Research ArticleHost-Pathogen Interactions

Leishmania GP63 Alters Host Signaling Through Cleavage-Activated Protein Tyrosine Phosphatases

See allHide authors and affiliations

Science Signaling  29 Sep 2009:
Vol. 2, Issue 90, pp. ra58
DOI: 10.1126/scisignal.2000213

Abstract

With more than 12 million people affected worldwide, 2 million new cases occurring per year, and the rapid emergence of drug resistance and treatment failure, leishmaniasis is an infectious disease for which research on drug and vaccine development, host-pathogen, and vector-parasite interactions are current international priorities. Upon Leishmania-macrophage interaction, activation of the protein tyrosine phosphatase (PTP) SHP-1 rapidly leads to the down-regulation of Janus kinase and mitogen-activated protein kinase signaling, resulting in the attenuation of host innate inflammatory responses and of various microbicidal macrophage functions. We report that, in addition to SHP-1, the PTPs PTP1B and TCPTP are activated and posttranslationally modified in infected macrophages, and we identify an essential role for PTP1B in the in vivo progression of Leishmania infection. The mechanism underlying PTP modulation involves the proteolytic activity of the Leishmania surface protease GP63. Access of GP63 to macrophage PTP1B, TCPTP, and SHP-1 is mediated in part by a lipid raft–dependent mechanism, resulting in PTP cleavage and stimulation of phosphatase activity. Collectively, our data present a mechanism of cleavage-dependent activation of macrophage PTPs by an obligate intracellular pathogen and show that internalization of GP63, a key Leishmania virulence factor, into host macrophages is a strategy the parasite uses to interact and survive within its host.

Introduction

The strategic modulation of host cell signaling pathways by Leishmania, in which parasite-induced activation of macrophage (MØ) protein tyrosine phosphatases (PTPs) plays a critical role, results in the inhibition of several phagocyte functions, providing a suitable environment for intraphagosomal parasite development and survival (1). In the early stages of Leishmania-MØ interactions, alterations in the JAK/STAT (Janus kinase/signal transducer and activator of transcription), MAPK (mitogen-activated protein kinase), and IRAK-1 [interleukin-1 (IL-1) receptor–associated kinase 1] signaling pathways occur as a result of parasite-induced activation of the host PTP SHP-1 (2, 3). This event leads to the attenuation of innate inflammatory responses and decreased nitric oxide (NO) production, two processes that are fundamental for controlling the intracellular parasite and its progression within the host (2, 4, 5). Despite the critical role of SHP-1 in the Leishmania-mediated inhibition of MØ signaling, additional PTP enzymatic activation is observed in SHP-1-deficient MØs infected with the parasite (2), suggesting that additional MØ PTPs could play a role in this down-regulation process.

PTP1B and T cell phosphatase (TCPTP) are also PTPs that, in addition to SHP-1, negatively regulate JAK/STAT and MAPK signaling (6, 7). The role of PTP1B in metabolic, oncogenic, and immunological processes (810) highlights its importance as a key regulator of signal transduction. PTP1B inhibits MØ activation in vivo; MØ NO production and the concentrations of IL-12 and interferon-γ (IFN-γ) in serum were higher in PTP1B−/− mice after bacterial lipopolysaccharide (LPS) stimulation than they were in wild-type mice (9). Because decreased production of IL-12, tumor necrosis factor–α (TNF-α), and NO are hallmarks of Leishmania infections (1), PTP1B is a rational candidate for mediating Leishmania-dependent inhibition of host functions.

The initial interaction between Leishmania and its host cell is largely mediated by abundant promastigote surface molecules, including lipophosphoglycan (LPG) and the surface protease GP63 (1). GP63 is a virulence factor that has been implicated in binding the parasite to MØs and enhancing parasite phagocytosis, parasite evasion of complement-mediated lysis, parasite migration through the extracellular matrix (11), and induction of the host T helper cell 1 (TH1)–type immune response, amongst other responses (1214). We identified GP63 as the key Leishmania virulence factor that modulates host PTPs and revealed an essential role for PTP1B in the progression of cutaneous leishmaniasis in infected mice. Furthermore, we report a mechanism whereby Leishmania GP63 accesses the MØ intracellular milieu in part by a lipid raft–dependent manner, allowing a direct interaction with host protein substrates. Collectively, our findings provide insight into strategies used by pathogens to exploit host cell negative regulatory mechanisms resulting in the evasion of innate immune response activation.

Results

Leishmania infection modulates multiple PTPs

Activation of SHP-1, a key negative regulator of MØ signaling, greatly contributes to the Leishmania-induced alterations of MØ JAK signaling, MAPK signaling, and TLR (Toll-like receptor) signaling mediated by IRAK-1 (1, 3, 15). Although SHP-1 activation seems to suffice for the parasite-mediated inhibition of NO production, induction of PTP activity is still observed in SHP-1−/− MØs (2), suggesting that other host PTPs are modulated by Leishmania. To address this point, we performed an in-gel PTP activity assay of MØ PTPs in infected and uninfected cells. Mouse B10R MØs were infected with Leishmania mexicana (0 to 4 hours) and protein extracts were subjected to in-gel PTP assay (Fig. 1A). Modulation of the activity of multiple PTPs was detected as early as 1 min after exposure of MØ to Leishmania and was sustained for up to 4 hours after infection, as evidenced by the appearance of new active PTP bands (~65, 45, 39, and 37 kD) and disappearance of basally active PTPs (~140, 70, and 48 kD). Infection with Leishmania major, Leishmania donovani, or L. mexicana showed that mammalian cutaneous and viscerotropic species modulate MØ PTPs, and the kinetics of PTP modulation were fastest in cells infected by L. mexicana (Fig. 1B). Although the lizard pathogen Leishmania tarentolae infected the cells, it had no effect on PTP activity (Fig. 1B). This suggests that the parasite-dependent PTP modulation is associated with Leishmania mammalian pathogenicity, but not with tropism within the host.

