A Trypanosoma cruzi Phosphatidylinositol 3-Kinase (TcVps34) Is Involved in Osmoregulation and Receptor-mediated Endocytosis*

Trypanosoma cruzi, the etiological agent of Chagas disease, has the ability to respond to a variety of environmental changes during its life cycle both in the insect vector and in the vertebrate host. Because regulation of transcription initiation seems to be nonfunctional in this parasite, it is important to investigate other regulatory mechanisms of adaptation. Regulatory mechanisms at the level of signal transduction pathways involving phosphoinositides are good candidates for this purpose. Here we report the identification of the first phosphatidylinositol 3-kinase (PI3K) in T. cruzi, with similarity with its yeast counterpart, Vps34p. TcVps34 specifically phosphorylates phosphatidylinositol to produce phosphatidylinositol 3-phosphate, thus confirming that it belongs to class III PI3K family. Overexpression of TcVps34 resulted in morphological and functional alterations related to vesicular trafficking. Although inhibition of TcVps34 with specific PI3K inhibitors, such as wortmannin and LY294,000, resulted in reduced regulatory volume decrease after hyposmotic stress, cells overexpressing this enzyme were resistant to these inhibitors. Furthermore, these cells were able to recover their original volume faster than wild type cells when they were submitted to severe hyposmotic stress. In addition, in TcVps34-overexpressing cells, the activities of vacuolar-H+-ATPase and vacuolar H+-pyrophosphatase were altered, suggesting defects in the acidification of intracellular compartments. Furthermore, receptor-mediated endocytosis was partially blocked although fluid phase endocytosis was not affected, confirming a function for TcVps34 in membrane trafficking. Taken together, these results strongly support that TcVps34 plays a prominent role in vital processes for T. cruzi survival such as osmoregulation, acidification, and vesicular trafficking.

Trypanosoma cruzi, the etiological agent of Chagas disease, has the ability to respond to a variety of environmental changes during its life cycle both in the insect vector and in the vertebrate host. Because regulation of transcription initiation seems to be nonfunctional in this parasite, it is important to investigate other regulatory mechanisms of adaptation. Regulatory mechanisms at the level of signal transduction pathways involving phosphoinositides are good candidates for this purpose. Here we report the identification of the first phosphatidylinositol 3-kinase (PI3K) in T. cruzi, with similarity with its yeast counterpart, Vps34p. TcVps34 specifically phosphorylates phosphatidylinositol to produce phosphatidylinositol 3-phosphate, thus confirming that it belongs to class III PI3K family. Overexpression of TcVps34 resulted in morphological and functional alterations related to vesicular trafficking. Although inhibition of TcVps34 with specific PI3K inhibitors, such as wortmannin and LY294,000, resulted in reduced regulatory volume decrease after hyposmotic stress, cells overexpressing this enzyme were resistant to these inhibitors. Furthermore, these cells were able to recover their original volume faster than wild type cells when they were submitted to severe hyposmotic stress. In addition, in TcVps34-overexpressing cells, the activities of vacuolar-H ؉ -ATPase and vacuolar H ؉ -pyrophosphatase were altered, suggesting defects in the acidification of intracellular compartments. Furthermore, receptor-mediated endocytosis was partially blocked although fluid phase endocytosis was not affected, confirming a function for TcVps34 in membrane trafficking. Taken together, these results strongly support that TcVps34 plays a prominent role in vital processes for T. cruzi survival such as osmoregulation, acidification, and vesicular trafficking.
Phosphatidylinositol 3-kinase (PI3K) 6 activities have been found in all eukaryotic cell types examined to date (1,2) and are linked to a diverse set of key cellular functions, including cell growth, survival, and intracellular trafficking. PI3Ks belong to a large family of enzymes that has been divided into three functional classes on the basis of their protein domain structure, lipid substrate specificity, and regulatory properties. Class I PI3Ks were the first ones to be identified and are important components of the signaling pathways that regulate eukaryotic cell growth (3,4). These PI3Ks have a 110-kDa catalytic subunit that exhibits a substrate preference for PI 4-phosphate and PI 4,5-bisphosphate (5). Class II PI3Ks are less well known but may also function in the regulation of cell growth (5,6) and, additionally, in clathrin-mediated endocytosis (7,8). These enzymes prefer phosphatidylinositol (PI) as substrate but may also utilize PI 4-phosphate. Finally, class III family of PI3Ks is related to the yeast vacuolar protein sorting 34, Vps34p, and their homologs from other eukaryotes. Vps34p-like kinases specifically phosphorylate PI to produce phosphatidylinositol 3-phosphate (PI 3-P) and are associated with a Vps15p-like protein kinase (6,9). In addition to protein sorting to the vacuole/ lysosome, class III PI3Ks have been implicated in several other membrane transport events, including endocytosis, vesicle acidification, and autophagy (10 -14).
