| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Section of Pharmacology (S.A., A.P., A.M., T.F.), Department of Oncology, Biology and Genetics University of Genova 16132 Genova, Italy; Institut National de la Santé et de la Recherche Médicale, Unité 531 (J.-P.E., C.S.), Institut Fédératif de Recherche 31, Institut Louis Bugnard, 31432 Toulouse Cedex 4, France; Department of Experimental and Clinical Medicine (R.I.), University "Magna Graecia" of Catanzaro, 88100 Catanzaro, Italy; and Department of Biology and Molecular and Cellular Pathology (A.F.), Consiglio Nazionale delle Ricerche Endocrinology and Experimental Oncology Center, University "Federico II" of Naples, 80131 Naples, Italy
Address all correspondence and requests for reprints to: Tullio Florio, M.D., Ph.D., Sezione Farmacologia, Dipartimento Oncologia, Biologia e Genetica, Università di Genova, Viale Benedetto XV, 2, 16132 Genova, Italy. E-mail: tullio.florio{at}unige.it.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
(PTP) is an important intracellular effector of the cytostatic action of SST. Here we characterize, in Chinese hamster ovary-k1 cells, the intracellular pathway that from somatostatin receptor 1 (SSTR1), leads to the activation of PTP and that involves, in a multimeric complex and sequential activation, the tyrosine kinases Janus kinase (JAK) 2 and Src, and the cytosolic phosphotyrosine phosphatase SHP-2. We show that inhibitors of JAK2 and Src and dominant-negative mutants of SHP-2 and Src abolished the SSTR1-mediated PTP activation, suggesting that all these effectors participate in the activation of PTP. In basal conditions, JAK2 forms a multimeric complex with SHP-2, Src and PTP. In response to SST, JAK2 is activated in a G protein-dependent manner, dissociates from and phosphorylates SHP-2, increasing its activity. Subsequently, SHP-2 dissociates from Src, dephosphorylates the Src inhibitory tyrosine-529, and causes an autocatalytical increase of the phosphorylation of Src tyrosine 418, located inside its kinase activation loop. Active Src, in turn, controls the activity of PTP, via a direct interaction and phosphorylation of the phosphatase. These data for the first time depict an intracellular pathway involving a precise sequence of interactions and cross-activation among tyrosine phosphatases and kinases acting upstream of PTP. In particular the sequential activation of JAK2, SHP-2, and Src conveys the molecular signaling from SSTR1 to the activation of this phosphatase that is responsible for the final biological effects of SST. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In the past years, different endogenous molecules, displaying antiproliferative effects, have been reported to regulate the activity of PTPs (1, 2, 3) and, among them, somatostatin (SST) was reported to induce many of its effects through the modulation of the activity of these enzymes (4, 5, 6, 7, 8).
SST is an endogenous regulator of proliferation in a great variety of cells, exerting its biological functions through the interaction with a family of G protein-coupled receptors, named SSTR1 to 5 (9). SSTRs, identified in both normal and tumor cells (10, 11), transduce the antiproliferative effects of SST acting indirectly, via endocrine and antiangiogenic mechanisms (12, 13, 14), or eliciting a direct antiproliferative activity (15).
The intracellular pathways regulated by the activation of SSTR have been deeply studied and, to date, PTPs are considered one of the main intracellular effectors responsible for the direct inhibition of cell growth induced by SST (6, 7, 16). For example, SST-activated PTPs were reported to dephosphorylate active epidermal growth factor receptor, inhibiting its proliferative effects (4). Molecular studies aimed to identify the SST-stimulated PTPs, demonstrated that these enzymes share many biochemical features with SHP-1 and -2, a subfamily of cytosolic PTPs containing a tandem of Src homology 2 (SH2) domains (8, 17, 18, 19). SHP-1 and SHP-2 have been involved in both protein-protein interaction, via their association with specific phosphotyrosine residues (20), and signal transduction of different growth factor or G protein-coupled receptors (21, 22, 23). It was reported that SHP-1 can dephosphorylate different signaling molecules, including growth factor receptors, promoting p27kip1 overexpression and cell cycle arrest (24, 25). Similarly, SHP-2 was shown to be associated with growth factor, cytokine, or immuno-receptors (26, 27) and with various signaling proteins (28, 29). Thus, in addition to their involvement in the cytostatic effects of SST, these PTPs play a more general role as adaptor or transducing molecules rather then specific effectors.
Interestingly, SHP-1 and SHP-2 may behave as either positive or negative regulators of intracellular signaling, depending on the specific cellular background and the substrates affected (30, 31, 32). Thus, it was proposed that other PTPs should participate in the antiproliferative activity of SST.
In previous studies, we and others demonstrated that SHP-1 and -2 activation by SST was very rapid and rapidly their activity returns to basal levels (22, 33). Conversely, a different, long-lasting, PTP activity is induced after SST treatment (6, 19).
In particular, the cytostatic effects of SST in PC Cl3 rat thyroid cells are mediated by the activation of a receptor-like PTP that, for its homology with the human density-enhanced phosphatase 1/human phosphotyrosine phosphatase (DEP-1/HPTP) gene, was named PTP (34). This ubiquitously expressed protein contains a unique intracellular catalytic domain, a short transmembrane domain and an extracellular region containing eight fibronectin type III-like repeats (35).