Fig. 1

Leishmania infection modulates multiple MØ PTPs. (A) MØs were infected with L. mexicana (0 to 240 min) or with (B) L. donovani, L. major, or L. tarentolae (15 and 60 min). Total protein lysates (40 μg) were loaded for in-gel PTP assay, where bands of dephosphorylation (clear bands) represent active PTPs. Modulation of MØ PTPs is represented by arrows, where black and white, respectively, represent increase in or loss of active PTP bands after Leishmania infection.

Modulation of host PTPs is associated with protein cleavage and is independent of Leishmania internalization

Because of the disappearance of PTP activity at higher molecular weights and the appearance of PTP activity at lower molecular weights, the PTP activity profile seen by in-gel PTP assay is suggestive of PTP cleavage. To investigate the nature of PTP modulation, we performed Western blot analysis of MØ PTPs that are implicated in the regulation of signaling pathways previously reported as modulated by Leishmania infection (1). Upon L. mexicana infection, a rapid and time-dependent cleavage of SHP-1, PTP1B, and TCPTP (Fig. 2A) was observed. Full-length TCPTP was completely lost within 5 min after infection, whereas disappearance of the full-length SHP-1 and PTP1B was apparent by 30 min after infection. Hierarchical cleavage of PTP1B and TCPTP occurred, because the appearance of a second smaller cleavage product was detected for each PTP at later time points (Fig. 2A). Cleavage was specific; no protein degradation or cleavage products were detected even 1 hour after infection for various other phosphatases: the PTP SHP-2, the lipid and protein phosphatase PTEN (phosphatase and tensin homolog), or the serine-threonine phosphatase PP2A (Fig. 2B).

Fig. 2

Modulation of MØ PTPs is associated with protein cleavage and independent of parasite internalization. The kinetics (5 to 60 min) of PTP cleavage was evaluated by Western blot (WB) of total protein lysates from Leishmania-infected or uninfected MØs. (A) TCPTP, PTP1B, and SHP-1 were cleaved as early as 5 min after infection. (B) PTEN, SHP-2, and PP2A were not cleaved or degraded upon infection. (C) MØs were incubated with stationary-phase L. mexicana supernatant for 1 to 60 min or (D) exposed for 1 hour to 0.8-μm latex beads or beads coated with parasite culture supernatant. PTPs in the total protein lysates were evaluated by Western blot. In all panels, black arrows represent full-length proteins and cleavage fragments (for the PTPs).

Leishmania internalization into MØs is rapidly completed within 10 min of initial MØ-parasite contact (16). In spite of this, parasite internalization was not required for PTP cleavage because incubation with parasite culture supernatant led to SHP-1, PTP1B, and TCPTP proteolysis (Fig. 2C). Although the extent of PTP cleavage was lower, evidenced by no completion of the cleavage reaction for SHP-1 and PTP1B (defined as disappearance of the full-length PTP), the reaction followed similar kinetics as in cells infected with the whole parasite; PTP cleavage fragments were evident as early as 1 min after incubation with parasite culture supernatant (Fig. 2C) and within 5 min after infection (Fig. 2A). PTP cleavage did not result nonspecifically from a phagocytosis-dependent mechanism, because it was undetectable in B10R MØs exposed to 0.8-μm latex beads for 1 hour. However, when the latex beads were coated with L. mexicana culture supernatant, complete PTP cleavage occurred (Fig. 2D). These findings suggest that a soluble Leishmania factor is responsible and sufficient for triggering MØ PTP cleavage. Although not requiring parasite phagocytosis, the factor may also be associated with the parasite, and Leishmania internalization may facilitate delivery of the proteolysis-stimulating factor and thus may contribute to the extent of PTP proteolysis.

Although SHP-1, PTP1B, and TCPTP were all cleaved in response to exposure to Leishmania, only the intracellular distribution of SHP-1 changed upon MØ infection (fig. S1). As previously reported (1) and shown in fig. S1, in Leishmania-infected MØs SHP-1 adopts a punctuated cytoplasmic distribution. PTP1B is present diffusely in the cytoplasm of both uninfected and infected MØs, and TCPTP is observed as nuclear punctuated clusters in infected and uninfected cells (fig. S1).

Leishmania GP63 is responsible for parasite-induced PTP cleavage

Leishmania surface molecules, including LPG, glycosylinositol phospholipids (GIPLs), and the protease GP63, are widely recognized by their abundance and importance in the interface of host-pathogen interactions. Given the importance of GP63 protease as a virulence factor and its nature as a protein that is bound to the parasite surface and released from the parasite (17, 18), we investigated whether GP63 was involved in MØ PTP cleavage. It is worth noting that the fraction of GP63 that is associated with the parasite represents approximately two-thirds of the total GP63 with the other one-third released by the parasite (19).

GP63 proteases are encoded by a family of genes: In L. mexicana, there are 10 genes encoding GP63 proteases (20), there are 7 in L. major, and at least 7 in L. donovani (13). Genetic knockout of all GP63 proteases (L. major GP63−/−) completely abrogates PTP cleavage in infected MØs, and reinsertion of L. major GP63 gene 1 into the GP63 knockout strain (L. major GP63R) rescues PTP cleavage in infected cells (Fig. 3A). Moreover, in-gel PTP activity assay of protein extracts of MØs infected with L. major GP63−/− shows that modulation of host PTP activity is abrogated in the absence of GP63 even upon prolonged times of infection (fig. S2); the pattern of PTP activity matches that of the uninfected control (Fig. 3B and fig. S2). Collectively, these data show that GP63 is necessary for the cleavage of target PTPs.

Fig. 3

Parasite-induced PTP modulation is dependent on Leishmania GP63. B10R MØs were infected with L. major, L. major GP63−/−, or L. major GP63R for 1 hour. Total MØ protein lysates were evaluated by (A) Western blot and (B) in-gel PTP activity assay. (C) GP63 directly cleaves PTPs. Purified GST-SHP-1, GST-TCPTP, or immunoprepicipitated (IP) PTP1B (2.5 μg) was incubated with 30 × 106 L. major, L. major GP63−/−, L. major GP63R, or 1 μg rGP63 for 1 hour at room temperature on a rocking platform. TCPTP, SHP-1, and PTP1B cleavage were evaluated by Western blot. Arrows indicate cleavage fragments.

