PTEN differentially regulates endocytosis, migration, and proliferation in the enteric protozoan parasite Entamoeba histolytica

PTEN is a lipid phosphatase that is highly conserved and involved in a broad range of biological processes including cytoskeletal reorganization, endocytosis, signal transduction, and cell migration in all eukaryotes. Although regulation of phosphatidylinositol (3,4,5)-trisphosphate [PtdIns(3,4,5)P3] signaling via PTEN has been well established in model organisms and mammals, it remains elusive in the parasitic protist E. histolytica, which heavily relies on PtdIns phosphate(s)-dependent membrane traffic, migration, and phago- and trogocytosis for its pathogenesis. In this study, we characterized the major PTEN from E. histolytica, EhPTEN1, which shows the highest expression at the transcript level in the trophozoite stage among 6 possible PTENs, to understand the significance of PtdIns(3,4,5)P3 signaling in this parasite. Live imaging of GFP-EhPTEN1 expressing amebic trophozoites showed localization mainly in the cytosol with a higher concentration at pseudopods and the extending edge of the phago- and trogocytic cups. Furthermore, quantitative analysis of phago- and trogocytosis using a confocal image cytometer showed that overexpression of EhPTEN1 caused reduction in trogo- and phagocytosis while transcriptional gene silencing of EhPTEN1 gene caused opposite phenotypes. These data suggest that EhPTEN1 has an inhibitory role in these biological processes. Conversely, EhPTEN1 acts as a positive regulator for fluid-phase and receptor-mediated endocytosis in E. histolytica trophozoites. Moreover, we showed that EhPTEN1 was required for optimal growth and migration of this parasite. Finally, the phosphatase activity of EhPTEN1 towards PtdIns(3,4,5)P3 was demonstrated, suggesting that the biological roles of EhPTEN1 are likely linked to its catalytic function. Taken together, these results indicate that EhPTEN1 differentially regulates multiple cellular activities essential for proliferation and pathogenesis of the organism, via PtdIns(3,4,5)P3 signaling. Elucidation of biological roles of PTEN and PtdIns(3,4,5)P3 signaling at the molecular levels promotes our understanding of the pathogenesis of this parasite.

Human amebiasis is caused by the infection of the enteric protozoan parasite Entamoeba histolytica. World Health Organization estimates 50 million people throughout the world suffers from amebic infections, resulting in around 100,000 deaths annually [13]. Infection by E. histolytica usually occurs via ingestion of fecally contaminated food or water with the infective cyst of this parasite [14]. Destruction of intestinal epithelial tissue by amoebic trophozoites causes colitis and amoebic dysentery while in some patients trophozoites can infect extraintestinal organs where they form abscesses [15]. It is known that the virulence mechanisms of E. histolytica are sustained by actin-associated processes such as migration, adhesion, and trogo-/ phagocytosis as well as vesicular traffic involved in the secretion of proteases [16][17][18][19]. A sufficient set of PI-kinases and phosphatases to generate 7 species of phosphoinositides appear to be conserved in E. histolytica [20]. AGC kinases have recently been identified as PtdIns(3,4,5) P 3 -binding proteins and shown to be involved in trogocytosis and phagocytosis in E. histolytica [21]. Physiological significance of PtdIns(3)P-and PtdIns(4)P-binding proteins including FYVE domain-containing proteins was shown [22]. In addition, the distinct roles of PtdIns(3) P-binding sorting nexins (SNXs) in trogocytosis have been demonstrated in E. histolytica [23]. Among them, PtdIns(3,4,5)P 3 -mediated signaling is assumed to have a pivotal role in E. histolytica virulence. Although physiological roles of PTEN have been well established in higher eukaryotes, the role of PTEN in E. histolytica in pathogenesis remains elusive.
In the present study, we characterized the biological roles of EhPTEN1, which shows the highest expression at the transcript level in the trophozoite stage among six putative PTENs encoded by the genome. We have shown that EhPTEN1 is enzymatically active against PtdIns (3,4,5)P 3 and is required for optimal growth of E. histolytica cells. We have also found that EhPTEN1 is involved in the regulation of different modes of endocytosis, namely fluid-phase endocytosis, receptor-mediated endocytosis, phago-, trogocytosis, and cell migration.

Identification and features of PTEN genes in E. histolytica
A genome-wide survey of PTEN in the genome of E. histolytica HM-1:IMSS reference strain (AmoebaDB, http://amoebadb.org) by BLASTP analysis using human PTEN (P60484) as a query, revealed that E. histolytica possesses 6 possible PTEN or PTEN-like proteins that contain PTEN phosphatase domain, and show different domain configurations (Fig 1A) [20]. We tentatively designated them firstly in an ascending order of the number of recognizable domains and secondly in a descending order of the overall length (EhPTEN1, EHI_197010; EhPTEN2, EHI_098450; EhPTEN3, EHI_131070; EhPTEN4, EHI_054460; EhPTEN5, EHI_041900; EhPTEN6, EHI_010360). Our previous transcriptome data [24][25][26] verified that one protein (EhPTEN1, EHI_197010) is highly expressed in the trophozoite stage in both E. histolytica HM-1: IMSS cl6 and G3 strains, while the 5 other PTENs are expressed at relatively low levels ( Fig 1B). RNA seq data from our laboratory [27] indicate the abundance (FPKM) of six EhPTEN is as follow: EhPTEN1-6, 173. 7, 25.8, 9.7, 6.7, 17.8, and 7.0 in HM-1:IMSS cl-6 transfected with HA control vector; 62.0, 11.6, 5.9, 2.0, 9.6, and 7.6 in G3 strain transfected with psAP2-Gunma mock vector. EhPTEN1 shows 39% mutual identity to human PTEN at the amino acid level (S1 Table). Multiple sequence alignment by ClustalW program (http:// clustalw.ddbj.nig.ac.jp) shows that the key catalytic residues in the phosphatase domain (H-C-K/R-A-G-K-G-R) needed for lipid and protein phosphatase activity [4,28] are well conserved in EhPTEN1 (Fig 1C). In addition, the PtdIns(4,5)P 2 -binding motif (K/R-x4-K/R-x-K/ R-K/R-R, PDM domain), which is predicted to regulate the recruitment of protein to the plasma membrane, located at the amino terminus, is also conserved in EhPTEN1 [6,29]. The cytosolic localization signal (D-G-F-x-L-D-L, CLS), where mutation of phenylalanine was shown to induce nuclear localization [30], as well as threonine and isoleucine responsible for TI loop formation in an extension of the active site pocket are also conserved ( Fig 1C) [28]. EhPTEN1 also possesses C2 domain which has an affinity for membrane phospholipids and helps PTEN to be recruited to the cell membrane [28]. The homology of the first 190 amino acids in the amino terminus of PTEN to the actin binding protein tensin may suggest the involvement of PTEN in the regulation of actin dynamics. InterPro domain search annotates the region of a.a. 550-680 of EhPTEN1 as a domain of unknown function (DUF457). The two PEST sequences, rich in proline, glutamic acid, serine, and threonine, at the C terminus of human PTEN, are not conserved in EhPTEN1. Although PEST sequences are known to enhance proteolytic sensitivity, the regulation of EhPTEN1 functions may differ from the human PTEN [31]. EhPTEN1 also lacks PDZ-binding motif (T/S-x-V) located at the C terminus in human PTEN and facilitates the protein-protein interactions [6,32]. The human PTEN was shown to interact with several PDZ domain-containing proteins, such as membrane-associated guanylate kinase inverted (MAGI) and Na + /H + exchanger regulatory factor (NHERF) in a PDZ-dependent manner [8,9]. Interaction of MAGI to the PTEN C-terminal PDZ domain facilitates targeting of PTEN to the plasma membrane, thereby enhancing its activity [9]. While NHERF regulates the activation of the PI3K-AKT pathway by binding and recruitment of PTEN to platelet-derived growth factor receptor (PDGFR) [8]. However, the E. histolytica genome apparently does not encode proteins with similarity to NHERF or MAGI, as indicated by BLAST search. Therefore, these PDZ domain binding proteins may be unique to mammals, which represents a major difference between mammalian PTEN and EhPTEN1. The~400 a.a. carboxyl-terminal extension which is absent in human ortholog and rich in charged amino acids could be involved in regulating its protein-protein interactions.