Many studies suggest that PTP is a negative regulator of cell growth (36, 37, 38, 39). In addition, DEP-1/PTP participates in the control of cell differentiation, being its expression increased in differentiated breast cancer cells (38) and induced by differentiating agents in normal thyroid cells (40). On the contrary, PTP expression is down-regulated after oncogene-dependent thyroid cell transformation, as well as in malignant human thyroid tumors (41, 42). Furthermore, we demonstrated that the stable transfection of PTP in transformed rat thyroid cells led to the recovery of a partially differentiated phenotype with a reduced proliferation rate, caused by the inhibition of the ERK1/2-dependent proteolysis of p27kip1 (34, 42).
ERK1/2 was identified as a major substrate of DEP-1/PTP, and MAPK inactivation was reported to be responsible for the inhibition of glioma cell proliferation induced by SST (43, 44).
An important evidence of the differential roles of PTP and SHP-2 in the antiproliferative effects of SST was the observation that the activation of SHP-2 occurred in both PTP expressing and nonexpressing transformed cells, but only when PTP was activated, SST was able to induce p27kip1-mediated cytostatic effects (34).
However, we reported that the activation of SSTR1, in stably transfected Chinese hamster ovary (CHO)-k1 cells (CHO-SSTR1), results in cell growth arrest via the activation of SHP-2 (19). Thus, we cannot exclude the involvement of other PTPs, such as SHP-1 or SHP-2, in the intracellular pathway leading to the activation of PTP by SST.
The aim of this work was to delve deeper in the intracellular pathway that from the activation of SSTR1 leads to the regulation of PTP activity, taking in account the possible participation of other PTPs, such as SHP-2, and the involvement of cytosolic tyrosine kinases, namely Src and JAK2.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
in CHO-SSTR1 Cells activation, we used a subclone of CHO-k1 cells stably transfected with SST receptor subtype 1 (CHO-SSTR1) (45).
We previously reported that the activation of SSTR1 in CHO-SSTR1 results in cell growth arrest via the activation of SHP-2 (19), whereas the expression of PTP was required to induce cell growth arrest in PC Cl3 and C6 cell lines (34, 43).
Thus, we initially confirmed in Western blot (WB) experiments that CHO-SSTR1 cells express both SHP-2 and PTP (data not shown). Conversely, SHP-1 is not expressed in these cells (22).
Figure 1
shows the powerful activation of PTP induced by SST treatment (1 µM), measured as para-nitrophenyl phosphate (pNPP) hydrolysis in an immunocomplex assay, as reported (22). This effect was clearly evident after 10 min of treatment, reached a maximum after 15–30 min and was still statistically significant after 60 min. This result indicates that PTP is an intracellular effector regulated by SST, in CHO-SSTR1 cells. However, using the surface plasmon resonance technique we demonstrated that the activation of this PTP, upon SSTR1 stimulation, was not mediated by a direct binding of PTP to specific sequences within the receptor (tyrosine-based inhibitory motif, ITIM) (data not shown), that were previously described to be involved in the SSTR2 activation of cytosolic PTPs such as SHP-2 (46). Thus, a more complex signal transduction may regulate the SSTR1-dependent activation of PTP.
|
Activity. Following published data on signaling pathways regulated by somatostatin receptors (46, 47), we evaluated the involvement of the kinases JAK2 and Src and the cytosolic PTP, SHP-2. JAK2, belonging to Janus kinase family, is a tyrosine kinase that is activated by a great variety of transmembrane receptors (48, 49), including several G protein-coupled receptors (50). JAK2 activation occurs after a rapid trans-phosphorylation of the activation loop of the two kinases that constitute the molecule itself (51). Activated JAK2 can phosphorylate multiple targets and, in particular, downstream signaling molecules containing SH2 domain, including SHP-2 (52) that is itself a recognized intracellular effector of SST (19, 22). Finally, we looked at the nonreceptor tyrosine kinase c-Src because related kinases have been demonstrated to be involved in G protein-coupled receptor signaling (53, 54) and, more importantly, we previously reported that this kinase was involved in SSTR1-induced cell growth inhibition (22).
Thus, using pharmacological inhibitors or dominant-negative mutants of these effectors, we evaluated their involvement in SST activation of PTP.
First, we tested the capability of SST to activate PTP in experimental conditions in which JAK2 activity is abolished by pretreating CHO-SSTR1 cells with the compound AG490.
PTP-specific activity was measured in an immunocomplex PTP assay after treating CHO-SSTR1 cells for 15 min with SST (1 µM) alone or after 10 min of pretreatment with AG490 (10 µM). The strong increase in PTP activity, caused by SST treatment, was totally abolished in the presence of the JAK2 inhibitor (Fig. 2A).
|
, we analyzed PTP activity in CHO-SSTR1 overexpressing SHP-2 wild-type (wt) in comparison to cells transfected with a dominant-negative mutant of SHP-2 [SHP-2(C/S)]. The obtained results are summarized in Fig. 2B
: PTP activity induced by SST treatment (1 µM) was completely abolished in cells transfected with SHP-2(C/S). In fact, although in CHO-SSTR1 expressing SHP-2 wt, SST increased basal PTP activity (+300%, vs. basal activity), no changes in phosphatase activity were observed after transfection of SHP-2(C/S).