We evaluated whether GP63 can directly interact and cleave target PTPs. GP63 protease recognizes a consensus site in its target substrates, P1↓P′1-P′2-P′3, where the arrow represents the site of cleavage. P1 corresponds preferentially to a polar amino acid, P′1 to a hydrophobic amino acid residue, and P′2 and P′3 to basic residues; however, substrate hydrolysis is not restricted to these residues (21). Sequence analysis of human SHP-1, PTP1B, and TCPTP predicted many putative GP63 cleavage sites (fig. S3 and table S1) and the predicted molecular weights of fragments resulting from cleavage at some of these putative motifs closely correspond to the cleavage fragments observed by Western blot. Localization of these motifs within the three-dimensional structure representation of the N-terminal domains of SHP-1, TCPTP, and PTP1B (fig. S3) revealed that putative cleavage sites 2, 3, 5, and 6 of SHP-1, 1 through 3 of TCPTP, and 1 through 5 of PTP1B were present in the periphery of the globular molecules, implying structural permissiveness for protein-protein (PTP-GP63) interactions. The unavailability of crystal structures for full-length PTPs prevented a similar structural analysis of putative SHP-1, TCPTP, and PTP1B C-terminal cleavage sites.

Molecular weight prediction of cleavage fragments generated by proteolysis at each of the putative cleavage sites (table S1) and targeted deletion of SHP-1 site 10 and PTP1B sites 1, 4, and 5 (fig. S4) allowed us to rule out sites 1, 3, 4, 10, and 11 of SHP-1, 1 through 5 of PTP1B, and 1 and 3 of TCPTP as putative cleavage sites of GP63. Thus, the most likely cleavage sites are 2 and 5 through 9 in SHP-1, 6 in PTP1B, and 2 and 4 through 11 in TCPTP.

The direct PTP-GP63 interaction was confirmed by assaying for cleavage in vitro with purified proteins. Purified glutathione S-transferase (GST)–tagged full-length SHP-1 (22) or GST-tagged TCPTP fusion proteins or immunoprecipitated PTP1B were incubated with 3 × 107 L. major, L. major GP63−/− or L. major GP63R, or with recombinant GP63 (rGP63; 1 μg) (23). Evaluation of cleavage by Western blot showed that cleavage occurred after incubation with wild-type or GP63R L. major strains and was absent with L. major GP63−/− treatment (Fig. 3C). The minor band noted on L. major GP63−/−–infected samples likely corresponds to unspecific cleavage of the immunoglobulin G chain by another parasite protease, because it is only present in immunoprecipitated samples, not in total protein extracts. Cleavage was also detected upon incubation of the PTPs with rGP63, confirming that the PTPs are substrates of GP63. These data show that there is a direct interaction between these PTPs and GP63 and that GP63 is likely to be the protease that cleaves PTPs in Leishmania-infected MØs.

Leishmania GP63 alters the profile of PTPs associated with JAK-2 at the host cell plasma membrane

TCPTP, PTP1B, and SHP-1 are negative regulators of the JAK/STAT signaling pathway and dephosphorylate JAK family members (7, 24, 25). Induction of JAK/STAT signaling follows activation of the IFN receptor (IFNR) by IFN, where the JAKs, which are constitutively bound to the IFNR, are autophosphorylated and activated (26). Negative regulation of signal transduction at this stage in the pathway requires PTPs to shuttle to or constitutively reside at the inner face of the plasma membrane to access their substrates, the phosphorylated JAKs. Given that upon Leishmania-MØ contact PTP cleavage occurs as fast as 5 min after infection (Fig. 2A), we hypothesized that a rapid interaction between GP63 and the target PTP should occur at the stage of initial parasite-cell contact. GP63 would then gain access, likely through internalization, to the shuttling or resident PTPs at the inner face of the MØ plasma membrane.

JAK2 was immunoprecipitated from Leishmania-infected or uninfected MØs. In-gel PTP activity assay of the proteins that immunoprecipitated with JAK2 shows the presence of multiple PTPs (Fig. 4A), two of which were identified as PTP1B and SHP-1 on the basis of their molecular masses when compared to PTP profiles of SHP-1−/− and PTP1B−/− MØs, and detection of the proteins by Western blot in JAK2 immunoprecipitates (Fig. 4B). In addition to SHP-1 and PTP1B, another active PTP (~120 kD) coimmunoprecipitated with JAK2 in uninfected MØs or MØs infected with L. major GP63−/−, suggesting that JAK signaling may also be regulated by other phosphatases in addition to those previously described. Furthermore, after L. major infection, the profile of PTPs associated with JAK2 was consistent with PTP cleavage; cleavage fragments of PTP1B were associated with JAK2 in L. major–infected cells (Fig. 4B), and the activity of the PTP at 120 kD was completely lost in L. major–infected cells, an event that did not occur in cells infected with L. major GP63−/− (Fig. 4A). Conversely, the SHP-1 that coimmunoprecipitated with JAK2 corresponded to the uncleaved fraction in L. major–infected cells (Fig. 4, A and B).

Fig. 4

Leishmania infection alters the association of PTPs with JAK2. B10R MØs were infected for 1 hour with L. major, L. major GP63−/−, or L. major GP63R. (A) JAK2 was immunoprecipitated and immune complexes were run in an in-gel PTP activity assay (left). Simultaneously, total protein lysates of MØs from SHP-1+/+, SHP-1−/−, PTP1B−/−, and PTP1B+/+ mice were run in an in-gel PTP activity assay to allow identification of corresponding SHP-1 and PTP1B bands (right). (B) Proteins immunoprecipitated with JAK2 were analyzed by Western blot to identify coimmunoprecipitated PTPs. Total protein lysates from infected and uninfected samples are shown as PTP cleavage controls.

