INTRODUCTION

Somatostatin is a widely distributed inhibitory hormone that exhibits various biological effects, including neurotransmission, inhibition of exocrine and endocrine secretions, and cell proliferation. The diverse biological effects of somatostatin are mediated through somatostatin receptors that are coupled to a variety of signal transduction pathways including adenylate cyclase, ionic conductance channels, and protein phosphatases (1, 2). Recently five somatostatin receptors have been cloned. They belong to the family of G protein-coupled receptors and can couple to diverse signal transduction pathways (3-7).

The ability of somatostatin and its stable analogues to promote inhibition of normal and tumor cell growth has been known for many years (8, 9). In pancreatic tumor cells, we and others have previously shown that somatostatin and analogues antagonize the mitogenic effect of growth factors acting on tyrosine kinase receptors such as epidermal growth factor (9, 10). Although the molecular events leading to the inhibition of cell proliferation are still poorly understood, it has been shown that, after binding to somatostatin receptors, somatostatin analogues cause a rapid stimulation of a membrane protein-tyrosine phosphatase (PTPase)¹ activity and dephosphorylate phosphorylated epidermal growth factor receptors (9, 11, 12) suggesting that a PTPase may participate in the somatostatin-induced inhibition of growth factor-mediated mitogenic signal. Recently, the expression of the sst2 somatostatin receptor subtype in NIH3T3 and Chinese hamster ovary (CHO) cells led us to the demonstration of the direct involvement of sst2 in both the antiproliferative effect of somatostatin and its stimulatory effect on PTPase activity (13, 14). Incubation of cells expressing sst2 with the PTPase inhibitor, vanadate, prevented both effects suggesting that a PTPase may be implicated in the negative growth signal induced by activation of sst2. In addition, we demonstrated that a PTPase of 70 kDa, identified as SHP-1, copurified with membrane somatostatin receptors (15) from pancreatic acinar cells that highly expressed sst2 receptor subtype (16). Taken together, these results suggest that SHP-1 may be a candidate for sst2-mediated early signaling events.

SHP-1 (also named SHPTP-1, SHP, or HCP) is a non-transmembrane PTPase that contains two Src homology 2 (SH2) domains involved in its association with multiple signaling molecules (17-20). SHP-1 associates in vivo with activated growth factor tyrosine kinase receptors such as epidermal growth factor receptor (21). SHP-1 has also been shown to associate with activated cytokine receptors, interleukin-3 receptor  chain (22), erythropoietin receptor (23), interferon-/ receptor (24), and also with B cell FcRIIB receptor (25). As a result of these interactions, SHP-1 becomes tyrosine-phosphorylated by these factors (22, 26, 27). SHP-1 is also tyrosine-phosphorylated in response to activation of G protein-coupled receptors (28, 29).

Recent studies have suggested a role of SHP-1 in terminating growth factor mitogenic signals by dephosphorylating critical molecules. SHP-1 dephosphorylates a variety of protein-tyrosine kinase receptors when coexpressed in 293 cells (30) and has been shown to down-regulate interleukin-3-induced tyrosine phosphorylation and mitogenesis in hematopoietic cells (22). Association of SHP-1 with the erythropoietin receptor causes inactivation of Janus kinase 2 and termination of erythropoietin-induced proliferation signal (23). Similarly, association of SHP-1 with interferon-/ receptor negatively regulates interferon-/-induced Janus kinase 1/Stat1 signaling pathway (24). SHP-1 has been also implicated as the mediator used by the FcRIIB1 receptor in B lymphocytes to turn off B cell antigen receptor signaling (25). The role of SHP-1 in the negative regulation of hematopoiesis is consistent with the various hematopoietic abnormalities and hypersensitivity to growth factors in mice carrying the lethal mutation in the SHP-1 gene that results in motheaten phenotype (31, 32).

