RET/PTC (Rearranged in Transformation/Papillary Thyroid Carcinomas) Tyrosine Kinase Phosphorylates and Activates Phosphoinositide-Dependent Kinase 1 (PDK1): An Alternative Phosphatidylinositol 3-Kinase-Independent Pathway to Activate PDK1

Abstract

Thyroid cancers are a leading cause of death due to endocrine malignancies. RET/PTC (rearranged in transformation/papillary thyroid carcinomas) gene rearrangements are the most frequent genetic alterations identified in papillary thyroid carcinoma. Although the oncogenic potential of RET/PTC is related to intrinsic tyrosine kinase activity, the substrates for this enzyme are yet to be identified. In this report, we show that phosphoinositide-dependent kinase 1 (PDK1), a pivotal serine/threonine kinase in growth factor-signaling pathways, is a target of RET/PTC. RET/PTC and PDK1 colocalize in the cytoplasm. RET/PTC phosphorylates a specific tyrosine (Y9) residue located in the N-terminal region of PDK1. Y9 phosphorylation of PDK1 by RET/PTC requires an intact catalytic kinase domain. The short (iso 9) and long forms (iso 51) of the RET/PTC kinases (RET/PTC1 and RET/PTC3) induce Y9 phosphorylation of PDK1. Moreover, Y9 phosphorylation of PDK1 by RET/PTC does not require phosphatidylinositol 3-kinase or Src activity. RET/PTC-induced phosphorylation of the Y9 residue results in increased PDK1 activity, decrease of cellular p53 levels, and repression of p53-dependent transactivation. In conclusion, RET/PTC-induced tyrosine phosphorylation of PDK1 may be one of the mechanisms by which it acts as an oncogenic tyrosine kinase in thyroid carcinogenesis.

RET/PTC (REARRANGED IN transformation/papillary thyroid carcinomas) as a fusion kinase from chromosomal translocation analyses (1, 2). The protein acts as a thyroid-specific oncogenic kinase in the development of spontaneous and postradiation papillary thyroid cancers (3, 4). The transforming potential of RET/PTC is related to the activation of multiple signaling pathways through interactions with several adapters and signaling molecules, and tyrosine kinase activity. Wild-type c-RET and RET/PTC are autophosphorylated, and the phosphotyrosine residues act as docking sites for Grb2/Grb10 (5), phospholipase Cγ (6), Shc (Src homology 2 domain containing) (7), and enigma (8). These interactions result in the activation of Ras/MAPK (9, 10) and phosphatidylinositol 3-kinase (PI3K)/Akt pathways (11). RET/PTC displays ligand-independent intrinsic tyrosine kinase activity. Although inhibition of tyrosine kinase activity of RET/PTC is related to reduction in oncogenic potential (12–14), the direct substrates of this kinase remain to be identified.

Altered PI3K and Akt/protein kinase B (PKB) signaling pathways are involved in tumorigenesis (15). Phosphatase and tensin homolog (PTEN), a dual-function lipid and protein phosphatase that dephosphorylates phosphatidylinositol 3,4,5-triphosphate at the D3 position, acts as a negative regulator of PI3K signaling and is frequently mutated in the advanced stages of a number of human cancers (16). The absence of phosphatase and tensin homolog (PTEN) strongly correlates with Akt/PKB activation in tumor cell lines (17). Akt/PKB gene amplification is observed in ovarian, pancreatic, gastric, and breast cancers (18). Activation of Akt/PKB is regulated by the phosphorylation of T308 and S473 residues by phosphoinositide-dependent kinase 1 (PDK1) and PDK2, respectively (19). The finding that mouse embryonic stem cells lacking PDK1 do not display Akt/PKB activation by IGF-I confirms that PDK1 is a critical regulator of activation through phosphorylation of the T308 residue of Akt/PKB (20). Although the role for PDK1 in Akt/PKB activation is established, the mechanism of regulation of PDK1 activity remains a subject of controversy (21). Previously, it was hypothesized that PDK1 is constitutively active in resting cells and not further activated by growth factor stimulation (21). In addition, subcellular localization of PDK1 appears to be growth factor insensitive (22), although conflicting studies report growth factor-dependent translocation of PDK1 to the plasma membrane (23). Although several serine sites on PDK1 (S25, S241, S393/396, S410) are phosphorylated, only phosphorylation of the activation loop serine (S241) is necessary for PDK1 activity (24). Recent studies show that stimulation of cells with pervanadate and hydrogen peroxide induces tyrosine phosphorylation, subcellular redistribution, and activation of PDK1 (25). In addition, coexpression of v-Src tyrosine kinase results in phosphorylation and activation of PDK1, and in vitro treatment with v-Src induces phosphorylation of Y9 and Y373 residues of PDK1 (25).

We show that RET/PTC acts as a tyrosine kinase, and that coexpression of RET/PTC1 and RET/PTC3 with PDK1 results in tyrosine phosphorylation and PDK1 activation. Both RET/PTC and PDK1 colocalize in the cytoplasm but not in the plasma membrane. RET/PTC-mediated phosphorylation of PDK1 is not blocked by the PI3K inhibitors, wortmannin and LY294002, or the c-Src kinase inhibitor, PP1. RET/PTC decreases the wild-type p53 protein concentration and the level of p53-mediated transactivation. Inhibition of p53 activity by RET/PTC is dependent on RET/PTC-mediated Y9 phosphorylation of PDK1. These results suggest that the chimeric oncogenic kinase, RET/PTC, activates the PI3K-independent tyrosine phosphorylation of PDK1.

