myo-Inositol Uptake Is Essential for Bulk Inositol Phospholipid but Not Glycosylphosphatidylinositol Synthesis in Trypanosoma brucei*

Background: Intracellular myo-inositol homeostasis involves both de novo synthesis and uptake of myo-inositol from the environment. Results: Down-regulation of the myo-inositol transporter in Trypanosoma brucei causes depletion of bulk inositol lipids, but not glycosylphosphatidylinositols, and leads to parasite death. Conclusion: De novo synthesis of myo-inositol is not sufficient to ensure bulk inositol lipid production. Significance: myo-Inositol metabolism in T. brucei is compartmentalized. myo-Inositol is an essential precursor for the production of inositol phosphates and inositol phospholipids in all eukaryotes. Intracellular myo-inositol is generated by de novo synthesis from glucose 6-phosphate or is provided from the environment via myo-inositol symporters. We show that in Trypanosoma brucei, the causative pathogen of human African sleeping sickness and nagana in domestic animals, myo-inositol is taken up via a specific proton-coupled electrogenic symport and that this transport is essential for parasite survival in culture. Down-regulation of the myo-inositol transporter using RNA interference inhibited uptake of myo-inositol and blocked the synthesis of the myo-inositol-containing phospholipids, phosphatidylinositol and inositol phosphorylceramide; in contrast, it had no effect on glycosylphosphatidylinositol production. This together with the unexpected localization of the myo-inositol transporter in both the plasma membrane and the Golgi demonstrate that metabolism of endogenous and exogenous myo-inositol in T. brucei is strictly segregated.


myo-Inositol is an essential precursor for the production of inositol phosphates and inositol phospholipids in all eukaryotes.
Intracellular myo-inositol is generated by de novo synthesis from glucose 6-phosphate or is provided from the environment via myo-inositol symporters. We show that in Trypanosoma brucei, the causative pathogen of human African sleeping sickness and nagana in domestic animals, myo-inositol is taken up via a specific proton-coupled electrogenic symport and that this transport is essential for parasite survival in culture. Down-regulation of the myo-inositol transporter using RNA interference inhibited uptake of myo-inositol and blocked the synthesis of the myo-inositol-containing phospholipids, phosphatidylinositol and inositol phosphorylceramide; in contrast, it had no effect on glycosylphosphatidylinositol production. This together with the unexpected localization of the myo-inositol transporter in both the plasma membrane and the Golgi demonstrate that metabolism of endogenous and exogenous myo-inositol in T. brucei is strictly segregated.
The polyol, myo-inositol, is the most common isomeric form of cyclohexane hexols, a group of molecules having diverse cellular functions (reviewed in Ref. 1). In eukaryotes, myo-inositol is used to synthesize the following: (i) phosphatidylinositol (PI), 2 a structural membrane component and precursor for other inositol-containing membrane lipids such as inositol phosphorylceramide (IPC); and (ii) glycosylphosphatidylinositol (GPI) for membrane anchoring of proteins (reviewed in Refs. [2][3][4]. In addition, PI is a precursor for a large and complex group of membrane-bound signaling molecules and soluble second messengers that are involved in the regulation of cellular processes, such as growth, membrane trafficking, cell migration, differentiation, and apoptosis. Furthermore, myo-inositol and its derivatives are considered compatible solutes for cytoprotection in various eukaryotic cells (reviewed in Refs. 1,5,6).
Intracellular myo-inositol can be acquired by two major pathways. First, it can be generated de novo in a two-step reaction by a mechanism that is conserved from bacteria to eukaryotes (1). Biosynthesis of myo-inositol has been well documented in eukaryotic organisms, including plants, yeast, protozoa, and mammals (1,(7)(8)(9), as well as in many archaea (reviewed in Ref. 10) and certain bacteria (reviewed in Refs. 11,12).
Second, myo-inositol can be imported from the environment via Na ϩ -or H ϩ -linked myo-inositol symporters, known as SMITs or HMITs, respectively. Whereas SMITs of mammals belong to the Na ϩ /glucose transporter family, SGLT (TC 2.A.21.3.1 (13)), HMITs of eukaryotes and prokaryotes are members of the major facilitator superfamily, MFS (TC 2.A.1 (13,14)), which includes proteins capable of transporting a wide range of substrates in response to chemiosmotic ion gradients (15). Although many myo-inositol symporters have been characterized biochemically, the essentiality of SMITs and HMITs for cell viability has only been addressed in a few cases. In Saccharomyces cerevisiae (16) and Candida albicans (17), myo-inositol uptake from the environment was not essential for growth, demonstrating that endogenous production was able to fully compensate for the lack of exogenous myo-inositol. In contrast, in Schizosaccharomyces pombe (18) and Arabidopsis thaliana (19), myo-inositol uptake was shown to be essential for normal development. Furthermore, in mammals, homozygous SMIT1 knock-out mice showing severe depletion of intracellular brain myo-inositol levels, but normal PI content, died shortly after birth (20). However, neonatal lethality could be prevented by prenatal maternal myo-inositol supplementation (21,22). Taken together, although both pathways (de novo myoinositol synthesis and uptake of myo-inositol from the extracellular medium) contribute to intracellular myo-inositol homeostasis, it is unclear in most organisms whether endogenous production and exogenous supply can complement each other or whether both pathways are essential, possibly providing myo-inositol for different cellular processes.
