR-amplified. The 317-bp fragment was amplified using the following set of primers: DB/VP/01 (5Ј-CCGGTTCTGCAGGGATTCATGAGTGT-CACGACAGCTAC-3Ј) and DB/VP/02 (5Ј-CCGGTTGAATTCCCGGC-CGTCTTTGCCTCC-3Ј). The 365-bp fragment was amplified with the two following primers, DB/VP/03 (5Ј-CCGGTTAAGCTTATGAGTGTC-ACGACAGCT-3Ј) and DB/VP/04 (5Ј-CCGGTTGAATTCAAGGGAAAA-TGCAGCATTC-3Ј). After precloning in the pGEMt vector (Promega), the resulting fragment was cloned between the BamH1 and HindIII restriction sites of pLew100. The resulting construction was named pLew 100-SVP. For stable transformation, T. brucei procyclic forms (29 -13 cell line) were harvested from a log-phase culture (5 ϫ 10 6 cells), washed once in Zimmerman post-fusion medium (ZPFM) (18), and resuspended to a cell density of 4 ϫ 10 7 cells/ml in ZPFM cells were electroporated with 10 g of DNA using the Eurogentech CellJect machine at 1600 V, infinite resistance, and 40 microfarads. Parasites were then transferred to SDM-79 medium containing 10 g/ml G418 and 5 g/ml hygromycin. The modified strain was named PFiVP1 and induced for 5 days with 1 g/ml tetracycline. Cells were diluted 10-fold when densities reached a minimum of 1 ϫ 10 6 cells/ml. They were not allowed to grow beyond 5 ϫ 10 6 cells/ml before diluting again.
For stable transfection of bloodstream forms, cells were harvested from a log-phase culture, washed once in ZPFM and 55 mM glucose and resuspended in ZPFMG containing 2 mM ATP and 5 mM glutathione to a cell density of 2 ϫ 10 7 cells/ml (27, 28). 10 7 cells were electroporated with 10 g of DNA at 1400 V, R-inf, and 20 microfarads using the Eurogentech CellJect machine. Surviving cells were then seeded in a microtiter plate as described by Wirtz et al. (27), and selection was applied the following day by adding phleomycin to a final concentration of 0.5 g/ml. Two clones were obtained: BFiVP1/1A1 and BFiVP1/2A2. Cells were diluted 20-fold when densities reached a minimum of 5 ϫ 10 5 cells/ml. They were not allowed to grow beyond 1 ϫ 10 6 cells/ml before diluting again. The genetically modified strain was induced for 4 days with 1 g/ml tetracycline.
Spectrofluorometric Determinations on Intact Cells-The [Ca 2ϩ ] i and pH i were measured according to Scott et al. (29) and Fraser-L'Hostis et al. (30) For pH i , we used the wavelengths 490 and 440 nm for excitation and 535 nm for emission. The spectrofluorometer fluoromax (SPEX Instruments) and software DM 3000 (SPEX Instruments) were used for data analysis. The ratio of intensities (F490/F440) was used to calculate the intracellular pH according to a calibration curve determined by the incubation of BCECF-loaded trypanosomes treated with 2 M nigericin in a buffer containing 116 mM KCl, 0.8 mM MgSO 4 , 5.5 mM glucose, 20 mM Hepes, 20 mM Mes covering the pH range from 6.5 to 8.
ATP-and PP i -driven Proton Transport in Permeabilized Procyclic Trypanosomes-Cells (0.2 mg of protein/ml equivalent to 2.5 ϫ 10 7 cells) were added to a buffer containing 2 mM MgSO 4 , 50 M EGTA, 10 mM K-Hepes, pH 7.2, 65 mM KCl, and 125 mM sucrose. Acridine Orange was used at 3 M, PP i at 0.1 mM, ATP at 1 mM, and digitonin at 16 M. The spectrophotometer Cary 100 (Varian) was used to follow Acridine Orange uptake as described previously (14).
Analysis of PolyP-PolyP levels were determined as described by Ruiz et al. (8) using the recombinant exopolyphosphatase from Saccharomyces cerevisiae. PFiVP1 procyclic trypomastigotes were induced with 1 g/ml tetracycline during 4 days. 2 ϫ 10 7 cells then were washed twice with Dulbecco's phosphate-buffered saline, and long and short chain polyP was quantified.
