Cellular and Molecular Remodeling of the Endocytic Pathway during Differentiation of Trypanosoma brucei Bloodstream Forms

Transcription

1 EUKARYOTIC CELL, Aug. 2010, p Vol. 9, No /10/$12.00 doi: /ec Copyright 2010, American Society for Microbiology. All Rights Reserved. Cellular and Molecular Remodeling of the Endocytic Pathway during Differentiation of Trypanosoma brucei Bloodstream Forms Benoit Vanhollebeke, 1,2 Pierrick Uzureau, 1 Daniel Monteyne, 1 David Pérez-Morga, 1 and Etienne Pays 1 * Laboratory of Molecular Parasitology, Institute of Molecular Biology and Medicine, Université Libre de Bruxelles, 12 rue des Professeurs Jeener et Brachet, B-6041 Gosselies, Belgium, 1 and Department of Biochemistry and Biophysics, UCSF, 1550 Fourth Street, San Francisco, California Received 26 March 2010/Accepted 15 June 2010 During the course of mammalian infection, African trypanosomes undergo extensive cellular differentiation, as actively dividing long slender (SL) forms progressively transform into intermediate (I) forms and finally quiescent G 1 /G 0 -locked short stumpy (ST) forms. ST forms maintain adaptations compatible with their survival in the mammalian bloodstream, such as high endocytic activity, but they already show preadaptations to the insect midgut conditions. The nutritional requirements of ST forms must differ from those of SL forms because the ST forms stop multiplying. We report that the uptake of several ligands was reduced in ST forms compared with that in SL forms. In particular, the haptoglobin-hemoglobin (Hp-Hb) complex was no longer taken up due to dramatic downregulation of its cognate receptor, TbHpHbR. As this receptor also allows uptake of trypanolytic particles from human serum, ST forms were resistant to trypanolysis by human serum lipoproteins. These observations allowed both flow cytometry analysis of SL-to-ST differentiation and the generation of homogeneous ST populations after positive selection upon exposure to trypanolytic particles. In addition, we observed that in ST forms the lysosome relocates anterior to the nucleus. Altogether, we identified novel morphological and molecular features that characterize SL-to-ST differentiation. Trypanosoma brucei is a protozoan parasite of clinical, veterinary, and economical importance. It develops in the bloodstream of a variety of mammals that invariably die if left untreated. Its division rate is high, with typical population doubling times ranging from 6 h in vitro to less than 3 h in vivo. If this process was uncontrolled, the parasite would rapidly kill its host, with impaired transmission through the tsetse fly vector. The parasite exhibits a surface coat made of an abundant and immunogenic antigen, termed VSG for variant surface glycoprotein (24). This coat triggers an efficient antitrypanosome humoral response, but this response is not sufficient to control infection, as revealed from the rapid death of mice infected with monomorphic trypanosomes in which growth is deregulated. Instead, it is the combination of parasite cell density self-regulation and the immune response that shapes the parasitemia in a succession of characteristic waves (26). The process underlying the transformation of rapidly dividing long slender (SL) forms into G 1 /G 0 -arrested short stumpy (ST) forms is only partially understood. It has been proposed to depend on diffusible small metabolites that might trigger cyclic AMP (camp)-dependent signaling (16). Different protein kinases appear to be involved in a cascade that results in strong overall transcriptional repression with the specific activation of some genes (1, 6, 8, 14, 31). In T. brucei bloodstream forms, the endocytic rate is very * Corresponding author. Mailing address: Laboratory of Molecular Parasitology, IBMM, Université Libre de Bruxelles, 12 rue des Professeurs Jeener et Brachet, B-6041 Gosselies, Belgium. Phone: Fax: epays@ulb.ac.be. Supplemental material for this article may be found at Published ahead of print on 25 June high (9). However, so far only a few surface receptors have been described. The transferrin (Tf) receptor is a heterodimer associating the proteins encoded by two genes present in genomic expression sites for VSG, termed ESAG7 and ESAG6 (for expression site-associated genes 7 and 6, respectively) (25). While ESAG6 is anchored to the plasma membrane by a glycosylphosphatidylinositol (GPI) moiety, ESAG7 is not. Bloodstream forms also possess surface receptors for serum lipoproteins, because these parasites require exogenous sterol, mainly in the form of cholesterol, as a bulk membrane sterol constituent (4, 32). Both low-density lipoprotein (LDL) and high-density lipoprotein (HDL), the major classes of serum lipoproteins, serve as primary sources of lipids (2, 4, 10, 32). However, the molecular identity of such receptors has remained elusive, and contradictory reports have been generated regarding their substrate specificity and diversity. While some favor a scenario in which HDL and LDL are engaged by distinct surface molecules (2), others have provided evidence that a single scavenger receptor might recognize both lipoprotein particles (10). Finally, a T. brucei haptoglobinhemoglobin (Hp-Hb) surface receptor was discovered and termed TbHpHbR (28). It is a GPI-anchored glycoprotein with unknown structure, except for the presence of linear chains of poly-n-acetyllactosamine (pnal), a hallmark of endocytic proteins in T. brucei (21). SL and ST forms are exposed to the same environmental challenges in terms of immune escape and antibody clearance. Therefore, in contrast with the marked differences distinguishing bloodstream and insect-specific procyclic forms (PFs), SL and ST forms share most endocytic characteristics (20). For instance, high Tf and membrane uptake rates were shown to be maintained and even increased in ST forms (7, 20). However, 1272