Cellular localization and dynamics of EhPTEN1 in the motile E. histolytica trophozoite
To examine the cellular localization of EhPTEN1 in trophozoites, we established a transformant line expressing EhPTEN1 with the GFP-tag at the amino terminus (GFP-EhPTEN1). The expression of GFP-EhPTEN1 or GFP (control) in transformant trophozoites was verified by immunoblot analysis using anti-GFP antibody. A single band corresponding to non-truncated GFP fusion protein with an expected molecular mass of GFP-EhPTEN1 (90 kDa plus 26 kDa for the GFP tag) was observed in the GFP-EhPTEN1-expressing transformant (Fig 2A). Live imaging analysis revealed that GFP-EhPTEN1 was localized throughout the cytosol ( Fig  2B and S1 Movie). The line intensity plots across the GFP-EhPTEN1 overexpressing trophozoites further demonstrated the enrichment of GFP-EhPTEN1 along the extended pseudopod of the motile trophozoite ( Fig 2B) (Two additional images are also show in S1A and S1B Fig). The normalized average fluorescence intensities at the leading regions of pseudopods were nearly 1.5-fold higher in GFP-EhPTEN1 overexpressing trophozoites compared to GFPexpressing mock transformants ( Fig 2C). Furthermore, the intensity line plot of GFP-expressing control strain showed no accumulation of GFP signal on pseudopods ( S2 Fig and S2 Movie). These findings confirm the enrichment of GFP-EhPTEN1 in the pseudopod-like protrusive structures. Although the levels of expression of GFP-EhPTEN1 and GFP varied in the respective population, the accumulation of GFP-EhPTEN1 in the pseudopod like structure at the initial stage of trogocytosis and phagocytosis (S4 Movie) was observed even in the cell that expressed GFP-EhPTEN1 at low levels (showing only dim fluorescence). Similarly, immunofluorescence imaging of HA-EhPTEN1 overexpressing trophozoites using anti-HA antibody revealed that HA-EhPTEN1 was localized in the cytoplasm in steady-state, and enriched in using the data from our previous work [24][25][26] where transcriptome of trophozoites of E. histolytica HM-1:IMSS cl6 reference strain was analyzed using DNA microarrays technology [25]. The levels of steady-state mRNA of PTEN isotypes are shown after normalization against that of RNA polymerase II. (C) Multiple amino acid sequence alignment of human PTEN (P60484), Xenopus laevis (AAD46165.1), Drosophila melanogaster (Q9Y0B6), Dictyostelium discoideum PTEN (Q8T9S), and EhPTEN1 (XP_653141.2) was constructed by using clustalw algorithm (http://clustalw.ddbj.nig.ac.jp). PTEN phosphatase domain and C2 domain are shown with blue and yellow backgrounds, respectively. The green rectangle corresponds to the PtdIns(4,5)P 2 -binding motif (PDM domain). Amino acid residues implicated for PtdIns(3,4,5)P 3 catalysis are marked with a red rectangle. Cytosolic localization signal and restudies important for TI loop formation are indicated in black and blue lines, respectively. Note that for human PTEN, only the amino terminal part is shown. https://doi.org/10.1371/journal.ppat.1010147.g001

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica pseudopods when the fluorescence signal across the line traversing the cell was measured (S3C Fig). HA-EhPTEN1 localization was also evaluated by measuring the fluorescence signal in the different portion of the cell, confirming that HA-EhPTEN1 is highly concentrated in the pseudopods, compared to other parts of the cell (S3D Fig). The migration (motility) of the GFP-EhPTEN1 overexpressing trophozoites using the montage of time-lapse imaging was 0.54±0.09 μm/sec (mean±S.D.), which was significantly greater than that of control GFP

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica expressing transformants (0.27±0.08 μm/sec) (Fig 2D). We also investigated the effect of repression of EhPTEN1 gene expression and found that EhPTEN1 gene silencing reduced migration (see below).

Localization of EhPTEN1 during trogocytosis and phagocytosis
The fact that EhPTEN1 was previously identified as a PtdIns(3)P-binding effector and suggested to be involved in the phagosome biogenesis [23], prompted further characterization of the role of EhPTEN1 in host cell internalization. To examine the role of EhPTEN1 in ingestion of mammalian cells, we first examined trogocytosis (i.e., nibbling or chewing of a part of a live cell) of Chinese hamster ovary (CHO) cells by GFP-EhPTEN1 and GFP expressing transformant lines. We co-cultured trophozoites of the two transformant lines with live CHO cells that had been stained with CellTracker Orange. Time-lapse imaging of trogocytosis of CHO cells by the amoebae (Fig 3A and S3 Movie) revealed that GFP-EhPTEN1 was accumulated in the region that covers, but not always in close proximity to, the tunnel-like structure, which is the extended neck (or tube)-like structure connecting the enclosed (or being enclosed) trogosome and the remaining portion of the target cell that is partially ingested (Fig 3A-3D). Upon completion of closure of the trogosome, GFP-EhPTEN1 appeared to be dissociated from the region around the trogosome and the tunnel-like structure (Fig 3A-3E). The quantification of the fluorescence intensity in a cross section of the cell confirmed the dynamism of GFP-EhPETN1 during trogocytosis (Fig 3B-3E). In contrast, at the very early phase of trogocytosis, GFP-EhP-TEN1 was not concentrated on the newly formed trogocytic cup.
The dynamics of GFP-EhPTEN1 in a course of phagocytosis (i.e., internalization with a single bite, not multiple bites) of dead CHO cells was also examined. The live imaging of GFP-EhPTEN1 expressing trophozoites co-cultured with pre-killed CHO cells (Fig 4 and  S4 Movie) showed an enrichment of GFP-EhPTEN1 at the tip of the leading edge of the phagocytic cup during the internalization of dead host cells until phagosome closure ( Fig  4A and 4B). Soon after closure of the phagosome, GFP-EhPTEN1 was concentrated on the closing side of the phagosome (Fig 4A-4C), and rapidly disappeared soon after ( Fig 4D). The fluorescence intensity line plot of a cross section (as indicated by arrows) of the cell also reinforced the observation (Fig 4A-4D). As control, GFP-expressing mock strain showed no observable concentration of GFP signal in a course of CHO ingestion (S4 Fig  and S5 Movie).