The protozoan parasite Trypanosoma cruzi is the causative agent of Chagas disease in Latin America and the main cause of cardiac death in endemic areas. Currently over 15 million people are infected, and another 28 million people are at risk of infection (15). As T. cruzi passes through its digenetic life cycle, it encounters extreme fluctuations in external osmolarity. These osmotic fluctuations occur both within the insect gut (16) and also when the parasite moves through the vertebrate host. Previous studies on the response of T. cruzi to hyposmotic stress have shown that both insect and vertebrate stages possess a robust regulatory volume decrease (RVD) mechanism that completely reverses cell swelling (17). This process is accomplished by the efflux of various ions and osmolytes (18) and the release of water by the contractile vacuole complex (19). In addition, it has been found that cyclic AMP levels increase when T. cruzi epimastigotes are subjected to hyposmotic stress (20). Recently, we described a membrane-associated cAMPspecific phosphodiesterase, TcrPDEC2, which is inhibited by the same inhibitors that affect the RVD of the parasites after hyposmotic stress (20,21). TcrPDEC2 is characterized by the presence of a FYVE domain, which is a phosphoinositide-binding motif. It is known that phosphoinositides serve as lipid signals required for membrane recruitment of several proteins implicated in the regulation of vesicular transport and intracellular protein sorting (22,23), and control the translocation and activity of proteins that contain phosphoinositide-binding motifs such as the FYVE domain, as well as the pleckstrin homology, PX, and ENTH domains (23,24).
Because inhibitors of TcrPDEC2 affect osmoregulation in T. cruzi, and TcrPDEC2 possesses a FYVE domain able to bind to PI 3-P, we hypothesized that PI 3-P production might have a role in osmoregulation. However, very little is known about the phosphoinositide signaling in trypanosomatids, and up to now only one trypanosome ortholog of Vps34, named TbVps34, has been described in Trypanosoma brucei (25). However, no functional analysis of this enzyme was performed, and it is not known whether that enzyme specifically catalyzes PI 3-P production (25). Here we report the identification and functional characterization of TcVps34, the first class III PI3K reported in T. cruzi. This study provides new insights into the physiological importance of the PI3K pathway and describes the role of TcVps34 in vital processes, such as osmoregulation, acidification, and receptor-mediated endocytosis.

EXPERIMENTAL PROCEDURES
Chemicals and Reagents-All radiochemicals used in this work were purchased from PerkinElmer Life Sciences, and restriction endonucleases were from New England Biolabs, Beverly, MA. Bacto-tryptose, yeast nitrogen base, and liver infusion were from Difco. All other reagents were purchased from Sigma.
For T. cruzi extracts, 10 8 epimastigotes were harvested by centrifugation at 1,500 ϫ g for 10 min and washed two times with phosphate-buffered saline (PBS). Cell pellets were then resuspended in lysis buffer (50 mM HEPES buffer, pH 7.3, containing 0.01 mg⅐ml Ϫ1 leupeptin, 25 units⅐ml Ϫ1 aprotinin, 0.5 mM phenylmethylsulfonyl fluoride, and 14 mM 2-mercaptoethanol) and lysed by six cycles of freezing in liquid N 2 and thawing at 4°C. The total extracts were further centrifuged for 1 h at 100,000 ϫ g to obtain P100 and S100 fractions.
The Saccharomyces cerevisiae strain, SEY6210⌬vps34:Trp1, was a kind gift of Dr. Yoshinori Ohsumi (26). Before transformation, this strain was grown at 30°C in synthetic selected media without tryptophan (0.17% (w/v) yeast nitrogen base without amino acids and ammonium sulfate, 0.5% (w/v) ammonium sulfate, and 2% (w/v) glucose, supplemented with the corresponding amino acid mixture). Transformant cells were selected in minimal medium without Trp and uracil.
Cloning of TcVps34 Gene-The gene sequence corresponding to the T. brucei TbVps34 PI3K (AC159405) was used to screen T. cruzi sequences in the GeneDB data base using the WU-Blast2 algorithm. Two oligonucleotides carrying hemi-restriction sites (PI3K-Fw-pENTR-KpnI 5Ј-GGTACCATGGC-CACCAACGAAGGCAACGT-3Ј and PI3K-Rv-pENTR-XhoI 5Ј-CTCGAGTTAGTGCCGTGTTGCCTGCGCT-3Ј) were designed from the identified sequence (Tc00.1047053511903. 160). PCR amplifications were carried out using 600 -800 ng of T. cruzi genomic DNA, 100 ng of each primer, 2.5 mM MgCl 2 , 0.2 mM dNTPs, and 2 units of Pfu Ultra High-Fidelity DNA polymerase (Stratagene, La Jolla, CA). Three PCR products were sequenced and compared with the original sequence found at GeneDB. Taken into account that some discrepancies were detected, we reported the obtained sequence to NCBI data bank (accession number EU276115).
Sequence Analysis-The search in a T. cruzi genome data base was performed with Wu-Blast2. Sequence identity was analyzed with the BLASTP (www.ncbi.nlm.nih.gov) and Clustal W. Multiple sequence alignment with default parameters setting was used to generate the alignments. Protein domains were determined using SMART and PROSITE software.