2 VOL. 9, 2010 ENDOCYTIC REMODELING 1273 the ability to divide clearly distinguishes these developmental forms, and this difference might result in different requirements for nutrients. In order to investigate the adaptation of the endocytic system to the cellular differentiation between SL and ST forms, we undertook a systematic study of the regulation of ligand uptake during this process. Our analyses revealed that while Tf and bulk phase uptake remain high in ST forms, the uptake of Hp-Hb and LDL was reduced. This was linked to reduction of cell labeling with tomato lectin (TL), the pnal-specific lectin that can be used as a general marker for receptor-mediated endocytosis in T. brucei (21). This inhibition of ligand uptake was particularly dramatic in the case of Hp-Hb, the receptor of which appeared to be absent from ST forms. Accordingly, ST forms were resistant to the trypanosome lytic factor 1 (TLF-1), a potent toxic complex of human blood that enters the parasite through this receptor (28). During the course of this analysis, we discovered another characteristic of the endocytic apparatus of ST forms. The lysosome was found to relocate to the anterior part of the cell, resulting in the surrounding of the nucleus by the endocytic system. Altogether, our results highlight important endocytic differences between SL and ST parasites, namely, remodeling of receptor-mediated endocytosis and relocation of the lysosome. These observations could be useful to accurately discriminate and isolate ST forms, particularly in the first steps of the differentiation process. MATERIALS AND METHODS Trypanosome growth and differentiation. The AnTat 1.1E clone of Trypanosoma brucei brucei was used in all experiments. Intraperitoneal injections (100 l of a suspension of 10 5 ml 1 parasites in phosphate-buffered saline glucose [PSGS]) were performed in 8-week-old female Swiss mice preinjected with cyclophosphamide (300 mg kg of body weight 1 ). Parasites were purified from infected blood by separation on DEAE cellulose. The procedure was miniaturized to allow the differentiation process to be followed in the same infected mice over several days. Sixty microliters of infected blood was harvested from the tail vein in capillary tubes with sodium heparin and applied to 1 ml of PSGSequilibrated DEAE resin bed. Elution was performed in 1 ml PSGS solution equilibrated at 4 C to minimize toxicity. Experiments were also validated after purification at 37 C to exclude cold shock-induced differentiation. Briefly, all materials and solutions were equilibrated at 37 C. Blood collection was performed, DEAE columns were used, and washes were performed in a 37 C heated room. Parasites were added to 37 C equilibrated medium for uptake experiments. Ligand preparation and labeling. Alexa Fluor 488-labeled transferrin (Tf- A488; human holotransferrin) and lysine-fixable 40,000-molecular-weight (MW) dextran-fluorescein isothiocyanate (FITC) were from Molecular Probes, and Hp and tomato lectin-fitc were purchased from Sigma-Aldrich. Human HDL and LDL were prepared by sequential flotation ultracentrifugation. The purity of the fractions was over 98% as revealed by Coomassie blue staining of ApoA-I and ApoB. Haptoglobin (Sigma-Aldrich), HDL, and LDL were labeled with the Alexa Fluor 488 protein labeling kit (Molecular Probes) according to the manufacturer s instructions. Flow cytometry and microscopic examinations. Cells (10 6 cells at 10 7 ml 1 )in serum-free HMI-9 growth medium (12) supplemented with 1% bovine serum albumin, 1% glucose, and 50 g ml 1 human hemoglobin (Sigma-Aldrich) were pretreated for 30 min with the lysosomal protease inhibitor FMK-024 (17) when indicated and incubated in the presence of FMK-024 and the indicated concentration of labeled ligand (25 g ml 1 TL-FITC; 20 g ml 1 Hp-A488; 100 g ml 1 LDL-A488 and HDL-A488; 20 g ml 1 LDL-A488 and HDL-A488 for competition assays; 5 g ml 1 Tf-A488 or 5 mg ml 1 dextran-fitc). After uptake, cells were washed in 1 ml phosphate-buffered saline (PBS) and fixed for 10 min at room temperature in 0.1 ml 3.7% paraformaldehyde in PBS. Fixed cells were resuspended at a density of cells ml 1 in 10 mm Tris, ph 8, 0.15 M NaCl solution and either analyzed with a FACSCanto II cytometer (BD Biosciences) or spread on poly-l-lysine-coated slides. Coated cells were permeabilized with 0.1% (vol/vol) Triton X-100 in Tris-buffered saline for 5 min at room temperature. DNA was stained with DAPI (4,6-diamidino-2-phenylindole). Lysosomes were immunodetected with monoclonal anti-p67 antibodies (mab280; J. Bangs and D. Russel). EP1 and PAD1 (5) immunodetection was performed as described previously. Primary antibodies were detected with an Alexa 488-conjugated goat anti-mouse secondary antibody (Molecular Probes). Image processing was applied to all parts of the image and identically to all images of the same figure assembly with Adobe Photoshop software. The distances between the kinetoplast, the lysosome, and the nucleus were established using ImageJ software after calibration with a micrometer slide. For live lysis phenotype examination, cells were immobilized by spreading concentrated suspensions of cells on a slide coated with a thin layer of HMI-9 containing 0.65% (wt/vol) agar at 30 C as described in reference 29. Differentiation in procyclic forms was triggered by incubating freshly isolated cells in SDM-79 for 6hat27 C. Morphological and biochemical analyses of slender, intermediate, and stumpy forms. All morphological and biochemical scoring assays were performed through blinded examinations. For NAD diaphorase assays, cells applied as a coating on poly-l-lysine slides were fixed for 5 min with 2.5% (vol/vol) glutaraldehyde in PBS, rinsed in distilled water, and incubated for 2 h with freshly prepared NAD diaphorase reaction mixture (1.3 mg ml 1 of reduced -NADH disodium salt and 0.3 mg ml 1 nitroblue tetrazolium dissolved in PBS) before being washed in PBS and processed for microscopic examination. Morphological discrimination of different forms relied on integration of the following criteria: overall cell morphology and volume, length of protruding flagellum, distance of the flagellum from the cell body, and position of the nucleus relative to the posterior end of the cell. Northern blot analysis. Total RNA was isolated using the Trizol reagent (Gibco BRL) from at least 10 8 cells freshly isolated from mice for bloodstream forms and from in vitro cultures for procyclic forms. Ten-microgram samples were electrophoresed on a 1% (wt/vol) agarose-formaldehyde gel and transferred to a Hybond-C nitrocellulose membrane (Amersham Biosciences). Blots were hybridized with 32 P-radiolabeled probes (Rediprime