Effect of overexpression of EhPTEN1 on trogocytosis and phagocytosis
The dynamism of GFP-EhPTEN1, as revealed by live imaging, suggests that EhPTEN1 plays a role in the early to middle stages of trogo-and phagocytosis. We examined the effect of GFP-EhPTEN1 overexpression on the efficiency (i.e., speed and volume of internalization of prey) of trogocytosis and phagocytosis. GFP-EhPTEN1 expressing and GFP-expressing mock transformant strains were incubated with either live or pre-killed CHO cells that had been stained with CellTracker Orange to allow trogocytosis or phagocytosis, respectively. Internalization of CHO cells by the amoebae was measured by CQ1 confocal quantitative image cytometer (Figs 5 and 6). Three parameters were measured and compared between GFP-EhPTEN1 expressing and GFP-expressing mock transformant strains: the number of CHO-containing trogosomes or phagosomes per ameba (Figs 5A and 6A), the volume of all CHO-containing trogosomes or phagosomes per ameba (Figs 5B and 6B), and the percentage of the amebae that ingested CHOs (Figs 5C and 6C). GFP-EhPTEN1 overexpression caused statistically significant reduction in all three parameters in both trogocytosis and phagocytosis.

Gene silencing of EhPTEN1 enhances trogocytosis and phagocytosis in E. histolytica
Conversely, we attempted to verify if repression of EhPTEN1 gene expression by antisense small RNA-mediated transcriptional gene silencing [33] causes reverse phenotypes: enhancement of trogocytosis and phagocytosis. Small antisense RNA-based transcriptional gene silencing, used in this study, was robustly usable only in G3, but not in its parental strain HM-1:

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica IMSS [33]. The silencing of the EhPTEN1 gene expression was confirmed by RT-PCR and the level of silencing was estimated to be approximately 77.0±9.2% compared to the mock control (G3 transfected with the empty psAP2-Gunma vector) (Fig 7A and 7B). Non-specific off-target gene silencing of other EhPTEN genes (EhPTEN2-6) was ruled out, except for EhPTEN2, which showed a slight reduction in EhPTN1 gene silenced strain, validating gene-specific silencing (S5 Fig). The RNA pol II transcript level was also unaffected. EhPTEN1 gene silenced and psAP2 mock transformants were cultivated with live or dead CHO cells and images were captured every 10 min for 1 hr. As expected, EhPTEN1 gene silenced strain showed an enhancement of trogocytosis and phagocytosis (Fig 8). All three parameters to evaluate trogocytosis and phagocytosis, as above, i.e., the number of CHO-containing trogosomes or phagosomes per ameba (Fig 8A and 8B), the volume of all CHO-containing trogosomes or phagosomes per ameba (Fig 8C and 8D), and the percentage of the amebae that ingested CHOs (Fig 8E and 8F) was significantly increased in EhPTEN1gene silenced strain compared to the psAP2 mock control stain. For instance, the volume of trogosomes and phagosomes increased by around 1.4-fold for trogocytosis and 2-fold for phagocytosis, respectively, in EhP-TEN1 gene silenced strain at later time points of coincubation (at 40-60 mins). Together with

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica the results of EhPTEN1 overexpression, shown above, these data indicate that EhPTEN1 serves as a negative regulator of trogocytosis and phagocytosis.

EhPTEN1 is a positive regulator for the fluid-phase and receptor-mediated endocytosis in E. histolytica
To investigate the role of EhPTEN1 in pinocytosis of the fluid-phase marker and receptormediated endocytosis, we examined the internalization of RITC dextran and transferrin. Pinocytosis was analyzed by measuring, on a fluorometer, the fluorescence intensity of fluid-phase marker, RITC dextran, which was internalized after incubation of amoebic transformants with RITC dextran at 35˚C for up to 1 hr. Overexpression of GFP-EhPTEN1 cause approximately 30% increase in pinocytosis in comparison to GFP mock control (53±5.3 or 28±13% at time 30
We next examined internalization of transferrin conjugated with AlexaFluor 568 by CQ1 image cytometer. Transferrin is presumed to be internalized via receptor-mediated endocytosis. The volume of endosomes that contained transferin-AlexaFluor 568 increased by 30-50% in GFP-EhPTEN1 overexpressing trophozoites compared to GFP mock control at all time points up to 1 hr (p<0.05, Fig 10A). Conversely, transferrin endocytosis decreased by 30-40% in EhPTEN1 gene silenced strain compared to the psAP2 mock strain at 40-60 mins (p<0.05, Fig 10B). These data indicate that EhPTEN1 positively regulates pinocytosis of the fluid-phase maker and receptor-mediated endocytosis in E. histolytica.

Fig 6. Effect of GFP-EhPTEN1 expression on phagocytosis. (A)
Trophozoites of GFP mock transfected and GFP-EhPTEN1 expressing strains were incubated with heat killed CHO cells that have been stained with CellTracker Orange to evaluate phagocytosis. The images were taken on CQ1 as described in Materials and methods and analyzed to calculate the average numbers of CHO cell-containing phagosomes per amoeba. (B) The volume of the ingested CHO cells was calculated using three-dimensionally reconstituted data. (C) The percentage of amoeba trophozoites that ingested pre-killed CHO cells. Experiments were conducted three times independently in triplicates and a representative data set is shown. Statistical significance was examined with t-test ( � P< 0.05). Error bars indicate standard deviations of two biological replicates. https://doi.org/10.1371/journal.ppat.1010147.g006

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica

EhPTEN1 is essential for optimum growth and migration of E. histolytica
The biological role of EhPTEN1 in trogo-, phagocytosis, and endocytosis was clearly demonstrated as above. To investigate other physiological roles of EhPTEN1 in E. histolytica, the growth kinetic was monitored in EhPTEN1 gene silenced and control strains. EhPTEN1 gene silencing caused significant growth defect: the population doubling time of EhPTEN1 gene silenced and control strains was 28.1±0.41 and 19.1±0.52 hr, respectively (P<0.05; Fig 7C). We next examined the migration of the trophozoites of EhPTEN1 gene silenced and control strains

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica using time lapse imaging by CQ1. The velocity of motility was >60% reduced in EhPTEN1 gene silenced strain (0.16±0.07 μm /min) compared to the psAP2 mock control (0.44±0.08 μm /min) (Fig 7D).  The effect of GFP-EhPTEN1 expression on endocytosis. Trophozoites of GFP mock transfected and GFP-EhPTEN1 expressing strains were incubated in BI-S-33 medium containing transferrin and images were taken every 10 min for 1 hr by CQ1 as described in Materials and methods. The volume of endosomes was calculated using three-dimensionally reconstituted data. (B) The effect of EhPTEN1 gene silencing on endocytosis. Images of psAP2 mock and EhPTEN1gs transformant trophozoites that have been co-cultivated with transferrin were taken every 10 min for 1 hr by CQ1 as described in Materials and methods. The volume of endosomes was calculated using threedimensionally reconstituted data. All experiments were conducted three times independently and statistical significance was examined with t-test ( � P<0.05). Error bars indicate standard deviations of three biological replicates. https://doi.org/10.1371/journal.ppat.1010147.g010