RESULTS

RET/PTC Phosphorylates a Tyrosine Residue of PDK1 in Vivo

RET/PTC1 and RET/PTC3 are the most frequently identified forms of RET/PTC rearrangement in papillary thyroid cancers. We employed three different forms of RET/PTC for expression: specifically, RET/PTC1 (iso 9), RET/PTC3 (iso 9), and RET/PTC3 (iso 51) (Fig. 1A). Initially, we tested for autophosphorylation of tyrosine residues of RET/PTC after transfection of RET/PTC1 and RET/PTC3 in Chinese hamster ovary (CHO) cells (Fig. 1B). A phosphotyrosine blot with 4G10 antibody revealed the corresponding 56-kDa and 76-kDa fragments of RET/PTC1 and RET/PTC3, respectively, suggesting that both isoforms have autokinase activity in expressed cells (Fig. 1B). In addition, we observed multiple bands of tyrosine-phosphorylated proteins in RET/PTC1 and RET/PTC3-transfected cells. These findings suggest that the RET/PTCs are autophosphorylated, and several cellular substrates are tyrosine phosphorylated by RET/PTC in vivo.

Figure 1.

RET/PTC-Induced Tyrosine Phosphorylation of Endogenous and Exogenous PDK1 A, Schematic representation of RET/PTC1 (iso 9), RET/PTC3 (iso 9), and RET/PTC3 (iso 51) isoforms. The vertical arrow indicates the length of RET/PTC. The fusion partner genes for RET/PTC1 and RET/PTC3 are H4 and ELE1, respectively. B, Autophosphorylation of tyrosine residues in RET/PTC isoforms. CHO cells were cultured in six-well plates until 80% confluence and transfected with RET/PTC1 (iso 9) (500 ng/well) and RET/PTC3 (iso 9) (500 ng/well). Lysates were analyzed by SDS-PAGE. Tyrosine phosphorylation of RET/PTC1 and RET/PTC3 was detected with an antiphosphotyrosine antibody (4G10). A repetition of these experiments gave similar results. C, RET/PTC3-induced tyrosine phosphorylation of PDK1. CHO cells were cultured in 6-cm dish plates until 80% confluence and transfected with RET/PTC3 (1 μg/dish) and Myc-PDK1 (1 μg/dish). At 24 h after transfection, whole-cell lysates were immunoprecipitated with monoclonal anti-Myc antibody (Cell Signaling Technology, Inc.) and blotted with antiphosphotyrosine antibody (4G10). The figure is representative of at least three separate experiments. D, Tyrosine phosphorylation of endogenous PDK1 by RET/PTC3. The CHO cells were cultured in 10-cm dish plates until 80% confluence, and transfected with RET/PTC3 (4 μg/dish). At 24 h after transfection, whole-cell lysates were immunoprecipitated with anti-PDK1 antibody (Upstate Biotechnology, Inc.) and blotted with antiphosphotyrosine antibody (4G10). The figure is representative of at least three separate experiments.

PDK1 was identified as one of the tyrosine-phosphorylated proteins in RET/PTC1- and RET/PTC3-transfected CHO cells. To confirm RET/PTC-induced tyrosine phosphorylation of PDK1, we immunoprecipitated Myc-tagged PDK1 in CHO cells cotransfected with RET/PTC3 and Myc-PDK1 and performed immunoblot analyses with the antiphosphotyrosine antibody, 4G10 (Fig. 1C). Myc-PDK1 immunoprecipitated from cells transfected with RET/PTC3 contained a broad phosphotyrosine band, suggesting migration retardation of the phosphorylated PDK1 protein (Fig. 1C, lane 2). Immunoprecipitated PDK1, which was not coexpressed with RET/PTC3, displayed minimal basal tyrosine phosphorylation (Fig. 1C, lane 3). PDK1 was immunoprecipitated by anti-PDK1 antibodies (Upstate Biotechnology, Inc., Lake Placid, NY), and immunoblot analyses were performed with the antiphosphotyrosine antibodies, 4G10 (Fig. 1D), to observe the tyrosine phosphorylation of endogenous PDK1 by RET/PTC3. Tyrosine phosphorylation of PDK1 could be detected in the immunoprecipitates obtained by anti-PDK1, but not in the control IgG (Fig. 1D). These observations suggest that both the endogenous and exogenous PDK1 is tyrosine phosphorylated by RET/PTC in vivo.

We next determined whether RET/PTC3 interacts with PDK1. After coexpression of RET/PTC3 and Myc-PDK1 in CHO cells, we immunoprecipitated RET/PTC3 with an anti-RET antibody specific for the C-terminal domain of RET/PTC. Immunoprecipitated complexes were separated by 10% SDS-PAGE and analyzed with anti-Myc and antiphosphotyrosine antibodies (Fig. 2A). Myc-PDK1 was detected in a complex that coimmunoprecipitated with the anti-RET antibody. Reprobing the anti-Myc-PDK1 blot with antiphosphotyrosine antibodies to determine the Myc-PDK1 phosphorylation status in the RET/PTC3-Myc-PDK1 coimmunoprecipitate indicated that Myc-PDK1 was tyrosine phosphorylated. These findings suggest that RET/PTC3 interacts with PDK1 and induces tyrosine phosphorylation of the protein (Fig. 2A).

Figure 2.