Southern, Northern, and Western blot Analyses-Genomic DNA was purified as described by Pereira et al. (27). Total cellular RNA was isolated from 10 8 epimastigote cells in the exponential growth phase using TRIzol reagents, as described by the manufacturer (Invitrogen).
Southern and Northern blot analyses were performed as described by Alonso et al. (28). The products were revealed with a specific 990-bp TcVps34 probe obtained by digestion of the 2,724-bp fragment with KpnI and PvuII. All probes were labeled with [␣-32 P]dCTP using the Prime-a-Gene kit (Promega, Madison, WI) following the manufacturer's instructions.
For Western blot analysis, 80 g of P100 and S100 T. cruzi fractions were loaded into 8% (w/v) SDS-PAGE as described by Laemmli (29) and electrotransferred to Hybond-C membranes (Amersham Biosciences). The transferred membranes were then incubated for 4 h with a 1:2,000 dilution of rabbit anti-His 6 (Gene Tex, Inc. San Antonio, TX). Detection was carried out by incubating with a 1:5,000 dilution of a goat anti-rabbit conjugated to peroxidase (Kirkegaard & Perry Laboratories, Gaithersburg, MD). The latter was developed with the ECL Plus TM Western blotting detection system (PerkinElmer Life Sciences).
Complementation Assay-The coding region of the TcVps34 gene was excised from the pDEST17 Escherichia coli expression vector (Invitrogen) and inserted into the KpnI and XhoI restriction sites in the galactose-inducible yeast expression vector pYES2 (URA3; Invitrogen). The yeast Vps34 knock-out strain SEY6210 ⌬vps34:Trp1 was transformed with either the empty vector or the vector carrying the TcVps34 gene, using the lithium acetate procedure (30). Transformed cells were selected in minimal medium (without Trp and uracil) at 30°C. For complementation assays, transformed cells were grown to an A 600 ϭ 1.8 -2 with 20% (w/v) glucose as carbon source and plated in the same selective media supplemented with 20% (w/v) galactose and 10% (w/v) raffinose at different dilutions. Plates were incubated at 37°C for 3 days.
To express the N-terminal His-tagged TcVps34 in E. coli, a recombination reaction was performed to transfer the coding sequence from pENTR3C into the pDEST17 expression vector (Invitrogen). The construct (pDEST17-TcVps34) was trans- , and the recombinant protein was induced with 500 M isopropyl-1-thio-␤-Dgalactopyranoside at 30°C for 4 h. Cells were harvested by centrifugation and resuspended in lysis buffer (10 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% (v/v) Triton X-100) containing a protease inhibitor mixture (Complete mini, Roche Applied Science). The protein extract was prepared by sonication, centrifuged, and resuspended in the same lysis buffer. This fraction was purified using a nickel-nitrilotriacetic acid-agarose resin (Invitrogen) and eluted with lysis buffer containing 80 mM imidazole.
For the substrate specificity experiments, the PI3K activity assays were carried out as described above except that PI, PI 4-P, or PI 4,5-P 2 were included as substrates (2 mg⅐ml Ϫ1 ).
For PI3K activity assays using T. cruzi epimastigote cell extracts, 25, 50, or 75 g of total protein were used. After a standard activity reaction as described above, radiolabeled phosphoinositides were measured in a liquid scintillation counter and identified by TLC followed by autoradiography. Unlabeled PI, PI 4-P, and PI 4,5-P 2 standards were visualized by staining with iodine vapors.
Parasite Transfection-The full-length TcVps34 gene was amplified by using the primers PI3K-Fw-pTREX-HindIII 5Ј-AAGCTTATGACCACCGAAAACGT-3Ј and PI3K-Rv-pENTR-XhoI 5Ј-CTCGAGTTAGTGCCGTGTTGCCTGC-GCT-3Ј. The TcVps34 gene fused to a His tag in the N-terminal was amplified from the construct pDEST17-TcVps34 by using the primer pDEST17-Fw-HindIII 5Ј-AAGCTTATGTCGTA-CTACCATCACCATCA-3Ј and the same reverse primer PI3K-Rv-pENTR-XhoI. Both 2,724-bp PCR products were then cloned into pGEM-T Easy plasmid and subcloned into the pTREX expression vector, which integrates in the ribosomal locus and allows high expression levels driven by the strong polymerase I rRNA promoter in T. cruzi (32).
T. cruzi epimastigotes of CL Brener strain were transfected with the pTREX-TcVps34 and pTREX-His-TcVps34 constructs as described previously (32). Parasites transfected with a pTREX-GFP construction were monitored by fluorescence microscopy as a selection control. Stable cell lines were achieved after 60 days of treatment with 500 g⅐ml Ϫ1 G418 (Invitrogen), and the transgenic condition was confirmed by Southern and Western blot analyses.