Thus, as we showed for JAK2, also SHP-2 activation is required for SST-induced stimulation of PTP, in CHO-SSTR1 cells.
To assess the role of Src in SST intracellular signaling, we transiently transfected CHO-SSTR1 cells with cDNA encoding c-Src wt or mutants generating constitutively active (Src CA, kinase activating mutation) or dominant-negative (Src DN, dominant negative) isoforms. As shown in Fig. 2C, overexpression of Src wt caused per se an increase in PTP activity (+50%) compared with the basal activity measured in control CHO-SSTR1 cells (transfected with the empty vector) that was further increased after SST treatment (+150% vs. control cells). More importantly, transfection with Src CA cDNA caused a massive increment of PTP activity (3-fold/over basal), whereas blocking Src activity by expressing Src DN in CHO-SSTR1 cells did not modify basal PTP activity and completely abolished SST-dependent activation of the phosphatase. We confirmed these data by evaluating PTP-specific activity in CHO-SSTR1 treated with SST (1 µM) in the presence or absence of the Src inhibitor PP1 (200 nM, 10 min before SST treatment). PTP activity greatly increased after treatment with SST whereas, in the presence of PP1, the PTP activity was partially but significantly reduced (data not shown). These results suggest that Src activity is necessary for SST to induce PTP activity.
SST Directly Regulates JAK2, SHP-2, and Src Activities
To directly demonstrate that JAK2 is involved in the signal transduction pathway by which SSTR1 activates PTP, we looked at JAK2 phosphorylation/activation, as a consequence of SST treatment. To this aim, cells treated with SST (1 µM) for 30 and 60 sec were probed in WB experiments using a phosphospecific anti-JAK2 antibody that detects the phosphorylation at Tyr1007 and Tyr1008 (51). The obtained results clearly demonstrate a massive increase in JAK2 phosphorylation that was already statistically significant after 30 sec of treatment with SST and was still detectable after 1 min (Fig. 3A). These data indicate that a very early activation of JAK2 occurs after SSTR1 stimulation.
|
, B–I).
The phosphorylation of SHP-2 is a prerequisite for its activation in different cell systems (55). The time-course of SHP-2 phosphorylation was analyzed by immunoprecipitating, with an anti-SHP-2 antibody, CHO-SSTR1 lysates treated with SST (1 µM) for 15, 30, and 60 sec and then probing the proteins with an antiphosphotyrosine antibody. The WB reported in Fig. 3B-II indicates that SSTR1 activation caused SHP-2 tyrosine phosphorylation starting at 15 sec (although not statistically significant), reaching a maximum after 30 sec of SST treatment and lasting up to 60 sec.
Finally, we evaluated whether SST could directly modulate Src activity. For this purpose, we measured the changes in Src tyrosine phosphorylation subsequent to SST treatment.
Src contains two fundamental phosphorylation sites: tyrosines 418 and 529. Tyrosine 529 is located near the carboxyl terminus of the protein and, when phosphorylated by a specific kinase, C-terminal Src kinase, acts as a negative regulator, keeping Src inactive through an intramolecular interaction between the SH2 domain and the carboxyl terminus of the protein. This conformation blocks phosphorylation of the catalytic domain residue, tyrosine 418, thereby preventing Src activation. When tyrosine 529 is dephosphorylated, tyrosine 418 is maximally phosphorylated, leading to kinase full activation. Using two specific antibodies directed against Src Tyr418 and Tyr529, we were able to evaluate, by WB, the changes in Src phosphorylation at these sites, that represent an index of the kinase activation.
Figure 3C depicts the effects of SST on pSrcY418 (panel I) and pSrcY529 (panel II). SST (1 µM) caused the phosphorylation of the activating domain of Src (Tyr418) that was already visible at 30 sec of treatment, reached a maximum after 2 min and a concomitant dephosphorylation of the inhibitory domain of Src (Tyr529). After 5 min of treatment, the phosphorylation state of both tyrosines returned to basal level. The kinetics of the changes in the phosphorylation state of the two tyrosines clearly confirms that SST activates Src with a very rapid mechanism, which clearly precedes PTP activation.
In Fig. 3D is shown the comparison of the time-course for SST-induced activation of JAK2, SHP-2, Src, and PTP. PTP activation clearly represents the final event caused by SST treatment, whereas JAK2 activation is the earliest kinase to be induced, followed by the activation of SHP-2 and Src that displayed a very similar time-course.
SST Activation of JAK2, SHP-2, Src, and PTP Is Mediated by a Pertussis Toxin (PTX)-Sensitive G Protein
SSTR1 is coupled to a PTX-sensitive Gi/o protein and PTX treatment is known to uncouple G
i/o from this receptor, independently from which final effector was evaluated: cAMP (45, 56), PTPs (16, 33), and eNOS (57).