A first indication that myo-inositol from the two pathways may be used for different cellular events came from studies in Trypanosoma brucei, a protozoan parasite causing sleeping sickness in humans and nagana in animals. Deletion of the gene for inositol-3-phosphate synthase in T. brucei bloodstream forms to block endogenous production of myo-inositol resulted in complete inhibition of GPI synthesis. Because the levels of bulk PI and IPC were unaffected, the report suggested that de novo synthesized myo-inositol is primarily used to generate PI for GPI synthesis and hypothesized that myo-inositol for membrane lipid formation might be taken up from the medium (23,24). In support of this hypothesis, the study found that T. brucei bloodstream forms contain two pools of PI synthase as follows: one pool localizing to the ER and possibly being involved in the formation of PI for GPI synthesis using de novo synthesized myo-inositol, and the other pool associating with the Golgi and possibly using myo-inositol taken up from the environment for inositol lipid production (25). However, it was not clear how the two pools of PI synthase recruit their respective substrates, endogenously produced and exogenous myo-inositol. In addition, the requirement for exogenous myo-inositol for lipid synthesis and parasite survival was not tested experimentally (reviewed in Refs. 9, 26).
In T. brucei, myo-inositol uptake has not been studied before. In contrast, in the related trypanosomatid, Leishmania donovani, myo-inositol import was characterized biochemically and is mediated by H ϩ -linked electrogenic symport (27,28). Disruption of the L. donovani myo-inositol transporter gene resulted in reduced parasite growth in culture, but the cells remained viable (29). In addition, in Trypanosoma cruzi, transport activity studies suggested that two distinct myo-inositol transporters (one Na ϩ -coupled and the other H ϩ -coupled) may mediate myo-inositol uptake (30), yet neither transporter was cloned nor experimentally characterized. Only one candidate gene has been identified in the T. cruzi genome (31). In this study, we describe for the first time the identification and functional characterization of an essential myo-inositol transporter in T. brucei. Its down-regulation revealed compartmentalization of myo-inositol metabolism.
RNAi-mediated Gene Silencing-Expression of Tb11.02.3020 (annotated in GeneDB/TriTrypDB as putative sugar transporter) was down-regulated by RNAi using a stem loop construct containing a puromycin resistance gene. Selection of the gene sequences for RNAi was done with RNAit, a prediction algorithm designed to prevent potential cross-talk and hence off-target effects (34). Cloning of the gene fragments into the tetracycline-inducible vector, pALC14 (a kind gift of André Schneider, University of Bern), was performed as described previously (35), using two separate PCR products obtained with primers Tb3020-F (5Ј-GCCCAAGCTTGGATCCCGCTG-CAATCAACACACTCT-3Ј) and Tb3020-R (5Ј-GCTCT-AGACTCGAGTGGGAACACCTGTGAAACAA-3Ј) (spanning nucleotides 842-1181 of Tb11.02.3020), resulting in plasmid pAG3020. Plasmid extraction was performed using the Qiagen plasmid midi kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Before transfection of T. brucei procyclic forms, plasmid DNA was linearized with NotI and precipitated with phenol and chloroform.
Stable Transfection of Trypanosomes and Selection of Clones-T. brucei 29-13 or 427 procyclic forms (4 -5 ϫ 10 7 cells) were harvested at mid-log phase (0.5-0.8 ϫ 10 7 cells/ml) by centrifugation at 1250 ϫ g for 10 min, washed once in buffer (132 mM NaCl, 8 mM KCl, 8 mM Na 2 HPO 4 , 1.5 mM KH 2 PO 4 , 0.5 mM magnesium acetate, 0.09 mM calcium acetate, pH 7.0), resuspended in 450 l of the same buffer, and mixed with 15 g of linearized plasmids pAG3020 or pAG3020-3. Electroporation was performed with a BTX Electroporation 600 System (Axon Lab, Baden, Switzerland) with one pulse (1.5 kV charging voltage, 2.5 kV resistance, 25-microfarad capacitance timing, and 186 ohm resistance timing) and using a 0.2-cm pulse cuvette (Bio-Rad). Electroporated cells were immediately inoculated in 10 ml of SDM-79, containing 15% heat-inactivated FBS, and, if required for selection, 25 g ml Ϫ1 hygromycin and 15 g ml Ϫ1 G418. Clones were obtained by limiting dilutions in 24-well plates in SDM-79, containing 20% conditioned medium, in the presence of 2 g ml Ϫ1 puromycin (RNAi selection) or 25 g ml Ϫ1 hygromycin (in situ tagging). Antibioticresistant clones were tested for the presence of the introduced genes by PCR. Expression of HA-tagged TbHMIT or induction of RNAi was started by addition of 1 g ml Ϫ1 tetracycline to parasite cultures.
