Neoplastic Transformation Induced by the gep Oncogenes Involves the Scaffold Protein JNK-Interacting Leucine Zipper Protein

Abstract

The activated mutants of the α-subunits of G proteins G₁₂ and G₁₃ have been designated as the gep oncogenes owing to their ability to stimulate diverse oncogenic signaling pathways that lead to neoplastic transformation of fibroblast cell lines and tumorigenesis in nude mice models. Studies from our laboratory as well as others have shown that the growth-promoting activities of Gα₁₂ and Gα₁₃ involve potent activation of c-Jun N-terminal kinases (JNKs). Our previous studies have indicated that the JNK-interacting leucine zipper protein (JLP), a scaffold protein involved in the structural and functional organization of the JNK/p38 mitogen-activated protein kinase module, tethers Gα₁₂ and Gα₁₃ to the JNK signaling module. In the present study, in addition to demonstrating the physical association between JLP and Gα₁₂, we show that this interaction is enhanced by the receptor- or mutation-mediated activation of Gα₁₂. We also establish that JLP interacts with Gα₁₂ through the C-terminal domain that has been previously identified to be involved in binding to Gα₁₃. Furthermore, using this C-terminal domain as a competitively inhibitor of JLP that can disrupt Gα₁₂-JLP interaction, we demonstrate that JLP is required for the stimulation of JNK by Gα₁₂. Our results also indicate that such JLP interaction is required for Gα₁₂ as well as Gα₁₃-mediated neoplastic transformation of JLP. These studies demonstrate for the first time a functional role for JLP in the gep oncogene-regulated neoplastic signaling pathway.

Introduction

The gep oncogenes, defined by the activated mutants of Gα₁₂ and Gα₁₃ [1,2], stimulate critical signaling pathways involved in cell growth, differentiation, and apoptosis [3–10]. Analyses of Gα_12/13-mediated cell proliferation, differentiation, and apoptosis have identified that the activation of c-Jun N-terminal kinase (JNK) plays a central role in all of the critical pathways activated by Gα_12/13 [2,3,7,11]. Our previous studies have shown that JNK-interacting leucine zipper protein (JLP) acts as a scaffold in linking Gα₁₂ and Gα₁₃ to the JNK signaling module [12]. Although these initial studies indicated the association between JLP and Gα₁₂ or Gα₁₃, the functional consequence of the interaction remains far from clear. Considering the observations that JNK plays a critical role in Gα₁₂-mediated cell proliferation [6,11,13,14], we sought to investigate the role of JLP in Gα₁₂-mediated activation of JNK and neoplastic transformation. In this report, we present our findings that JLP interacts with Gα₁₂ through a specific C-terminal domain-similar to Gα₁₃—and this interaction promotes Gα₁₂-mediated activation of JNK. Furthermore, we demonstrate that the coexpression of JLP along with Gα₁₂ greatly enhances the ability of Gα₁₂ to induce neoplastic transformation of NIH3T3 cells and the expression of the dominant negative inhibitory mutant of JLP drastically decreases the transforming ability of activated Gα₁₂ as well as Gα₁₃. Thus, our findings presented here for the first time demonstrate a unique role for JLP in the oncogenic pathway(s) activated by the gep oncogenes.

Materials and Methods

Cells, Plasmids, and Transfections

NIH3T3 and COS-7 cells were maintained in Dulbecco modified Eagle medium (Invitrogen Life Technologies, Inc, Carlsbad, CA) containing 10% fetal bovine serum (Invitrogen Life Technologies) and 1% penicillin/streptomycin at 37°C in a 5% CO₂ incubator. Expression vectors encoding S-epitope-tagged JLP, its mutant derivatives, Gα₁₂, and Gα₁₃ constructs have been previously described [3,10]. Transfection of COS-7 cells was carried out using FuGENE-6 reagent (Roche Applied Sciences, Indianapolis, IN) according to the manufacturer's protocol. Transfection of NIH3T3 cells was carried out using calcium phosphate precipitation method following the previously published procedures [9,12].