If one of the role of SHP-1 is a negative regulation of growth factor signaling, one might therefore speculate that SHP-1 could be activated by factors that negatively regulate cell growth such as somatostatin. In this study, the role of SHP-1 in signal transduction pathway of the G protein-coupled sst2 somatostatin receptor was investigated. Our results provide evidence of physical and functional association of SHP-1 with sst2 and demonstrate that somatostatin is a physiological modulator of SHP-1 that may be required in somatostatin-induced tyrosine phosphatase activation and antiproliferative signals initiated by sst2.

EXPERIMENTAL PROCEDURES

SMS 201-995 (SMS) and somatostatin were generous gifts of Dr. C. Bruns (Sandoz, Basel, Switzerland) and Dr. L. Moroder (Munich, Germany), respectively. [Tyr¹¹]somatostatin was purchased from Bachem. [-³³P]ATP (3,000 Ci/mmol) was purchased from Isotopchim (France). Enhanced chemiluminescence (ECL) immunodetection system and Hybond ECL nitrocellulose membrane were from Amersham Corp. Poly(Glu, Tyr), cholesterol hemisuccinate, geneticin (G418), and Sepharose-protein A beads were from Sigma. Minimal essential medium (MEM), fetal calf serum (FCS), and Lipofectin reagent were from Life Technologies, Inc. CHAPS was from Serva.

The 2.1-kilobase HindIII/NotI fragment of mouse SHP-1 cDNA (Dr. M. L. Thomas, Howard Hughes Medical Institute, Washington University, St. Louis, MO) was subcloned into the expression vector pcDNAneo vector (Invitrogen). CHO (DG44 variant) cells were grown to 50% confluency in 60-mm diameter dishes and were cotransfected using Lipofectin reagent with 1 µg of SHP-1 or SHP-1 (C453S) mutant in pcDNA I neo vector and 3 µg of 1.2-kilobase XbaI fragment of mouse sst2 gene in pCMV6c vector (Dr. G. I. Bell, Howard Hughes Medical Institute, University of Chicago, Chicago, IL) (3). SHP-1 (C453S) mutant was constructed with the oligonucleotide primer 5-GAT GCC AGC GCT GGA ATG CAC AAT-3 by using the method of Kunkel et al. (33). The mutation was confirmed by dideoxynucleotide sequencing. Stable colonies obtained by selection with G418 (600 µg/ml) were screened for somatostatin binding using [¹²⁵I-Tyr¹¹]somatostatin as described below. Cellular clones expressing somatostatin binding sites at a similar level were screened for the presence of SHP-1 or SHP-1 (mutant) using Western blot analysis as described below.

Polyclonal anti-sst2 antibodies were generated in rabbits immunized with a peptide corresponding to amino acid residues 191-206 of mouse sst2 as described previously (34). The monoclonal anti-SHP-1 antibodies were obtained from Transduction Laboratories (Medgene, France). Monoclonal antibodies raised against phosphotyrosine (PY-20) were purchased from Santa Cruz Biotechnology (Tebu, France). G_o1-2, G_i1-2, G_i2, and G_i3 antibodies were from Gramsch Laboratories (Germany).

CHO cells and its derivatives were cultured in MEM containing 10% FCS and G418 (200 µg/ml) as described previously (14). For cell treatment, cells were plated in 100-mm dishes (10⁶ cells/dish) in MEM containing 10% FCS, and after an overnight attachment phase, cells were cultured for 24 h, washed, and cultured overnight in serum-free MEM. Cells were rinsed and then incubated for indicated times in the presence of the somatostatin analogue, SMS, at 1 nM.

For cell growth assay, cells were cultured in MEM containing 10% FCS and plated in 35-mm dishes at 55 × 10³ cells/ml (2 ml/dish). After an overnight attachment phase (time 0), cells were cultured for 24 h in serum-free MEM or in MEM containing 10% FCS with or without 1 nM SMS. Cell growth was measured at the indicated times by counting cells with a Coulter counter model ZM (Coulter Electronics) as described elsewhere (13).