RNA Isolation and Northern Blot Analysis-Total RNA for Northern blotting was isolated using the SV Total RNA Isolation System (Protégé, Madison, WI), following the manufacturer's instructions. RNA (10 g) was separated on formaldehydeagarose gels (1% agarose, 2% formaldehyde in MOPS) and myo-Inositol Uptake in T. brucei transferred to GeneScreen Plus nylon membranes (Perkin-Elmer Life Sciences). 32 P-Labeled probes were made by random priming the same PCR products used as inserts in the stemloop vector using the Prime-a-Gene labeling system (Promega). Hybridization was performed overnight at 60°C in hybridization buffer containing 7% (w/v) SDS, 1% (w/v) bovine serum albumin, 0.9 mM EDTA, 0.5 M Na 2 HPO 4 , pH 7.2, and the membrane was analyzed by autoradiography using BioMax MS film and a TransScreen-HE intensifying screen. Ribosomal RNA was visualized on the same formaldehyde-agarose gel by ethidium bromide staining to control for equal loading.
myo-inositol Uptake Assays-T. brucei procyclic forms (1 ϫ 10 8 cells) at mid-log phase (0.9 -1.1 ϫ 10 7 cells/ml) were harvested by centrifugation at 1250 ϫ g for 10 min and resuspended in phosphate-buffered saline (PBS; 135 mM NaCl, 1.3 mM KCl, 3.2 mM Na 2 HPO 4 , 0.5 mM KH 2 PO 4 , pH 7.4) at 27°C. Uptake of myo-[ 3 H]inositol was measured by adding 50 nM myo-[ 3 H]inositol to cells at 27°C. At various time points, uptake of label was terminated by pelleting 1.5 ϫ 10 7 parasites by centrifugation (1500 ϫ g, 5 min, 4°C) and washing three times in ice-cold PBS. After resuspension of the pellet in 100 l of PBS, radioactivity was measured by scintillation counting using a Packard Tri-Carb 2100TR liquid scintillation analyzer (PerkinElmer Life Sciences). Aliquots of the parasite suspensions before centrifugation were used to determine the total amount of radioactivity in the assay. All measurements were made in triplicates. The uptake of myo-[ 3 H]inositol at each time point was calculated and plotted as a function of incubation time. Uptake of label was linear for 90 min, with regression coefficients of 0.977 or higher.
Substrate Specificity Studies-To study the substrate specificity of the putative myo-inositol transporter, D-glucose, D-mannose, D-sorbitol, D-mannitol, xylitol, scyllo-inositol, and myo-inositol 2-monophosphate-bis-cyclohexylammonium salt (myo-inositol 2-phosphate) were added to T. brucei procyclic forms together with myo-[ 3 H]inositol. Uptake was terminated after 60 min of incubation at 27°C. All measurements were made in triplicates. Statistical analysis of the data was performed by the paired sample t test with two-tailed p values.
Metabolic Labeling of Trypanosomes and Extraction Protocols-Metabolic labeling of trypanosomes was performed essentially as described before (36). Briefly, myo-[ 3 H]inositol or [ 3 H]Etn was added to procyclic form trypanosomes at a density of 0.7-1.0 ϫ 10 7 cells/ml, and incubation was continued for 16 h. Cells were harvested by centrifugation at 1750 ϫ g for 10 min and washed with ice-cold Tris-buffered saline (10 mM Tris, 144 mM NaCl, pH 7.4) to remove unincorporated label, and bulk phospholipids were extracted with 2ϫ 10 ml of chloroform/methanol (CM; 2:1, by volume). CM fractions were pooled, dried under nitrogen, and resuspended in a small volume of CM. For isolation of GPI precursors and free GPIs, the remaining pellet was extracted three times with 5 ml of chloroform/methanol/water (CMW; 10:10:3, by volume). The delipidated pellet was extracted two times with 0.5 ml of 9% (v/v) butan-1-ol in water (butanol fraction), followed by 0.1% (w/v) Triton X-100 in 20 mM Tris-HCl, pH 7.4 (TX fraction), to solubilize GPI-anchored procyclins. The remaining material was resuspended in 1% (w/v) SDS (SDS fraction). CMW fractions were pooled, dried under nitrogen, and partitioned between butan-1-ol (CMWbut) and water (CMWaqu) to separate GPI precursors and free GPIs, respectively (37).
Lipid Analysis by Thin Layer Chromatography (TLC)-TLC was performed on Silica Gel 60 plates (Merck). To analyze the major inositol phospholipid classes, CM extracts were separated by one-dimensional TLC using solvent system 1 composed of chloroform/methanol/acetone/acetic acid/water (40: 15:15:12:8, by volume). Complete separation of all phospholipid classes was achieved by two-dimensional TLC using solvent system 2 (chloroform/methanol/ammonia/water, 90:74:12:9; by volume) for the first and solvent system 1 for the second dimension (38). On each plate, appropriate lipid standards were run alongside the samples to be analyzed. Lipid spots on TLC plates were visualized by exposure to iodine vapor. 3 H-Labeled GPI precursors were analyzed by one-dimensional TLC using solvent system 3 (chloroform:methanol:water, 10:10:3; by volume). Radioactivity was detected by scanning the air-dried plate with a radioisotope detector (Berthold Technologies, Regensdorf, Switzerland) and quantified using the Rita Star software provided by the manufacturer. Alternatively, plates were sprayed with EN 3 HANCE (PerkinElmer Life Sciences) and exposed to MXB films at Ϫ70°C.