Coprecipitation and Immunoblot Analysis

Coprecipitation studies were carried out with COS-7 cells using S-protein agarose (Novagen, EMD Biosciences, Inc, Madison, WI) or antibodies specific to proteins of interest followed by protein A Sepharose beads (Amersham Biosciences Corp, Piscataway, NJ) as follows. For coprecipitation analyses with S-tagged JLP constructs, cells were lysed at 24 hours after transfection, and the lysate protein (500 µg each) was incubated with 30 µl of S-protein agarose for 4 hours at 4°C. After repeated washes with lysis buffer, the S-protein agarose-bound proteins were separated by SDS-PAGE and electroblotted on to polyvinylidene difluoride membranes for immunoblot analysis. Immunoblot analyses of lysates or immunoprecipitated proteins were carried out according to previously published procedures [9,12]. For immunoprecipitation analyses, cell lysate protein (500 µg each) was incubated with 1 µg of respective antibodies for 4 hours at 4°C followed by the addition of 20 µl of 50% slurry of protein A Sepharose. After repeated washes with cell lysis buffer, the immunoprecipitated proteins were separated by SDS-PAGE and electroblotted onto polyvinylidene difluoride membranes for immunoblot analysis. Antibodies to HA epitope (2362) and phosphorylated JNK (9251) were obtained from Cell Signaling Solutions (Beverly, MA), whereas antibodies to S-epitope (sc-802), JNK-1 (sc-474), and Gα₁₂ (sc-409) were obtained from Santa Cruz Biotechnology, Inc (Santa Cruz, CA). For immunoblot analyses with Gα₁₃, antibodies raised against the C-terminus of Gα₁₃ (AS1-89-2) that has been previously described were used [15]. After washing the immunoprecipitates twice with lysis buffer, the immunoprecipitated proteins were separated by SDS-PAGE and electroblotted on to PVDF membranes. Immunoblot analyses with specific antibodies were carried out following the previously published procedures [12].

Focus Formation Assay

NIH3T3 cell focus formation assay was carried out as previously described [9]. Parental NIH3T3 cells were transfected with pcDNA3 vectors encoding Gα₁₂QL or Gα₁₃QL (5 µg) along with or without pSG5 vector encoding JLP and different domain constructs of JLP including GID-JLP (5 µg) using the calcium phosphate transfection method. Control group included the transfections with empty pcDNA3 vector (10 µg). Transfected NIH3T3 cells were cultured in the presence of 5% serum, and transformed foci were stained and scored after 14 days.

Results

Interaction between the gep Oncogene Gα₁₂ and JLP

Our previous findings have shown that both Gα₁₂ and Gα₁₃ physically associate with JLP [12]. To analyze this further, especially the interaction between JLP and Gα₁₂, we investigated whether the interaction is dependent on the activated status of Gα₁₂. To test, COS-7 cells were cotransfected with pSG5-vector encoding S-protein epitope-tagged JLP (S-JLP) and pcDNA3 vector encoding Gα₁₂WT or constitutively activated, GTPase-deficient mutant of Gα₁₂QL. Transfectants were lysed at 48 hours, and the cell lysates were subjected to an S-protein agarose pull-down assay followed by immunoblot analysis with antibodies-specific to Gα₁₂. Results from such analysis indicated the presence of Gα₁₂ in JLP precipitates, thereby confirming our previous findings that Gα₁₂ physically associates with JLP (Figure 1A). More interestingly, the levels of JLP pulled down from cells expressing Gα₁₂QL showed an increase—although modest—in the associated Gα₁₂ levels compared to that of wild-type Gα₁₂ (lane 2 vs lane 3), indicating that the activated Gα₁₂ more preferentially interacts with JLP.

Figure 1.

Gα₁₂ interacts with JLP. (A) COS-7 cells were cotransfected with vectors encoding S-JLP along with wild-type form (Gα₁₂WT) or constitutively active form (Gα₁₂QL) of Gα₁₂ expression constructs. Lysates were collected after 48 hours, and the pull-down assay was carried out using S-agarose beads. Immunoblot analysis was carried out to assess the presence of Gα₁₂ in the S-JLP precipitates, using an antibody specific to Gα₁₂. Expression levels of JLP and Gα₁₂ were monitored by immunoblot analysis with Gα₁₂ or S-protein antibody. Results form a typical experiment presented here (n = 3). (B) COS-7 cells were cotransfected with vectors encoding S-JLP along with wild-type form (Gα₁₂WT). Transfectants were serum-starved for 24 hours after which the cells were stimulated with 20 µM of LPA for 10 minutes. After LPA stimulation, cells were lysed, and the lysates were subjected to S-agarose pull-down assay. Immunoblot analysis was carried out to assess the presence of Gα₁₂ in the S-JLP precipitates, using an antibody specific to Gα₁₂. Expression levels of JLP and Gα₁₂ were monitored by immunoblot analysis with Gα₁₂ or S-protein antibody. Results form a typical experiment from a set of three repeats.