Subcellular Fractionation-Subcellular fractionation was performed according to Rodrigues et al. (14). Procyclic cells were lysed in a lysis buffer (125 mM sucrose, 50 mM KCl, 4 mM MgCl 2 , 0.5 mM EDTA, 20 mM Na-Hepes, pH 7.4, 5 mM dithiothreitol) containing a protease inhibitor mixture with final concentrations of 1 M chemostatin, 1 M leupeptin, 1 M pepstatin, and 10 M phenylmethylsulfonyl fluoride. After lysis and removal of silicon carbide by centrifugation at 144 ϫ g for 5 min, the extract was centrifuged again at 580 ϫ g for 10 min. 5.76 ml of supernatant were mixed with 6.12 ml of Percoll and 6.12 ml of 0.5 M sucrose before centrifugation at 69,500 ϫ g for 50 min.

RESULTS AND DISCUSSION
Cloning and Sequence Analysis of T. brucei V-H ϩ -PPase-To screen for genes encoding plant-like V-H ϩ -PPases in T. brucei, the amino acid sequence of V-H ϩ -PPase from T. cruzi, TcPPase (21), was used to search all available sequence databases using TBLASTN. This search yielded a 1.3-kb gene fragment, which closely resembled the region encompassed by nucleotides encoding amino acid residues 10 -445 of TcPPase. On the basis of this information, we cloned TbVP1 by using the reverse transcriptase-PCR technique. This 1.2-kb cDNA enabled appropriate gene-specific primers to be designed for the generation of 5Ј-end and 3Ј-end DNA fragments by using the RACE method (22) and the reconstruction of a full-length cDNA. The nucleotide sequence of 3510 bp revealed an open reading frame of 2481 bp (nucleotides 136 -2616) that encodes a 826-amino acid protein with a relative molecular mass of 85.9 kDa and a calculated pI of 5.08.
To assess the copy number of TbVP1 in the nuclear genome of T. brucei, Southern blots were probed with a 1.2-kb cDNA fragment (nucleotides 168 -1387) (data not shown). Under stringent conditions, the blot contained a single band except when predicted internal restriction sites (PstI and SacI) were present. These data together with the reverse transcriptase-PCR, 5Ј-RACE, and 3Ј-RACE were consistent with a single copy number gene in the T. brucei genome.
While this paper was in preparation, a search of GenBank and the T. brucei genome project databases using the full- frame was identical to the full-length cDNA. When the cDNA sequence was compared with the genomic sequence from chromosome IV, the predicted translation initiation site of TbVP1 was preceded by 102 bp of 5Ј-untranslated sequence. The polyadenylation site of TbVP1 was preceded by 874 bp of 3Јuntranslated sequence.
Sequence and Structural Analysis of TbVP1-The HMMTOP software (www.enzim.hu/hmmtop) revealed the presence of the 15 potential transmembrane domains that we oriented as described previously (Fig. 1) (32). We have no experimental evidence for the membrane orientation of TbVP1. However, the N-terminal extremity analyzed by the sequence analysis PSORT (psort.nibb.ac.jp), may correspond to a 26-amino acid signal peptide with a predictable cleavage site at positions 26 -27. This last characteristic was previously reported by Hill et al. (21) for the T. cruzi V-H ϩ -PPase and was suggested to play a role in enzyme sorting.
The sequence comparison analysis shown in Fig. 1 confirmed the presence in TbVP1 of the five "tool box" sequences characteristic of the V-H ϩ -PPases (33). In particular, the cytosolic loop III containing the putative PP i binding motif (D/E)X 7 KXE is conserved (34).