Interactions and Colocalizations between RET/PTC and PDK1 A, Interactions of RET/PTC with PDK1. CHO cells were cultured in 6-cm dishes until 80% confluence and transfected with RET/PTC3 (1 μg/dish) and Myc-PDK1 (1 μg/dish). After 24 h transfection, whole-cell lysates were immunoprecipated with anti-RET antibody (Santa Cruz Biotechnology, Inc.) and blotted with anti-RET, anti-Myc, and anti-phosphotyrosine antibodies (4G10). B, Colocalizations between RET/PTC and PDK1. CHO cells were transiently transfected with RET/PTC3 and pEGFP-PDK1 by lipofectAMINE (Invitrogen) (see Materials and Methods). At 24 h after transfection, cells were fixed in 3.7% formaldehyde for 40 min. Fixed cells were mounted onto glass slides with PBS and observed with a laser-scanning confocal microscope. GFP-fused wild-type PDK1 was detected by autofluorescence, and RET/PTC3 was detected by staining with primary anti-RET antibody and rhodamine-conjugated secondary antibody. The yellow stain in the merged image depicts colocalization of RET/PTC3 and PDK1. Shown are representative cells from one of three independent experiments.

Immunofluorescence confocal microscopy was performed on the CHO cells cotransfected with pEGFP-PDK1 and pCDNA3-RET/PTC3 to determine whether RET/PTC and PDK1 are colocalized within the cells. Green fluorescent protein (GFP) fluorescence from expressed PDK1 was evenly distributed in the cytoplasm and plasma membrane (Fig. 2B). The RET/PTC3 fluorescence pattern was very similar to that of PDK1, except that no fluorescence was observed in the plasma membrane. RET/PTC3 colocalized with PDK1 in the cytoplasm, as depicted in the merged image (Fig. 2B). However, RET/PTC3 was not colocalized in the plasma membrane. The data collectively suggest that RET/PTC3 and PDK1 colocalize mainly in the cytoplasm, rather than in the plasma membrane.

Identification of the Tyrosine Residues in PDK1 Phosphorylated by RET/PTC3

Next, we identified the tyrosine residues of PDK1 phosphorylated by RET/PTC3. Previous experiments revealed that phosphorylation of Y9 and Y373/6 of PDK1 was induced by pervanadate (25). In addition, these residues were shown to be the phosphorylation sites by experiments using v-Src (25). Phospho- specific antibodies that react specifically with the phosphorylated Y9 and Y373/6 residues of PDK1 were used to determine the phosphorylated state of PDK1 after phosphorylation by RET/PTC3. The cell lysates obtained by pervanadate (100 μm) treatment were used as a positive control for the detection of Y9 and Y373/6 phosphorylation of PDK1. The phospho-specific antibodies for the phosphorylated Y9 and Y373/6 residues of PDK1 could detect the tyrosine phosphorylation of the Y9 and Y373/6 residues of the PDK1 in the CHO cell lysates treated with pervanadate (Fig. 3A, lanes 5 and 6). The cell lysate obtained from CHO cells cotransfected with RET/PTC3 and Myc-PDK1 reacted with pY9-specific antibodies, but did not react with pY373/6 antibodies (Fig. 3A, lane 4). This finding indicates that RET/PTC3 specifically phosphorylates the Y9 residue of PDK1, but it did not affect the tyrosine phosphorylation of the Y373/6 residues in PDK1. To confirm the Y9 phosphorylation of PDK1 by RET/PTC3, the RET/PTC3 was cotransfected with the wild-type PDK, as well as the PDK1-Y9F and PDK1-Y373/6F mutants into CHO cells, and the phosphorylation status was detected by immunoblotting. Again, coexpression of Myc-PDK1 and RET/PTC3 resulted in tyrosine phosphorylation of PDK1 (Fig. 3B, lane 2). As expected, tyrosine phosphorylation was markedly decreased upon coexpression of Myc-PDK1 Y9F and RET/PTC3 (Fig. 3B, lane 2 vs. lane 4). However, the tyrosine phosphorylation of PDK1-Y3 73/6F was not abolished but showed a tyrosine phosphorylation level that was similar to the wild-type PDK1 in RET/PTC3-transfected cells (Fig. 3B, lane 6). Our data conclusively show that the Y9 residue of PDK1 is a major phosphorylation target of RET/PTC3.

Figure 3.

Determination of the Tyrosine Residue in PDK1 Phosphorylated by RET/PTC A, Determination of the tyrosine residue in PDK1 phosphorylated by RET/PTC. CHO cells cultured in six-well plates were transfected with RET/PTC3 (500 ng/well) and Myc-PDK1 (500 ng/well). The transfected cells were treated with pervanadate (100 μm) for 15 min (lanes 5 and 6). The whole-cell lysates were prepared and blotted with anti-phospho-specific PDK1 antibodies (pY9-PDK1 and pY373/6-PDK1). The figure is representative of at least three separate experiments. B, Decreased tyrosine phosphorylation in PDK1 on Y9F mutation. CHO cells cultured in six-well plates were transfected with RET/PTC3, Myc-PDK1, and Myc-PDK1-Y9F mutants. After 24 h transfection, whole-cell lysates were prepared and analyzed by SDS-PAGE. Differences in the migration of tyrosine-phosphorylated PDK1 were observed by immunoblotting with anti-phospho-specific PDK1 antibodies (anti-pY9-PDK1 and anti-pY373/6-PDK1). The figure is representative of at least three separate experiments.