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica

Demonstration of phosphatase activity and substrate specificity of EhPTEN1
To see if EhPTEN1 possesses lipid phosphatase activity, bacterial recombinant EhPTEN1 with the histidine tag at the amino terminus was produced using the pCold I E. coli expression system. SDS-PAGE analysis followed by Coomassie Brilliant Blue staining showed that the purified recombinant EhPTEN1 was apparently homogenous with the predicted molecular mass of 96 kDa including the histidine tag (S6A Fig). Immunoblot analysis of the purified recombinant protein using His-tag antibody confirmed the absence of truncation (S6B Fig). We first examined the enzymatic activities of recombinant EhPTEN1 using a variety of phosphoinositides (PIs) as substrates. EhPTEN1 revealed reasonable activity in a broad pH range with maximum activity obtained at pH 6.0 when the reaction was performed with 50 μM PtdIns(3,4,5)P 3 at 37˚C for 40 min (S6C Fig). We then determined the substrate specificity of EhPTEN1, using a panel of di-C8 PIs. EhPTEN1 showed highest activity with PtdIns(3,4,5)P 3 with the apparent specific activity of 8.18±0.78 nmol/min/mg (Fig 11). EhPTEN1 also catalyzed dephosphorylation of PtdIns(3,4)P 2 and PtdIns(3,5)P 2 with 6-or 3-fold lower specific activities, respectively, compared to that toward PtdIns(3,4,5)P 3 . The activities against PI monophosphates and PtdIns(4,5)P 2 were relatively low. All these characteristics are similar to those of human PTEN [34,35]. A comparison of kinetic parameters of EhPTEN1 reveals a higher affinity towards PtdIns(3,4,5)P 3 (Km = 92.5 ± 4.72 μM) as compared to PtdIns(3,4)P 2 (292 ± 18.8 μM) and PtdIns(3,5)P 2 (161 ± 20.1 μM) demonstrating that in vitro is as well the preferred substrate (Table 1 and S8 Fig). Furthermore, we produced a mutant form of EhPTEN1, in which the cysteine residue implicated for catalysis is replaced with serine by site-directed mutagenesis (EhP-TEN1_C140S) (S10 Fig). Recombinant EhPTEN1_C140S showed markedly lower phosphatase activity towards PtdIns(3,4,5)P 3 when compared to wild type, with the low specific activity (0.77 ± 0.14 nmol/min/mg). Moreover, expression of the inactive mutant GFP-EhPTEN1_C140S did not cause significant changes in pinocytosis, trogocytosis, and

Demonstration of phospholipid binding of EhPTEN1
The lipid overlay assay using amebic lysates from GFP-EhPTEN1 expressing and GFP-expressing mock transformants showed that EhPTEN1 preferentially bound to PtdIns(3)P, PtdIns(4) P, PtdIns(5)P, and, to a lesser extent, PtdIns(3,5)P 2 , and PtdIns(4,5)P 2 (S7 Fig). Furthermore, recombinant His-tagged EhPTEN1 also revealed a similar binding affinity toward a panel of PIs on the membrane, which is similar to the data given for recombinant human PTEN [36]. The binding of EhPTEN1 with diverse immobilized phospholipids may regulate its localization to different cellular compartment where EhPTEN1 exerts its phosphatase activity as has been shown previously that the binding of PtdIns(3)P with C2 domain of human PTEN target its localization to endosomal membranes [36].

Discussion
PTEN regulates fundamental roles in higher eukaryotes including cell survival, metabolic changes, cell polarity, and migration [9]. In this study, we have characterized the pivotal functions of EhP-TEN1 in migration, endocytosis, and cellular proliferation in E. histolytica. Confocal live imaging demonstrated the involvement of EhPTEN1 in the initial and intermediate stages of trogocytosis and phagocytosis. In trogocytosis of a live mammalian cell, EhPTEN1 was enriched in the region where the trogocytic tunnel was newly formed. Similarly, EhPTEN1 was accumulated on the cell periphery close to the leading edge of the phagocytic cup during the internalization of a dead host cell. The recruitment of EhPTEN1 was transient as it gradually became dissociated from the region after the completion of ingestion. These results are in line with the previous study that showed PTEN was associated with forming IgG conjugated zymosan containing phagosomes but disappeared once particle ingestion was completed [37]. The biochemical analysis showed that GFP-EhPTEN1 overexpression caused a reduction in trogocytosis and phagocytosis. In good agreement with these results, knockdown of EhPTEN1 caused remarkable enhancement in phagocytosis of dead CHO cells while trogocytosis toward live CHO cells was slightly increased. These results match those observed in earlier studies where PTEN deficient macrophages displayed enhanced phagocytic ability both in vitro and in vivo, while overexpression of PTEN significantly inhibited phagocytosis in macrophages [12,[38][39][40]. It was previously demonstrated that the depletion of PTEN in macrophages resulted in elevated PtdIns(3,4,5)P 3 levels, leading to activation of Vav1 and subsequent activation of Rac1 GTPase, the latter of which induces F-actin polymerization, which in turn enhances the engulfment of targeted cells [41]. While PIPs-mediated signaling and downstream effector in E. histolytica is not yet well understood, it has been recently reported that two AGC kinases from E. histolytica have the ability to bind PtdIns(3,4,5)P 3 and are involved in a panel of endocytic events including trogo-, phago-, and pinocytosis [21]. In