[Tyr¹¹]somatostatin was radioiodinated and purified by reverse phase-high performance liquid chromatography as described previously (14). Cells were grown in 75-cm² flasks for 48 h and lysed by freezing in liquid nitrogen in 50 mM Tris-HCl (pH 7.8) containing soybean trypsin inhibitor (0.3 mg/ml). After thawing, the cell lyzate was centrifuged at 26,000 × g for 30 min at 4 °C, and the pellet was resuspended in the same buffer. Binding studies were performed on the resultant crude membranes as described previously (14). Briefly, 5 µg of membrane proteins were incubated with 30 pM [¹²⁵I-Tyr¹¹]somatostatin at 25 °C for 90 min in 50 mM Tris-HCl (pH 7.8) containing bovine serum albumin (1 mg/ml), soybean trypsin inhibitor (0.3 mg/ml), bacitracin (0.5 mg/ml), 5 mM MgSO₄ (binding buffer). Nonspecific binding was determined in the presence of 1 µM somatostatin. Samples were centrifuged at 10,000 × g for 10 min at 4 °C. The pellet was washed twice with ice-cold binding buffer, and the radioactivity in the pellet was measured. Specific binding was calculated as the difference between the amount of radioactivity bound in the absence and the presence of 1 µM somatostatin.

Cells were washed twice and collected in phosphate-buffered saline. After centrifugation at 1000 × g for 5 min at 4 °C, cells were solubilized with 50 mM Tris-HCl buffer (pH 7.5) containing 140 mM NaCl, 1 mM EDTA, 0.1 mg/ml soybean trypsin inhibitor, 0.1 mM phenylmethylsulfonyl fluoride (buffer A) in the presence of 1.5% CHAPS and 0.5 mM sodium orthovanadate. The mixture was gently agitated for 30 min at 4 °C and thereafter centrifuged at 13,000 × g for 20 min. Soluble proteins (300-500 µg) were incubated for 3 h at 4 °C with either anti-SHP-1, anti-sst2, anti-G protein antibodies, or preimmune serum prebound to Sepharose-protein A beads prewashed in buffer A. The beads were then washed twice with buffer A and resuspended in either 50 µl of sample buffer for immunoblotting or 180 µl of PTPase buffer for PTPase assay.

For immunoblotting, solubilized proteins (90 µg) or immunoprecipitated proteins (see above) were resolved through 7.5% SDS-polyacrylamide gels, transferred to a nitrocellulose membrane, and immunoblotted with anti-SHP-1, anti-sst2, anti-phosphotyrosine or anti-G protein antibodies as described previously (35). Immunoreactive proteins were visualized by the ECL immunodetection system and quantified by image analysis using a Biocom apparatus (Biocom, Paris, France).

Immunoprecipitated proteins were resuspended in PTPase buffer containing 50 mM Tris-HCl (pH 7), 1 mg/ml bovine serum albumin, and 5 mM dithiothreitol (PTPase buffer). The substrate poly(Glu, Tyr) was phosphorylated with [-³³P]ATP as described elsewhere (13). The reaction was initiated by the addition of 30,000 cpm of ³³P-labeled poly(Glu, Tyr) and allowed to proceed for 10 min at 30 °C as described previously (13). PTPase activity was expressed in picomoles of inorganic phosphate released per min at 30 °C from radiolabeled substrate.

Statistical comparison between SMS-treated and nontreated cells was performed using Student's paired t test.