Differential Interference Contrast and Video Microscopy-Differential interference contrast (DIC) and video microscopy images were taken using an Olympus IX-71 inverted fluorescence microscope with a Photometrix CoolSnapHQ CCD camera driven by DeltaVision software (Applied Precision, Seattle, WA). For analysis of contractile vacuole filling, video microscopy of epimastigotes adhered to poly-L-lysine-coated coverslips and bathed in isosmotic buffer was carried out. Videos were recorded for 5 min and reconstructed using the Delta-Vision Softworx software. For video microscopy of severe hyposmotic treatment, cells were adhered to poly-L-lysinecoated coverslips and bathed in 65 mosM buffer. Videos were recorded for 20 min.
Electron Microscopy-About 1 ϫ 10 8 epimastigote forms were harvested and washed twice with PBS. The parasites were fixed with freshly prepared 2.5% glutaraldehyde, 4% formaldehyde, 0.1 M sodium cacodylate buffer, pH 7.3, for 1 h and then embedded in epoxy resin, sectioned, and stained using standard methods for transmission electron microscopy. Images were obtained on a Zeiss 900 transmission electron microscope operating at 80 kV. For scanning electron microscopy, cells were fixed as before, adhered to poly-L-lysine-coated coverslips, dehydrated in ethanol series, critical point dried, and coated with gold in a Balzers gold sputtering system. Cells were Phosphatidylinositol 3-Kinase in T. cruzi NOVEMBER 14, 2008 • VOLUME 283 • NUMBER 46 observed in a JEOL JSM 6340F field emission scanning electron microscope operating at 5 kV.
Regulatory Volume Decrease-Epimastigotes were collected and washed twice with isotonic chloride buffer (Iso-Cl buffer, 137 mM NaCl, 4 mM KCl, 1.5 mM KH 2 PO 4 , 8.5 mM Na 2 PO 4 , 20 mM HEPES, 11 mM glucose, 1 mM CaCl 2 , 0.8 mM MgSO 4 , pH 7.4). The osmolarity of the buffer was adjusted to 300 Ϯ 5 mosM as verified by an Advanced Instruments 3D3 osmometer (Norwood, MA). The washed cells were resuspended in Iso-Cl buffer to a cell density of 1 ϫ 10 8 ⅐ml Ϫ1 cells. The cells were distributed in 96-well plates with 150 l per well in triplicate, and hyposmotic stress was induced by 1:1 dilution of cell suspension with sterile deionized water, resulting in a final osmolarity of 150 mosM. The absorbance changes at 550 nm were recorded every 20 s for 10 min using a SpectraMax M2e plate reader (Molecular Devices) (18).
Membrane Integrity and Amino Acid Release-After treatment with the inhibitors wortmannin and LY294,000, membrane integrity was determined by ethidium bromide exclusion as described by Rohloff et al. (18). Total amino acid analysis of the cell extracts and the supernatant fraction of epimastigote cells exposed to hypotonic or isotonic buffer were determined as described before (18).
H ϩ Transport Assays-Active H ϩ transport was assayed as described previously (33) by measuring changes in the absorbance of acridine orange (493-530 nm) in an OLIS-modified SLM-Aminco DW2000 dual wavelength spectrophotometer. Briefly, for H ϩ -ATPase activity, epimastigotes were incubated at 30°C in a 2.5-ml standard reaction medium containing 125 mM sucrose, 65 mM KCl, 1 mM MgCl 2 , 10 mM HEPES, pH 7.2, 2 mM KH 2 PO 4 , 450 M EGTA, 4 M acridine orange, and 1 M digitonin. At the indicated time the reaction was started by adding 1 mM ATP. Bafilomycin A 1 and sodium o-vanadate were added at 1 and 100 M, respectively. For H ϩ -proton translocating pyrophosphatase (PPase) activity, the standard reaction medium contained 130 mM KCl, 10 mM HEPES, pH 7.2, 2 mM KH 2 PO 4 , 450 M EGTA, 4 M acridine orange, and 1 M digitonin. At the indicated times the reaction was started by adding 100 M PP i , and the specific inhibitor aminomethylene diphosphonate (AMDP) was added at 100 M. Each experiment was repeated at least three times with different cell preparations.
Transferrin and Fluid Phase Uptake-For FITC-transferrin (Molecular Probes, Invitrogen) uptake, epimastigotes were harvested and washed twice in serum-free LIT medium (sfLIT). The cells were incubated at 28°C for 10 min in sfLIT. FITCtransferrin (50 g⅐ml Ϫ1 ) was added, and the cells were further incubated at 28°C for various periods of time. Uptake was stopped by the addition of ice-cold sfLIT, and the cells were washed once at 4°C and fixed for 30 min in 4% formaldehyde in PBS. The cells were adhered to slides, mounted, and examined immediately.
Fluid phase uptake was monitored with FITC-BSA (Sigma). Uptake was performed as described above using 50 g⅐ml Ϫ1 of FITC-BSA.
For fluorescence quantification, the images were taken under nonsaturating conditions using identical exposure times. Fluorescence intensity was determined using Data Inspector program of the Delta Vision restoration system.