To establish further that the activated conformation of Gα₁₂ interacts more efficiently with JLP, we investigated whether receptor-mediated activation of Gα₁₂ promotes an increase in its association with JLP. Therefore, we tested whether lysophosphatidic acid (LPA), which activates Gα_12/13 through specific LPA receptors, can stimulate the association of Gα₁₂ and JLP. Because COS-7 cells that are known to express LPARs [16], these cells were transfected with vectors encoding S-JLP and Gα₁₂WT. At 24 hours after transfection, the transfectants were serum starved for 24 hours after which they were stimulated with 20 µM LPA for 10 minutes. The lysates from these transfectants were subjected to S-protein agarose precipitation. The S-protein agarose precipitates were analyzed for the presence of Gα₁₂ by immunoblot analyses using Gα₁₂-specific antibodies. As shown in Figure 1B, stimulating the transfectants with LPA increased the levels of Gα₁₂ that get co-precipitated with S-JLP. Together with the increased association seen with mutationally activated Gα₁₂ (Figure 1A), these results suggest that both the mutational and receptor-mediated activation of Gα₁₂ can enhance its interaction with JLP.

Mapping the Gα₁₂-Interacting Domain of JLP

As a scaffold protein involved in JNK/p38 mitogen-activated protein kinase (MAPK) signaling, JLP contain specific motifs such as SH2, SH3, and leucine zipper domain to tether the functional constituents of the JNK module [17]. Previously, we have shown that the JNK-, KLC-1-, Gα₁₃-, and SCG10-interacting domains of JLP can be mapped to distinct nonoverlapping regions of JLP [12,17–19]. Therefore, we were interested in determining whether JLP interacts with Gα₁₂ through any of these known domains or a distinctly different one. To test, we analyzed its interactions with Gα₁₂ using S-tagged constructs encoding distinct domains of JLP (Figure 2A). COS-7 cells were transfected with vector encoding the constitutively active Gα₁₂QL along with S-tagged mutant constructs with different JLP domains for 48 hours. The transfectants were lysed and the lysates were precipitated with S-protein agarose beads,. and the presence of Gα₁₂ in the precipitates was examined by immunoblot analysis using an antibody specific to Gα₁₂. Results from this analysis indicated that Gα₁₂ could be coprecipitated with S-JLP only from the transfectants expressing the extreme C-terminal domain of JLP spanning amino acids 1165–1307 (Figure 2B). Thus, these results demonstrate that JLP interacts with Gα₁₂ through its extreme C-terminal domain, spanning 1165 to 1307 amino acids (domain IV). Interestingly, this is the same domain that was previously identified to be involved in JLP's interaction with Gα₁₃ [12]. Thus, the C-terminal domain encompassing 1165–1307 amino acids of JLP that was previously referred as Gα₁₃-interacting domain of JLP (GID-JLP) turns out to be an interacting domain for both Gα₁₂ and Gα₁₃.

Figure 2.

Gα₁₂ interacts with the C-terminal domain of JLP. (A) Schematic diagram representing JLP deletion mutant constructs encoding different domains. These constructs encode S-tagged JLP-domain I (amino acids 1–164), II (amino acids 165–473), IIΔ (amino acids 210–398), III (amino acids 465–1008), C (amino acids 997–1170), and IV (amino acids 1165–1307). (B) Vector constructs expressing Gα₁₂QL were coexpressed with the different domain mutants of S-JLP in COS-7 cells. After 48 hours, the lysates were subjected to the pulldown assay with the S-protein agarose beads. The coprecipitated Gα₁₂ was identified by immunoblot analysis using antibodies to Gα₁₂. (C) The expression levels of transfected Gα₁₂QL and JLP constructs were monitored by immunoblot analyses using antibodies to Gα₁₂ (upper panel) or S-tag (lower panel).