II random prime labeling system; Amersham Biosciences) corresponding to the 580-bp TbHpHbR open reading frame (ORF) 5 extremity. After hybridization, the blots were washed three times with 0.1 SSC (1 SSC is 0.15 M NaCl, M sodium citrate) for 45 min at 65 C. Ethidium bromide-stained rrna served as loading control. Western blot analysis. Western blots (Hybond-C Extra membranes; Amersham Biosciences) were incubated for 2 h with a rat polyclonal anti-tbhphbr antibody (described in reference 28) or a mouse monoclonal antitubulin antibody (gift from K. Gull, Oxford University) in 150 mm NaCl, 0.5% (wt/vol) Tween 20, 20 mm Tris-HCl (ph 7.5), and 1% nonfat milk. The secondary antibodies, peroxidase-conjugated monoclonal mouse anti-rat IgG (Serotec) and peroxidase-conjugated sheep anti-mouse IgG (Amersham Biosciences), respectively, were diluted in the same buffer, and the bound antibodies were detected by chemiluminescence (Amersham Biosciences). For anti-tubulin staining, each lane was loaded with cells. For anti-tbhphbr staining, each lane was loaded with protein fractions purified by binding to Hp-Hb resins and equivalent to cells, as described in reference 28. In vitro trypanolysis assays. TLF-1 and TLF-2 were prepared by gel filtration on Superdex 200 (GE Healthcare). Trypanosomes isolated from mice were incubated at densities between and ml 1 in HMI-9 medium (with 10% fetal calf serum [FCS]) at 37 C in a CO 2 -equilibrated incubator. Cell densities were established by triplicate counting using a Neubauer hemocytometer. Electron microscopy studies. Bloodstream trypanosomes were fixed overnight at 4 C in 2.5% glutaraldehyde, 0.1 M cacodylate buffer (ph 7.2), and postfixed in 2% OsO 4 in the same buffer. After serial dehydration in increasing ethanol concentrations, samples were embedded in agar 100 (Agar Scientific Ltd., United Kingdom) and left to polymerize for 3 days at 60 C. Ultrathin sections (50 to 70 nm thick) were collected in Formvar-carbon-coated copper grids by using a Leica EM UC6 ultramicrotome and stained with uranyl acetate and lead citrate. Observations were made on a Tecnai 10 electron microscope (FEI), and images were captured with a MegaView II camera and processed with AnalySIS and Adobe Photoshop software. RESULTS Endocytosis rate of the known ligands of T. brucei. We labeled the known T. brucei-specific ligands with fluorescent

3 1274 VANHOLLEBEKE ET AL. EUKARYOT. CELL Downloaded from FIG. 1. Developmentally regulated ligand accumulation in T. brucei. (A) Ligand accumulation in SL and ST forms. Trypanosomes isolated from mice (SL, 3 dpi; ST, 8 dpi) were pretreated with the lysosomal cathepsin B inhibitor FMK-024 and incubated for 120 min with the indicated fluorescently labeled ligands (for concentrations, see Materials and Methods), before being fixed and observed by epifluorescence (DAPI [DNA counterstain] in blue; ligand in green). (B) Ligand accumulation rate in ST versus SL forms. FMK-024-pretreated cells were incubated for 5, 45, 90, and 135 min with the labeled ligands. At the different time points and in triplicate, mean cell-associated fluorescence was measured by flow cytometry and the average fluorescence accumulation rate was calculated by linear regression. The numbers report the ratios of accumulation rates of ST to those of SL. The arrowheads and arrows point to the lysosomes and the flagellar pockets, respectively. N, nucleus; k, kinetoplast. tags and compared their relative accumulation rates in SL and ST forms in the presence of the lysosomal cathepsin B inhibitor FMK-024 (17) (Fig. 1). Under the experimental conditions used, uptake of the ligands was linear over 150 min and inhibition of lysosomal degradation did not appear to block endocytosis (linear regression R 2 value over 95% for all ligands except for TL, where 80% values were recorded). So, time course assays of ligand accumulation were performed within this time interval (5, 45, 90, and 135 min) and the average accumulation rate was computed from three independent experiments. In accordance with previous reports (20), Tf accumulation occurred in ST forms (Fig. 1). We even observed an increase in lysosomal accumulation rate (2.3-fold) that paralleled a higher accumulation rate of the fluid phase marker dextran (1.7-fold). The magnitude of the increase was comparable to the 2-fold-increased membrane turnover rate measured in those forms (7). However, the accumulation level of the other markers was reduced (Fig. 1). HDL uptake was also slightly lower in ST forms (15% reduction). While Hp-Hb accumulation became barely detectable, 70% and 40% reductions were observed for LDL and TL, respectively. Differential regulation of HDL and LDL uptake. While the HDL uptake rate was only slightly influenced by the differentiation process, LDL particles accumulated in largely reduced amounts ( 70%) (Fig. 1). We performed competition assays with an excess of unlabeled lipoprotein particles to make sure that the relatively constitutive HDL accumulation did not result from fluid phase uptake of those particles. In both SL and ST forms either excess LDL or excess HDL reduced both LDL and HDL accumulation (see Fig. S1 in the supplemental material). Therefore, it is not clear if different receptors are involved, and the differential lipoprotein uptake cannot be explained at this stage. Developmentally regulated TbHpHbR expression. The complete absence of Hp-Hb accumulation in ST forms resulted from drastic downregulation of TbHpHbR. As shown in Fig. 2A and B, the TbHpHbR transcripts and protein were undetectable in extracts from a mixed population of ST and interme- on November 8, 2018 by guest