We further investigated whether RET/PTC variants, specifically RET/PTC1 and the long form of RET/PTC3, designated RET/PTC3 (iso 51), phosphorylate Y9 of PDK1 (Fig. 4). Cells were cotransfected with Myc-PDK1 and either RET/PTC1, RET/PTC3, or RET/PTC3 (iso 51) and Y9 phosphorylation was analyzed using an anti-pY9 phospho-specific antibody. Both RET/PTC3 and RET/PTC1 phosphorylated the Y9 residue of PDK1 (Fig. 4A). The short (iso 9) and long (iso 51) forms of RET, generated by alternative splicing of the wild-type c-RET gene, displayed different signaling characteristics (45). However, as observed in Fig. 4B, the long form, RET/PTC3 (iso 51), induced the Y9 phosphorylation of PDK1, and the level of Y9 phosphorylation of PDK1 by RET/PTC3 (iso 51) was similar to the level induced by the short form, RET/PTC (iso 9). These observations suggest that the N-terminal and C-terminal regions of RET/PTC, which are generated by the ret-fused genes and by alternative splicing, respectively, were not major determinants for the Y9 phosphorylation of PDK1 in vivo. Collectively, these observations indicate that the kinase domain of RET/PTC is critical for the tyrosine phosphorylation of PDK1.

Figure 4.

Tyrosine Phosphorylation of PDK1 by RET/PTC Variants and Isoforms A, PDK1 tyrosine phosphorylation by RET/PTC variants. CHO cells cultured in six-well plates were transiently transfected with RET/PTC1 (500 ng/well), RET/PTC3 (500 ng/well), and Myc-PDK1 (500 ng/well). At 24 h after transfection, total lysates were prepared and blotted with anti-phospho- specific PDK1 antibody (pY9-PDK1) and monoclonal anti-Myc and anti-RET antibodies. B, PDK1 tyrosine phosphorylation by RET/PTC3 isoforms. CHO cells cultured in six-well plates were transfected with RET/PTC3 (iso 9) (500 ng/well), HA-tagged RET/PTC3 (iso 51) (500 ng/well), and Myc-PDK1 (500 ng/well). After 24 h transfection, total lysates were prepared and blotted with anti-phospho-specific PDK1 (pY9-PDK1), anti-Myc, and anti-HA antibodies. RET/PTC3 expression (iso 51) was detected using an anti-HA antibody. The figure is representative of at least three separate experiments.

Tyrosine Phosphorylation of PDK1 by RET/PTC Requires Intrinsic Tyrosine Kinase Activity, But Not Src or PI3K Activities

We next determined the role of intrinsic tyrosine kinase activity of RET/PTC in the phosphorylation of PDK1. We generated a construct, RET/PTC3 K284M, with no tyrosine kinase activity as a result of mutations in the kinase catalytic domain. Transfection of the kinase-inactive RET/PTC3 K284M mutant failed to induce Y9 phosphorylation (Fig. 5, lane 2 vs. lane 3). The results strongly indicate that this Y9 residue of PDK1 is a specific substrate of RET/PTC3, and that Y9 phosphorylation of PDK1 requires the intrinsic tyrosine kinase activity of RET/PTC. Y588 of RET/PTC3 corresponds to Y1062 of wild-type c-RET receptor tyrosine kinase, which recruits phosphotyrosine binding domain-containing proteins, such as Shc (5). This phosphotyrosine residue is critical for the activation of PI3K and Shc-mediated signaling pathways (11). To determine whether Y588 plays a role in the Y9 phosphorylation of PDK1, experiments were performed by expressing the mutant RET/PTC3 Y588F. RET/PTC3 Y588F induced Y9 phosphorylation of PDK1 in a manner similar to that of RET/PTC3 (Fig. 5, lane 4). Our data indicate that Shc-mediated binding of Grb2, Gab2, and the p85-regulatory subunit of PI3K to Y588 is not a prerequisite for Y9 phosphorylation of PDK1.

Figure 5.

Requirement of Intrinsic Tyrosine Kinase Activity of RET/PTC for Phosphorylation of PDK1 CHO cells cultured in six-well plates were transfected with RET/PTC3 (500 ng/well), RET/PTC3 (K284M) (500 ng/well), RET/PTC3 (Y588M) (500 ng/well), and Myc-PDK1 (500 ng/well). Total cell lysates were prepared and blotted with anti-phospho-specific PDK1 antibody (pY9-PDK1), monoclonal anti-Myc, antiphosphotyrosine (4G10), and anti-RET antibodies. The figure is representative of at least three separate experiments.

The finding that RET/PTC3 Y588F induces Y9 phosphorylation suggests that PI3K is not involved in this process, because Y588 (Y1062 in wild-type c-RET) is critical for PI3K activation (11). We determined the roles of PI3K in RET/PTC-induced Y9 phosphorylation of PDK1 using the inhibitors, wortmannin and LY294002. The selected inhibitors did not affect RET/PTC3-induced Y9 phosphorylation, as shown in Fig. 6A (lanes 4 and 5). In addition, the Src inhibitor (PP1), PDK1 inhibitor (N-α-tosyl-l-phenylalanyl chloromethyl ketone), and the mammalian target of rapamycin (mTOR) kinase inhibitor (rapamycin) did not hinder Y9 phosphorylation induced by RET/PTC3 (Fig. 6A, lanes 7 and 8).

Figure 6.