RESULTS
Cloning and Characterization of the TcVps34 Gene-A search of T. cruzi data bases using the TbVps34 gene from T. brucei as query resulted in the identification of an orthologous sequence in this parasite (Tc00.1047053511903.160). Two oligonucleotides designed from this sequence were used to amplify the T. cruzi gene using genomic DNA as template. After sequencing TcVps34, the 2,724-bp open reading frame was identified. The sequence contained several differences from that reported at the GeneDB data base and was therefore annotated at the NCBI under the accession number EU276115.
The predicted amino acid sequence of TcVps34 presented the phosphoinositide 3-kinase class III accessory domain (amino acids 318 -477), which has been suggested to be involved in substrate presentation, and the phosphoinositide 3-kinase catalytic domain (amino acids 513-906). In addition, TcVps34 possesses a C2 domain at the N-terminal region (amino acids 97-222), which is present in several proteins related to signal transduction (supplemental data S1). Amino acid sequence analysis showed that TcVps34 has the highest identity (57.4%) with T. brucei Vps34 (XP_847451), followed by the Leishmania major and S. cerevisiae orthologs (35.1 and 25.7%, respectively) (supplemental data S1). In addition, TcVps34 presented only 27.5% of identity with the human class III PI3K (NP_002638).
TcVps34 Complements a Yeast Vps34 Knock-out Strain-To analyze whether the TcVps34 gene encodes a functional PI3K, the TcVps34 coding region was subcloned into pYES2 expression vector and used to complement a S. cerevisiae Vps34 knock-out strain (SEY6210⌬vps34:Trp1), which possesses a temperature-sensitive phenotype. Expression of TcVps34 restored the ability of mutant cells to grow at 37°C, whereas transformation with the empty vector did not, indicating that the TcVps34 gene encodes a catalytically active PI3K (Fig. 1).
TcVps34 Is a Functional Class III Phosphatidylinositol 3-Kinase-The formation of PI 3-P by TcVps34 was tested using the partial affinity-purified recombinant protein ( Fig.  2A). After a standard PI3K radioactive assay, phosphorylated lipids were resolved by TLC and revealed by autoradiography. The recombinant protein was able to phosphorylate PI giving PI 3-P as the major product, as evidenced by the position of the standards run in a borate buffer system (34). The detection of this product was dependent on the protein concentration and was very weak in noninduced E. coli extracts (Fig. 2B), indicating that endogenous E. coli proteins were not responsible for the PI3K activity. In addition, the recombinant TcVps34 substrate specificity was studied using PI, PI 4-P, and PI 4,5-P 2 as substrates. The enzyme only phosphorylated PI producing PI 3-P and was not able to use either PI 4-P or PI 4,5-P 2 as substrate (Fig. 2C). These findings, together with the sequence data, led us to conclude that TcVps34 belongs to the class III PI3K family.
TcVps34-overexpressing Epimastigotes Show a Singular Phenotype-Because Vps34 is essential in other trypanosomatids studied (25), we investigated the phenotypic changes induced by TcVps34 overexpression. A His-tagged version of the TcVps34 gene was subcloned into the pTREX integration vector (pTREX-His-TcVps34) and transfected into CL Brener epimastigote cells. Stable cell lines were achieved after 60 days of selection in the presence of G418.
Southern blot analysis indicated that TcVps34 is a singlecopy gene. In addition, the presence of an extra copy of TcVps34, integrated into the genome of the transfected overexpressing cells, was evidenced as an extra hybridization band (Fig. 3A). TcVps34 expression in the transgenic cells was analyzed by Northern and Western blot, showing increased levels for the specific mRNA in pTREX-His-TcVps34-transfected parasites when compared with wild type CL Brener controls and the presence of the tagged protein only in pTREX-His-TcVps34-transfected cells (Fig. 3, B and C). Coomassie Blue staining showed equivalent amounts of protein in the P100 and S100 fractions (results not shown). Furthermore, when PI3K activity assays were carried out using 25 and 75 g of total epimastigote extracts, pTREX-His-TcVps34-overexpressing cells showed a 2.5-fold increase when compared with control cells, thus confirming that activity was linearly dependent on protein concentration and overexpression of a functional enzyme (Fig. 3D).