Host lipid raft integrity is important for efficient PTP cleavage by GP63

Western blot analysis of Leishmania-infected or uninfected MØ crude membrane and cytosolic fractions showed that Leishmania GP63 is found both associated with membrane fractions and as a soluble GP63 pool (Fig. 5A). To test whether posttranslational modifications of GP63, such as the glycosylphosphatidylinositol (GPI) anchor, were important for GP63 internalization and PTP proteolytic cleavage, we incubated MØs with rGP63 (1 μg/ml), which lacks the GPI anchor or other posttranslational modifications and is devoid of any parasite-derived or host cell-derived interacting proteins. Western blot (Fig. 5B) and confocal microscopy (Fig. 5C) showed that rGP63 was internalized into MØs, albeit much less efficiently than the uptake of GP63 occurring with parasite infection or incubation with parasite culture supernatant. Confocal microscopy of L. major–infected MØs showed that after 1 hour of Leishmania infection, GP63 is present within the host cell and localizes to the cytosolic face of the plasma membrane and is present in puncta throughout the cytoplasm (Fig. 5C).

Fig. 5

MØs internalize GP63. (A) MØs were infected for 1 hour, cells were lysed, protein extraction was performed, and proteins were separated into crude membrane and cytosolic fractions. Membrane pellets were washed five times with lysis buffer and solubilized in SDS-PAGE sample loading buffer. Fractions were run on gel and blotted for Leishmania GP63. GP63 run as a band at the expected mass of 63 kD (arrowhead) and as a lower 50-kD band (*), likely representing a proteolytic fragment as previously reported (56). (B and C) Lipid rafts contribute to GP63 internalization and access to PTPs. B10R MØs were incubated with 1 μg of rGP63 or were infected with L. major for 1 to 3 hours. Intracellular GP63 was evaluated by Western blot (B) and confocal microscopy (C). Arrowheads indicate GP63 protein (B) and Leishmania parasites (C). rGP63§ is 60 ng of rGP63 loaded directly onto the gel to observe the migration pattern of non–GPI-GP63. Actin was used as a loading control (B). (D) PTP cleavage by rGP63 is limited. Proteolytic cleavage of PTP1B, TCPTP, and SHP-1 was assessed by Western blot after incubation of MØs with 1 μg rGP63 or infection with L. major.

Consistent with the limited uptake of non–GPI-linked rGP63, cleavage of PTP1B or SHP-1 was only marginally detected after 3 hours of MØ incubation with rGP63 (Fig. 5D). Conversely, TCPTP cleavage was efficiently observed after 30 min of incubation (Fig. 5D), which may reflect its nature as an exceptionally avid substrate of GP63, compared to SHP-1 and PTP1B (fig. S5).

In Leishmania parasites, GP63 is found in three different pools: The major population is a C-terminal GPI-anchored protein associated with the cell surface, and smaller pools are present inside the parasite and are released from the parasite (17, 18, 27). Lipid rafts, dynamic membrane domains enriched in cholesterol, sphingolipids, and GPI-anchored proteins, mediate the internalization of cargo proteins and lipids from the cell surface of mammalian cells and act as platforms of cell signaling molecules (28). To test whether host lipid raft domains were required for GP63 entry into the host cell and PTP cleavage, we pretreated MØs with a subcytotoxic dose (fig. S6) of the cholesterol-chelating compound methyl-β-cyclodextrin (MβCD) and performed a time course Leishmania infection. After MβCD-mediated lipid raft disruption, infection-induced PTP1B and SHP-1 cleavage were markedly reduced (Fig. 6A). Similar to cells treated with rGP63, which is poorly internalized, the first cleavage event of TCPTP occurred even in cells exposed to MβCD, but the generation of a second cleavage product of TCPTP was prevented by lipid raft disruption.

Fig. 6

MØ lipid raft integrity is important for GP63 internalization and PTP cleavage. (A) Western blotting shows that PTP cleavage is limited in MØs with MβCD-disrupted lipid rafts. Arrowheads indicate full-length PTPs and cleavage fragments. (B) Confocal microscopy shows that GP63 (green) partially colocalizes with the lipid raft marker CTxB (red). DNA was stained with DAPI (blue). Arrowheads represent areas of GP63-CTxB colocalization; the asterisk indicates Leishmania parasites. In both (A) and (B), cells were pretreated for 1 hour with 20 mM MβCD in serum-free medium and subsequently infected. Noninternalized parasites were removed by washing before sample collection. DIC, differential interference contrast microscopy.

Confocal microscopy revealed that GP63 was located in punctuated structures along the cell periphery in infected MØs, with partial colocalization with the lipid raft marker cholera toxin B (CTxB) (Fig. 6B and fig. S7); whereas in cells pretreated with MβCD GP63 was undetectable within MØs, but remained detectable in the Leishmania parasite. MØs exposed to L. major in the absence of MβCD would be expected to have only a fraction of the GP63 associated with the MØ surface because some would already have been internalized. Although many of the CTxB-positive sites were also positive for GP63, there was also some GP63 at the MØ surface that did not colocalize with the lipid raft marker.

PTP cleavage fragments are enzymatically active

In-gel PTP activity assay of immunoprecipitated PTP1B, SHP-1, and TCPTP from naïve and Leishmania-infected MØs showed that PTP cleavage fragments are enzymatically active (Fig. 7A). After L. major infection, cleavage-induced PTP activation—measured by hydrolysis of the substrate p-nitrophenyl phosphate (pNPP) by protein lysates—was increased to 40% above the basal activity. Induction of PTP activity was markedly reduced when the MØs were infected with L. major GP63−/−, and cells infected with L. major GP63R showed similar PTP activity to that of cells infected with wild-type L. major (Fig. 7B). Furthermore, Western blotting for phosphorylated tyrosine residues showed a strong dephosphorylation of MØ proteins when the cells were infected with L. major wild-type or GP63R strain (Fig. 7C). However, when cells were infected with L. major GP63−/−, PTP activation and its consequent substrate dephosphorylation were abrogated.