To demonstrate that the intracellular pathway associated to SSTR1 depends on the activation of a PTX-sensitive GTP-binding protein, CHO-SSTR1 cells were pretreated with PTX (180 ng/ml, 18 h), before treatment with SST (1 µM) and then JAK2 (Fig. 4A), SHP-2 (Fig. 4B), Src (Fig. 4C), and PTP (Fig. 4D) activations were evaluated. The blockade of the G protein resulted in a complete abolishment of SST-mediated SHP-2 phosphorylation/activation. Similarly, although PTX alone did not modify basal SHP-2 activity, it completely abolished SST-mediated SHP-2 activation. We confirmed this observation by evaluating Src dephosphorylation at Tyr529 after PTX treatment. Src dephosphorylation of Tyr529 induced by SST treatment was completely abolished by PTX inhibition of the G protein activity. Finally, as expected from the previous data, we show that also the increase in PTP activity, induced by SST, was completely prevented by blocking SSTR1-coupled G protein. The specificity of these effects was demonstrated by the observation that PTX was unable to revert PTP activation induced by overexpression of Src CA, that obviously does not require G protein activation (data not shown).
|
Activation, we used the same pharmacological and molecular tools, we used in the previous experiments. In light of the early kinetics of JAK2 activation after SST treatment, we first tried to revert SST signaling using the JAK2 inhibitor AG490.
The increase in JAK2 phosphorylation, subsequent to SST treatment, was completely prevented in the presence AG490 (10 µM, 10 min pretreatment) (Fig. 5A).
|
). Because JAK2 is thought to directly phosphorylate SHP-2 (52) and we showed that SST induced phosphorylation of SHP-2 to activate this phosphatase (Fig. 3B-II), we then examined SHP-2 phosphorylation, induced by SST treatment (1 µM, 30–60 sec), in the presence or absence of AG490. As shown in Fig. 5C, AG490 strongly reduced SHP-2 phosphorylation stimulated by SST. Altogether, these findings indicate that JAK2 activity is a prerequisite for SHP-2 activation, likely through a direct phosphorylation. The role of JAK2 in this signal transduction pathway was further demonstrated studying the effect of inhibition of JAK2 activity on Src activation. To this purpose, we measured the phosphorylation of SrcY418 after SST (1 µM, 30 and 60 sec) stimulation in the presence or absence of AG490. In the experimental conditions in which JAK2 activity was inhibited, SST was not able to induce a significant activation of Src (Fig. 5D). Altogether, these data demonstrate that JAK2 represents the earliest step in the intracellular pathway that controls PTP activity.
Thus, we evaluated whether Src was a substrate for SHP-2 or vice versa (46). To delve deeper into this issue, we modulated Src activity by transfecting CHO-SSTR1 cells with the cDNA of Src CA and Src DN and then measured SHP-2 activity in the immunocomplex PTP assay. In basal conditions, SHP-2 activity did not change in cells expressing Src CA or Src DN compared with control cells (Fig. 6A). Moreover, after treating the cells with SST (1 µM, 2 min), we observed an increase in SHP-2 activity that was not abolished when cells were transfected with Src DN (Fig. 6A).
|
), was completely abolished in the presence of the dominant-negative mutant of SHP-2 (Fig. 6B-II). Thus, in our experimental model, Src activation induced by SST requires a functional SHP-2. We propose that Src dephosphorylation/phosphorylation, as index of its activation, is tightly dependent on SHP-2 activation that likely causes a dephosphorylation of the inhibitory Tyr529 of Src, allowing its activation.
Thus, the intracellular signal transduction pathway, leading from SSTR1 stimulation to PTP activation, implies a SHP-2-mediated activation of Src.
A Multi-Effector Complex Transduces the SST Signal from SSTR1 to PTP
To determine whether all the components activated in this intracellular pathway (JAK2, SHP-2, Src, and PTP), are also physically linked, we examined whether these molecules coimmunoprecipitate using anti-PTP antibody. Interestingly, we found that in basal conditions both JAK2, SHP-2, and Src were coimmunoprecipitated using anti-PTP antibody, indicating that a large multiprotein complex is assembled nearby SSTR1 (Fig. 7A).
|
(intracellular portion) and Src (SH2 domain), we tested whether Src regulation of PTP activity involved a physical interaction between the two proteins. Increasing concentrations of recombinant PTP (30 nM–2 µM), obtained as glutathione-S-transferase (GST) fusion protein, were put on a proper plate coated with GST-Src(SH2). Then we analyzed the time-course of the resonance signal emitted in case of molecular interaction between the two proteins (as described in Materials and Methods). Figure 7B shows a representative sensogram of the binding obtained: the real-time association and dissociation kinetics between PTP and Src was concentration-dependent showing a high affinity (KD 408 nM). Conversely, performing surface plasmon resonance following the same protocol as described before and using the same chip covered with PTP, but collocating GST-SHP-1(SH2) or GST-SHP-2(SH2) on the plate, no resonance signal was measured (Fig. 7C), thus indicating the absence of association between PTP and SHP-1 or SHP-2 SH2 domains.
This result provides a direct evidence of the interaction between PTP and Src but excludes a physical link between PTP and SHP-2.