P13K- and Src-Independent Tyrosine Phosphorylation of PDK1 by RET/PTC A, Effects of PI3K inhibitors on RET/PTC-mediated tyrosine phosphorylation of PDK1. CHO cells cultured in six-well plates were transfected with RET/PTC3 (500 ng/well) and Myc-PDK1 (500 ng/well). After 24 h transfection, we treated PI3K inhibitors [wortmannin (W, 100 nm) and LY294002 (LY, 1 μm)], Src kinase inhibitor, PP1 (5 μm), PDK1 inhibitor, N-α-tosyl-l-phenylalanyl chloromethyl ketone (TK, 50 μm), and mammalian target of rapamycin (mTOR) inhibitor, rapamycin (10 ng/ml), for 45 min. The figure is representative of at least three separate experiments. B, Noninvolvement of Src kinase in RET/PTC-induced tyrosine phosphorylation of PDK1. Src(−/−) MEF cells were cultured in six-well plates until 80% confluence and transfected with RET/PTC3 and Myc-PDK1. After 24 h transfection, total lysates were prepared and blotted with anti-phospho-specific PDK1 antibody (pY9-PDK1), monoclonal anti-Myc, and anti-RET antibodies. The figure is representative of at least three separate experiments.

Earlier reports suggest that tyrosine phosphorylation of PDK1 by pervanadate is a result of Src tyrosine kinase activity (25). The role of Src in RET/PTC-mediated tyrosine phosphorylation of PDK1 was investigated in this study. We employed c-Src-deficient mouse embryo fibroblasts (MEFs) from c-Src null mice to determine the involvement of Src in RET/PTC3-induced Y9 phosphorylation of PDK1. Wild-type Src (+/+) MEFs and Src (−/−) MEFs were cotransfected with RET/PTC3 and Myc-PDK1, and Y9 phosphorylation of PDK1 was observed. RET/PTC3 induced Y9 phosphorylation in both Src (+/+) (data not shown) and Src (−/−) cells (Fig. 6B, lane 4). These results and the findings that the Src inhibitor, PP1, did not affect RET/PTC3-induced Y9 phosphorylation (Fig. 6A, lane 6) support the hypothesis that Y9 phosphorylation of PDK1 by RET/PTC3 results from direct phosphorylation events that do not require endogenous Src tyrosine kinase.

RET/PTC3 Increases Y9 Phosphorylation-Dependent PDK1 Activity and Modulates Akt/PKB Kinase

The regulation of PDK1 activity is poorly understood. Phosphorylation of multiple serine and tyrosine residues and subcellular distribution are suggested as the major regulatory mechanisms for PDK1 activity (21). To determine whether RET/PTC affects PDK1 activity, we measured phosphotransferase activities of immunoprecipitated Myc-PDK1 using a synthetic peptide (suntide) encompassing the T308 phosphorylation site of Akt/PKB (Fig. 7A) (25). When the Myc-PDK1 transfected cells were treated with pervanadate (100 μm), a well-characterized activator of PDK1 (25), the immunoprecipitated Myc-PDK1 activity was increased approximately 2.6-fold compared with the untreated Myc-PDK1 transfected cells (Fig. 7A). In addition, pervanadate increased the Myc-PDK1 activity in the CHO cells transfected with the Myc-PDK1 Y9F construct. These observations were similar to those reported previously in HEK 293 cells: pervanadate increased the kinase activities of Myc-PDK1, wild type, and the Y9F mutant (25). Coexpression with RET/PTC3 caused up to a 2.1-fold increase in Myc-PDK1 (compared with the Myc-PDK1 cells without the RET/PTC3 transfection) activity accompanied by RET/PTC3 transfection without any cell stimulation in three independent experiments (Fig. 7A). Coexpression with the kinase-inactive RET/PTC3 K284M mutant (see Fig. 5) did not affect PDK1 activity. These findings suggest that tyrosine phosphorylation by RET/PTC3 increases PDK1 activity in vivo. To further confirm the involvement of RET/PTC3-induced Y9 phosphorylation in the regulation of PDK1 activity, cells were cotransfected with RET/PTC3 and the PDK1 mutant Y9F construct, and measured for PDK1 activity. Mutation of the Y9 residue markedly inhibited the RET/PTC3-induced increase in PDK1 activity (Fig. 7A). Therefore phosphorylation of the Y9 residue by RET/PTC3 independently increases PDK1 activity in vivo.

Figure 7.

Regulation of PDK1 Activity by RET/PTC A, Increased total PDK1 kinase activity via tyrosine phosphorylation by RET/PTC3. CHO cells were cultured in 6-cm dishes until 80% confluence and transfected with RET/PTC3 (1 μg/dish), RET/PTC3 (K758M) (1 μg/dish), Myc-PDK1 (1 μg/dish), and Myc-PDK1-Y9F (1 μg/dish). The transfected cells were treated with pervanadate (100 μm) for 15 min. Total lysates were prepared and immunoprecipitated with monoclonal anti-Myc antibody. Immunoprecipitated wild-type PDK1 activity was measured using substrates such as suntide (RRKDGATMKTFCGTPE, 100 μm) encompassing the Thr-308 phosphorylation site of AKT/PKB (described in Materials and Methods). The activity of the immunoprecipitated PDK1 in the cells not transfected with RET/PTC3 was taken as 1. The kinase activity is the average ± sd of three independent experiments done in duplicate. B, Increased membrane PDK1 kinase activity by RET/PTC3 transfection. CHO cells were cultured in 10-cm dishes until 80% confluence and transfected with RET/PTC3 (4 μg/dish) and/or Myc-PDK1 (4 μg/dish). The Myc-PDK1 transfected cells were treated with pervanadate (100 μm) for 15 min. The plasma membrane fractions were prepared as described in Materials and Methods. The activity of the immunoprecipitated Myc-PDK1 is an average (±sd) of three independent experiments with duplicate immunoprecipitates. The activity of the immunoprecipitated PDK1 in the cells not transfected with RET/PTC3 was taken as 1.