RESULTS

We had previously reported that, in CHO cells expressing sst2 somatostatin receptors, somatostatin analogues stimulated PTPase activity (14). Furthermore, in pancreatic cells that highly expressed endogenous sst2, SHP-1 copurified with somatostatin receptors (15). To investigate whether SHP-1 interacts with sst2, we stably coexpressed SHP-1 and sst2 in CHO cells. CHO/sst2-SHP-1 clones expressed sst2 as a protein of 95 kDa detected by immunoblotting as previously reported (34) and SHP-1 as a protein of 68 kDa, whereas these proteins were barely detectable in wild CHO cells (Fig. 1). sst2 immunoprecipitates prepared from CHO/sst2-SHP-1 cells were examined by immunoblotting for the presence of SHP-1. As shown in Fig. 2, the 68-kDa SHP-1 protein was coprecipitated with the 95-kDa sst2 protein. Similarly, SHP-1 immunoprecipitation resulted in the coimmunoprecipitation of SHP-1 and sst2. These protein bands were not seen when immunoprecipitations were performed in the presence of preimmune serum instead of immune serum. Quantification of immunoblots revealed that the amount of sst2 present in the SHP-1 immunoprecipitates (lane ip SHP-1-blot sst2) represents approximately 17-20% of immunoprecipitated sst2 (ip sst2-blot sst2). Taking into account the efficiency of sst2 immunoprecipitation, this represents about 5% of total cellular sst2. From these results, we conclude that SHP-1 physically associates with sst2 and that sst2·SHP-1 complexes are preformed in resting cells.

To determine whether the interaction of SHP-1 with sst2 was affected by somatostatin treatment, CHO/sst2-SHP-1 cells were incubated in the presence of the somatostatin analogue, SMS, for various times prior to solubilization and immunoprecipitation with anti-sst2 antibodies. The amount of SHP-1 associated with sst2 was then analyzed by immunoblotting with anti-SHP-1 antibodies and the blots were reprobed with anti-sst2 antibodies to ensure that comparable amounts of sst2 molecules were immunoprecipitated at each time point of SMS treatment. As observed in Fig. 3, somatostatin treatment resulted in a transient increase of the amount of SHP-1 immunoprecipitated with anti-sst2 antibodies within the first 30 s after which the rate of immunodetected SHP-1 in sst2 immunoprecipitates decreased, about 80% of the preformed sst2·SHP-1 complexes being dissociated after 10 min of SMS treatment. Similarly, the amount of immunodetected sst2 in SHP-1 immunoprecipitates was decreased by 60 ± 6% following SMS treatment of cells for 3 min.

We further investigated the effect of SMS on SHP-1 activity in CHO/sst2-SHP-1 cells. Cells were incubated in the presence of 1 nM SMS for various times, after which they were solubilized, and SHP-1 activity was measured in SHP-1 immunoprecipitates. As shown in Fig. 4, SHP-1 activity was increased upon treatment with SMS. The stimulation of SHP-1 activity was maximal after 30 s of SMS treatment and slightly declined up to 10 min.

We then examined the effect of SMS treatment on the level of tyrosine phosphorylation of SHP-1. SMS-treated or untreated cells were subjected to immunoprecipitation with anti-SHP-1 antibodies and immunoblotted with either anti-phosphotyrosine or anti-SHP-1 antibodies. The immunoblot revealed that SHP-1 was tyrosine phosphorylated in untreated cells and that SMS induced a rapid and transient dephosphorylation of SHP-1 (Fig. 5). The dephosphorylation of SHP-1 was observed as early as 30 s after SMS treatment, was maximal at 1 min and declined subsequently until control levels. All these results indicate that SMS induced a transient increase and a subsequent dissociation of preformed sst2·SHP-1 complexes, which was associated with the stimulation and the transient dephosphorylation of the enzyme.