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica addition, it has also been shown that PtdIns(4,5)P 2 is localized on the plasma membrane whereas PtdIns(3,4,5)P 3 is localized on the phagocytic cup and the extended pseudopodia in E. histolytica trophozoites [42,43]. Furthermore, it was previously reported that treatment with wortmannin, the PI3K inhibitor, impaired phagocytosis of several particles including bacteria, mucin-coated beads, and hRBCs by E. histolytica trophozoites [42,44]. These data may be consistent with the mutual antagonistic role of PTEN and PI3K. These observations suggest that the control of PtdIns(3,4,5)P 3 synthesis and decomposition are important for the regulation of endocytic events in E. histolytica. We have clearly demonstrated phosphatase activity and preferred substrate specificity toward PtdIns(3,4,5)P 3 of EhPTEN1. Hence, it is highly conceivable that EhPTEN1 can regulate the local concentrations of PtdIns(4,5)P 2 and PtdIns(3,4,5)P 3 at the target sites during trogoand phagocytic processes. It seems conceivable that EhPTEN1 negatively regulates trogo-and phagocytosis by reducing the local PtdIns(3,4,5)P 3 concentration, leading to the suppression of actin-dependent cytoskeletal reorganization needed for trogo-and phagocytosis. Indeed, the concentration of EhPTEN1 is swiftly reduced on and close to trogo-and phagocytic cups (not-yetenclosed) and trogosomes and phagosomes (enclosed) soon after the completion of ingestion. We have shown that EhPTEN1 is involved in receptor-mediated endocytosis and macropinocytosis of the fluid-phase marker in an opposite fashion as in trogo-and phagocytosis. GFP-EhPTEN1 overexpression enhanced transferrin uptake while EhPTEN1 gene silencing decreased it. It was shown that at least two concentration-dependent mechanisms for transferrin endocytosis exist in E. histolytica [45]: Receptor-mediated endocytosis active at low transferrin concentrations [46] and receptor-independent internalization at high transferrin concentrations [47]. As previously demonstrated, receptor-mediated endocytosis of transferrin in E. histolytica is indeed clathrin-mediated [clathrin-mediated endocytosis (CME)] [48], and receptor-mediated endocytosis is in general clathrin-mediated and actin independent [49]. Unlike trogo-and phagocytosis, CME is also distinct in that it depends on PtdIns(4,5)P 2 and does not require PtdIns(3,4,5)P 3 [49,50]. PtdIns(4,5)P 2 binds and recruits several proteins associated with CME formation thus depleting cells of PtdIns(4,5)P 2 prevents transferrin receptor endocytosis [1,[49][50][51]. Taken together, EhPTEN1 possibly facilitates the transferrin internalization through augmentation of PtdIns(4,5)P 2 synthesis. Furthermore, EhPTEN1 showed similar phenotypes toward pinocytosis of the fluid-phase marker. As stated above, transferrin, when present at high concentrations, is internalized by receptor independent fashion in E. histolytica [47]. Thus, it is consistent with our observation that E. histolytica internalization of the fluid-phase marker and transferrin by actin-dependent macropinocytosis (S7 and S8 Movies), also as previously shown [52]. We also previously showed that EhAGCK2, which preferentially binds PtdIns(3,4,5)P 3 over PtdIns(4,5)P 2 , is involved in pinocytosis of the fluidphase marker [21], supporting the actin-dependence of macropinocytosis. However, it was shown that the local production of PtdIns(4,5)P 2 in the early stages of macropinocytosis is essential for the formation of ruffles and is partly responsible for the remodeling of the actin cytoskeleton [53]. Furthermore, it has been reported previously that deletion of PTEN in D. discoideum caused a reduction in fluid uptake [54]. Nevertheless, the lipid rafts in the plasma membrane of E. histolytica is highly enriched with PtdIns(4,5)P 2 [43] and disruption of lipid rafts with cholesterol-binding agents significantly inhibited fluid-phase pinocytosis of E. histolytica [55]. Altogether, we assume that EhPTEN1 accelerates transient synthesis of PtdIns (4,5)P 2 on the plasma membrane which in turn facilitates the formation of actin-associated macropinocytic cup. This study has provided the first observation that PTEN is differentially involved in multiple actin-related cytoskeletal activities.
We have shown that repression of gene expression of EhPTEN1 caused significant growth defect. This phenotype can be possibly explained by reduced ability in nutrient uptake where macropinocytosis is the primary and widely used method for feeding in amoebae trophozoites [56]. Furthermore, it was previously demonstrated that the growth defect in E. histolytica in low iron medium was rescued by the addition of iron-loaded holo-transferrin, and that holotransferrin was recognized by an amoebic transferrin receptor and endocytosed via clathrincoated vesicles [46,57]. These data, taken together, underscore the importance of macropinocytosis and transferrin endocytosis for the proliferation of amoebae. On the other hand, it was previously shown that loss of PTEN significantly lowered growth in D. discoideum, possibly attributable to mislocalization of myosin II during cytokinesis [58,59]. Similarly, myosin II mutants caused reduction in growth and multinucleation in E. histolytica [60]. These observations likely support the premise that EhPTEN1 regulates amoebic cell proliferation by regulation of cytokinesis and/or nutrient uptake by macropinocytosis. Furthermore, the requirement of EhPTEN1 for optimum proliferation indicates that E. histolytica apparently does not possess compensatory mechanisms for the PIPs dysregulation caused by the loss of EhPTEN1, and thus have posed it as rational drug target.
We have shown that GFP-EhPTEN1 enhances cell migration while repression of EhPTEN1 gene expression causes inhibition of motility. These observations agree well with the fact that EhPTEN1 was transiently concentrated in newly formed pseudopods. The GFP-EhPTEN1 distribution in E. histolytica is similar to the localization of mammalian PTEN, which predominantly shows cytosolic localization and mediates conversion of PtdIns(3,4,5)P 3 to PtdIns(4,5) P 2 through dynamic interaction with the inner face of the plasma membrane [61,62]. In D. discoideum, PTEN was implicated in cell migration as a positive regulator of motility, because an ameba strain lacking PTEN showed a reduction in migration speed and defect in chemotactic efficiency due to disruption of PtdIns(3,4,5)P 3 / PtdIns(4,5)P 2 concentration gradient throughout the cell [63][64][65]. In contrast, in mammalian cell types including B cells, glioma cells, and fibroblasts, PTEN was shown to inhibit migration [66,8]. Loss of PTEN also resulted in dysregulation of myosin II assembly at the cell cortex, where PTEN prevents the formation of lateral pseudopodia and promotes cell body contraction and posterior retraction in D. discoideum [63,67]. In addition, PtdIns(4,5)P 2 , produced by PTEN, can recruit and activate a wide variety of actin regulatory proteins at the plasma membrane, thereby controlling motility [1,50]. For example, PtdIns(4,5)P 2 activates N-WASP directly or indirectly through interaction with IQGAP1 which result in promoting actin polymerization by activation of N-WASP-Arp2/3 complex [1,50,68]. Among them, myosin II and Arp2/3 complex are conserved in E. histolytica, where myosin II plays a critical role in movement [60,69] and Arp2/3 complex is involved in actin nucleation [70]. Thus, it is conceivable that EhPTEN1 mediates signaling for pseudopod formation and migration through regulation of PtdIns(3,4,5)P 3 metabolism.
In conclusion, we have shown the biological significance of EhPTEN1 in different forms of endocytosis including trogocytosis, phagocytosis, pinocytosis, and clathrin-mediated endocytosis. We have also demonstrated the essentiality of EhPTEN1 in pseudopod formation, motility, and optimal growth of E. histolytica. Taken together, these findings emphasize the importance of EhPTEN1 in modulating a plethora of functions in E. histolytica. Exploring PTEN functions in E. histolytica will hopefully increase our knowledge on the regulation of cellular processes related to actin remodeling through the PIPs signaling pathway.

Identification and comparison of PTEN sequences
Amino acid sequences of PTEN from E. histolytica and other organisms were gained from AmoebaDB (http://amoebadb.org/amoeba/) and NCBI (https://www.ncbi.nlm.nih.gov) respectively, and aligned using CLUSTAL W program (http://clustalw.ddbj.nig.ac.jp/) to examine the domain configuration and the key residues for phosphatase activity [71].