Examination of cultures by DIC microscopy (Fig. 4A), transmission electron microscopy (Fig. 4B), and video microscopy ( Fig. 4C and supplemental data S2) revealed striking morpho-logical differences between pTREX-His-TcVps34-overexpressing cells and wild type cells. Cells overexpressing His-TcVsp34 showed enlarged contractile vacuoles when compared with wild type cells (Fig. 4A). At the electron microscopy level (Fig.  4B), contractile vacuoles were visible in most cells overexpressing TcVsp34, although this was not the case in wild type cells, as also observed by DIC (Fig. 4A). To exclude the possibility that the histidine tag was playing a role in the observed phenotype, transgenic parasites overexpressing the full-length TcVps34 gene without the tag were generated, and their phenotype was evaluated as described above for pTREX-His-TcVps34. These parasites showed the same morphological alterations when analyzed by DIC and transmission electron microscopy (Fig.  4A, right panel, and Fig. 4B, respectively) indicating that these alterations are because of the enhanced expression of TcVps34. Moreover, these enlarged organelles were functional, as evidenced by following their periodic filling under video micros-  After incubation with the indicated amounts of protein, the reaction products were separated by TLC, and labeled lipids were detected by autoradiography. The main product is PI 3-P; the migration of standards PI, PI 4-P, PI 4,5-P 2 , and the origin are identified by arrows. C, for substrate specificity, PI3K assays were performed as above in the presence of PI, PI 4-P, and PI 4,5-P 2 as substrates. PI 3-P; the migration of standards: PI, PI 4-P, PI 4,5-P 2 , and the origin are identified by arrows. NOVEMBER 14, 2008 • VOLUME 283 • NUMBER 46 JOURNAL OF BIOLOGICAL CHEMISTRY 31545 copy, when pTREX-His-TcVps34-overexpressing epimastigotes were incubated in isosmotic buffer (300 mosM; Fig. 4C and supplemental data S2). Furthermore, most pTREX-His-TcVps34-and pTREX-TcVps34-overexpressing cells (ϳ95%) displayed a hypertrophic alteration of the plasma membrane between the cytostome and the flagellar pocket, as detected by scanning electron microscopy (Fig. 5).

Phosphatidylinositol 3-Kinase in T. cruzi
Effect of PI3K Inhibitors on the Regulatory Volume Decrease in Control and pTREX-His-TcVps34-overexpressing Cells-Because of the distinctive phenotype with enlarged contractile vacuoles observed in the pTREX-His-TcVps34-overexpressing cells, a possible role for TcVps34 in the adaptation to hyposmotic stress was next investigated. Wild type CL Brener and pTREX-His-TcVps34-overexpressing epimastigotes were preincubated for 30 min with 3 M wortmannin or 60 M LY294,000 and then subjected to hyposmotic stress. RVD was followed by light scattering and compared with untreated controls. These inhibitors considerably reduced the RVD in wild type cells, although they had no effect on pTREX-His-TcVps34-overexpressing parasites (Fig. 6, A and  B). To exclude the possibility of toxic side effects because of the treatment with the inhibitors, we confirmed that the plasma membrane remained intact and that the hyposmotically induced amino acid efflux was not affected (data not shown). Because overexpression of TcVps34 could be limiting the effects of the PI3K inhibitors with other possibly present lipid kinases that these inhibitors can affect, we investigated the lipid kinase activity in wild type and pTREX-His-TcVps34-overexpressing parasites in the presence of 3 M wortmannin and 60 M LY294,000. Similarly, we observed that both compounds inhibited the lipid kinase activity in wild type cells (41 and 30% approximately for wortmannin and LY294,000, respectively), whereas they had a minor effect in pTREX-His-TcVps34-overexpressing parasites (9 and 5% approximately for wortmannin and LY294,000 respectively). Video microscopy of both pTREX-TcVps34-and pTREX-His-TcVps34-overexpressing epimastigotes submitted to severe hyposmotic stress (65 mosM) showed that, in contrast to wild type cells, after 20 min of incubation, TcVps34-overexpressing cells were able to recover their original volume ( Fig. 6C and supplemental data S3, S4, and S5). Taken together, these results suggest an active role for TcVps34 in the parasite response to hyposmotic stress. It is important to remark that no differences were observed when wild type cells were compared with those transfected with the pTREX-GFP vector (data not shown).