4 VOL. 9, 2010 ENDOCYTIC REMODELING 1275 Downloaded from FIG. 2. Developmental control of TbHpHbR expression. (A) Northern blot analysis of TbHpHbR transcripts. The rrna serves as loading control. (B) Immunodetection of TbHpHbR in protein samples enriched after Hp-Hb affinity chromatography. Equal loading was assessed by tubulin detection. (C) Monitoring of Hp-Hb uptake. Populations of cells harvested at different points postinfection were mixed and incubated with Hp-A488 Hb for 2 h in the presence of FMK-024. (D) Time course of cellular differentiation occurring in two independent mice, as evaluated by measurement of parasitemia (top panel), morphological criteria (second panel), NAD diaphorase staining (third panel), and Hp-Hb accumulation (bottom panel). (E) Flow cytometry analysis corresponding to one of the two infections reported in panel D. The plots in the first column illustrate the evolution of side scatter (SSC) versus forward scatter (FSC). Those in the second column report Hp-A488 Hb accumulation (FITC) versus FSC, and the histograms of the last column depict the cell frequency distribution (Count) versus Hp-A488 Hb accumulation. Red and green circles surround SL and I or ST cells, respectively. BF, bloodstream form. on November 8, 2018 by guest diate (I) forms. In a population with mixed SL, I, and ST forms, Hp-A488 Hb accumulation was detectable only in parasites exhibiting an SL morphology (Fig. 2C), indicating that TbHpHbR downregulation occurs between SL and I forms. In order to analyze the developmentally regulated Hp-Hb uptake during T. brucei differentiation, cyclophosphamidetreated mice were challenged with a homogeneous population of SL parasites, and cellular differentiation was followed over time in individual mice by using both established developmental markers (cellular morphology and diaphorase activity) and Hp-A488 Hb lysosomal accumulation in the presence of FMK-024 (Fig. 2D and E). Three days postinjection (dpi), a homogeneous population of SL trypanosomes was observed. Morphologically, they were recognized through their elon-

5 1276 VANHOLLEBEKE ET AL. EUKARYOT. CELL FIG. 3. TbHpHbR downregulation does not result from cold shock and does not occur in actively dividing cells. (A) Northern blot analysis of TbHpHbR, PAD1, and EP1 transcripts after trypanosome purification at 37 C or 4 C. The rrna serves as loading control. (B) Flow cytometry analysis of Hp-Hb accumulation versus DNA content (propidium iodine staining) at different time points of the infection as illustrated in Fig. 2. gated cell shape, with a flagellum tightly apposed to the cell body and protruding largely from the cell anterior part. Biochemically, they were negative for diaphorase staining. As reported earlier (28), those parasites accumulated high levels of Hp-Hb (Fig. 2E). As infection proceeded, the proportion of SL parasites gradually decreased in favor of I forms (peak on 6 dpi) that transformed into fully differentiated ST forms by 7 and 8 dpi. Diaphorase staining became detectable in I forms (Fig. 2D). As shown in Fig. 2C to E, I forms were negative for Hp-Hb staining. Accumulation of Hp-Hb allowed the early steps of cellular differentiation to be followed quantitatively by flow cytometry. At 4 dpi a homogeneous population of strongly labeled cells was observed. At 5 dpi (thus, 24 h before the appearance of fully differentiated ST forms), two resolved clusters could be recognized when cell size and fluorescence were probed on the flow cytometer (Fig. 2E). The percentage of unlabeled cells expanded in the following days, to represent almost 100% of the population by 7 dpi. This observation paves the way toward the purification of homogeneous populations of I forms by fluorescence-activated cell sorting (FACS). Indeed, while the side scatter-forward scatter (SSC-FSC) scatter plot allows the detection of the gradual increase in cell complexity and cell size along the differentiation process (the mean SSC and FSC values of SL forms are lower than those of ST forms), it does not allow discrimination between the two populations (Fig. 2E, first column). When the sole Hp-A488 Hb lysosomal accumulation is considered, two overlapping peaks can be observed (Fig. 2E, last column). But when both cell size (FSC channel) and ligand accumulation (FITC channel) are plotted, then long SL forms (red circles) can be safely discriminated from either I or ST forms (green circles) (Fig. 2E, middle column). TbHpHbR downregulation is specific for intermediate and stumpy forms and does not result from cold shock. In order to evaluate if TbHpHbR downregulation was due to cold shock, we measured the levels of TbHPHbR mrna in ST forms entirely isolated at either 37 C or 4 C. As controls, the levels of procyclin EP1 and PAD1 transcripts were also measured, as cold shock is known to trigger expression of procyclins (23) and PAD1 is a marker for cold shock induction of differentiation between SL and ST forms (5). As shown in Fig. 3A, TbHpHbR transcripts were detected only in SL forms, even if these cells were isolated at 4 C. Therefore, cold shock does not seem to be involved in inhibition of transcription of TbHpHbR. TbHpHbR downregulation does not occur in actively dividing cells. We monitored Hp-Hb accumulation along with DNA staining during the cellular differentiation between SL and ST forms. As shown in Fig. 3B, at 7 dpi all cells stained negatively for Hp-Hb and resided in the G 1 /G 0 phase of the cell cycle. Intermediate and stumpy forms are resistant to TLF-1-mediated lysis. TbHpHbR allows heme delivery to bloodstreamform hemoproteins that help the parasite to grow in mice (28). In the absence of this receptor, no cell-associated heme could be detected. In normal human serum (NHS), another ligand for this receptor circulates as part of a potent toxic complex called the trypanosome lytic factor 1 (TLF-1). A distinct form of toxic complex, termed TLF-2, targets the parasites independently of TbHpHbR (28, 30). Therefore, the absence of TbHpHbR in ST forms was expected to confer specific resistance of these parasites to TLF-1-mediated lysis. As shown in