JLP Enhances Gα₁₂ Activation of JNK

Based on the findings that 1) Gα₁₂ interacts with JLP through the extreme C-terminal GID domain and 2) JLP interacts with JNK through the N-terminal domain [17], it can be hypothesized that JLP tethers Gα₁₂ to the JNK module, thereby promoting its ability to stimulate the JNK module. If this hypothesis is true, cotransfection of JLP should enhance Gα₁₂-mediated JNK response, whereas the inhibition of such interaction should attenuate Gα₁₂-mediated activation of JNK. In fact, our previous studies have shown that the coexpression of JLP stimulates LPA as well as Gα₁₃-mediated activation of JNK, whereas the coexpression of GID-JLP inhibits such activation [12]. To test whether the same holds true for Gα₁₂, COS-7 cells were transfected with the activated mutant of Gα₁₂ (Gα₁₂QL) in the absence or presence of JLP for 48 hours. The lysates from these transfectants were resolved by SDS-PAGE and subjected to immunoblot analysis using antibodies specific to the phosphorylated forms of JNK to monitor their activation profile. Results indicated that Gα₁₂ stimulated an increase in the activation of JNK (Figure 3, lane 3), and this could be further enhanced by the coexpression of JLP (Figure 3A, lanes 5 and 6), thus suggesting that JLP plays a potentiating role in the Gα₁₂-mediated activation of JNK signaling module. In contrast, when COS-7 cells were transiently transfected with vectors encoding Gα₁₂QL along with S-epitope-tagged GID-JLP for 48 hours, Gα₁₂-stimulated increase in the activation of JNK (Figure 3B, lane 2) was attenuated by the coexpression of GID-JLP (Figure 3B, lane 6). Taken together, these results establish a cardinal role for JLP in Gα₁₂-mediated stimulation of JNK.

Figure 3.

JLP enhances Gα₁₂-mediated activation of JNK. (A) COS-7 cells were transiently transfected with pcDNA3 vector control, wild-type (Gα₁₂WT), or constitutively active (Gα₁₂QL) forms of Gα₁₂ alone or along with JLP. Cells were collected after 48 hours of transfection. These lysates were subjected to immunoblot analysis with antibodies to phospho-specific JNK to assess its activation profile. After stripping, the blot was reprobed with JNK antibody to monitor equal expression levels of JNK. (B) COS-7 cells were cotransfected with constructs expressing JNK, Gα₁₂QL, and full-length S-JLP or mutant GID-JLP (bearing amino acids 1165–1307). Total cell lysates were subjected to immunoblot analysis and assessed for JNK phosphorylation using phospho-JNK antibody. Cell lysates were also analyzed for the expression of HA-JNK, Gα₁₂, full-length S-JLP, and mutant GID-JLP with their respective antibodies.

JLP Is Involved in gep Oncogene-Mediated Neoplastic Transformation

Our findings that JLP potentiates Gα₁₂-mediated activation of JNK gains greater significance in light of the previous observation from our laboratory as well as others that JNK plays a critical role in Gα₁₂-mediated mitogenic signaling pathways [3,5,6,10]. Therefore, we focused on evaluating whether the inhibition of Gα₁₂-mediated JNK response by the coexpression of GID-JLP would attenuate the neoplastic transformation mediated by the either of the gep oncogenes. Neoplastic transformation of NIH3T3 fibroblast cells and subsequent foci formation have been widely used to monitor the oncogenic potential of different oncogenes including Gα_12/13 [20–24]. To carry out this analysis, NIH3T3 cells were transfected with vectors encoding Gα₁₂QL, full-length JLP, and GID-JLP alone or in combinations. At 14 days after transfection, the number of foci formed was scored by fixing the cells using glutaraldehyde followed by crystal violet staining. Our results indicated that the expression of Gα₁₂QL induces foci formation in NIH3T3 cells confirming the previous results from several laboratories including ours [22–24]. The foci formation was greatly increased by the coexpression of JLP. Quite significantly, the coexpression of GID-JLP drastically inhibited the number of foci formed compared with Gα₁₂QL alone (Figure 4). More interestingly, the inhibitory effect of GID-JLP can be neutralized by the coexpression of full-length JLP (Figure 4), thereby establishing the critical role of JLP inGα₁₂-mediated neoplastic transformation. Quantification of the foci from repeat experiments indicated that the number of foci induced by Gα₁₂QL was increased by the coexpression of by 90%. In contrast, the coexpression of the dominant negative GID-JLP reduced Gα₁₂QL-induced foci formation by 70%. Interestingly, the coexpression of JLP along with GID-JLP rescued such GID-JLP mediated inhibition. Similar results were obtained using Gα₁₃QL, which is known to involve JNK signaling interacting with JLP (Figure 5A). Taken together, these results suggest that a functional interaction between Gα_12/13 and JLP is involved in neoplastic transformation mediated by the gep oncogenes. These results for the first time identify a novel role for JLP in the oncogenic pathways induced by the gep oncogenes Gα₁₂QL and Gα₁₃QL.