Lipid Phosphorus Determination-Phospholipids in chromatographic fractions were scraped from TLC plates and digested by boiling in 70 -72% perchloric acid for 30 -45 min. The released inorganic phosphate was reacted with ammonium molybdate and quantified photometrically (39). Each determination was accompanied by a series of inorganic phosphate standards. The assay was linear between 0 and 200 nmol of phosphate per sample (linear correlation coefficients were Ͼ0.99 in all experiments). Statistical analysis of the data were performed as mentioned above.
SDS-PAGE and Fluorography-Extracted proteins were separated on 12% polyacrylamide gels under reducing conditions (40). For detection of 3 H-labeled proteins, gels were fixed, stained with Coomassie Brilliant Blue, immersed in Amplify (GE Healthcare), dried, and exposed to MBX films at Ϫ70°C.
Functional Characterization in Xenopus Oocytes-Electrophysiological experiments were performed using an OC-725C two-electrode voltage clamp amplifier (Warner Instruments Corp., Hamden, CT) in combination with an XY-recorder, or the signal was digitized at 100 Hz with MacLab/200 (AD Instruments, Spechbach, Germany). Experiments were carried out at a holding potential of Ϫ80 mV. The perfusion medium contained 90 mM NaCl, 1 mM KCl, 1 mM MgCl 2 , 1 mM CaCl 2 , and 5 mM HEPES-NaOH, pH 7.4. The perfusion solution (6 ml/min) was applied through a glass capillary with an inner diameter of 1.35 mm, the mouth of which was placed about 0.5 mm from the surface of the oocyte (43). Concentration-response curves were fitted with the equation I(c) ϭ I max /(1 ϩ (EC 50 /c)), where c is the concentration of myo-inositol; EC 50 is the concentration of myo-inositol eliciting the half-maximal current amplitude; I max is the maximal current amplitude, and I is the current amplitude. myo-Inositol was applied for 20 s for each measurement.
Generation of HA-tagged TbHMIT-To generate in situ HAtagged TbHMIT, one allele of Tb11.02.3020 was C-terminally tagged with 3ϫ HA using the one-step PCR strategy described elsewhere (44). Alternatively, to overexpress HA-tagged TbH-MIT using the inducible vector pALC14, genomic DNA was amplified using primers Tb3020-3-F (5Ј-GCGCCCAAGCT-TATGAAGTGGCGCGTGAAGATA-3Ј) and Tb3020-3-R (5Ј-CCCCGCGGATCCCTAAATGGGCGCACGGGCGGT-3Ј), containing a 3ϫ HA tag and a stop codon at the 5Ј-end. The PCR product was ligated between HindIII and BamHI sites of pALC14, resulting in plasmid pAG3020-3. Before transfection of T. brucei procyclic forms, plasmid DNA was linearized with NotI and precipitated with phenol and chloroform.
Microscopy-For immunolocalization of in situ HA-tagged TbHMIT, trypanosomes were collected by centrifugation at 1250 ϫ g for 10 min, washed once with PBS, and allowed to settle onto Superfrost Plus Microscope Slides (Thermo Scientific, Braunschweig, Germany). Parasites were fixed with 4% paraformaldehyde in PBS for 10 min, washed three times with PBS, permeabilized with 0.2% Triton X-100 in PBS for 5 min, and blocked with 2% (w/v) bovine serum albumin in PBS. Trypanosomes harboring the inducible vector for HA-tagged TbHMIT were cultured in the presence of tetracycline for 24 h to induce protein expression and processed for immunolocalization as described above. HA-tagged proteins were detected using monoclonal mouse anti-HA antiserum (Covance, Munich, Germany) at a dilution of 1:250 in PBS for 1 h at room temperature. Golgi was labeled by incubating fixed parasites for 1 h at room temperature with rabbit anti-TbGRASP antibody (45) (kindly provided by G. Warren, University of Vienna; used at a dilution of 1:1000). Subsequently, the slides were washed three times with PBS and incubated with the corresponding secondary antibodies, Alexa Fluor 594 goat anti-mouse IgG or Alexa Fluor 488 goat anti-rabbit IgG (Invitrogen) at a dilution of 1:1000 for 1 h at room temperature. Slides were washed three times with PBS and mounted using Vectashield containing 4Ј,6Ј-diamidino-2-phenylindole (DAPI; Vector Laboratories). Fluorescence microscopy was performed on a Leica AF6000 microscope (Leica Microsystems, Heerbrugg, Switzerland), using the software provided by the manufacturer.
In Silico Methods-Trypanosomatid myo-inositol transporter orthologs were retrieved from UniProt Knowledgebase (UniProtKB release 2011_04) by a blastp search using the L. donovani H ϩ -myo-inositol transporter LdMIT (UniProtKB E9BGT1) as query. The obtained hits were supplemented with all annotated polyol and inositol transporters of the manually curated Swiss-Prot section of UniProtKB. The sequences were aligned with ClustalW2 (46), and an unrooted tree was constructed with Dendroscope (47).