6 VOL. 9, 2010 ENDOCYTIC REMODELING 1277 FIG. 4. Resistance of ST and I forms to TLF-1-mediated lysis. (A) Survival of SL (3 dpi) and ST (8 dpi) forms after 4 h of incubation of 10 6 cells ml 1 with purified TLF-1 or TLF-2. (B) Survival of SL and ST trypanosomes after an overnight incubation of cells ml 1 in log dilutions of NHS (0 is 100% NHS) in the presence or absence of 200 g ml 1 Hp-Hb. (C) Survival of cells from a 5-dpi population composed of SL and I forms incubated for 24 h with the indicated concentrations of NHS. (D) Flow cytometry analysis of three cell populations depicted in panel C: the initial population before treatment (5 dpi), the population obtained after 24 h of in vitro cultivation in the absence of NHS, and the population resulting from a 24-h treatment with 0.1% NHS. The y axis reports the cell frequency distribution, and the x axis reports fluorescence intensity, i.e., Hp-A488 Hb lysosomal accumulation. Fig. 4A, while both SL and ST forms were lysed by TLF-2, ST forms were remarkably unaffected by physiological concentrations of TLF-1. This could further be recognized by incubating the cells in NHS saturated or not with Hp-Hb, which competes with TLF-1 for binding to TbHpHbR. As shown in Fig. 4B, overnight incubations of cells with 0.1% NHS completely killed SL populations while ST forms remained unaffected. The killing of SL forms was largely blocked by saturating Hp-Hb levels, indicating that killing was linked to TLF-1 and not TLF-2. The Hp-Hb concentration-dependent process was analyzed in greater detail (Fig. 4C and D). Trypanosomes isolated from mice at 5 dpi, and thus a mixture of SL ( 60%) and I ( 40%) forms, were incubated overnight with various NHS concentrations. After incubation, the resulting population was analyzed both by morphological criteria (Fig. 4C) and by flow cytometry (Fig. 4D). Over 1 log of dilutions (0.5% to 0.025%), the resulting population was composed exclusively of cells exhibiting morphological characteristics of either I or ST forms. As expected, the resulting population stained negatively for Hp-Hb accumulation, confirming again that Hp-Hb and TLF-1 target the same cells via TbHpHbR. The enrichment in ST forms was close to 1,000-fold as judged by flow cytometry. The possibility that the NHS treatment generated SL forms downregulating TbHpHbR was evaluated through morphological examination of the resulting population by a blinded approach. The occurrence of SL-like forms was below 0.2%. In the absence of NHS treatment, the resulting population contained more than 70% SL forms. Intracellular rearrangements during T. brucei differentiation. During the course of this study, we came to observe that, in reference to the anteroposterior cell axis, the position of the lysosome and the nucleus inverted upon differentiation from SL to ST forms, i.e., the lysosome became located anterior to the nucleus as evidenced by ligand accumulation (Fig. 1 and 5A) or by anti-p67 immunostaining (Fig. 5A). Figure 5A illustrates the subcellular organization of three different cells with decreasing kinetoplast-to-nucleus distances harvested at 6 dpi. When they were assessed systematically, a clear relationship could be drawn between the differentiation status and the respective locations of the nucleus, the kinetoplast, and the lysosome. While homogeneous populations of SL forms (3 dpi) exhibited a single lysosome located halfway between the kinetoplast and the nucleus (Fig. 5B), 6-dpi populations (mostly a mixture of ST and I forms with some SL forms) were characterized by a shorter kinetoplast-to-nucleus distance and frequently at least one lysosome located anterior to the nucleus (Fig. 5A and C). This positioning did not result from FMK-024

7 1278 VANHOLLEBEKE ET AL. EUKARYOT. CELL Downloaded from FIG. 5. Subcellular reorganization during T. brucei differentiation. (A) Tf accumulation (5 g ml 1, 120 min) and p67 (lysosomal marker) immunolocalization in a 6-dpi FMK-024-treated trypanosome population. d(k-n) represents the distance between the kinetoplast and the nucleus, along the virtual curve depicted by the white dotted line. (B) Distribution of the distance between the kinetoplast and the lysosome [d(k-l); y axis] and the distance between the kinetoplast and the nucleus (x axis) of a population of SL cells harvested at 3 dpi. Each cross represents an individual cell. (C) Same as in panel B, with a population harvested at 6 dpi. Each cross represents an individual cell. The colored circles refer to the cells displaying two lysosomes as revealed by anti-p67 staining and Tf accumulation [two d(k-l) values of identical color for a single d(k-n) value]. (D) Phase-contrast imaging of immobilized live ST forms incubated or not with TLF-2. Arrows and arrowheads point to the swelling lysosome (L) and nucleus (N), respectively. treatment, as was confirmed by anti-p67 staining on untreated cells (data not shown). In some developmental forms where the kinetoplast-to-nucleus distance was intermediate (around 2.5 m), two vesicles were decorated with anti-p67 antibodies and accumulated Tf; thus, two apparent lysosomes were observed (Fig. 5A). The possibility that the observed repositioning of the lysosome resulted from fixation artifacts was ruled out through the observation of the swelling lysosome of live but immobilized ST forms treated with TLF-2 (Fig. 5D). Under this experimental setting, the swollen lysosome appeared anterior to the nucleus, in contrast with observations of SL forms (29). The evolution of lysosomal localization was followed over time during mouse infection. As shown in Fig. 6A, cells with anterior lysosome localization clearly accumulated along with the differentiation process, finally representing the majority of the differentiated population. Such cells ( 95%) acquired STspecific PAD1 expression (Fig. 6B). Moreover, upon triggering of cellular differentiation into PFs, cells with anterior lysosome localization exhibited even better expression of EP1 procyclin than did cells with posterior localization (Fig. 6C and D). Altogether, these results indicated that the movement of the lysosome toward the anterior compartment does not result from experimental artifact and is a genuine feature of the cellular differentiation into ST forms. The lysosome reorganization was detailed using transmission electron microscopy analysis (15). On the way from its original location in SL forms to the cell anterior compartment in ST forms, the lysosome traveled all around the nucleus (Fig. 7). Interestingly, during this movement the lysosome membrane seemed to closely associate with that of the expanding mitochondrion (Fig. 7B). This analysis also confirmed the differentiation-linked increase of the lysosome size, which was already noted (3). DISCUSSION In order to remain durably in their mammalian hosts, trypanosomes keep their densities within limits acceptable for on November 8, 2018 by guest