Fig. 7

PTP cleavage products are enzymatically active. (A) MØs were infected for 1 hour. PTP1B, TCPTP, and SHP-1 were immunoprecipitated, and immune complexes were run in an in-gel PTP activity assay. Black and white arrows represent full-length PTPs and cleavage fragments, respectively. Asterisks mark coimmunoprecipitated PTPs; U.C., unspecific control/isotype control. (B) Total phosphatase activity was evaluated by hydrolysis of pNPP by cell lysates. Optical density readings were taken at 405 nm. Data are presented as percentage values ± SEM of three independent experiments performed in duplicate. (C) Total tyrosine phosphorylation of infected or uninfected MØs was evaluated by Western blot with the 4G10 antibody against phosphotyrosine. Loading control was performed with an antibody against β-actin. Arrows indicate modulated phosphorylated proteins. (D and E) Importance of SHP-1 in MØ responses. Top panels: (D) SHP-1+/+ and (E) SHP-1−/− MØs were primed for 1 hour with IFN-γ (250 and 50 U/ml), respectively, followed by 1 hour of Leishmania infection. JAK2 phosphorylation was evaluated by Western blotting. Bottom panels: SHP-1+/+ and SHP-1−/− MØs were primed for 1 hour with IFN-γ [125 (D) or 25 to 100 U/ml (E)], respectively, and subsequently infected for 24 hours. NO production was quantified by nitrite accumulation in the supernatant of cultured cells using Griess reaction (2). Data are representative of three individual experiments performed in duplicate ± SEM. Statistically significant differences (*) were considered when P < 0.05.

Using MØs stimulated with IFN-γ for 1 hour before infection (IFN-γ–primed MØs), we studied the impact of GP63-dependent PTP activation on the tyrosine phosphorylation of JAK2 and its functional outcome. JAK2 was dephosphorylated upon infection with GP63-expressing Leishmania strains, whereas JAK2 hyperphosphorylation was evident in cells infected with L. major GP63−/− (Fig. 7D). In contrast to the “silent response” of naïve Leishmania-infected MØs, characterized by the lack of a strong pro-inflammatory response and down-regulation of MØ functions, IFN-γ–primed and infected cells induce NO production (29). L. major GP63−/− infection induced significantly higher NO production compared to IFN-γ–primed MØs infected with L. major or L. major GP63R (Fig. 7D, lower panel). Despite cleavage of PTP1B (fig. S8), dephosphorylation of JAK2 and modulation of NO production (Fig. 7E) were not observed in IFN-γ–primed and infected SHP-1−/− MØs, suggesting that SHP-1 plays a more important role than does PTP1B in the modulation of JAK2 signaling in the context of Leishmania infection.

PTP1B is important during early disease progression

Among the functions attributed to PTP1B are the down-regulation of both IFN-γ signaling (25) and MØ activation (9). Having identified PTP1B as a target of Leishmania parasites, we evaluated its role in the in vivo progression of Leishmania infection. L. major–infected PTP1B−/− mice showed a delay in the onset and progression of footpad inflammation (Fig. 8A) and reduced parasite burden in the early stages of disease (Fig. 8B) compared to wild-type mice. Significant differences were observed within the first 5 weeks after infection, suggesting that similarly to SHP-1 (5), PTP1B is important for disease progression during the early, but not late, stages of infection. L. major also induces inflammatory cell recruitment in a murine air pouch model (30). Correlating with delayed pathology progression and reduced parasite burden, basal and L. major–induced leukocyte recruitment in PTP1B−/− mice was, respectively, five- and threefold higher than in wild-type mice (Fig. 8C). Consistently, bone marrow–derived MØ (BMMØ) from PTP1B−/− mice exhibited higher potential for activation as seen by an increased production of NO in response to LPS (Fig. 8D) and, consequently, a stronger capacity to control the intracellular parasite (Fig. 8E).

Fig. 8

PTP1B is important for early disease progression. (A) PTP1B−/− and littermate wild-type (WT) BALB/c control mice were infected by injection into the right hind footpad with 5 × 106 stationary-phase Lm-LUC promastigotes (L. major engineered to express luciferase). Lesion progression was evaluated by measurement of footpad swelling. Measurements are expressed as the footpad thickness difference (Δ) between the infected footpad and the uninfected control. Results are presented as the mean ± SEM (n = 7). (B) Parasite burden was measured by luciferase activity in the footpads 4 or 9 weeks after infection. Results are presented as the mean ± SEM (n = 5). (C) In vivo leukocyte recruitment was evaluated by injection of endotoxin-free PBS or L. major promastigotes into the air pouch of PTP1B−/− and WT mice. Total leukocytes were enumerated by microscopic count. (D) BMMØs were stimulated overnight with LPS (100 ng/ml) or left untreated. NO production was quantified by nitrite accumulation in the supernatant of cultured cells. (E) BMMØs were infected for 6 hours with stationary L. donovani–LUC promastigotes (10:1 Ld-MØ ratio). Cells were washed to remove noninternalized parasites. LPS (20 ng/ml) was added and infection was allowed to progress for 24 hours. Intracellular parasite survival was determined by luciferase activity after 24 hours of infection. Statistically significant differences were considered when P < 0.01 (**) and 0.05 (*).

Collectively, these data imply a key role for PTP1B during the initial stages of Leishmania infection, suggesting that Leishmania exploits this host PTP in concert with SHP-1, promoting maximal down-regulation of host cell functions for the successful establishment of infection.

Discussion

The balance of protein tyrosine phosphorylation, maintained by the concerted action of protein tyrosine kinases (PTKs) and PTPs, modulates signaling pathways fundamental to determining the outcome of multiple cellular functions. Tight regulation of PTP and PTK activities is crucial for proper signal transduction. We show that the Leishmania surface protease GP63, a key virulence factor important for lesion development (14) and initial interaction with its host cell (31), is internalized by MØs where it interacts with and activates the PTPs SHP-1, PTP1B, and TCPTP.

Rapidly upon Leishmania-MØ interaction, SHP-1 activation leads to the down-regulation of JAK/STAT and MAPK signaling and downstream inhibition of MØ functions, including NO production (2, 4, 5, 15, 24). To date, the exact mechanism whereby Leishmania parasites induce PTP activation is not completely understood. Recent reports propose that the Leishmania elongation factor 1 α (EF1α) and fructose-1,6-bisphosphate aldolase are SHP-1 activators (32, 33). However, the mechanism by which these proteins interact with and activate host PTPs remains elusive.