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica

Establishment of E. histolytica transformants
To construct a plasmid to express EhPTEN1 fused with HA or GFP tag fused at the amino terminus, a DNA fragment corresponding to cDNA encoding EhPTEN1 was amplified by polymerase chain reaction (PCR) from E. histolytica cDNA using a pair of primers listed in S2 Table. The PCR-amplified fragments were digested with XmaI and XhoI and cloned into pEhEx-HA and pEhEx-GFP vectors [21,22] that had been predigested with XmaI and XhoI, to produce pEhExHA-EhPTEN1 and pEhExGFP-EhPTEN1. For antisense small RNA-mediated transcriptional silencing of EhPTEN1 gene, a 420 bp fragment of the protein coding region of EhPTEN1 gene, corresponding to the amino terminus of the protein, was amplified by PCR from cDNA with sense and antisense oligonucleotides containing StuI and SacI restriction sites (S2 Table). The amplified product was digested with StuI and SacI and ligated into the compatible sites of the double digested psAP2-Gunma plasmid [75] to synthesize a gene silencing plasmid designated as psAP2-EhPTEN1. Two plasmids, pEhExHA-EhPTEN1 and pEhExGFP-EhPTEN1, were introduced into the trophozoites of E. histolytica HM-1:IMSS cl6 strain, whereas psAP2-EhPTEN1 was introduced into G3 strain by lipofection as described previously [76]. Transformants were initially selected in the presence of 1 μg/ml G418 until the drug concentration was gradually increased to 10 μg/ml for the EhPTEN1 gene silenced stain and 20 μg/ml for the GFP-and HA-EhPTEN1 overexpressing stains. Finally, all transformants were maintained at 10 or 20 μg/ml G418 in BIS medium.

Reverse transcriptase PCR
Reverse transcriptase PCR was performed to check mRNA levels of EhPTEN1 in EhPTEN1 gene silenced and control strains. Total RNA was extracted from trophozoites of EhPTEN1 gene silenced and control strains that were cultivated in the logarithmic phase using TRIZOL reagent (Life Technologies, California, USA). Approximately one μg of DNase treated total RNA was used for cDNA synthesis using Superscript III First -Strand Synthesis System (Thermo Fisher Scientific, Massachusetts, USA) with reverse transcriptase and oligo (dT) primer according to the manufacture's protocol. Ex Taq PCR system was used to amplify DNA from the cDNA template using the primer pairs listed in S2 Table. The PCR conditions were as follow: initial denaturation at 98˚C for 10 sec; then 25 cycles at 98˚C for 10 sec, 55˚C

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica for 30 sec, and 72˚C for 20 sec; and a final extension at 72˚C for 7 min. The PCR products obtained were resolved by agarose gel electrophoresis.

Immunoblot analysis
Trophozoites of amoeba transformants expressing HA-EhPTEN1 or GFP-EhPTEN1 grown in the exponential growth phase were harvested and washed three times with phosphate buffer saline (PBS). After resuspension in lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% Triton-X 100, 0.5 mg/ml E-64, and protease inhibitor), the trophozoites were kept on ice for 30 min, followed by centrifugation at 500 × g for 5 min. Approximately 20 μg of the total cell lysates were separated on 10% SDS-PAGE and subsequently electrotransferred onto nitrocellulose membranes. The membranes were incubated in 5% non-fat dried milk in Tris-Buffered Saline and Tween-20 (TBST; 50 mM Tris-HCl, pH 8.0, 150 mM NaCl, and 0.05% Tween-20) for 1 hr at room temperature to block non-specific protein. The blots were reacted with one of the following primary antibodies diluted as indicated: anti-HA 16B12 monoclonal mouse antibody at a dilution of 1:1,000, anti-GFP mouse monoclonal antibody (1:100), and anti-CS1 rabbit polyclonal antisera [77] (1:1,000) at 4˚C overnight. The membranes were washed with TBST and further reacted with horseradish peroxidase-conjugated (HRP) anti-mouse or antirabbit IgG antisera (1:10,000) at room temperature for 1 hr. After washings with TBST, the specific proteins were visualized with a chemiluminescence HRP Substrate system (Millipore, Massachusetts, USA) using LAS 4000 (Fujifilm Life Science, Cambridge, USA) according to the manufacture's protocol.

Live cell imaging
Approximately 5×10 5 trophozoites of the transformant strain expressing GFP-EhPTEN1 were cultured on a 35 mm (in diameter) collagen-coated glass-bottom dish (MatTek Corporation, Massachusetts, USA) in 3 ml of BIS medium under anaerobic conditions. CHO cells were stained with BIS medium containing 10 μM CellTracker Orange for 40 min followed by washing three times with PBS. Approximately 2×10 4 prestained CHO cells 200 μl BIS were gently overlayed to trophozoites grown on the glass-bottom dish as prepared above. The central part of the dish was then carefully covered with a 1 cm square coverslip and the edge of the coverslip on the slide glass was sealed with nail polish. Live images were captured on Zeiss LSM780 confocal microscope (Carl-Zeiss, Oberkochen, Germany) using a 63× oil immersion objective with default settings on the time series mode and analyzed by ZEN software (Carl-Zeiss). For live imaging of trogocytosis and phagocytosis, approximately twenty cells were analyzed per each cell line in an experiment of three independent experiments. GFP-EhPTEN1 recruitment was estimated by quantitating the average fluorescent intensity per pixel across the line drawn over the cell or that of 11-13 regions of interest on the pseudopods and other parts of the cell.

Indirect immunofluorescence assay (IFA)
Approximately 5×10 3 trophozoites in 50 μl BIS were transferred to an 8 mm round well on a slide glass (Matsunami Glass Ind, Osaka, Japan). After 30 min incubation in an anaerobic chamber at 35.5˚C, 5×10 4 CHO cells that had been pre-stained with 10 μM CellTracker Blue in 50 μl BIS were added to the well and the mixture was incubated for 15 min. After removing the medium, cells were fixed with PBS containing 3.7% paraformaldehyde at room temperature for 10 min, and subsequently permeabilized with PBS containing 0.2% Triton 100-X and 1% bovine serum albumin (BSA) for 10 min each at room temperature. The cells were then reacted with anti-HA mouse monoclonal antibody (1:1000) for 1 hr at room temperature. Then the sample was reacted with Alexa Flouor-488 conjugated anti-mouse IgG (1:1000)

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica antibody (Thermo Fisher, Massachusetts, USA). The images were then captured using LSM 780 confocal microscope and analyzed by ZEN software (Carl-Zeiss, Oberkochen, Germany).