8 VOL. 9, 2010 ENDOCYTIC REMODELING 1279 FIG. 6. Characterization of stumpy cells with anterior lysosomal localization. (A) Evolution of lysosomal localization over time during mouse infection. (B) ST populations isolated at 8 dpi were stained for the differentiation marker PAD1. (C) ST populations isolated at 8 dpi were stained for procyclin (EP1) after 6 h of incubation in SDM-79 at 27 C. (D) Typical staining after treatment as in panel C. host survival. Part of this strategy relies on their ability to respond to a so-far-uncharacterized quorum-sensing component. Upon sensing this signal, the actively dividing parasite halts its cell cycle progression and enters into a differentiation process that contributes to both stopping host colonization and preparing the cells for the next step of the developmental cycle, which takes place in their insect vector, the tsetse fly. Those differentiated ST forms will either be taken up by the fly and contribute to the completion of the life cycle or survive for some days in the bloodstream of the mammal before being destroyed by the developing immune reaction. ST forms share with SL forms the need to evade immediate destruction by the immune system. The outstanding endocytic capacity of this parasite is part of this strategy. It has indeed been shown previously that high endomembrane turnover rates characterize both SL and ST forms (7). Somewhat more surprisingly, Tf uptake was shown to follow the same developmental pattern despite the supposed lack of requirement for growth factors in the quiescent ST forms (20). The capacity of T. brucei to build up and accommodate large intracellular iron stores could alleviate the need for tight control of Tf uptake (27). Nevertheless, it might be anticipated that the nutritional requirements (iron, sterol, and heme) of ST forms differ greatly from those of SL forms. While the latter need not only to satisfy homeostasis but also to compensate for the rapid dilution resulting from cell division, ST forms are locked in the G 1 /G 0 phase and hence do not require accumulation of nutrients for offspring cells. In agreement with this idea, we report that, except for Tf and TL, other ligands for receptor-mediated endocytosis accumulated in reduced amounts in ST forms compared with SL forms. The lipoprotein uptake pattern was found to differ between SL and ST forms. While LDL uptake appeared to be strongly reduced in ST forms, HDL uptake was almost identical in the two forms. Either the receptors expressed at the ST and SL surface are distinct, or differential modifications impact the ligand specificity of a single scavenger receptor (10). ST-specific inhibition of ligand uptake was particularly striking for the Hp-Hb complex. No accumulation in excess of that entering by bulk phase could be evidenced in ST forms. This was linked to the apparent complete absence of the Hp-Hb receptor TbHpHbR, in keeping with previous observations resulting from global gene expression profiling (13). Given the toxic and oxidative properties of the associated heme moiety, it is possible that this downregulation contributes to protect the parasite against potential cytotoxicity. Either ST forms do not require hemoproteins any longer or the half-life of those proteins is long enough to sustain heme-based electron transfer reactions for the time during which they circulate. Heme uptake has been shown to contribute to both growth rate and resistance to oxidative burst of the host (28). It is obvious that the growth-promoting contribution of heme becomes dispensable in those quiescent forms. The possibility that ST forms are more sensitive to host oxidative stress will be difficult to evaluate, as this contribution of heme-based metabolism has been observed only in vivo as a reduction of growth rate, an experimental setup inapplicable to nondividing ST forms. However, as ST forms strikingly differ from SL forms in their general increased resistance to various environmental or experimental stresses, including acidic ph, proteases, or complement-mediated killing (18, 22), it is possible that these parasites use hemoprotein-independent processes to resist environmental aggression.

9 1280 VANHOLLEBEKE ET AL. EUKARYOT. CELL Downloaded from FIG. 7. Transmission electron microscope analysis of SL-to-ST differentiation. Trypanosomes at 8 dpi were processed for transmission electron microscopy analysis. The posterior end of the cell is to the right of the pictures. (A) Representative micrographs of cells with an anterior lysosome, the dominant cellular organization (80.6% of the population, n 31). (B) Less frequent cellular organizations. (Left) Posterior localization (probably relapsing SL cell; 3.2% of the population). (Center) Nuclear proximal posterior lysosome (3.2% of the population). (Right) Cell harboring both an anterior and a posterior lysosome (12.9% of the population). Arrows point to the mitochondrion, closely associated with the lysosome. L, lysosome; N, nucleus; FP, flagellar pocket. Scale bar, 1 m. on November 8, 2018 by guest The absence of TbHpHbR explains the observed resistance of early and late ST forms to the action of trypanolytic HDLs from NHS, since it is known that these particles enter the parasite through TbHpHbR (28). The documented sensitivity of ST forms to NHS appears to be entirely due to the TLF-2 component, which enters trypanosomes independently from TbHpHbR. Previous studies have reported on the increased resistance of ST forms to TLF-1-mediated lysis (19). In that report, while in some assays ST forms were resistant to TLF- 1-mediated lysis at 37 C as reported here, in others, ST forms were found to remain sensitive, unless forced to differentiate further by a mild cold shock treatment. In our analysis we did not observe this instability of the TLF-1 phenotype, as in all experiments ST forms were found to be fully resistant even when the whole experiment from cell isolation from mice to cell fixation was entirely performed at 37 C, with preheated solutions and in thermostatic rooms (data not shown). We speculate that for some strains like TREU 667, used in reference 19, a mild cold shock might sometimes be required to achieve full downregulation of TbHpHbR, while in other strains like AnTat 1.1, used in this study, the downregulation could be more robust. The downregulation is likely to occur in vivo, as TLF-1 injection in mice killed SL but not ST forms (data not shown).