Previous studies, in addition to our current findings, have reported the GP63-dependent cleavage of intracellular MØ proteins, including the myristoylated alanine-rich C kinase substrate (MARKS), MARKS-related protein (MRP) (34), and the nuclear factor κB (NF-κB) p65RelA subunit (35). Herein, we report the internalization of GP63 into host MØs, providing mechanistic evidence for direct GP63–host protein interactions. Partial colocalization of the lipid raft marker CTxB and GP63 in Leishmania-infected MØs (Fig. 6B), together with decreased PTP1B and TCPTP and abrogation of SHP-1 cleavage after MβCD-mediated membrane cholesterol depletion, suggests that lipid rafts are important for mediating access of GP63 to the PTPs. Although we found that many lipid raft foci colocalized with GP63, there was also a proportion of GP63 that was at the MØ surface, but that did not colocalize with lipid rafts (fig. S7); thus a lipid raft–independent mechanism may also contribute to GP63 internalization. We hypothesize that the efficient internalization and intracellular localization of GP63 within MØs may be in part mediated by its GPI anchor. Similarly, the Trypanosoma brucei variant surface glycoprotein (VSG)—an abundant GPI-anchored protein that is important for parasite evasion of the immune response—is released and then internalized by MØs (36). VSG internalization is mediated by a component of its GPI core, and internalization is required for the activation of MØ NF-κB signal transduction (36). Although lipid rafts and the GPI anchor appear to mediate efficient GP63 internalization and access to the PTPs, GP63 lacking a GPI anchor can also be internalized (Fig. 5, B and C) to a limited extent, further suggesting that a lipid raft–independent mechanism may contribute to the delivery of GP63 into the MØ intracellular milieu. This observation, in addition to the nature of TCPTP as an exceptionally fit substrate of GP63 (fig. S5), may explain why lipid raft disruption with MβCD largely abrogates SHP-1 and PTP1B cleavage, but not cleavage of TCPTP. In vitro, cleavage of SHP-1 and PTP1B requires at least five times more GP63 compared to the amount needed for TCPTP cleavage; completion of the cleavage reaction for TCPTP, PTP1B, and SHP-1 occurs at a GP63 concentration of 0.5, 2.5, and 5 ng/μg, respectively (fig. S5). Thus, it is not surprising to see that lipid raft disruption completely prevents SHP-1 cleavage, partially prevents PTP1B proteolysis, but does not affect TCPTP cleavage, because lipid raft–independent GP63 internalization may deliver the required amount of protease to cleave TCPTP and, to a much lesser extent, PTP1B.

Caspase-dependent cleavage of PTP-PEST (37) and calpain-mediated cleavage of PTP1B and SHP-1 (38, 39) have been associated with regulation of PTP activity. PTP cleavage and activation is likely to be important in the initial interaction and “silent entry” of Leishmania promastigotes into its host cell. This is largely based on the rapid appearance of PTP cleavage fragments after Leishmania-MØ contact and the partial restoration of the PTP activity profile at 24 hours after infection (fig. S9), by which time GP63 abundance has dramatically decreased due to promastigote-to-amastigote differentiation. The importance of PTP cleavage in the initial stages of infection is further supported by the enhanced MØ phosphotyrosine dephosphorylation, down-regulation of JAK/STAT signaling due to JAK2 dephosphorylation, and inhibition of NO production (Fig. 7, B to D). Interestingly, the kinetic differences of PTP cleavage between L. major, L. mexicana, and L. donovani infection and the inability of the lizard pathogen L. tarentolae to modulate PTP activity correlate with the abundance of GP63 in each individual species (L. mexicana > L. donovani and L. major) and the absence of an active GP63 surface protease in L. tarentolae (40).

In vitro cleavage assays are consistent with GP63 cleavage sites identified by sequence analysis. Putative cleavage site 2 of TCPTP falls within the α1 helix of the PTP domain, which is important for substrate specificity (41). Cleavage at site 2 would not disrupt the PTP domain motif 1 (residues 42 through 48) and would maintain the integrity of the active site and phosphorylated tyrosine (p-Tyr) recognition loop, which is consistent with the PTP enzymatic activity of the cleavage product (Fig. 7A). Interestingly, the molecular region between amino acid residues 349 and 381 displayed at least seven putative GP63 cleavage sites, in agreement with previous observations that identified a proteolytic sensitive area between residues 353 and 387 of its C-terminal tail (42).

The C-terminal tail of SHP-1 has been associated with the regulation of enzymatic activity. Protein kinase C–mediated phosphorylation of Ser591 inhibits SHP-1 (43). In T cells, a portion of SHP-1 associates with lipid rafts through its C-terminal tail where it is thought to interact with some of its substrates, including the tyrosine kinases Lck and ZAP70 (44). If SHP-1 is cleaved at identified putative cleavage site 9 by GP63, this region responsible for lipid raft association would be removed, which, in addition to enhancing its enzymatic activity by removal of the inhibitory phosphorylation site, would potentially allow for interaction with alternate substrates by modifying its cellular localization. In the in-gel PTP activity assay of immunoprecipitated SHP-1 (Fig. 7A), one additional PTP band was coimmunoprecipitated from Leishmania-infected MØs, supporting interaction with alternate proteins. In-depth studies on this line are the subject of our current research.

Sequence and structural analyses of SHP-1 localize putative cleavage sites 2 and 3 at accessible regions in the N- and C-terminal Src homology 2 (SH2) domains, respectively. SHP-1 SH2 domains are essential for regulation of activity and substrate specificity (44). Molecular mass prediction of cleavage at site 2 generates fragments of 4 and 63 kD, which closely correspond to those observed experimentally. Site 2 is located within the N-terminal SH2-domain (N-SH2), which interacts with the catalytic site of the enzyme forming a “closed” conformation and impeding enzymatic activity. After interaction of the N-SH2 with a tyrosine phosphopeptide, conformational changes result in the dissociation of the N-SH2 from the catalytic cleft, allowing enzymatic activity (45). Thus, GP63-mediated cleavage at site 2 would be expected to promote PTP activation. Moreover, cleavage at this site should result in the inability to coimmunoprecipitate JAK2 and the SHP-1 cleavage fragment as observed in Fig. 4B, because additional sites for protein-protein interaction (N-SH2) would be unavailable, which would rapidly liberate SHP-1 from the protein complex upon dephosphorylation of JAK2. These data suggest that putative cleavage site 2 is a likely candidate for GP63-mediated proteolysis.