Association of sst2 with SHP-1 Involves the Pertussis Toxin-sensitive G Protein, G_i3

In CHO cells expressing sst2, we previously reported that stimulation of PTPase activity by somatostatin was suppressed by pretreatment of cells with pertussis toxin indicating that a pertussis toxin sensitive G protein was involved in this effect (13). Furthermore, in these cells, sst2 has been demonstrated to be coupled to pertussis toxin-sensitive G proteins, G_i3 and G_o2, but not to G_i1 and G_i2 (36). Preincubation of CHO/sst2-SHP-1 cells with pertussis toxin for 18 h at 100 ng/ml abolished the stimulatory effect of SMS on SHP-1 activity in SHP-1 immunoprecipitates (not shown) suggesting that sst2-coupled activation of SHP-1 was mediated by a pertussis toxin-sensitive G protein. To identify the G protein involved in the sst2·SHP-1 complexes, SHP-1 and sst2 immunoprecipitates from CHO/sst2-SHP-1 cells were analyzed by immunoblotting with antibodies directed against G_i1, 2, 3 and G_o1-2 subunits. G_i3 was immunoprecipitated with anti-sst2 antibodies as well as anti-SHP-1 antibodies, suggesting that G_i3 was present in the sst2·SHP-1 complexes. In contrast, G_i1, G_i2, and G_o were never detected in the sst2·SHP-1 immunoprecipitates (Fig. 6). In addition PTPase activity can be immunoprecipitated by anti-G_i3 antibodies but not by G_o antibodies (not shown). All these results argue in favor of a role for G_i3 in the formation of the sst2·SHP-1 complexes.

i3

o1-2

i3

i2

i1-2

To investigate whether the association of G_i3 with sst2 can be modified by somatostatin, CHO/sst2-SHP-1 cells were treated for various times with 1 nM SMS and G_i3 was identified in sst2 and SHP-1 immunoprecipitates. As observed in Fig. 6, SMS treatment induced a transient increase at 30 s of the amount of G_i3 immunoprecipitated either by anti-sst2 or anti-SHP-1 antibodies which was followed by a rapid decrease of the amount of immunodetected G_i3. Only 30% of sst2-associated G_i3 was detected after 10 min of SMS treatment. The time course of G_i3 dissociation from sst2 and SHP-1 paralleled that of dissociation of SHP-1 from sst2, suggesting that these events can be linked.

To obtain direct evidence of the role of SHP-1 in the inhibitory effect of somatostatin on cell proliferation, we generated a catalytically inactive SHP-1 by introducing a point mutation into the conserved catalytic residue, cysteine 453, which is crucial for catalytic activity of the enzyme. Cys⁴⁵³ was mutated to Ser. This mutation completely abolished the phosphatase activity of the SHP-1 mutant transiently expressed in COS-7 cells.² We stably coexpressed the SHP-1 mutant and sst2 in CHO cells and we selected clones (CHO/sst2-SHP-1(C453S)) that expressed the two proteins at a similar level with that observed in CHO/sst2-SHP-1 cells as demonstrated by immunoblotting (Fig. 7). CHO/sst2-SHP-1(C453S) cells were incubated with SMS after which SHP-1 was immunoprecipitated with anti-SHP-1 antibodies, and PTPase activity was measured. SMS no more stimulated SHP-1 activity in cells expressing the SHP-1 mutant (Fig. 8). Expression of sst2 in CHO cells did not modify the serum-stimulated cell growth whereas coexpression of sst2 and SHP-1 inhibited by 39 ± 6% (n = 3) serum-activated cell growth after 3 days of culture when compared with control cells. These results are consistent with the negative role of SHP-1 on CHO cell growth (37). Incubation of wild CHO cells for 24 h in the presence of 1 nM SMS did not modify the serum-stimulated cell growth, whereas SMS inhibited by 36% the growth of CHO/sst2-SHP-1 cells (Fig. 9), in agreement with the growth inhibitory effect of somatostatin analogues in cells expressing sst2 (13, 14). In contrast, the SMS-induced inhibition of cell proliferation was abolished in CHO cells coexpressing sst2 and the SHP-1 mutant. This indicates that the expression of the catalytic inactive SHP-1 blocks the negative regulation of cell growth induced by SMS and demonstrates that SHP-1 may play an important role in the transduction of the negative growth signal promoted by activation of sst2.