Identification of an Essential Gene in T. brucei Encoding a
Putative myo-Inositol Transporter-Using the translated protein sequence of the experimentally confirmed H ϩ -myo-inositol transporter from L. donovani (UniProtKB E9BGT1 (48)), we searched the predicted proteome of T. brucei and identified a gene encoding a putative T. brucei myo-inositol transporter (GeneDB Tb11.02.3020). The gene had been annotated as putative sugar transporter and showed 54 and 64% identity at the protein level to the L. donovani transporter and to a putative T. cruzi myo-inositol transporter (TriTrypDB Tc00. 1047053505183.130 (31)), respectively. Using Phobius (49) and TMHMM (50) algorithmic predictions, the translated T. brucei protein, named TbHMIT, consists of 12 transmembrane regions. Homologous proteins could also be found in other trypanosomatids, such as Trypanosoma congolense, Trypanosoma vivax, Leishmania braziliensis, Leishmania major, Leishmania mexicana, and Leishmania infantum. Amino acid sequence comparison shows that the highest degree of identity between the homologous proteins of the different species resides in transmembrane domains 1, 4, and 11 and in a short loop between transmembrane domains 11 and 12 (data not shown).
To study if TbHMIT is essential in T. brucei procyclic forms in culture, we generated tetracycline-inducible RNAi cell lines against Tb11.02.3020. Transfection of T. brucei procyclic forms with plasmid pAG3020 and selection by resistance to puromycin resulted in several mutant clones, two of which (A3 and A4) were selected for all subsequent experiments. After induction of RNAi by addition of tetracycline, both clones showed a clear growth reduction compared with uninduced (control) cells, i.e. parasite growth stopped after 4 days of culture (Fig. 1A). Northern blot analysis showed that after 2 days of RNAi, Tb11.02.3020 mRNA could no longer be detected (Fig. 1A,  inset). Together, these results demonstrate that expression of Tb11.02.3020 is essential for normal growth of T. brucei procyclic forms in culture.
RNAi against TbHMIT Inhibits myo-Inositol Uptake into T. brucei-To investigate if TbHMIT is involved in uptake of myoinositol into T. brucei, trace amounts of myo-[ 3 H]inositol were added to RNAi clone A3 before and after down-regulation of TbHMIT, and time-dependent incorporation of label into parasites was determined by measuring radioactivity in the cell myo-Inositol Uptake in T. brucei pellet after centrifugation. The results in Fig. 1B show that uptake of myo-[ 3 H]inositol into control trypanosomes increased linearly over 90 min. In contrast, in the RNAi clones after down-regulation of TbHMIT, uptake of myo-[ 3 H]inositol was reduced by Ͼ90% compared with control cells, demonstrating that TbHMIT is involved in myo-inositol uptake into T. brucei procyclic forms.
TbHMIT Is a myo-Inositol Transporter-To study if TbH-MIT itself mediates myo-inositol transport, a cRNA coding for TbHMIT was microinjected into Xenopus oocytes. After 3 days, perfusion of oocytes with 10 mM myo-inositol resulted in an inward current of 55 Ϯ 15 nA (mean value Ϯ S.D. using 13 oocytes from four independent batches). In control oocytes injected with water instead of cRNA, no current was induced upon perfusion with 10 mM myo-inositol (n ϭ 9). To determine the apparent affinity of the transport system, oocytes were exposed to increasing concentrations of myo-inositol ( Fig. 2A). The data from four experiments were averaged and indicated an EC 50 value of the current elicited by myo-inositol of 61 Ϯ 5 M (Fig. 2B). When the measurements were repeated in a medium, in which all Na ϩ had been replaced by K ϩ , the current amplitudes induced by myo-inositol were not changed significantly, indicating that uptake of myo-inositol was independent of the extracellular Na ϩ concentration. In contrast, a change in pH of the assay medium from 7.4 to 6.5 increased the inward current in response to 10 mM myo-inositol to 152 Ϯ 11%, whereas a change in pH to 8.5 decreased it to 42 Ϯ 3% of the value observed at pH 7.4 (means Ϯ S.D. using nine oocytes from three independent batches; Fig. 2C). These results indicate that uptake of myo-inositol is coupled to H ϩ transport.
Transport Specificity of TbHMIT-The substrate specificity of TbHMIT was analyzed in T. brucei procyclic forms by measuring uptake of myo-[ 3 H]inositol (50 nM final concentration) in the absence or presence of high concentrations of unlabeled potential substrates (Fig. 3). The results in Fig. 3 show that uptake of myo-[ 3 H]inositol was not affected by the presence of the polyols D-sorbitol, D-mannitol, or xylitol (5 mM, final concentrations). Similarly, no effect on myo-[ 3 H]inositol uptake was observed in the presence of the two sugars, D-glucose and D-mannose (5 mM, final concentrations). In contrast, uptake of myo-[ 3 H]inositol was blocked by the presence of 5 mM unlabeled myo-inositol (Ͼ98% inhibition; positive control), whereas scyllo-inositol and myo-inositol 2-phosphate (5 mM, final concentrations) inhibited uptake of myo-[ 3 H]inositol by 75 and 95%, respectively. This, together with the observation that scyllo-inositol and myo-inositol-2-phosphate also elicited small current responses in TbHMIT-expressing oocytes, indicates that the two inositol isomers are also substrates of TbHMIT. When uptake of myo-[ 3 H]inositol in T. brucei procyclic forms was measured in the presence of increasing concentrations of unlabeled myo-inositol (0 -5 mM, final concentrations), an EC 50 value similar to that determined in TbHMIT-expressing Xenopus oocytes was obtained (data not shown).