In PTP1B, site-directed mutagenesis and molecular weight predictions narrow possible cleavage sites to the C-terminal site 6 (table S1 and fig. S4). Regulation of PTP1B function by its C-terminal domain has been proposed because it harbors the endoplasmic reticulum localization signal and multiple sites for serine and threonine phosphorylation (46). Calpain-mediated cleavage within the C-terminal tail has been associated with enhanced enzymatic activation (38), thus providing support that GP63-mediated cleavage may also stimulate the activity of this PTP.

A major finding of our investigation was the Leishmania-dependent modulation of PTP1B and its role in disease progression in vivo. Previous reports have shown that in response to LPS PTP1B−/−, mice exhibit higher concentrations of serum IL-12 and IFN-γ and enhanced NO production by MØs (9). Together with NO, IL-12 and IFN-γ play a central role in the control of leishmaniasis by favoring a TH1 immune response, which results in protection of the host (47). Not surprisingly, the strategic down-regulation of IL-12 and NO by Leishmania parasites allows for its development within the mammalian host. We show that the absence of PTP1B delays footpad thickening, correlating with reduced parasite burden and greater leukocyte recruitment. Additionally, PTP1B−/− BMMØ presented an enhanced Leishmania killing capacity, paralleled by increased NO production (Fig. 8). The in vivo data suggest that the GP63-dependent modulation of PTP1B plays an important role during the early, but not late, stages of the disease. These observations are very similar to what has been previously shown for SHP-1−/− mice (5): Delay in footpad lesion development and progression, decreased parasite burden during the initial stages of the disease, and increased leukocyte recruitment at basal level and during infection are also features of L. major infection in the SHP-1–deficient mouse model. Interestingly, deletion of either SHP-1 or PTP1B alone only affects the early stages of the disease, suggesting that both PTPs may cooperate toward the maintenance and complete down-regulation of host pro-inflammatory functions. Modulation of PTP activity may promote a permissive environment for the initial establishment and development of Leishmania parasites within the mammalian host. The activation and concerted action of PTP1B, SHP-1, and presumably TCPTP during the early phase of murine cutaneous leishmaniasis result in the parasite-induced down-regulation of signaling pathways, ultimately leading to disease progression. Thus, MØ PTP1B and SHP-1 are targets that could abrogate disease development. Future investigations with mice containing specific targeted deletion of these PTPs in MØs should clarify the therapeutic potential of PTP1B and SHP-1 in treating leishmaniasis.

Collectively, our data show that, in addition to SHP-1, TCPTP and PTP1B are activated upon Leishmania infection by GP63-dependent cleavage, and we identify PTP1B as necessary for the initial stages of disease development. We also provide evidence that a Leishmania surface molecule can access the MØ intracellular “milieu” through a process mediated in part by host lipid rafts. Thus, GP63 functions as a key Leishmania virulence factor, allowing its successful development as an obligate intracellular parasite.

Materials and Methods

Cell culture and in vitro infections

B10R murine MØ cell line was cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Gibco-BRL) supplemented with 10% heat-inactivated fetal bovine serum (FBS), streptomycin (100 μg/ml), penicillin (100 U/ml), and 2 mM l-glutamine at 37°C and 5% CO2. L. mexicana, L. donovani 2211, L. major strain stably transfected with the luciferase reporter gene (Lm-LUC) (48), L. major, L. major GP63 knockout (GP63−/−), and L. major GP63−/− recomplemented (L. major GP63R) (14) promastigotes were kept in SDM-79 medium supplemented with 10% heat-inactivated FBS and hemin (5 mg/ml) at 25°C (49). MØs were infected with stationary-phase promastigotes in a 20:1 Leishmania-MØ ratio for various time periods. Unattached and noninternalized parasites were removed by washing the plates with phosphate-buffered saline (PBS). Parasite culture supernatant was prepared by two consecutive rounds of high-speed centrifugation of stationary-phase Leishmania promastigote cultures. Supernatants were collected and immediately used for subsequent experiments.

In vivo infections and BMMØ differentiation

Ptpn1−/− mice (PTP1B−/−) (8) and wild-type BALB/c littermates were kept in pathogen-free housing. All animal work was carried out according to the regulations of the Canadian Council of Animal Care and approved by the McGill Animal Care Committee. 5 × 106 stationary-phase Lm-LUC parasites were injected into the right hind footpad of 6- to 8-week-old mice and footpad swelling was evaluated weekly for up to 10 weeks after infection. Parasite burden was evaluated by luciferase activity in infected footpads (48). BMMØs were differentiated from bone marrows of 6- to 8-week-old uninfected wild-type and PTP1B−/− mice. Bone marrows were cultured for 5 days in DMEM containing 30% L929-cell conditioning medium (LCCM) and 10% FBS and an additional 2 days with refreshed LCCM.

Air pouch and leukocyte migration

Air pouches were raised on the dorsum of wild-type and PTP1B−/− mice by subcutaneous. injection of 3 ml of sterile air on day 0 and 2.5 ml on day 3 as previously described (50). At day 7, 1 ml of endotoxin-free PBS with or without 5 × 106 L. major stationary-phase promastigotes was injected into the pouch. Six hours after inoculation, pouches were washed with 5 ml of PBS to collect leukocytes recruited into the exudate. Cells were enumerated by direct count with a hematocytometer.

