Report

Stat3-Mediated Transformation of NIH-3T3 Cells by the Constitutively Active Q205L Gαo Protein

See allHide authors and affiliations

Science  07 Jan 2000:
Vol. 287, Issue 5450, pp. 142-144
DOI: 10.1126/science.287.5450.142

Abstract

Expression of Q205L Gαo(Gαo*), an alpha subunit of heterotrimeric guanine nucleotide–binding proteins (G proteins) that lacks guanosine triphosphatase (GTPase) activity in NIH-3T3 cells, results in transformation. Expression of Gαo* in NIH-3T3 cells activated signal transducer and activator of transcription 3 (Stat3) but not mitogen-activated protein (MAP) kinases 1 or 2. Coexpression of dominant negative Stat3 inhibited Gαo*-induced transformation of NIH-3T3 cells and activation of endogenous Stat3. Furthermore, Gαo* expression increased activity of the tyrosine kinase c-Src, and the Gαo*-induced activation of Stat3 was blocked by expression of Csk (carboxyl-terminal Src kinase), which inactivates c-Src. The results indicate that Stat3 can function as a downstream effector for Gαo* and mediate its biological effects.

Although downstream effectors that mediate the actions of G protein α subunits Gαs, Gαi, and Gαq have been elucidated, little is known about signaling pathways activated by Gαo. Expression of the GTPase-deficient (and thus constitutively active) mutant of Gαo in which Gly205 is changed to Leu (Gαo*) in NIH-3T3 cells results in transformation (1), but the molecular mechanisms underlying this phenomenon are not known. MAP kinases 1 and 2 participate in stimulation of proliferation and transformation of NIH-3T3 cells (2). The transcription factor Stat3 is activated and required for transformation of NIH-3T3 cells by the v-Src oncogene (3). Hence, the roles of MAP kinases and Stat3 in transformation of NIH-3T3 cells by Gαo* were investigated.

The Stat family of proteins is implicated in the functions of a wide range of cells (4). When activated, Stat3 becomes phosphorylated, dimerizes, and translocates to the nucleus, where it binds DNA and modulates gene expression. To determine the effects of Gαo* on the phosphorylation state of native Stat3, we transiently transfected NIH-3T3 cells with a Gαo* expression vector, extracted the proteins, resolved them by SDS–polyacrylamide gel electrophoresis (SDS-PAGE), and blotted them with antibodies specific for Stat3 phosphorylated on Tyr705 (5). Expression of Gαo* led to phosphorylation of Tyr705 on endogenous Stat3 proteins in NIH-3T3 cells (Fig. 1A). Expression of Gαo* did not lead to phosphorylation of Stat1 (6). To further determine if the Gαo*-induced phosphorylation of Stat3 reflected an increase in Stat3 transcriptional activity, we did a transcriptional activation assay in cells transfected with a Stat3-responsive luciferase reporter construct (7). Expression of Gαo* resulted in activation of endogenous Stat3 (Fig. 1B), as evidenced by increased reporter gene expression. To determine that the reporter gene activity was in fact due to Stat3 activation, we coexpressed mutant Stat3 proteins that are not phosphorylated or fail to bind DNA and act in a dominant negative manner (3). Activation of Stat3 in cells expressing Gαo* was inhibited by the coexpression of dominant negative Stat3 proteins (Fig. 1B). Additionally, expression of wild-type Gαo or G protein β and γ subunits had no effect on Stat3 activity (Fig. 1C). Expression of Gα12* gave a small but consistent twofold increase in Stat3 activity (Fig. 1C). Activation of Stat3 by Gαq* was varied (Fig. 1C), and cell density appears to play a role in this activation; further work is being done. Thus, Gαo* appears to activate Stat3 in a specific manner in NIH-3T3 cells.

Figure 1

Phosphorylation and activation of Stat3 induced by Q205L Gαo. (A) Soluble proteins from cell lysates of NIH-3T3 cells treated with Il-6 or transfected with Gαo* were probed with antibody specific for phosphorylated Stat3 at Tyr705 (top) or with antibody to Stat3 (bottom). (B) NIH-3T3 cells transfected with the Ly6E Stat3-responsive luciferase reporter construct were transfected with either Gαo* alone or with Gαo* and dominant negative Stat3 constructs VVV461-463AAA/EE434-435AA (DN 1) or Y705-F (DN 2). (Statistical differences between the treatments were assessed with the Newman-Keuls test: P < 0.001; †, compared with control; ††, compared with Gαo*;n = 4 to 5 in duplicate transfections each time.) (C) Stat3 transcriptional assay performed as in (B). Cells were transfected with either Gαo*, wild-type (wt) Gαo, Gβ2γ112*, or Gα9*. (Newman-Keuls test:P < 0.001; †, compared with control;n = 2 in duplicate transfections each time.)

The role of MAP kinases 1 and 2 in cell proliferation and transformation has been extensively studied, and activation of these enzymes can transform NIH-3T3 cells (2). Receptor-mediated activation of Gαo leads to increased activity of MAP kinases 1 and 2 in Chinese hamster ovary cells (8). Therefore, we examined the effects of Gαo* expression on MAP kinase activity in NIH-3T3 cells (9). Expression of Gαo* did not activate MAP kinases 1 or 2 in NIH-3T3 cells (Fig. 2), as measured by immunoblot analysis with antibodies that specifically recognize the active forms of MAP kinases 1 and 2, which are phosphorylated on Thr202 and Tyr204. Thus, the Stat3 signaling pathway and not the MAP kinase 1 and 2 pathway is activated by Gαo* in NIH-3T3 cells.