Effect of TbHMIT RNAi on PI and IPC Synthesis in T. brucei Procyclic Forms-De novo synthesis of inositol-containing lipids in T. brucei RNAi cells before and after down-regulation of
TbHMIT was studied by labeling parasites with myo-[ 3 H]inositol for 18 h. We found that lipid extracts from RNAi cells after 3 days of induction with tetracycline incorporated Ͻ11% of radioactivity of control uninduced cells (data from three experiments using clones A3 and A4). Subsequent analysis of 3 H-labeled lipids by one-dimensional TLC and fluorography demonstrated that most radioactivity co-migrated with PI and IPC (Fig. 4A). Quantification of the spots after scraping and liquid scintillation counting showed that the amounts of radioactivity in PI and IPC from induced cells were Ͻ10% (values from two independent TLC analyses) of those in the respective lipids from uninduced cells (Fig. 4A, compare lanes A3ϩ and A4ϩ  versus lane A3Ϫ). The identification of the labeled lipids as PI and IPC was confirmed by two-dimensional TLC and mass spectrometry. Together, these results demonstrate that downregulation of TbHMIT inhibits incorporation of exogenous myo-[ 3

H]inositol into PI and IPC.
Down-regulation of TbHMIT Changes Parasite Phospholipid Composition-Possible changes in steady-state levels of T. brucei phospholipids after RNAi-mediated down-regulation of TbHMIT were analyzed by lipid quantification after two-dimensional TLC. Total lipid extracts from TbHMIT clones A3 and A4 grown in the absence or presence of tetracycline for 4 days revealed a Ͼ92 and Ͼ83% reduction in IPC and PI, respectively, in TbHMIT knockdown parasites compared with untreated cells (Table 1). In contrast, RNAi against TbHMIT resulted in an increase in the sphingomyelin content by Ͼ2.3fold as compared with uninduced cells. In addition, we observed a small increase in phosphatidylcholine (PC) and phosphatidylserine levels and a small decrease in phosphatidylethanolamine (PE). The cardiolipin content was not significantly changed in RNAi cells compared with control parasites ( Table 1).
Effect of TbHMIT Down-regulation on GPI Synthesis-Because myo-inositol is a component of the GPI core structure and thus is essential for GPI synthesis, we studied a possible effect of TbHMIT down-regulation on de novo GPI formation in T. brucei procyclic forms. RNAi cells were cultured in the absence or presence of tetracycline for 2 days and subsequently incubated with [ 3 H]Etn for 16 h to label GPI precursors, free GPIs and GPI-anchored proteins (36). Sequential extraction of GPI molecules revealed that the amounts of radioactivity recovered in the fractions containing the GPI precursors and free GPIs were similar in RNAi cells incubated in the absence or presence of tetracycline (data not shown). Analysis of the GPI precursor fraction by one-dimensional TLC followed by radio- Current amplitudes obtained at a given myo-inositol concentration were averaged. The data points represent mean values Ϯ S.D. and were fitted as indicated under "Experimental Procedures." C, currents elicited in an oocyte by 10 mM myo-inositol were measured as in B. During the application of myoinositol, the pH of the perfusion medium was changed from pH 7.4 to 6.5 or pH 8.5 and back to pH 7.4, as indicated below the bars. Acidification resulted in an increase and alkalinization in a decrease in current amplitude.

myo-Inositol Uptake in T. brucei
activity scanning showed no major differences between extracts from induced and noninduced cells (Fig. 4B). In both samples, Ͼ65% of radioactivity associated with the major GPI precursor lipid, designated PP1 (51), with the rest being in GPI precursors migrating between the origin and PP1. In addition, each lane showed a small amount of residual [ 3 H]PE that was not extracted together with bulk phospholipids. The radioactivity scans were similar to those from a previous study, in which the individual GPI lipids had been characterized in detail (51). In addition, similar amounts of [ 3 H]Etn were incorporated into the major GPI-anchored protein of T. brucei procyclic forms, EP procyclin, in uninduced and induced cells (Fig. 4B,  inset), indicating that de novo GPI protein synthesis was not affected by RNAi against TbHMIT. In control experiments, phospholipid analysis of these [ 3 H]Etn-labeled parasites confirmed that TbHMIT expression was down-regulated in all samples used for GPI lipid and protein analysis, i.e. the phospholipid composition in [ 3 H]Etn-labeled TbHMIT-depleted cells showed the same alterations compared with control cells (see Table 1). Together, these results demonstrate that downregulation of TbHMIT by RNAi has no effect on the biosynthesis of GPI molecules in T. brucei procyclic forms.