Phosphatase assays

As previously described (24), infected and noninfected MØs were collected, lysed in PTP lysis buffer [50 mM tris (pH 7.0), 0.1 mM EDTA, 0.1 mM EGTA, 0.1% 2-mercaptoethanol (2-ME), 1% Igepal, aprotinin (25 μg/ml), and leupeptin (25 μg/ml)] and kept on ice for 45 min. Lysates were cleared by centrifugation and protein content was determined by Bradford’s method. Protein extract (10 μg) was incubated in phosphatase reaction buffer [50 mM Hepes pH 7.5, 0.1% β-ME, 10 mM 4-nitrophenylphosphate disodium salt hexahydrate -pNPP-) for 30 minutes. OD was read at 405 nm.

In-gel PTP assay

In-gel PTP assay was performed as described (51, 52). Briefly, Poly(Glu,Tyr) substrate was tyrosine phosphorylated by overnight (O/N) incubation with GST-FER protein kinase (10 μg) and 150 μCi of [γ-32P]deoxyadenosine 5′-triphosphate. The substrate was then incorporated in a 10 to 12% SDS–polyacrylamide gel mixture at a concentration of 2 × 105 cpm/ml. MØ protein extracts, prepared as described above, were denatured for SDS–polyacrylamide gel electrophoresis (SDS-PAGE) and loaded onto the gel. After electrophoresis, the gel was incubated O/N in buffer A [50 mM tris-HCl (pH 8.0), 20% isopropanol], washed twice with buffer B [50 mM tris-HCl (pH 8.0), 0.3% β-ME] followed by full protein denaturation in buffer B containing 6 M guanidine hydrochloride and 1 mM EDTA. Gels were washed twice in buffer C [50 mM tris-HCl (pH 8.0), 1 mM EDTA, 0.3% β-ME and 0.04% Tween 20] and final renaturation O/N in buffer C. Gels were dried and exposed to X-ray film. Active PTPs were detected as clear bands on the film.

Immunoprecipitation and Western blotting

Primary antibodies used were α-SHP-1 MAB1128 and α-PTP1B (Millipore), α-SHP-2 (Santa Cruz Biotechnology), α-PTEN, α-JAK-2 (Cell Signaling), α-PP2A (06-222, Upstate), α-TCPTP 3E2 [obtained as described (53)], and monoclonal antibody clone #253 against GP63 (54). Protein immunoprecipitation was performed by preclearing 1 mg of protein extract from Leishmania-infected or uninfected MØ protein lysates with 1 μg of rabbit antibody against rat and protein A/G beads for 1 hour at 4°C. Precleared lysates were then incubated O/N with 3 μg of antibody against SHP-1, PTP1B, TCPTP, or JAK2 and 2 hours with protein A/G beads. Immunoprecipitates were washed with protein lysis buffer and denatured in SDS-PAGE sample loading buffer. Western blots were performed as previously described (2).

GST pull-down and in vitro cleavage assay

GST-tagged full-length TCPTP and SHP-1 constructs were expressed in Escherichia coli BL21. Fusion proteins were isolated by 1-hour incubation of bacterial lysates and glutathione-Sepharose beads (GE Healthcare) at 4°C. Fusion proteins were incubated with 3 × 107 L. major, L. major GP63−/−, L. major GP63R, or with 1 μg recombinant GP63 (rGP63) (23) for 1 hour at room temperature in a final volume of 200 μl. GST-tagged proteins were reprecipitated by centrifugation and denatured in SDS sample loading buffer. PTP cleavage was evaluated by Western blot.

Confocal microscopy

Cells were plated on glass coverslips and infected for 1 hour. MØs were washed with ice-cold PBS, fixed with 4% formaldehyde at 4°C, and permeabilized for 5 min in PBS containing 1% bovine serum albumin and 0.05% NP-40. After blocking in 5% nonfat evaporated milk in PBS, the coverslips were incubated with α-GP63 mouse monoclonal antibody clone # 96 (55). Slides were washed with PBS and incubated with Alexa Fluor 488 α-mouse antibody (Molecular Probes). CTxB (10 μg/ml)–Alexa 594 (Molecular Probes) was used to identify lipid raft domains. DAPI (4,6′-diamidino-2-phenylindole) (Molecular Probes) was used to stain DNA (nuclei). After mounting, cells were visualized by confocal microscopy with a Zeiss LSM 510 system.

Statistical analysis

Data were analyzed by one-way ANOVA (analysis of variance). Statistically significant difference between groups was considered when P < 0.05 or P < 0.01. All data are presented as the mean ± SEM.

Acknowledgments

This study was supported by a Canadian Institute of Health Research (CIHR) operating grant to M.O. M.O. is member of a CIHR group on Host-Pathogen Interaction and the FQRNT (Fonds Québécois de la Recherche sur la Nature et les Technologies) Centre for Host-Parasite Interaction. M.O. is a CIHR Investigator and a Burrough Wellcome Fund Fellow in Molecular Parasitology. M.A.G. is the recipient of a studentship from the Research Institute of the McGill University Health Centre and the Faculty of Medicine at McGill University. We thank K. Burridge (University of North Carolina, Chapel Hill) for providing the GST-FER construct. We thank J. Laliberté, M. Stuible, and M. Jaramillo (McGill University) for technical assistance with confocal microscopy, PTP1B constructs, and critical reading of our manuscript, respectively.

Supplementary Materials

www.sciencesignaling.org/cgi/content/full/2/90/ra58/DC1

Methods

Fig. S1. Distribution of PTPs after Leishmania infection.

Fig. S2. Pattern of PTP activity upon infection with different Leishmania species.

Fig. S3. Putative cleavage sites in the PTPs.

Fig. S4. Targeted deletion of putative cleavage sites.

Fig. S5. Recombinant GP63 mediates PTP cleavage in cell lysates.

Fig. S6. MβCD cytotoxicity.

Fig. S7. Colocalization of GP63 and the lipid raft marker CTxB.

Fig. S8. Leishmania infection triggers PTP1B cleavage in SHP-1−/− MØs.

Fig. S9. Cleaved PTPs partially disappear after 24 hours of infection.

Table S1. Putative GP63 cleavage sites.

References

References and Notes

View Abstract

Stay Connected to Science Signaling

Navigate This Article