Trogocytosis and phagocytosis assay using CQ1
We quantified trogocytosis and phagocytosis efficiency using an image cytometer as described previously [27]. Briefly, trophozoites of E. histolytica were incubated in BIS containing 10 μM CellTracker Blue (Thermo Fisher) at 35.5˚C for 1 hr. After staining, ameba trophozoites were washed 3 times with PBS and resuspended in OPTI-MEM medium (Thermo Fisher, Massachusetts, USA) containing 15% adult bovine serum (Sigma Aldrich). Approximately 2×10 4 ameba trophozoites were seeded into a well on a 96-well glass bottom plate (IWAKI, Shizuoka, Japan) and incubated in anaerobic chamber for 40 min. After incubation, about 1×10 5 live or heat killed CHO cells that have been stained with 10 μM CellTracker Orange were added to the well containing amebae. The images of trogocytosis and phagocytosis were captured on a Confocal Quantitative image cytometer CQ1 (Yokogawa Electric Corporation, Tokyo, Japan) using a 20× objective every 10 min for 1 hr. Five fields each per well for three wells were chosen for image analysis. Images on five planes with 2 μm intervals in Z axis were obtained to calculate the volume of the ingested CHO cells using three-dimensional reconstructed images. For analysis of motility, images of approximately 400 cells were captured from each well and analyzed using CellPathfinder software (Yokogawa Electric Corporation, Tokyo, Japan) according to the manufacture's protocol. The protocol for quantitative analysis of the images was optimized based on the cell morphology and fluorescence intensity to identify E. histolytica trophozoites and CHO cells in each field. The multiple parameters were measured to evaluate the efficiency of trogocytosis and phagocytosis: the average number of internalized CHO cells per amoeba, the combined volume of internalized CHO cells per amoeba, and the percentage of amebic trophozoites that ingested the target cells in the whole population.

Measurement of fluid-phase and receptor-mediated endocytosis
Approximately 2.5×10 5 amebic transformants were incubated in BIS medium containing 2 mg/ml fluorescent fluid-phase marker RITC dextran at 35˚C for indicated time points. The cells were collected, washed three times with PBS, and resuspended in 250 μl of lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, and 0.1% Triton-X 100). Fluorescence intensity was measured using a plate reader (SpectraMax Paradigm Multi-Mode, Molecular Devices, California, USA) at an excitation wavelength of 570 nm and an emission wavelength of 610 nm.
Approximately 2×10 4 amebic transformants were incubated in BIS containing 20 μM Cell-Tracker Blue (Thermo Fisher) at 35.5˚C for 1 hr. After staining, approximately 10 4 amebic transformants resuspended in 100 μl of BIS medium were transferred to a well on a 96-well glass-bottom plate. After incubation at 35.5˚C in an anaerobic chamber for 40 min, 0.5 mg/ml of transferrin conjugate with Alexa Fluor 568 was added to the well and images were acquired by CQ1 and analyzed as above.

Migration (motility) assay
Amoebic trophozoites grown in the logarithmic growth phase were harvested and labelled with 20 μM CellTracker Green for 1 hr at 35.5˚C. After washing 3 times with PBS, cells were transferred to a well on a 96-well glass-bottom plate and time lapse images were captured on CQ1. The motility of the cells was measured using CellPathfinder software.

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica

Quantitative real-time (qRT) PCR
The relative levels of mRNA of EhPTEN1 gene and RNA polymerase II gene, as an internal standard, were measured by qRT-PCR. PCR reaction was prepared using Fast SYBR Master Mix (Applied Biosystems, California, USA) with 100 ng cDNA and a primer set shown in S2 Table. PCR was conducted using the StepOne Plus Real-Time PCR system (Applied Biosystems, California, USA) with the following cycling conditions: an initial step of denaturation at 95˚C for 20 sec, followed by 40 cycles of denaturation at 95˚C for 3 sec, annealing and extension at 60˚C for 30 sec. The mRNA expression level of EhPTEN1 gene in the transformants was expressed as relative to that in the control transfected with psAP2.

Growth assay of E. histolytica trophozoites
Approximately 10 4 trophozoites of E. histolytica G3 strain transformed with psAP2-EhPTEN1 and psAP2 (control), grown in the logarithmic phase, were inoculated into 6 ml of fresh BI-S-33 medium containing 10 μg/mL G418, and the parasites were counted every 24 hr on a hemocytometer.

Production of EhPTEN1 recombinant protein
To construct the plasmid for the production of recombinant EhPTEN1 containing a histidinetag at the amino terminus, the full-length protein coding sequence of EhPTEN1 gene was amplified by PCR using oligonucleotide primers listed in S2 Table. PCR was performed with PrimeSTAR Max DNA polymerase (Takara Bio Inc, Shiga, Japan) with the following parameters: initial incubation at 95˚C for 1 min; followed by 30 cycles of denaturation at 98˚C for 10 sec; annealing at 55˚C for 5 sec; and elongation at 72˚C for 15 sec; and a final extension at 72˚C for 30 sec. The PCR fragment was digested with BamHI and SalI and ligated into BamHI and SalI double digested pCold-1 vector (Takara Bio Inc, Shiga, Japan) to produce pCold1-EhPTEN1 plasmid. The pCold-1-EhPTEN1 was introduced into E. coli BL21 (DE3) cells by heat shock at 42˚C for 45 sec. E. coli BL21 (DE3) strain harboring pCold-1-EhPTEN1 was grown at 37˚C in 50 ml of LB medium (BD Difco, New Jersey, USA) in the presence of 100 μg/ ml ampicillin. The overnight culture was used to inoculate 500 ml of fresh medium, and the culture was further continued at 37˚C with shaking at 220 rpm for approximately 2 hr. When A 600 absorbance reached 0.6, then 1mM of isopropyl β-D-thio galactopyranoside (IPTG) was added, and cultivation was continued for another 24 hr at 15˚C. The E. coli cells from the induced culture were harvested by centrifugation at 75,000 rpm for 20 min at 4˚C. The cell pellet was washed three times with PBS, re-suspended in 30 ml of the lysis buffer (50 mM Tris-HCl, pH 8.0, 300 mM NaCl, and 0.1% Triton X-100) containing 100 μg/ml lysozyme, and 1 mM phenylmethyl sulfonyl fluoride (PMSF), and incubated at room temperature for 30 min. After incubation, the mixture was sonicated on ice and centrifuged at 13,000 rpm for 20 min at 4˚C. The supernatant was mixed with 1 ml of 50% Ni 2+ -NTA His-bind resin (Qiagen, Hilden, Germany), incubated for 1 hr at 4˚C with mild rotatory shaking. The resin that recombinant His-EhPTEN1 bound was washed in a disposal column three times with 5 ml of lysis buffer containing 10-30 mM of imidazole. Bound proteins were eluted with 3 ml each of lysis buffer containing 100-300 mM imidazole to obtain recombinant EhPTEN1. The integrity and the purity of the recombinant protein were confirmed with 10% SDS-PAGE analysis, followed by Coomassie Brilliant Blue staining. Then the protein was concentrated, and the buffer was replaced with 50 mM Tris-HCl, 150 mM NaCl, pH 8.0 using Amicon Ultra 50K centrifugal device (Millipore, Massachusetts, USA). The protein was stored at -30˚C with 50% glycerol in small aliquots until further use.