Localization of TbHMIT in T. brucei Procyclic Forms-Localization of in situ HA-tagged TbHMIT in T. brucei procyclic forms showed staining of an intracellular structure located between the nucleus and the kinetoplast (Fig. 5A). Similar staining using immunofluorescence microscopy and immunolabeling at the electron microscope level has been reported before in T. brucei procyclic forms labeled with an antibody against the Golgi marker TbGRASP (45). Co-staining of parasites expressing in situ HA-tagged TbHMIT with anti-HA and anti-Tb-GRASP showed that the two signals completely co-localized (Fig. 5A). Similar results were obtained in parasites in which HA-tagged TbHMIT was overexpressed (Fig. 5B); although in these cells the area stained by anti-HA was larger than that stained by anti-TbGRASP, the two signals always co-localized, even in cells undergoing Golgi duplication (Fig. 5B, 2nd row  (45)). In addition, anti-HA clearly revealed surface staining in parasites overexpressing HA-TbHMIT (Fig. 5B); similar staining could also be seen in cells expressing in situ HA-tagged TbHMIT; however, the signal only became apparent when the slides were overexposed (data not shown). The localization of in   APRIL 13, 2012 • VOLUME 287 • NUMBER 16 situ tagged HA-TbHMIT did not change when parasites were incubated in low myo-inositol-containing medium for 24 h (data not shown).

myo-Inositol Uptake in T. brucei
Phylogenetic Positioning of TbHMIT-To determine the phylogenetic position of TbHMIT relative to other myo-inositol/ polyol transporters, we generated an unrooted tree based on the multiple alignment of all trypanosomatid myo-inositol transporter orthologs with various known or predicted myo-inositol/polyol transporters (Fig. 6). In this phylogram, the trypanosomatid myo-inositol transporters group as a separate clade clearly distinct from their nearest neighbors, the HMITs from plants and mammals (e.g. 38% amino acid identity between TbHMIT and Homo sapiens HMIT Q96QE2), although they share considerably less similarity with mammalian SMITs (e.g. 13% amino acid identity between TbHMIT and H. sapiens SMIT P53794). Interestingly, the T. cruzi Brener Esmeraldo haplotype carries an myo-inositol transporter pseudogene (TriTrypDB Tc00.1047053504125.100) containing multiple inframe stop codons at the syntenic position, whereas the T. cruzi Brener non-Esmeraldo-like haplotype possesses a full-length myo-inositol transporter ortholog (Tc00.1047053505183.130).

DISCUSSION
Most organisms able to synthesize myo-inositol de novo from glucose 6-phosphate also express transporters for uptake of myo-inositol from the environment, suggesting that both processes may be required to fulfill the cellular demand for myoinositol or that de novo synthesized myo-inositol is used for different cellular processes than myo-inositol imported from the surrounding medium.
In trypanosomatids, myo-inositol transport has been documented in L. donovani (27,29,48) and T. cruzi (30,31). However, although an H ϩ -myo-inositol transporter from L. donovani has been functionally characterized, the dependence of trypanosomatids on uptake of extracellular myo-inositol for growth has not been studied in detail. By using RNAi in combination with biochemical transport assays and phospholipid analyses in T. brucei procyclic forms and functional expression studies in Xenopus oocytes, we describe the first identification of an essential H ϩ -coupled myo-inositol symporter in T. brucei.
Recombinant TbHMIT expressed in Xenopus oocytes mediated an electrogenic transport of myo-inositol with micromolar affinity. Although the transport was clearly not dependent on the presence of Na ϩ ions, the pH sensitivity indicated that the charge is contributed by H ϩ , strongly suggesting that TbHMIT represents a H ϩ -coupled myo-inositol symporter. This classification is consistent with the close phylogenetic relationship of TbHMIT to plant and mammalian HMITs and more distant relationship to SMITs. In addition, the observed specificity of TbHMIT for myo-inositol, and certain inositol derivatives, is in line with previous studies on HMITs from L. donovani, T. cruzi, A. thaliana, and C. albicans (29,30,52,53), showing that other polyols and hexoses are not transported by HMITs, in contrast to SMITs, which readily transport glucose (54 -56). Our observation that scyllo-inositol and myo-inositol phosphate inhibited FIGURE 5. Immunofluorescence microscopy of T. brucei procyclic forms. A, trypanosomes expressing in situ HA-tagged TbHMIT were washed, allowed to settle onto microscope slides, fixed with paraformaldehyde, and permeabilized with Triton X-100. B, trypanosomes were cultured in the presence of tetracycline for 24 h to induce overexpression of HA-tagged TbHMIT and treated as described in A. Proteins were detected using antibodies against HA and TbGRASP, in combination with the corresponding fluorescent secondary antibodies. DNA (shown in the merged pictures) was stained with DAPI. DIC, differential interference contrast.

myo-Inositol Uptake in T. brucei
myo-[ 3 H]inositol uptake in T. brucei procyclic forms and elicited small currents in Xenopus oocytes indicates that these derivatives are also substrates for TbHMIT. To our knowledge, this is the first study in a protozoan parasite showing that myoinositol uptake is essential. The specificity of TbHMIT (and likewise other trypanosomatid HMITs) could potentially be exploited to design toxic myo-inositol analogs as trypanocidal drugs. In addition, the strict dependence of T. brucei from exogenous myo-inositol validates TbHMIT as a potential drug target.