Localization of TbVP1 in Procyclic and Bloodstream Form Trypanosomes-In a first step toward the characterization of TbVP1, we analyzed its subcellular location in bloodstream and procyclic trypomastigote forms. An histidine-tagged recombinant protein (A259 to D377), comprising loop III and the two adjacent regions, was expressed in E. coli, purified by Ni 2ϩ HisBind TM chromatography and used to immunize mice. Indirect immunofluorescence using this antiserum as the probe revealed vesicular-like structures of varying sizes in procyclic and bloodstream forms indicating expression in both stages. These structures are mainly distributed in the central region, with some additional vesicles at the anterior region of the bloodstream forms ( Fig. 2A). A similar distribution of acidocalcisomes in T. brucei was previously reported (14). To further analyze the subcellular location of TbVP1, we obtained procyclic acidocalcisomes from a high density Percoll gradient fraction of trypanosomes as described previously (14). We observed by immunofluorescence that 80% of the purified high density vesicles were recognized by the anti-TbVP1 mouse antiserum, indicating an association of this V-H ϩ -PPase with acidocalcisomes (data not shown). TbVP1 was also detected by Western blots of this high density fraction (Fig. 2B). No reaction was detected when the antiserum was pre-adsorbed with the Histagged TbVP1L3 protein (Fig 2B). The recognized polypeptide had an apparent molecular mass of 55 kDa. A size discrepancy between the expected (80 kDa) and the observed molecular mass was already reported for the V-H ϩ -PPases of plants (35,36) and trypanosomatids (21). It could be attributed either to the usual anomalous migration of hydrophobic proteins on SDS gels (37) or to partial degradation. In agreement with these results, Fig. 2B shows that the 55-kDa protein band is absent in a mutant deficient in V-H ϩ -PPase (see below). We conclude that TbVP1 is associated with vesicles of size, distribution, and density that correspond to the acidocalcisomal compartment.
Characterization of the H ϩ -translocating Activity of TbVP1-PP i -driven H ϩ transport activity was shown previously to be associated with acidocalcisomes (14). Using RNA interference methodology to silence TbVP1 expression, we tested whether in the absence of TbVP1 acidocalcisomes failed to display PP i -dependent H ϩ -translocating activity. Because TbVP1 is expressed and associated with acidocalcisomes in both procyclic and bloodstream trypomastigote forms, the gene-silencing experiment was performed initially on the former stages. After 5 days of double-stranded RNA expression, TbVP1 protein in the acidocalcisomes completely disappeared (Fig. 3A, photograph  b), indicating high stability and slow turnover of the protein.
Nevertheless, the presence of acidocalcisomes was confirmed by electron microscopy (data not shown).
Procyclic trypomastigotes permeabilized with digitonin were shown to accumulate and retain the weak base Acridine Orange after the addition of 0.1 mM PP i . The resultant H ϩ gradient collapsed completely after the addition of 1 M nigericin, a H ϩ /K ϩ ionophore, or the alkalinizing agent NH 4 Cl, confirming that the Acridine Orange absorbance decrease was attributed to its accumulation in an acidic compartment (Fig. 3B) as previously demonstrated (14). In the induced PFiVP1 parasites, the establishment of the proton gradient across the acidocalcisomal membrane and the corresponding increase in acidity did not occur after PP i addition (Fig. 3C). Therefore, the absence of the proton-translocating process associated with the V-H ϩ -PPase catalytic activity was correlated with the absence of the TbVP1 enzyme. By this genetic analysis, we confirmed that we were characterizing the acidocalcisomal V-H ϩ -PPase. Interestingly, digitonin-permeabilized PFiVP1 parasites were able to accumulate Acridine Orange when ATP was added to the reaction medium (Fig. 3C). This accumulation was inhibited when concanamycin A was added prior to the beginning of the reaction (Fig. 3C, dashed line) (14). Moreover, the addition of 40 mM NaCl induced partial release of Acridine Orange accumulated after ATP addition, and subsequent addition of ADP stimulated this release (Fig. 3C). Also, as previously shown (38), the hydrophobic antioxidant 3,5-dibutyl-4-hydroxy toluene inhibited the Na ϩ /H ϩ exchanger activity (data not shown). These data suggest the presence on these mutated acidocalcisomes of a functional H ϩ -ATPase pump and a Na ϩ /H ϩ exchanger, which were shown to be important for acidification and calcium uptake and release from acidocalcisomes (9,38). In permeabilized procyclic trypomastigotes of T. brucei (14) and in isolated acidocalcisomes of T. cruzi (15), the H ϩ -translocating pyrophosphatase activity is sufficient for the acidification of acidocalcisomes. However, in vivo acidification by the V-H ϩ -ATPase pump in the absence of the H ϩtranslocating PPase activity had never been tested. Therefore, we evaluated the state of this acidic compartment in the induced PFiVP1 and the bloodstream RNA interference mutant (BFiVP1) strains.