DISCUSSION

Somatostatin acts as a growth inhibitory factor in a variety of normal and tumor cells (2, 9, 10). We and others have demonstrated that somatostatin and analogues induce the stimulation of a membrane PTPase, which may be involved in the inhibitory effect of these peptides on cell proliferation (9, 11, 12, 38). More recently, we have established the role of the somatostatin receptor sst2 in mediating the stimulatory effect of somatostatin on PTPase activity and its negative effect on cell growth (13, 14). The identification of involved PTPase is therefore important for the understanding of the negative growth signal transduction promoted by sst2. In the present study, we have shown that the phosphotyrosine phosphatase SHP-1 associates with the sst2 somatostatin receptor subtype and established that SHP-1 is involved in the growth inhibitory signal transduction pathway of sst2.

SHP-1 is a cytoplasmic PTPase containing two SH2 domains that enable it to bind specific tyrosine residues of phosphorylated proteins. SHP-1 has been reported to associate with a variety of activated growth factor tyrosine kinase receptors, cytokine receptors, and also with the B cell FcRIIB receptor (21-25). We have established that SHP-1 associates with another class of receptor, the G protein-coupled receptor sst2. Results from coprecipitation of SHP-1 with sst2 in CHO cells coexpressing sst2 and SHP-1 provide evidence that SHP-1 associates with sst2 in basal conditions. Specific components mediating this association remain to be determined. However, the demonstration that the G_i3 subunit, which is known to couple sst2 receptors at the resting level (36), is present in the sst2·SHP-1 complex and can immunoprecipitate PTPase activity strongly suggests that G_i3 could achieve a direct coupling between sst2 and SHP-1. Such a receptor-G protein coupling which is evident in overexpression system (39) has been recently reported for the 5-HT_1A receptor which interacts with G_z in the absence of agonist (40). The molecular base of interaction between G_i3 and SHP-1 is not known, but preliminary results suggest that G_i3 could be tyrosyl-phosphorylated and therefore interact with the SHP-1 SH2 domains.³ However, our results do not preclude the possibility that another protein may contribute to the interaction of SHP-1 with sst2.

Furthermore, we demonstrated that the occupancy of sst2 promotes a rapid and transient increase (at 0.5 min) of the recruitment of SHP-1 and to a lesser extent that of G_i3 to sst2, which is followed by a rapid dissociation of these molecules from sst2. The weak increase of the interaction between sst2 and either G_i3 or SHP-1 following sst2 stimulation may be related to the known transient nature of the association of receptors with G proteins driven by receptor occupancy (41). A similar dynamic interaction was recently observed with IL-8 receptors that interacted with G_i2 in a time-dependent manner, attaining a maximal level by 1 min and then declining until control levels after 10 min of IL-8 stimulation (42). Moreover, the observation that the interaction of sst2 with G_i3 and SHP-1 was decreased following sst2 activation indicates that somatostatin stimulates the dissociation of the majority of preformed complexes. We also demonstrated that somatostatin activation of sst2 leads to a rapid increase of SHP-1 activity, which is maximal at 0.5 min, maintained as a plateau until 3 min, and then decreased at 10 min. These results argue in favor of the role of SHP-1 in the early events of somatostatin action. One might speculate that, following exposure of cells to SMS, the transient increase of association of SHP-1 to sst2 via G_i3 was accompanied by an activating conformational change of the enzyme, the activation of the enzyme being able to result from the relief of autoinhibition by the SH2 domains (43). The fact that somatostatin-induced stimulation of the enzyme was inhibited by pertussis toxin emphasizes the implication of a G_i/o-like protein in this effect and supports the idea that stimulation of SHP-1 by somatostatin results from activation of G_i3. Consistent with these results, we and others have previously reported that the activation of PTPase activity by somatostatin involved a pertussis toxin-sensitive G protein (11, 14). On the other hand, we found that somatostatin induces a rapid dissociation (at 1 min) of sst2·G_i3·SHP-1 complexes. When dissociation of the complexes start, the enzyme is still active. The dissociation may allow SHP-1 to associate with specific substrates, such as phosphorylated growth factor receptors and downstream molecules, with this probably leading to negative regulation of mitogenic signals by dephosphorylation.