Because we show that Src directly interacts with and activates PTP, we verified whether this activation may involve a tyrosine phosphorylation of this PTP. Thus, we treated the cells with SST, immunoprecipitated the cell lysate with anti-PTP antibody, and performed WB using anti-phosphotyrosine antibody or vice versa. However, we did not identify any phosphorylation of the phosphatase (data not shown). However, we reasoned that the PTP phosphorylation state, due to the high basal PTP activity, could be very transient because, in these experimental approach, we cannot treat the cells with vanadate, as PTP inhibitor, because it would also abolish the activity of SHP-2 that, on the contrary, is required for SST effects. To bypass the requirement of SHP-2, we evaluated the possible phosphorylation of PTP by Src in cells transfected with Src CA, pretreated with vanadate (10 µM). In untreated Src CA-transfected cells, we performed a WB for PTP after immunoprecipitation with -phosphotyrosine antibody. As shown in Fig. 7D (upper panel), a PTP immunoreactive band was observed only in the transfected cells. As a control, we performed also a WB with -phosphotyrosine antibody. on cell lysates immunoprecipated with -PTP antibody (Fig. 7D, middle panel), that confirmed that a phosphotyrosine immunoreactive PTP band was present only in cells expressing active Src, Thus, using this approach we demonstrated a significant tyrosine phosphorylation of PTP, suggesting that this event may be related to the increased activity of this PTP in conditions of Src activation.
Moreover, we analyzed also the dynamics of the interaction between the other components of this transduction pathway: JAK2/SHP-2 and SHP2/Src.
To investigate whether SHP-2 and JAK2 are directly bound, as reported (47) and described also for SHP-2/JAK1 (63), we analyzed the interaction between these proteins by coimmunoprecipitation experiments. Lysates obtained from CHO-SSTR1 treated or not with SST (1 µM) for 30 sec, 1 or 2 min, were immunoprecipitated with anti SHP-2 antibody and then the immunoprecipitated proteins were immunoblotted with anti-JAK2 antibody. As shown in Fig. 8A, a JAK2/SHP-2 complex was identified in resting cells, whereas the two proteins time-dependently dissociated after SST treatment, starting after 30 sec and the reassociation of the proteins being observed after 2 min.
|
demonstrates that, in basal conditions, Src and SHP-2 coimmunoprecipitated, indicating that, in the absence of SSTR1 activation, they are associated in a multimeric complex. However, after SST treatment, the two proteins dissociated.
Evaluating these results together with the data we obtained about the association of JAK2 and SHP-2, and Src and PTP, we propose, in resting CHO-SSTR1 cells, the existence of a JAK2/SHP-2/Src/PTP complex that dissociates after single protein activation.
The Same Intracellular Pathway Is Responsible for PTP Activation in Cells Endogenously Expressing SSTR1
Finally, we verified whether the results described above can be reproduced also in cells that endogenously expressed SSTR1. We used C6 rat glioma cells that were previously shown to express SSTR1, 2, 3, and 5 (43). In these cells, we selectively activated SSTR1 using the specific agonist BIM23926 and evaluated the activation of the individual component of the pathway, as described for CHO-SSTR1 cells. The results we obtained demonstrated that the activation of SSTR1 in C6 cells (by BIM23926, 100 nM) caused an increase in the phosphorylation of JAK2 (Fig. 9A), an increase in SHP-2 activity (+85% over basal) (Fig. 9B), an activation of Src (evaluated by both dephosphorylation of Tyr 529 and phosphorylation of Tyr 418) (Fig. 9C) and an increase in PTP activity (+122%) (Fig. 9D). Moreover, these effects were reverted by the preincubation with the JAK2 inhibitor AG490, confirming that also the activation of endogenous SSTR1 induces the same sequence of effects we described in detail in CHO-SSTR1 cells (Fig. 9, A–C).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
in rodents, that, when activated by SST, can interact and directly dephosphorylate ERK1/2, causing an up-regulation of p27Kip1 (34, 43, 44). Importantly, we demonstrated that, in different cell models, including primary cultures of human glioblastoma cells, the expression and activation of PTP is a necessary prerequisite for the achievement of SST effects (34, 43). However, although it is now widely accepted that PTPs represent a novel enzymatic system regulated by G protein-coupled receptors and that their activity mediate antiproliferative signals, the intracellular pathways by which activated receptors may control PTP activity have not been completely identified.
In the present study, we delved deeper into the signal transduction cascade that from the stimulation of SSTR1 by its natural ligand, SST, ends, as final target, to the activation of PTP.
We previously reported that, in CHO-SSTR1 cells, SST antiproliferative effects involve the activation of SHP-2 showing a very rapid kinetics of activation (19, 22) (this paper).
Here, we extend those data showing that, in the same cells, PTP was strongly activated by SST, although displaying a regulation much slower than SHP-2. In fact, we report that, whereas the activation of SHP-2 is an extremely rapid process (starting after 30 sec of SST treatment), the activation of PTP was slower and long lasting (10–60 min). Moreover, we demonstrate that the inhibition of SHP-2 activity completely prevented SST-dependent regulation of PTP. These data suggest that the SST-dependent regulation of the two enzymes may be sequential, laying down on the same intracellular pathway.