Figure 4.

Effect of JLP and GID-JLP on Gα₁₂QL-mediated foci formation. Parental NIH3T3 cells were transfected with empty vector pcDNA3 (VC), Gα₁₂QL, full-length JLP (FL-JLP), Gα₁₂QL + FL-JLP, GID-JLP, Gα₁₂QL + GID-JLP, or Gα₁₂QL + GID-JLP + FL-JLP using the calcium phosphate transfection method. (A) The representative foci formed at 14 days were stained photographed (4x). Results presented are from a typical experiment (n = 4). (B) The foci formed from four independent experiments were scored and plotted (mean ± SD).

Figure 5.

Effect of JLP and GID-JLP on Gα₁₃QL-mediated foci formation. (A) NIH3T3 cells were transfected with empty vector pcDNA3 (vector), full-length JLP (FL-JLP), GID-JLP, Gα₁₃QL, Gα₁₃QL + FL-JLP, Gα₁₃QL + GID-JLP, or Gα₁₃QL + GID-JLP + FL-JLP using the calcium phosphate transfection method. After 14 days, the number of foci from three independent experiments was scored and plotted (mean ± SD). (B) At 14 days after transfection, expressions of the encoded proteins by the transfected complementary DNA constructs were verified by immunoblot analysis using a parallel set of transfectants.

Discussion

Members of the gep oncogene family, defined by the activated mutants of Gα₁₂ and Gα₁₃, are the most potent G protein subunits that have been observed to promote oncogenic signaling [2]. It has been demonstrated that the activation JNK is critically required for the G₁-S transition stimulated by Gα_12/13 [6,10]. It has also been observed that Gα₁₂ differentially regulates JNK, extracellular signal-regulated kinase, and p38 MAPK modules [25]. Whereas these findings suggested the role for MAPKs, specifically JNK in the oncogenic signaling by the gep oncogenes, the possible role of a specific scaffold protein that can tether Gα₁₂ or Gα₁₃ to MAPK signaling is largely not understood until now. Our observation presented here demonstrating a novel interaction between Gα₁₂ and the JNK scaffold protein JLP and (Figures 1–3) as well as the critical role of JLP in Gα_12/13-mediated oncogenic signaling pathway (Figure 4) confirms and identifies JLP as a scaffold protein involved in G₁₂/13-mediated oncogenic signaling.