10 VOL. 9, 2010 ENDOCYTIC REMODELING 1281 The downregulation of TbHpHbR was shown to result from decreased mrna levels. This paralleled the transition between dividing SL forms and the initiation of differentiation into ST forms. In particular, I forms did not seem to express the receptor any longer. This dramatic downregulation might be linked to the location of the TbHpHbR gene at the very end of a transcriptional unit that immediately precedes a transcription unit for the procyclic form-specific procyclin genes (28). Indeed, strong transcriptional changes are linked to SL-to-ST differentiation, including partial activation of procyclin loci (1, 23). Therefore, it is possible that enhanced recruitment of RNA polymerase at the procyclin promoter could hinder proper transcription elongation in the TbHpHbR gene located in the close upstream environment. However, although in ST cold shock triggered the synthesis of procyclin mrnas, the downregulation of TbHpHbR did not require cold shock. As this downregulation was not observed in dividing cells, it could be linked to the cell cycle blockade typical of early SL-to-ST differentiation. The absence of TbHpHbR in early ST forms offers new avenues to study the events triggering T. brucei differentiation in the bloodstream. So far, this is poorly characterized due to the impossibility of isolating and profiling homogeneous populations of cells at the initiation of this process. Currently available techniques require plating SL forms at high densities in vitro, a situation that might not fully reproduce the endogenous conditions. In vivo, the differentiation is not fully synchronous, as can be appreciated from Fig. 2D (except for 3 dpi, the population is always composed of at least two different forms). This leaves the investigator with two types of relatively homogeneous differentiated populations: either a population rich in I forms but containing both SL and ST forms (6 dpi) or a population of ST forms (8 dpi) sometimes contaminated with relapsing SL forms. In both cases, cell analysis at the onset of the differentiation process is intractable for a population. mrna profiling of homogeneous trypanosome populations between ST and procyclic forms has allowed major advances in our understanding of the differentiation between bloodstream and insect-specific forms (5, 14). As we have shown that a homogeneous population of differentiated forms (most interestingly, early I forms) can be obtained either by exposure to carefully titrated amounts of NHS or through FACS, the tools developed in this work should allow a similar approach to be undertaken for the SL-to-ST differentiation. Two options are available, either the FACS of the cells after Hp-A488 Hb uptake or the counterselection of SL forms with TLF-1. Although ST and I forms seemed unaffected by the overnight treatment with TLF-1, we favor the first option because it is not known whether mild TLF-1 treatment or overnight culture might affect cell development, and conversely the Hp-Hb accumulation can be completed in less than 2 h. More generally, Hp-Hb uptake represents a new and easy developmental marker, which together with diaphorase activity, mitochondrial elaboration, and morphological modifications could be used to assess early steps of the differentiation process. Finally, we observed a so-far-undisclosed feature of the SLto-ST differentiation process. Reciprocal translocation of the lysosome and the nucleus coincided with the appearance of typical features of ST forms, such as characteristic flagellum length, PAD1 immunoreactivity, and diaphorase staining intensity. Therefore, we propose that this inversion is a hallmark of the terminally differentiated ST forms. Curiously, the lysosome movement seemed to be in concert with that of the expanding mitochondrion, as the two organelles appeared to be closely associated. Presumably concerted, rather than independent, segregation is more efficient and energetically more favorable. This coupled translocation might result from the increased size of the two organelles, which might not fit any longer the narrowing space between the kinetoplast and the nucleus. The significance of this cytological reorganization is not obvious, but one possibility is that this positioning could increase the rate of intracellular organelle autophagy upon initiation of the differentiation into procyclic forms (11). ACKNOWLEDGMENTS This work was supported by the Belgian Fonds National de la Recherche Scientifique (FNRS) and the Interuniversity Attraction Poles Programme-Belgian Science Policy. During the course of this work B.V. was a research fellow of the FNRS. We thank K. Matthews (University of Edinburgh) for anti-pad1 antibodies and the Brachet Fund for help. REFERENCES 1. Amiguet-Vercher, A., D. Pérez-Morga, A. Pays, P. Poelvoorde, H. V. Xong, P. Tebabi, L. Vanhamme, and E. Pays Loss of the mono-allelic control of the VSG expression sites during the development of Trypanosoma brucei in the bloodstream. Mol. Microbiol. 51: Bastin, P., A. Stephan, J. Raper, J. M. Saint-Remy, F. R. Opperdoes, and P. J. Courtoy An M(r) 145,000 low-density lipoprotein (LDL)-binding protein is conserved throughout the Kinetoplastida order. Mol. Biochem. Parasitol. 