This study showed that Myc-PDK1 and RET/PTC are mainly colocalized in the cytoplasm, but not in the plasma membrane. It has been reported that the subcellular fractionation of cells revealed a marked recruitment of PDK1 to the plasma membrane after the pervanadate treatment (25). The fractions of the crude plasma membrane from the CHO cells transfected with or without RET/PTC3 and Myc-PDK1 were obtained, and the PDK1 activity was measured. Interestingly, the PDK1 activity in the membrane fraction of the RET/PTC3-transfected cells was significantly higher than that of the RET/PTC3-untransfected cells (Fig. 7B). These observations suggest that PDK1 may translocate into the plasma membrane after the tyrosine phosphorylation by RET/PTC3 in the cytoplasm.

We detected T308 phosphorylation in Akt/PKB after coexpression of RET/PTC3, Myc-PDK1, and hemagglutinin (HA)-Akt/PKB in the CHO cells using phospho-specific antibodies (Cell Signaling Technology, Beverly, MA). Cells transfected with Akt/PKB and PDK1 in the absence of RET/PTC3 did not contain detectable levels of phosphorylated Akt/PKB (Fig. 8). However, after cotransfection of Akt/PKB, PDK1, and RET/PTC3, T308-phosphorylated Akt/PKB was observed in CHO cells. These observations indicate that RET/PTC3-mediated tyrosine phosphorylation induces the activation of the PDK1 downstream signaling molecule, Akt/PKB.

Figure 8.

RET/PTC3 Dose-Dependent AKT Thr-308 Phosphorylation CHO cells cultured in six-well plates were transfected with HA-AKT (500 ng/well), Myc-PDK1 (500 ng/well), and RET/PTC3 at the indicated doses. After a 24-h transfection, total cell lysates were prepared and blotted with anti-Akt/PKB, monoclonal anti-Myc, and anti-RET antibodies. The figure is representative of at least three separate experiments.

RET/PTC3-Induced PDK1 Activation Modulates p53 Levels and Transactivating Activities

Mutations in p53 are infrequently observed in papillary thyroid cancers. However, to date, qualitative defects in p53 function have not been evaluated in papillary thyroid cancers, which are known to have RET/PTC rearrangements. Recent studies show that activation of Akt/PKB, which lies downstream of PDK1, is involved in ubiquitination and degradation of p53 (26). We observed that RET/PTC3 affected p53 levels in HCT116 cells carrying the wild-type p53 tumor suppressor protein. RET/PTC3 was introduced in the HCT116 cell by transfection using lipofectAMINE PLUS (Invitrogen, San Diego, CA). The transfection efficiency was evaluated using pCMV-SPORT-βgal (500 ng) and pEGFP-PDK1 (1500 ng) in a six-well culture plate. This transfection method typically showed up to 30–40% transfection efficiency with lipofectAMINE PLUS in the HCT 116 cells (data not shown). As shown in Fig. 9A, PTC3 expression was evident with the transfection of 500 ng of RET/PTC3, and the expressed level was further increased by escalating the RET/PTC3 DNA doses to 1500 ng/well. Interestingly, the p53 levels decreased with increasing doses of the transfected RET/PTC3, as shown in the cells transfected with 500 ng/well of RET/PTC3 (Fig. 9A). Because the p53 level in HCT116 cells that are transiently transfected with RET/PTC3 reflects the p53 levels in both the transfected and nontransfected cells, we observed the p53 level in the HCT116 cells that stably expressed RET/PTC3. The p53 levels in the five clones of the RET/PTC3-expressing stable cells were significantly lower than the control stable cell lines (data not shown).

Figure 9.

RET/PTC3-Induced PDK1 Activation Modulates p53 Levels and Transactivating Activities A, Regulation of endogenous P53 protein level by RET/PTC expression. Wild-type p53-expressing HCT116 cells were maintained in DMEM with 10% fetal bovine serum. HCT116 cells were cultured in six-well plates until 80% confluence and transfected with RET/PTC3 at the indicated doses. The equivalent dose of the empty vector pcDNA3 was used for the control experiments. Total cell lysates were prepared and blotted with anti-p53, anti-RET, and anti-Actin antibodies. The figure is representative of at least three separate experiments. B, Decreased p53 transactivation by RET/PTC3. HCT116 and HCT116 (p53(−/−) cells were cultured in 12-well plates until 80% confluence and transfected with RET/PTC3 (500 ng/well) and p53-luc (100 ng/well). To normalize gene expression, we used pRL-SV40 plasmid encoding Renilla luciferase. After 24 h transfection, total cells were washed with PBS and lysed. Luciferase activity was measured with a luminometer. The luciferase activity is the average ± sd of three independent experiments done in triplicate. C, Inhibition of p53 transactivation by RET/PTC3. HCT116 cells were cultured in 12-well plates until 80% confluence and transfected with p53 (100 ng/well), pRL-SV40 plasmid (50 ng/well), and RET/PTC3 at the indicated doses. After 24 h transfection, total cell lysates were prepared and luciferase activity measured. The luciferase activity is the average ± sd of three independent experiments done in triplicate.