Effects of the Inhibition of TbVP1 Expression on the Acidity and PolyP Content of Procyclic and Bloodstream Form Acidocalcisomes-The BFiVP1 strain was obtained as described under "Experimental Procedures." After 4 days of doublestranded RNA expression, TbVP1 protein completely disappeared as occurred with PFiVP1 (Fig. 3A) from the acidocalcisomes of BFiVP1 (Fig. 4A). Concomitantly, in both forms, the amount of accumulated Acridine Orange in acidocalcisomes was drastically reduced (Figs. 4B and 5A). Although some vesicles located around the nucleus (Figs. 4B and 5A) were associated with the perinuclear acidic compartment (39), others corresponded to less acidic acidocalcisomes. These results suggest that TbVP1 was essential to maintain acidocalcisomal acidity and consequently that the H ϩ -ATPase alone could not fulfill that function. In this regard, we have previously reported (40) that only very high concentrations (Ͼ5 M) of bafilomycin A 1 , a specific inhibitor of the V-H ϩ -ATPase when used at nanomolar concentrations, can release Acridine Orange from acidocalcisomes of permeabilized trypanosomes. We also reported that neither bafilomycin A 1 nor concanamycin A inhibit PP idriven Acridine Orange transport into acidocalcisomes (14).
In T. cruzi, acidocalcisomal alkanization by NH 4 Cl was followed by progressive decrease in the levels of both short and long chain poly(P) (8). A loss of acidity in the acidocalcisomes of induced PFiVP1 and BFiVP1 parasites might have affected the polyP content of the cells. The results shown in Fig. 5B indicated a 10-fold reduction in short and long chain polyP in the acidocalcisomes of the tetracycline-induced PFiVP1. DAPI staining, which was shown to allow the detection of polyphosphate in vesicles (8), did not reveal an accumulation of polyphosphate in the BFiVP1 cells (Fig. 4C).
In trypanosomes, Ca 2ϩ uptake by acidocalcisomes is mediated by a Ca 2ϩ -ATPase acting as a Ca 2ϩ /H ϩ -exchanging ATPase (14), and polyP because of its high calcium chelation capacity appears to play a central role in the regulation of cellular Ca 2ϩ (8,41). Furthermore, a correlation between the short and long chain polyP content in the acidocalcisomes and changes in the pH i was suggested by Ruiz et al. (8). Therefore, we first analyzed the calcium content of acidocalcisomes and its mobilization and next studied the pH i regulation of the induced PFiVP1 parasites.

Ca 2ϩ and pH i Homeostasis in the Induced PFiVP1
Trypanosomes-In a first step, we analyzed the intracellular calcium storage in the acidocalcisomes. This study was performed in intact cells loaded with the fluorescent Ca 2ϩ indicator Fura-2. In the induced PFiVP1 strain, the starting [Ca 2ϩ ] i was identical to that of the non-induced strain. We have previously described that in T. brucei, the combination ionomycin ϩ nigericin is able to release Ca 2ϩ from the acidocalcisomes (29). Ionomycin binds essentially no Ca 2ϩ below pH 7.0, and it cannot move Ca 2ϩ out of acidic compartments because of the competition of protons at the inside face of the membrane (42). In the absence of nigericin or NH 4 Cl, ionomycin releases a relatively small amount of Ca 2ϩ only from neutral or alkaline compartments (Fig. 6A), but significantly more Ca 2ϩ is released after the addition of either nigericin (Fig. 6B) or NH 4 Cl (Fig. 6C). The addition of nigericin did not raise the [Ca 2ϩ ] i in the induced parasites (Fig. 6A). Moreover, a subsequent addi- tion of ionomycin released much less calcium from induced cells (Fig. 6A). Reverse addition of these two ionophores confirmed that Ca 2ϩ release was not induced by nigericin (Fig. 6B). The results in Fig. 6, A and B, suggest that the mutant strain has very little Ca 2ϩ in the acidocalcisomes (sensitive to ionomycin ϩ nigericin). In contrast, the alkalinization of acidic compartments by NH 4 Cl caused equivalent release of calcium by the induced and non-induced parasites (Fig. 6C). Furthermore, the subsequent addition of ionomycin also caused the release of equal amounts of Ca 2ϩ from the two types of parasites (Fig.  6C). The reverse addition, ionomycin/NH 4 Cl, showed that previous alkalinization was necessary for calcium release by ionomycin (Fig. 6C). After 5 days of induction, approximately 1 mM long chain polyP and 10 mM short chain polyP remain, which could chelate considerable amounts of calcium (1). By stimulating hydrolysis of the remaining polyP, NH 4 Cl might release bound calcium. In contrast, nigericin would not release this Ca 2ϩ because it is bound to polyP. Although nigericin can stimulate polyP hydrolysis, it would act more slowly than NH 4 Cl and would be less effective under the reduced acidity of the induced acidocalcisomes (8). A new equilibrium between polyP hydrolysis and synthesis may have been reached to release H ϩ and to chelate Ca 2ϩ . It has been reported that some polyphosphatases have alkaline pH optima (8), and we recently detected pyrophosphatase activity at an alkaline pH localized to that compartment (data not shown).