Figure 2

Activation of MAP kinases 1 and 2 in cells expressing Gαo*. Proteins from soluble cell lysates of NIH-3T3 cells treated with either IL-6 or PDGF or transfected with Gαo* were probed with antibody specific for phosphorylated MAP kinases 1 and 2 on Thr202 and Tyr204 (top) or antibody to MAP kinases 1 and 2 (bottom).

Expression of Gαo* transforms NIH-3T3 cells and leads to colony formation in soft agar (1). Expression of dominant negative Stat3 inhibits transformation of NIH-3T3 cells by v-Src but not transformation by H-Ras (3). To test the hypothesis that activation of Stat3 is necessary for transformation of NIH-3T3 cells by Gαo*, we prepared transfected cells that expressed dominant negative Stat3 and Gαo* and assayed colony formation (10). Expression of dominant negative Stat3 inhibited Gαo* transformation of NIH-3T3 cells (Fig. 3). Thus, activation of Stat3 is necessary for transformation of NIH-3T3 cells by Gαo*.

Figure 3

Inhibition of Gαo* transformation of NIH-3T3 cells by dominant negative Stat3. Colony formation in soft agar was assayed with Gαo*-transformed cells that were cotransfected with either one of the two dominant negative Stat3 constructs. (Newman-Keuls test: P < 0.001; †, compared with Gαo*-expressing cells.) The experiment was performed twice, with duplicate transfections each time.

Stat3 is activated in response to cytokines and is tyrosine phosphorylated by Janus kinases (JAK) (11). We found that expression of Gαo* did not activate JAK2 (6). However, Stat3 can also be directly phosphorylated and activated by c-Src or v-Src, which results in the proliferation and transformation of several cell types (3, 12). Therefore, we tested whether Gαo* activated endogenous c-Src. Gαo* was expressed in NIH-3T3 cells. The cells were lysed, endogenous c-Src was immunoprecipitated, and an in vitro kinase assay was done with the immune complex (13). Expression of Gαo* increased endogenous c-Src activity in NIH-3T3 cells (Fig. 4A). We then examined the role of endogenous Src in activation of Stat3 in response to Gαo* and in the transformation of NIH-3T3 cells. Csk inhibited activation of Stat3 in cells expressing Gαo* (Fig. 4B). Furthermore, expression of Csk also significantly inhibited transformation of NIH-3T3 cells expressing Gαo* (Fig. 4C). These data show that c-Src is activated in cells expressing Gαo* and that c-Src acts upstream of Stat3.

Figure 4

Role of c-Src in activation of Stat3 in cells expressing Gαo*. (A) Endogenous c-Src activity in NIH-3T3 cells expressing Gαo*. In vitro kinase assays of immunoprecipitated c-Src (†,t test, P < 0.01, n = 2 done in duplicate both times). C, control. (B) Stat3 transcriptional activation assay as described in Fig. 1B was done with NIH-3T3 cells expressing either Gαo* alone or Gαo* and Csk. (Newman- Keuls test: P < 0.001; †, compared with control; ††, compared with Gαo*-expressing cells; n = 3 in duplicate each time.) (C) Colony formation in soft agar (Fig. 3) of Gαo*-transformed cells also transfected with or without the Csk expression plasmid (†, ttest, P < 0.01). The experiment was performed twice, with duplicate transfections each time.

The activation of Stat3 by a G protein α-subunit, leading to expression of the transformed phenotype, demonstrates a connection between G protein and Stat pathways. What would be the biological relevance of such a connection? Most effects of Stat have been characterized in differentiated cells such as the effects of growth hormone, prolactin, and leptin. However, at least in NIH-3T3 cells, Stat-3 can trigger transformation (3). Proliferation and transformation of NIH-3T3 cells can be induced by many stimuli. Indeed, signaling components that trigger transformation in NIH-3T3 cells, when expressed in other cell types, can induce differentiation. One prominent example is the small GTPase Ras, which transforms NIH-3T3 cells but triggers neurite outgrowth in PC-12 cells (14), both effects being mediated by activation of MAP kinases 1 and 2. Gαo is abundantly expressed in growth cones of neurons (15), and expression of activated Gαo induces neurite outgrowth in PC-12 and NE-115 cells (16). Large amounts of c-Src are also found in nerve growth cones (17), and amounts of c-Src increase severalfold during neuronal differentiation (18). The molecular mechanisms involved in triggering of neurite outgrowth are currently not known. Several signaling pathways have been implicated, including the MAP kinase pathways in the effects of NGF and the Stat pathway in the effects of interleukin-6 (IL-6) (19,20). Thus, it is possible that Gαo might also stimulate Stat3 transcriptional activity in differentiated cells such as neurons to regulate neuronal plasticity.

It is unlikely that Gαo activation of c-Src and Stat3 represents the sole effector pathway of Gαo signaling. Studies from our laboratory show that wild-type Gαodirectly interacts with Rap1-GTPase–activating protein (Rap1-GAP) to modulate Rap1 activity (21). Thus, Gαomay be able to engage several distinct signaling pathways to elicit its biological effects.

  • * To whom correspondence should be addressed. E-mail: ramp01{at}doc.mssm.edu

REFERENCES AND NOTES

View Abstract

Stay Connected to Science

Navigate This Article