In resting CHO-SSTR1 cells, a SST-regulated multieffector complex was identified, involving JAK2, SHP-2, Src and PTP. In fact, in unstimulated cells, a coimmunoprecipitation of PTP with JAK2, SHP-2, and Src was detected. Moreover, also JAK2/SHP-2 and SHP-2/Src complexes were observed. In recent studies, a similar complex was observed in pancreatic cells in which the association of JAK2 and SHP-1 was reported (47). Interestingly, in the latter report the SST receptor subtype SSTR2 was also identified as component of the complex. In another study using CHO cells that overexpress SSTR2, this receptor was associated in resting conditions with SHP-2 and Src, and only after a ligand-dependent activation of SSTR2, SHP-1 was recruited to the complex via the increased activity of Src and SHP-2 (46). In those studies, it was reported that also SSTR2 belongs to this large multieffector complex. Although we do not have data supporting the participation of SSTR1 to a unique complex with JAK2, SHP-2 and Src, the occurrence of such condition is verisimilar. Indeed, it was demonstrated that the binding of SHP-2 to SSTR2 occurs via a phosphorylated tyrosine belonging to a specific ITIM consensus sequence (I\/V\/L\/S-x-Y-x-x-L\/V\/I) (46) and similar functional motifs are also present in the third intracellular loop (L-C-Y243-V-L-I) and in the C-tail (I-L-Y323-G-F-L) of SSTR1. Importantly, we also show that, among the proteins identified in the complex, PTP directly interacts with Src but not with SHP-2 (as demonstrated by surface plasmon resonance analysis), suggesting a precise sequence of interaction among the different partners of this transduction pathway.
Thus, our data, in line with these previous works, support the notion that beside classical signal transduction effectors, SSTR may interact with a large number of effector molecules, including tyrosine kinases and PTPs.
In our study, after SST binding to SSTR1, JAK2, SHP-2 and Src are sequentially activated, a process that, similarly to previous reports (47), requires the dissociation of the active component of the complex. Importantly, in the novel pathway we describe, the first player after the activation of the receptor was a PTX-sensitive G protein (Gi/o family) that was able to block the subsequent activation of all the kinases and phosphatases. In the multieffector complex, JAK2 is the first element to be activated. In fact, we demonstrate that, after SST treatment JAK2 phosphorylation/activation occurred in 30 sec and the specific blockade of its kinase activity, using the compound AG490, caused the arrest of all the signaling cascade (inhibition of SHP-2, Src, and PTP activation). Thus, the G protein-dependent activation of JAK2 represents, in our cell model, the initial signal required to activate PTP. Recent reports outlined that JAK2 represents a component of the intracellular pathway downstream several G protein-coupled receptors (64, 65). Interestingly, although a number of reports have proposed JAK2 as a positive regulator of cell proliferation (51), more recent studies have proposed that, when the activation of this kinase involves inhibitory receptors, JAK2 may also play an antimitogenic activity, as was reported for SSTR2 activation (47) and, in this paper, for SSTR1. SHP-1 and -2 represent important substrates for the kinase activity of JAK2 (52, 64). We show in CHO-SSTR1 cells that, upon SST treatment, the activation of JAK2 causes a tyrosine phosphorylation of SHP-2 and subsequent dissociation of the two molecules. It was previously reported that tyrosine phosphorylation represents an important mechanism of SHP-2 activation (55), and we show, indeed, that SST caused a JAK2-dependent increase in SHP-2 phosphorylation and activity. Moreover, it was reported that also the simple interaction of SHP-2 with tyrosine phosphorylated proteins is able to increase its phosphatase activity (66, 67). Thus, we cannot exclude that also the phosphorylated JAK2-SHP-2 interaction itself may contribute to the increased SHP-2 activity, favoring dephosphorylation of Src.
SHP-2 is among the first PTPs to be identified as intracellular effectors of SSTRs (18) and recently, this PTP was reported to participate in a SSTR2-activated multieffector complex with Src and SHP-1 (46). In our study, we confirm the pivotal role of SHP-2 in the signal transduction of SST. However, our data are partially discrepant with the previous study. In particular, we demonstrate that SHP-2 regulates Src activity likely through a direct dephosphorylation of the Src C-terminal inhibitory tyrosine. These data are in agreement with other works that demonstrate Src to be a direct substrate of SHP-2 (60, 61, 62). Conversely, in the paper from Ferjoux et al. (46), Src activation induced by SSTR2 caused an increase in SHP-2 activity and a subsequent regulation of SHP-1. The reasons for this discrepancy are not completely understood. However, it is now well known that the activity of SHP-2 is strictly dependent on the cell context, as far as substrates and biological effects. Thus, we can hypothesize that, as observed for other cellular responses (25, 57), the use of different cell types may justify the results obtained. For example in the subclone of CHO cells we used (CHO-k1), SHP-1 is not expressed (22), being, on the contrary, an important final effector in the study of Ferjoux et al. (46).
In our model, a definite molecular ordering in the SSTR1-activated pathway was identified in which SHP-2 controls Src activity. In any case, it is interesting to note that, in agreement with the above mentioned work (46), the molecular complex involving JAK2, SHP-2 and Src is dynamic in nature. In fact, our data show that SST-mediated early activation of JAK2 is concomitant with its dissociation from the other molecules. Moreover, JAK2 activation is related to the phosphorylation/activation of SHP-2 that, once active, rapidly dissociates from Src. Simultaneously, we observe a decrease in the phosphorylation of the inhibitory site of the kinase that causes an increased capability of Src to phosphorylate its substrates.