H ϩ -ATPase and H ϩ -PPase Activities in pTREX-His-TcVps34-overexpressing Cells-It has been reported that in Candida albicans, Vps34p interacts with Vma7p, a subunit of the vacuolar H ϩ -ATPase. In addition, C. albicans Vps34-null mutants show vacuolar acidification defects, as a result of a defective proton transport into the vacuole, because of the lack of Vps34p-Vma7p interaction (35). To investigate the influence of TcVps34 in intracellular acidification, we studied the H ϩ -ATPase and H ϩ -PPase activities in pTREX-His-TcVps34overexpressing cells. In this assay, vesicle acidification induced by ATP or pyrophosphate (PP i ) was measured as the decrease in absorbance of acridine orange at 493-530 nm in permeabilized parasites (33). Results showed that ATP-dependent proton uptake was higher in pTREX-His-TcVps34-overexpressing cells (Fig. 7, trace b) than in wild type cells (Fig. 7, trace a). This activity was strongly inhibited by 1 M of bafilomycin A 1 , a specific inhibitor of the vacuolar-type H ϩ -ATPases (36) (Fig.  7A), and less affected by 100 M of sodium o-vanadate, an inhibitor of P-type H ϩ -ATPases (37) (Fig. 7B). This suggests that proton transport by intracellular compartments in both wild type and pTREX-His-TcVps34-overexpressing cells is mainly achieved by the vacuolar-type H ϩ -ATPase. Interestingly, in contrast to that observed for ATP-driven proton uptake, PP idependent proton transport was higher in wild type cells (Fig.  7C, trace a) than in pTREX-His-TcVps34-overexpressing cells (Fig. 7C, trace b). This transport was strongly inhibited by 100 M of AMDP, a pyrophosphate analog that specifically inhibits vacuolar proton pyrophosphatases (38) (Fig. 7C). These results indicate that overexpression of TcVps34 produces alterations in vesicle acidification, and suggest that changes in acidification mediated by these two pumps might be coordinated. In all Total RNA (30 g) from wild type or pTREX-His-TcVps34-transfected parasites was electrophoresed in agarose-formaldehyde gels, blotted, and hybridized with the same probe described above. The migration position and loading control corresponding to the three ribosomal RNA bands are indicated. C, Western blot analysis of wild type and transfected parasites. Western blot were performed using soluble (S100) or membrane (P100) fractions (80 g) from wild type (WT) or pTREX-His-TcVps34-transfected parasites (His-TcVps34) and the recombinant (Rec) protein purified from E. coli (5 g) as control, resolved by SDS-PAGE (8% gels), electrotransferred on to Hybond C membranes, and revealed with an anti-His 6 antibody. D, enhanced PI3K activity in pTREX-His-TcVps34-transfected parasites. PI3K activity assays were performed as described under "Experimental Procedures" using 25 and 75 g of wild type (WT) and pTREX-His-TcVps34 (His-TcVps34) total parasite extracts. Bars represent the standard deviation of duplicates. experiments, proton gradient was collapsed and intraorganellar pH was neutralized with the addition of 1 M of the K ϩ /H ϩ ionophore nigericin.
Receptor-mediated and Fluid Phase Endocytosis in pTREX-His-TcVps34-overexpressing Cells-To study whether TcVps34 participates in the endocytic pathway, fluid phase uptake of BSA and receptor-mediated endocytosis of transferrin were analyzed and quantified by fluorescence microscopy. Uptake of the fluid phase marker FITC-BSA was not affected in pTREX-His-TcVps34-overexpressing cells when compared with wild type parasites (data not shown). However, receptor-mediated endocytosis measured with FITC-transferrin was significantly blocked in pTREX-His-TcVps34-overexpressing cells observed at 10 and 30 min (Fig. 8A and B, respectively). Quantification of fluorescence indicates an important reduction of internalized transferrin in pTREX-His-TcVps34-overexpressing cells (p Ͻ 0.0001) ( Table 1). These results suggest that TcVps34 is involved mainly in the transport of cargo taken up by receptormediated endocytosis to the lysosome, and they are in agreement with previous observations reported for other lower eukaryotic cells such as T. brucei and S. cerevisiae (23,25).

DISCUSSION
Little is known about the response mechanisms that allow survival of trypanosomatids when subjected to the environmental changes that occur throughout their life cycles. A very important aspect is their capacity to adapt to different osmotic conditions. A previous work has demonstrated a relationship between the RVD that occurs after hyposmotic stress and the cAMP signaling pathway (20). More recently, we characterized a membrane-associated cAMP-specific phosphodiesterase that possesses a FYVE domain, which is known to bind to membranes enriched in PI 3-P and whose activity is modulated by the same inhibitors that affect RVD during the hyposmotic stress (39). These results led us to investigate the importance of PI 3-P production in the osmoregulation of trypanosomatids.
In this work, we characterized TcVps34, the first class III PI3K described in T. cruzi, and we found that this PI3K participates in receptor-mediated endocytosis and osmoregulation in this parasite. TcVps34 is a single-copy gene and shares a high degree of similarity with its mammalian, S. cerevisiae, and T. brucei orthologs. Moreover, its protein sequence conserves the class III PI3K domain structure with an N-terminal C2 domain, an accessory domain, and a C-terminal kinase domain. The enzyme is functional as indicated by the reversion of the thermosensitive phenotype of a S. cerevisiae Vps34 knock-out   NOVEMBER 14, 2008 • VOLUME 283 • NUMBER 46 strain. We also demonstrated that the recombinant protein exclusively phosphorylated PI to produce PI 3-P, supporting our analysis that TcVps34 belongs to the group of class III PI3Ks.