PLOS PATHOGENS
PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica

Lipid phosphatase assay
EhPTEN1 enzymatic activity was determined by the method previously described [78]. Di-C8 phosphatidylinositol phosphate(s) (PIPs) (Echelon Bioscience, Salt Laken City, USA) were dissolved in 100 mM MOPS, pH 6.0, solution, flash frozen in liquid nitrogen, and stored at -20˚C between uses. For determination of pH optimum, the following buffers were used, 100 mM acetate buffer (pH 4.0, pH 4.5, pH 5.0, pH 5,5), 100 mM MOPS buffer (pH 6.0, pH 6.5, pH 7.0), and 100 mM Tris-HCl (pH 7.5, pH 8.0, pH 8.5, pH 9.0). For determination of substrate specificity, a reaction mixture was composed of 25 μl of 100 mM MOPS pH 6.0 containing 5 μg of recombinant EhPTEN1 and 100 μM PIPs. The reaction was carried out at 37˚C for 40 min and the produced phosphate was measured using Malachite Green Reagent (Cell Signaling technology, Massachusetts, USA). After incubation for 15 min at room temperature, the absorbance was measured at a wavelength of 630 nm. The Lineweaver Burk Plot was used to calculate the kinetic parameters of EhPTEN1.

Lipid membrane overlay assay
Approximately 6×10 6 trophozoites of GFP-EhPTEN1 expressing strain were harvested, washed with PBS, and concentrated by centrifugation. Approximately one hundred μl of lysis buffer (50 mM Tris-HCl, pH 7.5 150 mM NaCl, and 0.1% Triton X-100, 1× complete mini, and 0.5 mg/ml E64) was added to the cell pellet. The mixture was incubated on ice for 30 min and centrifuged at 13000 rpm for 5 min. The supernatant was collected and used as the total lysate. GFP-EhPTEN1 was immunoprecipitated using GFP-Trap Agarose Kit (ChromoTek, Planegg, Germany) according to the manufacturer's instruction, eluted, and confirmed by immunoblot. Lipid membranes on which a panel of phospholipids were immobilized (PIP strips: P-6001, Echelon Bioscience, Salt Laken City, USA) were blocked with 1% fat free BSA in PBS-T (PBS containing 0.05% Tween 20) for 1 hr at room temperature. The membranes were then incubated with 2 ml of lipid binding solution (1% fat free BSA in PBS-T, 1x complete mini, 0.05 mg/ml E64, 20 μl of eluted lysate) for 3 hr at 4C. After the membrane were washed twice with PBS-T at 4˚C, they were reacted with anti-GFP at 1:100 dilution with PBS-T containing 1% fat free BSA (PBS-TB) in for 3 hr at 4C. After incubation with the first antibody, the membranes were further reacted with HRP conjugated anti-mouse IgG rabbit antiserum at 1:10,000 dilution with PBS-TB in at 4˚C for 1 hr. The membranes were washed three times with PBS-T at 4˚C and the specific proteins were visualized with a chemiluminescence HRP substrate system (Millipore, Massachusetts, USA) using LAS 4000 (Fujifilm Life Science, Cambridge, USA) according to the manufactures' protocol. Recombinant EhPTEN1 protein was also used except that 1 μg/ml of recombinant protein were incubated on the membrane over night at 4˚C and anti-His antibody was used as the first antibody with a dilution of 1:1,000.

Site directed mutagenesis of EhPTEN1
Overlapping primers were designed to replace cysteine at amino acid 140 of the pEhExGF-P-EhPTEN1 and pCold1-EhPTEN1 plasmids with serine. The primer sets used are listed in S2 Table. PCR reaction was carried out using 0.2 μM/μl each of the primers, 25 μl of PrimeSTAR Max DNA polymerase (Takara Bio Inc, Shiga, Japan) and pEhExGFP-EhPTEN1 or pCold1-EhPTEN1as a template. The PCR parameters were: incubation at 95˚C for 1 min; followed by 30 cycles of denaturation at 98˚C for 10 sec, annealing at 55˚C for 15 sec, and elongation at 72˚C for 50 sec; and a final extension at 72˚C for 30 sec. The PCR fragment was evaluated by agarose gel electrophoresis and introduced into DH5α competent E. coli cells by heat shock at 42˚C for 45 sec. Bacterial colonies containing the plasmid that encodes mutated EhPTEN1 (pEhExGFP-EhPTEN1_C140S and pCold-1-EhPTEN1_C140S) were selected from LB plate PLOS PATHOGENS PTEN differentially regulates endocytosis, migration, and proliferation in Entamoeba histolytica containing 100 μg/mL ampicillin by DNA sequencing of the plasmids. The pEhExGFP-EhPTEN1_C140S plasmid was introduced into the trophozoites of E. histolytica HM-1:IMSS cl6 strain as described previously [76]. Transformants were initially selected in the presence of 1 μg/ml G418 until the drug concentration was gradually increased to 20 μg/ml. The pCold-1-EhPTEN1_C140S was introduced into E. coli BL21 (DE3) cells to produce the inactive mutant recombinant protein as described previously. Trophozoites of GFP mock transfected and GFP-EhPTEN1_C140S expressing strains were mixed with RITC dextran and uptake was monitored as described in Materials and methods. Experiments were conducted twice independently and error bars indicating standard errors of two biological replicates. (E-F) Effect of GFP-EhPTEN1_C140S expression on trogocytosis (E) and phagocytosis (F). Trophozoites of GFP mock transfected and GFP-EhPTEN1_C140S expressing strains were incubated with live CHO cells (E) or heat killed CHO cells (F) that have been stained with CellTracker Orange to evaluate trogocytosis and phagocytosis, respectively. The images were captured on CQ1 as described in Materials and methods and analyzed to calculate the numbers of CHO cell-containing trogosomes or phagosomes per amoeba. Experiments were conducted twice independently in duplicates and error bars indicating standard errors of two biological replicates. (TIF) (XP_656532.1), and EhPTEN6 (XP_654831.1) was constructed by using clustalw algorithm (http://clustalw.ddbj.nig.ac.jp). PTEN phosphatase domain and C2 domain are shown with blue and yellow backgrounds, respectively. The green rectangle corresponds to the PtdIns(4,5) P 2 -binding motif (PDM domain). Amino acid residues implicated for PtdIns(3,4,5)P 3 catalysis are marked with a red rectangle. Cytosolic localization signal and restudies important for TI loop formation are indicated in black and blue lines, respectively. The DUF457 domain is shown in magenta background, whereas PEST and PDZ-BM are shown in green and cyan background, respectively. Note that for human PTEN, only the amino terminal part is shown. (TIF) S12 Fig. Schematic diagram represents potential function of EhPTEN1 in E. histolytica during trogocytosis, phagocytosis, clathrin-mediated endocytosis (CME), micropinocytosis, and migration. EhPTEN1 localizes mainly in the cytosol with some enrichment in pseudopod-like structures and in the early stage of trogo-and phagocytosis, probably due to changes in the lipid content of the membrane during these events. EhPTEN1 can negatively regulates trogocytosis and phagocytosis by reducing the local PtdIns(3,4,5)P 3 concentration leading to the suppression of actin dependent cytoskeletal organization needed for trogo-and phagocytosis. In contrast, EhPTEN1 serves as a positive regulator of fluid-phase and receptormediated endocytosis in E. histolytica possibly via augmentation of PtdIns(4,5)P 2 synthesis which causes recruitment of clathrin regulator and enhances formation of lipid rafts that needed for micropinocytosis. (TIF) S1