Concerning the involvement of Src in this signal transduction pathway, it was already reported the participation of the Src family kinases in SSTR1 and cAMP-mediated antiproliferative effects (22, 68). In our study, we report that Src activation, as already discussed for its upstream regulators JAK2 and SHP-2, is an absolute requirement for SST to regulate PTP activity. In fact, in the presence of Src inhibitors or expressing Src DN mutants, SST was unable to activate PTP. Moreover, we report, using the plasmon surface resonance assay, that PTP and Src may directly interact showing a high affinity binding. Interestingly, we found an interaction between the intracellular domain of PTP and the SH2 domain of Src in the absence of phosphorylation of the PTP, as it was reported for other SH2 domains (69). This interaction was rather specific because it was not demonstrated using other proteins including SHP-1 and SHP-2 or SSTR ITIMs.
Importantly, we show that after Src activation (i.e. transfection with Src CA mutants), a significant tyrosine phosphorylation of PTP was observed. Thus, it is possible that this posttransductional change may be responsible for the increase in PTP activity, after SST treatment.
To date there are very few reports showing a regulation of receptor-like PTP by tyrosine phosphorylation, mainly involving PTP (70). Thus, further studies are in progress to better address this issue.
There was recently confirmed an association between PTP and Src, although in these studies a positive regulation of PTP on Src activity via the dephosphorylation of tyrosine 529 (71, 72) was demonstrated. In our study, we do not have data on PTP modulation of Src activity but, although differences in the signaling may arise from the different cells used, we cannot exclude the occurrence of an activatory loop between Src and PTP in which the two molecules control each other activities.
Importantly, we also demonstrated that the same intracellular pathway described in transfected CHO-k1 cells, was also activated in a different cell line that expresses endogenous SSTR1. This observation is very relevant and clearly suggests that, although many discrepancies can be observed in different signal transduction studies according to the cell model used, the pathway we describe can likely be reproducible in different cell types, and thus it may represent a general mechanism of activation of PTP.
Here, we did not analyze possible intracellular targets of PTP. This issue was indeed already deeply studied in different previous papers that identified tyrosine kinase receptors (73, 74), phospholipase C
(39), ERK1/2 (34, 43), or Src (71, 72) as specific target of PTP activity, according to the cell type analyzed. However, in addition to these different substrates, all these studies confirmed the central role of PTP as a negative regulator of cell proliferation.
In summary, as described in Fig. 10A, we report that a multimeric complex is present in resting cells including JAK2, SHP-2, Src, and PTP. However, although we do not provide evidences, in analogy with previous reports, we can hypothesize that also the SSTR1 and the G protein are likely participating to this signaling complex.
|
) via a Gi/o protein, JAK2 is activated causing, in turn, phosphorylation/activation of SHP-2. The activation of SHP-2 induces Src dephosphorylation at the inhibitory site, followed by its activation via phosphorylation of the tyrosine 418. Finally, active Src, directly induces PTP phosphorylation and activation, being this event the final effector of SST antiproliferative activity.
In conclusion, our study, for the first time, describes a precise intracellular pathway upstream PTP activation, that from the activation of a specific SSTR mediates the antiproliferative effects of SST.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Antibodies
Anti-Src-phosphoY418, anti-Src-phosphoY529, and anti-Src were purchased from BioSource International (Nivelle, Belgium), anti-phosphoJAK2 and anti-JAK2 were from Upstate Cell Signaling Solutions (Lake Placid, NY), anti-SHP-2 was from Santa Cruz Biotechnology (Santa Cruz, CA), antiphosphotyrosine was from Cell Signaling Technology (Danvers, MA). For PTP detection, we used antibodies raised against the intracellular region of PTP expressed as a recombinant protein fused to GST, and affinity purified (42).
Methods
Cell Culture and Transfection.
The generation of CHO-k1 cells stably expressing r-SSTR1 (CHO-SSTR1) has been previously described (45). Both CHO-SSTR1 and C6 rat glioma cells (obtained from the Interlab Cell Line Collection Genova, Italy) were cultured under sterile conditions in F-12 Ham’s medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal calf serum (FCS) (Invitrogen). Transfections were performed using the Fugene reagent (Roche Molecular Biochemicals, Indianapolis, IN) according to the manufacturer’s instructions. Cells were transfected with pUSEamp vector containing the cDNA encoding Src CA (Y529F mutation), Src DN (K296R/Y528F mutations) and Src wt (Upstate Biotech, Lake Placid, NY) and the with cDNA for myc-SHP-2 wt and myc-SHP-2(C/S), subcloned in pCDNA3 vector (22).
Immunoprecipitation.
Cells, plated in 100-mm2 Petri dishes and treated following the experimental protocols, were lysed in a buffer containing 20 mM Tris-HCl (pH 7.4), 140 mM NaCl, 2 mM EDTA, 2 mM EGTA, 10% glycerol, 1% Nonidet P-40, 1 mM dithiothreitol, 1 mM sodium orthovanadate, 1 mM, and the "Cømplete" protease inhibitor cocktail (Roche). Total cell lysates were then incubated with the appropriate antibody (1 µg/mg of proteins) for 2 h at 4 C, in the lysis buffer, then 50 µl of protein A-Sepharose were added for an additional 1 h. After three washes with lysis buffer, the immunocomplexes were precipitated and analyzed in WB.
Immunoblotting.