Phosphatidylinositol 3-Kinase in T. cruzi
TcVps34 overexpression in epimastigotes produced distinctive morphological and functional changes related to mem-brane trafficking. The presence of enlarged and functional contractile vacuoles was evident in most TcVps34-overexpressing parasites. In addition, some of these cells showed hypertrophic alterations in the region near the cytostome and the flagellar pocket. This phenotype could originate by a higher membrane trafficking from the enlarged contractile vacuole to these sites, where T. cruzi exchanges material with the environment (40). Furthermore, preincubation of nontransfected epimastigotes with the PI3K inhibitors wortmannin or LY294,000 reduced the regulatory volume decrease of the parasites subjected to hyposmotic stress. Previous studies have shown that the physiological increase in cell volume is a potent stimulus for class I PI3K activation (41,42); however, there is little information about the activation of class III PI3Ks under hyposmotic stress. Although wortmannin and LY294,000 inhibit all PI3Ks to some extent, the observation that plants (43) possess only one class III PI3K and S. cerevisiae (44) and T. brucei (25) possess only one genuine PI3K that appears to be class III by sequence comparison, together with our analysis of the T. cruzi genome, supports the idea that the observed effects are related to the inhibition of TcVps34. We conclude that TcVps34 therefore has a role in osmoregulation. It is interesting to note that Braga and de Souza (45) have reported that epimastigotes of T. cruzi (Y strain) treated with wortmannin also show morphological alterations in the flagellar pocket and membrane detachment from the flagellum. These observations might indicate a connection between TcVps34 and membrane trafficking, because the flagellar pocket together with the cytostome are the main sites where this process takes place in trypanosomatids (40,46). Similar observations have been made upon knockdown of TbVps34 expression by RNA interference in T. brucei (25).
A previous work has demonstrated that antibodies against human Vps34 interfere with sorting of endocytosed plateletderived growth factor receptors and weakly inhibit transferrin FIGURE 6. Effect of PI3K inhibitors on regulatory volume decrease. Wild type (A) and transfected pTREX-His-TcVps34-overexpressing cells (B) were preincubated with either 3 M wortmannin (triangles) or 60 M LY294,000 (squares) and then subjected to hyposmotic stress. Volume recovery compared with untreated controls (circles) was followed by light scattering. Results are representative of those obtained from at least three independent experiments. C, time lapse snapshots of wild type (WT) and transfected pTREX-His-TcVps34overexpressing cells (TcVps34) subjected to 65 mosM hyposmotic stress. Note that in contrast to wild type cells, pTREX-His-TcVps34-overexpressing cells were able to recover their original volume. recycling (47). Furthermore, the existence of an alternative recycling pathway dependent on PI3K activity has been reported (48). In this regard, our results indicate that fluid phase endocytosis is not affected in pTREX-His-TcVps34overexpressing cells, but surprisingly, as it occurs after the knockdown of TbVps34 in T. brucei, overexpression of TcVps34 in T. cruzi produces a blockage of receptor-mediated endocytosis. This could be due to a possible down-regulation of the endocytic pathway, to a defect in recycling of surface receptors, or to a dominant negative effect of the overexpressed protein. Furthermore, in mammalian cells it has been described that PI3K activity regulates the sorting of the epidermal growth factor receptor, whereas the transport of fluid phase markers remains unaffected (49).
Recently, the interaction of Vps34p and a subunit of the vacuolar-type H ϩ -ATPase has been reported in C. albicans (35). In addition, C. albicans vps34-null mutants show acidification defects (31). Therefore, the functional data obtained in this organism resemble the alterations observed in pTREX-His-TcVps34-overexpressing epimastigotes, although the hypothesis of an interaction of TcVps34 and T. cruzi vacuolar-type H ϩ -ATPase remains to be investigated. Interestingly, the observation that pTREX-His-TcVps34-overexpressing cells showed higher V-H ϩ -ATPase and lower H ϩ -PPase activities than wild type parasites suggests that both pumps might be coordinated to regulate the pH levels of intracellular compartments.
It is worth mentioning that in T. cruzi both V-type H ϩ -ATPase and H ϩ -PPase localize to the acidocalcisomes among other cellular locations (19,36,38). Acidocalcisomes are acidic calcium-containing organelles that, together with the contractile vacuole, play a very important role in osmoregulation in T. cruzi (20). This might suggest that the alterations in osmoregulation observed in the presence of PI3K inhibitors could be linked to the defects in proton uptake in the pTREX-His-TcVps34-overexpressing cells. In contrast, H ϩ uptake was poorly affected by vanadate, a specific inhibitor of the P-type H ϩ -ATPase, which is mainly localized in reservosomes (50), thus suggesting that this pump is scarcely altered by TcVps34 overexpression.
In summary, this work provides the first evidence for the presence of a class III PI3K in T. cruzi and describes its role in osmoregulation. Our results also suggest a role for TcVps34 in protein trafficking and vesicle acidification as it occurs with its orthologs in other organisms (Fig. 9). The unique features of T. cruzi vesicular trafficking lead us to propose that this pathway could provide promising targets for the chemotherapy of Chagas disease.  . Multiple roles for TcVps34. Schematic representation of the pathways in which TcVps34 is involved. TcVps34 plays a role in osmoregulation, and on this basis the PI3K inhibitors wortmannin and LY294,000 reduce the RVD of epimastigotes subjected to hyposmotic stress, resulting in a less water movement outside the cell by the contractile vacuole complex (CV) (I). In addition, TcVps34 is implicated in vesicle acidification, as suggested by the changes in the activities of the V-H ϩ -ATPase and the V-H ϩ -PPase proton pumps localized in the acidocalcisomes (Ac) after its overexpression (II). Finally, TcVps34 is involved in membrane trafficking, which is evidenced by a decrease in receptor-mediated endocytosis in cells overexpressing this enzyme (III).