It has been shown in T. brucei bloodstream forms that de novo synthesis of myo-inositol is essential for parasite growth, providing myo-inositol for GPI synthesis, and that this endogenously produced pool of myo-inositol cannot be substituted by uptake of myo-inositol from the extracellular environment (23,25). We now show that the converse is true as well, i.e. that uptake of myo-inositol from the medium is essential for T. brucei growth and that exogenous myo-inositol is used for the synthesis of membrane phospholipids but not for GPI formation. RNAi-mediated ablation of TbHMIT expression resulted in inhibition of myo-inositol uptake into T. brucei procyclic forms and caused a growth arrest. In addition, in vivo labeling experiments using myo-[ 3 H]inositol and quantitative analysis of the phospholipid composition demonstrated that down-regulation of TbHMIT inhibited de novo formation of the myo-inositolcontaining phospholipids, PI and IPC, resulting in a grossly altered trypanosome membrane lipid composition. Based on previous studies showing that inhibition of de novo phospholipid synthesis in T. brucei is lethal, by affecting cell cycle progression and organelle integrity (23, 25, 57-59), it is likely that the reduced levels of inositol-containing phospholipids in TbHMIT-depleted parasites are the major cause for the observed growth arrest. In contrast, by analyzing GPI formation using [ 3 H]Etn labeling, we found that de novo synthesis of GPI lipids and GPI-anchored proteins is not affected by downregulation of TbHMIT expression, i.e. GPI formation is independent of the supply of myo-inositol from the extracellular medium.
Besides depletion of PI and IPC, we noticed an increase in the SM content in T. brucei procyclic forms after down-regulation of TbHMIT. Because the decrease in IPC matched the increase in SM, it is likely that the lack of myo-inositol prompted a shift in ceramide usage from IPC production to SM synthesis, perhaps in an attempt to maintain the levels of ceramide-based membrane lipids. In addition, TbHMIT-depleted parasites showed an increase in both PC and phosphatidylserine and a decrease in PE. Although the higher relative amounts of PC and phosphatidylserine can be considered compensatory increases due to the reduction in PI, the relative decrease in PE is difficult to explain and may indicate a connection between PI and PE turnover.
The localization of TbHMIT in the Golgi was unexpected. It is generally believed that PI synthesis from exogenous inositol, mediated by PI synthase, occurs on the cytosolic side of the ER. Experiments using purified PI synthase from S. cerevisiae showed that the enzyme was asymmetrically incorporated into artificial vesicles with the active site facing outward (60). However, to our knowledge, no detailed studies have been done to determine the in vivo topology of PI synthase in any organism. Based on our results, we propose that, at least in T. brucei, the active site of PI synthase may be exposed to the lumen of the ER, catalyzing the synthesis of PI for GPI synthesis from endogenously produced inositol. In contrast, PI synthase in the Golgi exclusively uses inositol taken up from the environment into the Golgi via TbHMIT for PI production. Such a model is consistent with the following experimental observations: (i) PI in T. brucei shows a dual localization in the ER and Golgi (25); (ii) endogenously produced inositol from glucose 6-phosphate is used for PI synthesis in the ER for GPI anchor production (24); (iii) inositol taken up from the environment is used for bulk PI production but not GPI anchor synthesis (this study); (iv) extracellular inositol is taken up via TbHMIT localized in both the plasma membrane and the Golgi (this study); and (v) downregulation of TbHMIT blocks bulk PI and IPC production (this study). Because IPC is produced from PI via transfer of the inositol head group, mediated by sphingolipid synthase 1 (TbSLS1) (61), our results predict that TbSLS1 would also localize to the Golgi. Interestingly, the related enzyme, TbSLS4, catalyzing the production of sphingomyelin and ethanolamine phosphorylceramide from PC and PE, respectively, has been shown to be in the Golgi in T. brucei bloodstream forms (62). Furthermore, the proposed compartmentalization of myo-inositol and PI synthesis in T. brucei is consistent with the previous observation that only a subset of PI molecular species, presumably the subset produced from endogenous myo-inositol, is used for GPI anchor synthesis in T. brucei procyclic forms (63).
Interestingly, in a recent report, HMIT has been found to co-localize with a Golgi marker in primary cultured rat neurons (64). Although the study did not address the metabolism of myo-inositol, the observation suggests that membrane PI synthesis in mammalian cells may also occur in the Golgi. This in turn would explain why in mammalian cells myo-[ 3 H]inositol is often a poor precursor to label (and identify) GPI-anchored proteins; in analogy to T. brucei, exogenous inositol transported into the Golgi would not be available for GPI anchor synthesis in the ER. In contrast, myo-[ 3 H]inositol is the preferred substrate to label GPI anchors in S. cerevisiae (reviewed in Ref. 65), suggesting that yeast may be able to transport myoinositol into the ER or that the active site of yeast PI synthase faces the cytosol.