Our results also identify the C-terminal region of JLP, spanning amino acids 1165–1307, as the critical domain of interaction with Gα₁₂ (Figure 2). It is significant to note here that we have previously established this region as the domain that interacts with Gα₁₃ and hence referred to it as the Gα₁₃-interacting domain of JLP. Our present finding redefines this region as the domain that can interact with both Gα₁₂ and Gα₁₃. Similar to our previous results obtained with Gα₁₃, we have shown here that the coexpression of JLP enhances Gα₁₂-mediated JNK activation, whereas the overexpression of GID-JLP inhibits JNK activation. The fact that JLP interacts with both Gα₁₂ and Gα₁₃ through a single domain may be indicative that JLP binding to these α-subunits is mutually exclusive. If so, this would mean that the JLP interaction with Gα₁₂ or Gα₁₃ might be cell type- and/or context-specific. In this context, it should be noted here that JLP has been shown to interact with kinesin light chain 1 and this interaction has been ascribed to play a role in spatial regulation of JLP-associated signaling molecules [18]. This raises an interesting possibility that the spatiotemporal context involving JLP and its interacting proteins may determine whether JLP interacts with Gα₁₂ or Gα₁₃. Further studies should clarify whether such context-specific events determine which member of the gep oncogenes interacts JLP in vivo. Another set of observations that can support such context-specific signaling is our present observation that JLP is involved in Gα_12/13-mediated neoplastic transformation of NIH3T3 cells (Figures 4 and 5) in conjunction with our previous findings demonstrating that JLP is involved in Gα₁₃-specific retinoic acid-mediated endodermal differentiation of P19 embryonic carcinoma cells [26]. In this regard, the scaffolding role of JLP can be equated to the context-specific signaling role observed with the yeast MAPK scaffold protein Ste5 [27–29]. In yeast, upon stimulated by the mating pheromone, the scaffold protein Ste5 nucleates the assembly of the β-subunit of heterotrimeric G protein GPA (Gβ) to a kinase module consisting of MAP4K, MAP3K, MAP2K, defined by Ste20, Ste 11, Ste7 to activate Fus3, a MAPK [30,31]. Interestingly, it has been shown that, in the absence of any pheromone, Ste5 can also facilitate the assembly of the same Gβ-Ste20-Ste11-Ste7 complex to activate Kss1, a different MAPK, to elicit invasive growth response of haploid cells or vegetative growth of diploid cells [27]. Thus, in this model system, the differential context-specific recruitment of the downstream MAPKs, namely, Fus3 and Kss1 can drastically alter the final signaling outcome of the Gprotein-MAPK scaffold [30]. It is likely that JLP provide such context-specific scaffolding role for both oncogenic and differentiation by effectively tethering different downstream signaling components.

In addition to these observations, careful analysis of JNK activation by Gα₁₂ in the presence of JLP provides novel insight into the dynamic manner in which the scaffold protein such as JLP modulates the activity of the associated kinase. Whereas the expression of Gα₁₂ or Gα₁₂QL along with JLP leads to the maximal activation of JNK (Figure 3A, lanes 5 and 6), coexpression of epitope-tagged JNK seems to elicit only a moderate activation of JNK by Gα₁₂QL (Figure 3B, lane 2 vs lane 5). It is quite possible that the overexpression of JNK and the resultant increase in the basal levels of activated JNK (Figure 3B, lane 1)— possibly forming endogenous competing signaling complexes—leads to the squelching of Gα₁₂QL-JLP signaling. Further characterization of JLP and its interacting proteins should provide finer details on the dynamics of JLP and Gα₁₂ interaction in the activation of JNK. Nevertheless, the observation that the coexpression of GID-JLP attenuates Gα₁₂QL-stimulated JNK activity to basal levels (Figure 3B, lanes 3 vs 6) further confirms the crucial role played by JLP in Gα₁₂-mediated activation of JNK.

Finally, our results presented here point to a critical pathobiological role JLP can play in the oncogenic pathway stimulated by LPA. It has been well established that LPA plays a critical role in the genesis and progression of many different cancers [32–35]. It has also been shown that LPAR-deficient mouse embryonal fibroblasts show deficiency in the activation of JNKs. In this context, our observation that LPA stimulates the association of JLP to Gα₁₂ (Figure 1B) and our previous findings that LPA-stimulated JNK activity can be inhibited by the expression of GID-JLP that can inhibit JLP-Gα_12/13 interaction [12] gain greater significance. The observations that 1) LPA signaling is involved in the pathophysiology of many cancers [32–35], 2) LPA potently stimulates the gep oncogenes Gα_12/13 [13], 3) LPA stimulates the association between Gα₁₂ and JLP (Figures 1 and 2), and 4) both LPA- [12] and Gα₁₂-mediated stimulation of JNK is inhibited by GID-JLP (Figures 3 and 4) suggest a key role JLP in LPA-Gα_12/13-mediated oncogenic pathway. The observation that JLP plays a critical role in Gα_12/13-mediated neoplastic transformation (Figures 4 and 5) provides additional support for this view. Now that such basic paradigm is established, further studies should clarify the critical role of Gα_12/13 and JLP in LPA-promoted mediated tumorigenesis and tumor progression.

Abbreviations

JNK

c-Jun N-terminal kinase

JLP

JNK-interacting leucine zipper protein

MAPK

mitogen-activated protein kinase