In a second series of experiments, we analyzed the effect of TbVP1 expression inhibition on the intracellular pH regulation of PFiVP1. Nolan and Voorheis (43) pointed out that in trypanosomes, the cytoplasmic pH was determined by the resulting H ϩ concentrations in a three-compartment system consisting of the cytoplasm, mitochondrial matrix, and acidic compartments. The overall acidic capacity of PFiVP1 trypanosomes was estimated by measuring the uptake of Neutral Red, a dye commonly used for acidic compartment assays of living cells (31). Fig. 6D shows that inhibition of TbVP1 expression decreased total Neutral Red uptake in the acidic compartments of living cells by 70%. These data indicated that the acidic capacity of cells deprived of the TbVP1 was significantly reduced. Therefore, we analyzed the steady-state pH i of these mutants under different external pH (pH e ). In normal cells, the pH i remained close to 7.1 in the pH e ranges from 6.5 to 8 as reported previously (30,44). However, while the pH i of the induced mutant cells was maintained at acidic pH e , it increased almost 0.2 pH unit at basic pH e (Fig. 6E). The pH i values of 7.11 Ϯ 0.08 and 7.32 Ϯ 0.04 were observed for non-induced and induced cells, respectively, at a pH e of 8.
To further investigate the role of the V-H ϩ -PPase in pH i regulation, procyclic trypomastigotes were acidified to a pH of ϳ6.0 using the NH 4 Cl prepulse technique (44). Fig. 6F shows that induced PFiVP1 parasites recovered their pH i more slowly and reached a lower final pH i than non-induced cells. Taken together, there results suggest that when the pH i is acidic, the V-H ϩ -PPase is involved in removing H ϩ from the cytosol into the acidocalcisomes. On the other hand, when the pH i is alkaline, the acidocalcisomes would release H ϩ to the cytosol. In the absence of an acidocalcisomal V-H ϩ -PPase, there would be less protons available to be released under alkaline conditions and the pH i would not be maintained as in control cells.
TbVP1 Was Essential for the Growth of PFiVP1 and BFiVP1 in Culture-Although the mutation was not lethal for either procyclic or bloodstream trypomastigotes growing in culture, their growth was decreased (Fig 7A). The doubling time of induced PFiVP1 parasites was estimated at 27 h, whereas in the absence of tetracycline, the doubling time was around 15 h. The difference in the doubling time of induced and non-induced BFiVP1 trypanosomes was not as important (10 versus 7.5 h, respectively). Interestingly, induced BFiVP1 parasites could not reach the same cell density than the non-induced ones (5 ϫ 10 5 /ml instead of 2.5 ϫ 10 6 /ml, respectively) ( Fig 7B). These data showed the importance of the acidocalcisomes in regulating the growth of procyclic and bloodstream forms.
In conclusion, TbVP1 expression inhibition induced a loss of functional acidocalcisomes and a lower capacity of PFiVP1 cells to adapt to environmental pH stress. Because the cells were maintained under exponential growth conditions to perform the RNA interference experiments, no fundamental changes in the culture medium occurred during growth. In their life cycle, trypanosomes encounter many environmental changes including pH e variations to which they must adapt in order to survive and multiply (45). In this regard, polyP has been shown to be important to support the survival of bacteria in the stationary phase (46). Functional acidocalcisomes may be essential for these adaptations. We are currently studying the effect of nonfunctional acidocalcisomes on the survival of procyclic and bloodstream forms in their insect and mammalian hosts, respectively. A, PFiVP1 cells were maintained in the exponential phase of growth, and tetracycline (1 g/ml) was added at t ϭ 0 (black line). B, BFiVP1 non-induced (gray line) and induced (black line and square) cultures were transferred where indicated by the arrows. The black diamond growth curve corresponds to the induced BFiVP1 culture, which has not been transferred. Cells were induced for 3 days before starting the experiment. Similar results were obtained with different clones and with different mutant constructions (S. Luo and R. Docampo, unpublished results).