To determine the effects of RET/PTC3 on the transcriptional activity of p53, HCT116 was cotransfected with p53-responsive reporter constructs and RET/PTC3 (500 ng/well). In HCT 116 cells transfected with RET/PCT3 alone, about 30% of p53-responsive reporter activity was observed, compared with cells transfected with the pcDNA3 vector (Fig. 9B). HCT116 cells with no wild-type p53 (HCT116-null p53) displayed very low p53 transactivation and no RET/PTC3-induced decrease of p53-responsive reporter activity (Fig. 9B). To confirm the inhibitory effects of RET/PTC3 on p53 activity, we simultaneously expressed p53 and RET/PTC3 with p53-responsive reporter constructs. Increasing doses of RET/PTC3 resulted in decreased p53-mediated transactivation in a dose-dependent manner in HCT116 cells (Fig. 9C). These observations suggest that RET/PTC3 decreases intracellular levels and inactivates endogenous and exogenous p53.

As shown in Fig. 3, RET/PTC3 induced Y9 phosphorylation in PDK1. Next, the specific role of Y9 in RET/PTC-mediated repression of p53 activity was investigated (Fig. 10). Coexpression of RET/PTC3 and wild-type PDK1 enhanced repression of p53. However, coexpression of mutant PDK1-Y9F with RET/PTC3 did not affect p53 activity. PDK1 Y393/6F construct still mediated RET/PTC3-induced inhibition of p53-mediated transactivation. These findings strongly suggest that Y9 phosphorylation of PDK1 by RET/PTC is involved in the down-regulation of p53 activity.

Figure 10.

Requirements of PDK1-Y9 in the RET/PTC-Mediated Repression of p53 Activity HCT116 cells were cultured in 12-well plates until 80% confluence and transfected with RET/PTC3 (100 ng/well), p53-luc (100 ng/well), Myc-PDK1 (400 ng/well), Myc-PDK1-Y9F (400 ng/well), Myc-PDK1-Y373/6F (400 ng/well), and pRL-SV40 plasmids (50 ng/well). Total lysates were washed with PBS and lysed. Luciferase activity was measured with a luminometer. The luciferase activity is the average ± sd of three independent experiments done in triplicate.

DISCUSSION

The thyroid-specific oncogenic kinase, RET/PTC, is generated by the fusion of the RET tyrosine kinase domain to the 5′-terminal region of heterologous genes (27). The enzyme was first described in 1987, when a rearranged form of the RET protooncogene was shown to transform NIH3T3 cells (28). RET/PTC1 and RET/PTC3 are the most frequent forms of rearrangement arising from chromosome 10 inversions (29, 30). The rearrangements result in constitutive activation of RET tyrosine kinase. The oncogenic potential of RET/PTC is a result of its intrinsic tyrosine kinase activity and activation of cellular signaling pathways involved in transformation. In vivo oncogenic effects of RET/PTC in thyroid carcinogenesis have been demonstrated in transgenic mouse models expressing RET/PTC1 and RET/PTC3 tyrosine kinases under the control of thyroid-specific promoters (31). These studies, collectively, suggest that RET/PTC rearrangements are sufficient for the transformation of thyroid cells, and that intrinsic tyrosine kinase activity is required for RET/PTC-mediated transformation.

Although the specific phosphorylation motifs and substrates of RET/PTC tyrosine kinase are yet to be identified, the enzyme shares some features with Src tyrosine kinase. Tyrosine phosphorylation of numerous intracellular target substrates by c-Src tyrosine kinase results in cellular differentiation, transformation, motility, and adhesion (32, 33). A recent report revealed phosphorylation of two tyrosine residues, Y9 and Y373, in PDK1 by Src tyrosine kinase (25). PDK1 plays a central role in numerous cell signaling pathways. Several substrates of PDK1 have been identified, including protein kinase B (Akt/PKB) (19, 34, 35), p70 ribosomal protein S6 kinase (p70S6K1) (36, 37), cyclic A-dependent protein kinase A (38), protein kinase C (39), serum and glucocorticoid-inducible kinase (40), p90 ribosomal protein S6 kinase (41, 42), and p21-activated kinase-1 (PAK1) (43).

Here we provide evidence that RET/PTC phosphorylates a specific tyrosine residue in PDK1, resulting in increased PDK1 activity, both in vitro and in vivo. Our data show that Y9 of PDK1 is a preferential target of RET/PTC-induced phosphorylation. Earlier reports show that this Y9 residue is an inducible phosphorylation site of pervanadate in HEK 293 cells (25). Pervanadate treatment in HEK293 cells results in the phosphorylation of not only Y9, but also Y373/6, in PDK1 (25). Y9 and Y373 are also phosphorylated by v-Src in vitro (25). Coexpression of v-Src leads to tyrosine phosphorylation of PDK1 without growth factor stimulation, as well as an increase in PDK1 activity (25). Y9 is located at the N terminus of PDK1, whereas Y373/6 are located in the linker region between the catalytic and pleckstrin homology domains. RET/PTC displayed differential activity on the tyrosine residues of PDK1, unlike Src kinase. Although v-Src induced tyrosine phosphorylation on two sites (Y9 and Y373), RET/PTC phosphorylated only Y9. Mutation of Y9 abolished most of the RET/PTC-induced tyrosine-phosphorylated forms of PDK1. Phospho-specific antibodies to p373/6 did not detect the phosphorylated PDK1 coexpressed with RET/PTC3, whereas this antibody reacted with the lysate obtained from the cells treated with pervanadate (Fig. 3). These findings raise the possibility that Src is not involved in RET/PTC-induced Y9 phosphorylation of PDK1. In concurrence with these results, the Src inhibitor, PP1, did not block RET/PTC-induced Y9 phosphorylation. Additionally, RET/PTC actively phosphorylated Y9 in Src-deficient MEF cells. These data suggest that c-Src is not involved in RET/PTC-mediated Y9 phosphorylation of PDK1. Y9 is not conserved in PDK1 homologs from Drosophila melanogaster, Saccharomyces cerevisiae, Schizosacharomyces pombe, and Caenorhabditis elegans. Thus, regulation of PDK1 by Y9 phosphorylation appears to be a relatively recent event during evolution.