76: Brown, R. C., D. A. Evans, and K. Vickerman Developmental changes in ultrastructure and physiology of Trypanosoma brucei. Trans. R. Soc. Trop. Med. Hyg. 66: Coppens, I., P. Baudhuin, F. R. Opperdoes, and P. J. Courtoy Receptors for the host low density lipoproteins on the hemoflagellate Trypanosoma brucei: purification and involvement in the growth of the parasite. Proc. Natl. Acad. Sci. U. S. A. 85: Dean, S., R. Marchetti, K. Kirk, and K. R. Matthews A surface transporter family conveys the trypanosome differentiation signal. Nature 459: Domenicali Pfister, D., G. Burkard, S. Morand, C. K. Renggli, I. Roditi, and E. Vassella A mitogen-activated protein kinase controls differentiation of bloodstream forms of Trypanosoma brucei. Eukaryot. Cell 5: Engstler, M., T. Pfohl, S. Herminghaus, M. Boshart, G. Wiegertjes, N. Heddergott, and P. Overath Hydrodynamic flow-mediated protein sorting on the cell surface of trypanosomes. Cell 131: Fenn, K., and K. R. Matthews The cell biology of Trypanosoma brucei differentiation. Curr. Opin. Microbiol. 10: Field, M. C., J. H. Lumb, V. O. Adung a, N. G. Jones, and M. Engstler Macromolecular trafficking and immune evasion in African trypanosomes. Int. Rev. Cell Mol. Biol. 278: Green, H. P., M. Del Pilar Molina Portela, E. N. St. Jean, E. B. Lugli, and J. Raper Evidence for a Trypanosoma brucei lipoprotein scavenger receptor. J. Biol. Chem. 278: Herman, M., D. Pérez-Morga, N. Schtickzelle, and P. A. Michels Turnover of glycosomes during life-cycle differentiation of Trypanosoma brucei. Autophagy 4: Hirumi, H., and K. Hirumi Continuous cultivation of Trypanosoma brucei blood stream forms in a medium containing a low concentration of serum protein without feeder cell layers. J. Parasitol. 75: Jensen, B. C., D. Sivam, C. T. Kifer, P. J. Myler, and M. Parsons Widespread variation in transcript abundance within and across developmental stages of Trypanosoma brucei. BMC Genomics 10: Kabani, S., K. Fenn, A. Ross, A. Ivens, T. K. Smith, P. Ghazal, and K. Matthews Genome-wide expression profiling of in vivo-derived bloodstream parasite stages and dynamic analysis of mrna alterations during synchronous differentiation in Trypanosoma brucei. BMC Genomics 10: Langreth, S. G., and A. E. Balber Protein uptake and digestion in bloodstream and culture forms of Trypanosoma brucei. J. Protozool. 22: Laxman, S., A. Riechers, M. Sadilek, F. Schwede, and J. A. Beavo Hydrolysis products of camp analogs cause transformation of Trypanosoma brucei from slender to stumpy-like forms. Proc. Natl. Acad. Sci. U. S. A. 103:

11 1282 VANHOLLEBEKE ET AL. EUKARYOT. CELL 17. Lonsdale-Eccles, J. D., and D. J. Grab Lysosomal and non-lysosomal peptidyl hydrolases of the bloodstream forms of Trypanosoma brucei brucei. Eur. J. Biochem. 169: McLintock, L. M., C. M. Turner, and K. Vickerman Comparison of the effects of immune killing mechanisms on Trypanosoma brucei parasites of slender and stumpy morphology. Parasite Immunol. 15: Moore, D. R., A. Smith, K. M. Hager, R. Waldon, J. D. Esko, and S. L. Hajduk Developmentally regulated sensitivity of Trypanosoma brucei to the cytotoxic effects of human high-density lipoprotein. Exp. Parasitol. 81: Natesan, S. K., L. Peacock, K. Matthews, W. Gibson, and M. C. Field Activation of endocytosis as an adaptation to the mammalian host by trypanosomes. Eukaryot. Cell 6: Nolan, D. P., M. Geuskens, and E. Pays N-linked glycans containing linear poly-n-acetyllactosamine as sorting signals in endocytosis in Trypanosoma brucei. Curr. Biol. 9: Nolan, D. P., S. Rolin, J. Rueda Rodriguez, J. Van Den Abbeele, and E. Pays Slender and stumpy bloodstream forms of Trypanosoma brucei display a differential response to extracellular acidic and proteolytic stress. Eur. J. Biochem. 267: Pays, E Regulation of antigen gene expression in Trypanosoma brucei. Trends Parasitol. 21: Pays, E., L. Vanhamme, and D. Pérez-Morga Antigenic variation in Trypanosoma brucei. Facts, challenges and mysteries. Curr. Opin. Microbiol. 7: Salmon, D., M. Geuskens, F. Hanocq, J. Hanocq-Quertier, D. Nolan, L. Ruben, and E. Pays A novel heterodimeric transferrin receptor encoded by a pair of VSG expression site-associated genes in Trypanosoma brucei. Cell 78: Savill, N. J., and J. R. Seed Mathematical and statistical analysis of the Trypanosoma brucei slender to stumpy transition. Parasitology 128: Steverding, D Bloodstream forms of Trypanosoma brucei require only small amounts of iron for growth. Parasitol. Res. 84: Vanhollebeke, B., G. De Muylder, M. J. Nielsen, A. Pays, P. Tebabi, M. Dieu, M. Raes, S. K. Moestrup, and E. Pays A haptoglobin-hemoglobin receptor conveys innate immunity to Trypanosoma brucei in humans. Science 320: Vanhollebeke, B., L. Lecordier, D. Pérez-Morga, A. Amiguet-Vercher, and E. Pays Human serum lyses Trypanosoma brucei by triggering uncontrolled swelling of the parasite lysosome. J. Eukaryot. Microbiol. 54: Vanhollebeke, B., and E. Pays The trypanolytic factor of human serum: many ways to enter the parasite, a single way to kill. Mol. Microbiol. 76: Vassella, E., R. Krämer, C. M. Turner, M. Wankell, C. Modes, M. van den Bogaard, and M. Boshart Deletion of a novel protein kinase with PX and FYVE-related domains increases the rate of differentiation of Trypanosoma brucei. Mol. Microbiol. 41: Zhou, W., G. A. Cross, and W. D. Nes Cholesterol import fails to prevent catalyst-based inhibition of ergosterol synthesis and cell proliferation of Trypanosoma brucei. J. Lipid Res. 48: Downloaded from on November 8, 2018 by guest