We also demonstrated that, concomitantly with the increase of SHP-1 activity, somatostatin induces a transient dephosphorylation of SHP-1 on tyrosine residues that is maximal at 1 min and returns to basal level at 10 min. These results suggest that activated SHP-1 undergoes a rapid tyrosine autodephosphorylation due to the enzyme activation. This hypothesis is strengthened by the observation that SHP-1 was rapidly autodephosphorylated in vitro, whereas the catalytically inactive mutant SHP-1 was stably tyrosine-phosphorylated (44, 45). In contrast, SHP-1 has been reported to be tyrosine-phophorylated in response to various growth factors and cytokines receptors as well as mitogenic factors acting on G protein-coupled receptors (22, 26, 27, 29, 46). The role of tyrosine phosphorylation/dephosphorylation of SHP-1 has to be elucidated. Different sites of SHP-1 tyrosine phosphorylation have been identified and may play specific role (44, 47).

SHP-1 was found to play a major role in negatively regulating signaling pathways. For instance, SHP-1 has also been demonstrated to terminate IL-3 and EPO growth signals, its recruitment to the activated receptors causing dephosphorylation and inactivation of specifically associated signaling molecules (22, 23). In CHO cells overexpressing the insulin receptor, a negative effect of overexpressed SHP-1 on cell proliferation has been also reported (37). The demonstration that the negative effect of somatostatin on CHO cell growth can be suppressed in cells coexpressing sst2 and the inactive SHP-1 mutant argues in favor of the role of SHP-1 in the negative growth signal induced by somatostatin-activated sst2. Further, the rapid activation of SHP-1 following somatostatin addition in CHO cells coexpressing sst2 and SHP-1 raises the possibility that activation of SHP-1 is the initiating step for sst2 signal transduction leading to inhibition of cell proliferation. These hypothesis are strengthened by the observation that rat pancreatic tumor AR42J cells are sensitive to the antiproliferative effect of somatostatin analogs and highly express sst2 receptors and SHP-1 (10, 16, 35). Examining SHP-1 expression in cells sensitive to somatostatin analogs would add significantly to these conclusions. It is well known that growth factors transduce cell proliferation signal through activation of receptor tyrosine kinases that phosphorylate downstream enzymes and/or adaptator proteins leading to phosphorylation and activation of mitogen-activated protein kinase. Recent data provide evidence that mitogen peptides acting on G protein-coupled receptors can also promote phosphorylation and activation of mitogen-activated protein kinase via activation of tyrosine kinase (48). Conversely, one could be expected that growth inhibitory peptides acting via G protein-coupled receptors may induce growth inhibition through activation of PTPase resulting in tyrosine dephosphorylation of mitogen-induced tyrosine phosphorylation of signaling molecules. Such a stimulation of PTPase by growth inhibitory peptides acting via G protein-coupled receptors has been previously reported for angiotensin II AT2 receptors (49). The effector molecules that act downstream of sst2·SHP-1 complexes are not known. However, it has been shown that somatostatin dephosphorylates tyrosine phosphorylated EGF receptors (9), suggesting that growth factor receptors could be one of the substrates of SHP-1. sst1 and sst3 somatostatin receptor subtypes expressed in heterologous cells have been also demonstrated to mediate somatostatin stimulation of PTPase activity (14, 50, 51). Whether SHP-1 binds to other somatostatin receptor subtypes has yet to be investigated. If this was the case, then activation of SHP-1 could be an early signal pathway shared by the somatostatin receptor family.