Cells were lysed in a buffer containing 20 mM Tris-HCl (pH 7.4), 140 mM NaCl, 2 mM EDTA, 2 mM EGTA, 10% glycerol, 1% Nonidet P-40, 1 mM dithiothreitol, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride and the "Cømplete" protease inhibitor cocktail (Roche). An equal amount of proteins from each sample was size-fractionated by 6–10% SDS-PAGE and transferred to a polyvinylidene difluoride membrane (Bio-Rad, Milano Italy). Membranes were blocked with 25 mM Tris-HCl (pH 7.4), 140 mM NaCl, 2.5 mM KCl, Tween 20 0.1% containing 5% nonfat milk and probed with primary rabbit or mouse antibodies. The secondary antibody was a horseradish peroxidase-linked antirabbit or antimouse IgG antiserum (GE Healthcare, Milano Italy). The antibody-reactive bands were visualized by ECL (GE Healthcare).
PTP Assay.
PTP assay was performed on immunocomplexes using the synthetic substrate pNPP in a spectrophotometric assay (4). Samples were incubated at 30 C in 80 µl volume containing 20 µl of a 5x reaction buffer [250 mM HEPES (pH 7.2), 50 mM dithiothreitol, 25 mM EDTA, 500 nM microcystin-leucine-arginine (Alamone Labs, Jerusalem Israel)], and the reaction was started by adding 20 µl of 50 mM pNPP, carried out for 60 min and stopped by adding 900 µl of 0.1 N NaOH. The absorbance of the sample, directly proportional to the amount of the cleaved substrate, was measured at 410 nm (75). For PTP immunocomplex assay, the precipitation was done with IgG-coupled magnetic beads (Dynabeads, Dynal ASA, Oslo, Norway) because the protein A-Sepharose caused, per se, hydrolysis of pNPP (34).
Surface Plasmon Resonance Analysis.
GST-fused recombinant Src(SH2) and PTP were purified by Sepharose affinity chromatography (GE Healthcare) as previously described (76). Binding kinetics of GST-fusion proteins to immobilized proteins were measured with a BIAcore 3000 biosensor instrument (Biacore AB, Paris, France). For this purpose, GST-PTP was linked on a carboxymethyl dextran matrix of a CM5 chip (BIAcore) using a standard amine coupling protocol (BIAcore). 3600 Resonance Units of GST-PTP corresponding to 3.6 ng protein were coupled to the matrix on one flow cell of the instrument; a second flow cell without protein was used as reference. To measure interactions, GST-Src (SH2), GST-SHP-1 (SH2) or GST-SHP-2 (SH2) diluted in the running buffer [HEPES 10 mM (pH 7.4), NaCl 150 mM, EDTA 3 mM, surfactant P20 0.005%] was perfused at different concentrations (from 30–2000 nM) over the immobilized GST-PTP at a flow rate of 20 µl/min at 25 C. After each binding assay, flow cells were regenerated by short pulses of 0.005% sodium dodecyl sulfate. Binding constants (ka, kd, KD) were calculated using the global fitting algorithm provided by Biacore (Biaevaluation 4.0.1).
Statistical Analysis
Experiments were performed in quadruplicate and repeated at least three times. Statistical analysis was performed by means of one-way ANOVA. P < 0.05 was considered statistically significant.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author Disclosure Summary: All authors report that they have no disclosures to make, as they have nothing to declare.
, Activation; —
, inhibition basal values. WB was performed on aliquots of each sample using anti-SHP-2 antibody to demonstrate that equal amount of SHP-2 was immunoprecipitated (lower panel). **, P < 0.01 vs. control values. II, Time-course of SHP-2 phosphorylation after SST treatment. Cells were treated with SST (1 µM) for 15, 30, or 60 sec, then an equal amount of proteins from total cell lysates was immunoprecipitated with anti-SHP-2 antibody and probed with an antiphosphotyrosine antibody (upper panel). SST-induced SHP-2 phosphorylation started after 15 sec of treatment (although non statistically significant), reached the maximum after 30 sec and was still detectable at 60 sec. WB was then performed using anti-SHP-2 antibody, to demonstrate that equal amount of SHP-2 was immunoprecipitated (lower panel). The histogram reports the densitometric analysis of pTyr SHP-2/SHP-2 ratio expressed as percentage of the control for each time of treatment. Data are expressed as percentage of basal values from three independent experiments performed in quadruplicate. *, P < 0.05 vs. control values. C, SST regulation of Src activity in CHO-SSTR1 cells. I, Time-course of Src phosphorylation-activation subsequent to SST treatment (1 µM) in CHO-SSTR1 cells, detected using pSrcY418 antibody in WB. Src phosphorylation on tyrosine 418 induced by SST treatment was already detectable after 30 sec, reached a maximum after 120 sec and returned to basal level after 300 sec. II, Time-course of Src dephosphorylation-activation subsequent to SST treatment (1 µM) in CHO-SSTR1 cells, detected using pSrcY529 antibody in WB. Src dephophorylation on tyrosine 529 induced by SST treatment was detectable after 30 sec, decreased until 120 sec and returned to basal level after 300 sec. III, Equal protein loading was demonstrated probing the membrane with an antibody directed against the total Src. D, Comparison of the time-courses of SST-induced activation of JAK2, SHP2, Src and PTP