The receptor tyrosine kinase, RET, is alternatively spliced to yield at least three functional isoforms (iso 9, iso 43, and iso 51), which differ only in their carboxyl termini. The signaling complex associated with iso 9 was markedly different from that of iso 51 (44, 45). These data provide a biochemical basis for the functional differences between iso 9 and iso 51 RET/PTC proteins in vivo. However, in our experiments, short (iso 9) and long (iso 51) forms of RET/PTC3 were equally effective in inducing Y9 phosphorylation in PDK1. Furthermore, RET/PTC1 variants similarly induced phosphorylation of Y9 in PDK1, compared with RET/PTC3. These observations suggest that the conserved catalytic domain of RET/PTC proteins is mainly involved in the interactions and catalytic processes for tyrosine phosphorylation in PDK1.

Wild-type c-RET activates PI3K-dependent signaling pathways through interactions with several adapters and the regulatory p85 subunit of PI3K (46, 47). However, the RET/PTC3-induced Y9 phosphorylation of PDK1 was not affected by the PI3K inhibitors. We conclude that PI3K activation is not a prerequisite condition for tyrosine phosphorylation of PDK1. In fact, RET/PTC3 is not located in the plasma membrane. Moreover, RET/PTC3 and PDK1 colocalize only to the cytoplasm and not to the plasma membrane (Fig. 2B). Mutant PDK1 with no functional pleckstrin homology domain induces tyrosine phosphorylation (data not shown). Our observations suggest that phosphatidylinositol, which is produced by RET/PTC- induced PI3K activation, is not involved in RET/PTC-mediated tyrosine phosphorylation within cells.

We determined the role of Y9 phosphorylation in the regulation of PDK1 activity. An in vitro kinase assay with a synthetic peptide (suntide), identified as a suitable assay substrate in previous studies (25), was performed with immunoprecipitated PDK1. Immunoprecipitated PDK1 coexpressed with RET/PTC3 displayed an increase in phosphotransferase activity. However, phosphotransferase activity of the mutant construct, PDK1-Y9F, was not enhanced by RET/PTC3. These findings, collectively, suggest that Y9 is one of the regulatory residues for phosphorylation and activation of PDK1 by RET/PTC3. In addition, the transfected Myc-PDK1 activities were higher in the crude membrane fractions in the cells transfected with RET/PTC3. Therefore, it may be possible that the Y9 phosphorylation of PDK1 may facilitate the targeting the PDK1 to the plasma membrane where it activates substrates such as Akt/PKB.

Phosphorylation of T308 in Akt/PKB, the downstream kinase substrate of PDK1, was also observed after coexpression of PDK1 and RET/PTC3. The observation that the kinase-inactive RET/PTC3 K284M mutant neither activates PDK1 (Fig. 7A) nor induces T308 phosphorylation of Akt/PKB (data not shown) indicates that RET/PTC3-induced tyrosine phosphorylation triggers signaling pathways downstream of PDK1 in vivo.

Mutations in the p53 protein are not usually observed in papillary thyroid carcinoma. Few studies have been performed to determine the effects of RET/PTC on the function or stability of p53. The function of p53 is controlled by several mechanisms, including the regulation of protein stability. Central to this process is Mdm2, a ubiquitin ligase that targets p53 for ubiquitination and facilitates export of the tumor suppressor protein from the nucleus to the cytoplasm, where degradation by proteasomes takes place. Recent analyses demonstrate that PI3K signaling pathways are involved in the regulation of p53 stability via phosphorylation of Mdm2 (26, 48). In our experiments, introduction of RET/PTC3 led to a decrease in the level of wild-type p53 in HCT116 cells. The decrease in the p53-mediated transactivation may result from activation of the endogenous PDK1. This is because RET/PTC3 expression alone (500 ng/well) without PDK1 decreases its transactivating activities (Fig. 9B). Moreover, RET/PTC3 suppressed exogenous p53-mediated transactivation (Fig. 9C). The suppression of p53 transactivation was more prominent upon coexpression with PDK1, as shown in Fig. 10. The submaximal dose of RET/PTC3 (100 ng/well) was sufficient to cause the repression of p53-mediated transactivation, but the repressive effects of RET/PTC3 were potentiated with the coexpression of Myc-PDK1. However, expression of the PDK1-Y9F mutant construct with RET/PTC3 did not lead to repression of p53 transactivation. These observations support the hypothesis that RET/PTC3-induced Y9 phosphorylation-dependent PDK1 signaling pathways control the function of p53 in vivo. Although, the RET/PTC3-mediated repression of p53 transactivation requires the Y9 phosphorylation of PDK1, the exact mechanism for the down-regulation of the intracellular p53 level was not solved in this study. There may be many possibilities for how the RET/PTC3-mediated activation of PDK1 results in a decrease in the intracellular level of p53. The RET/PTC-induced Y9 phosphorylation of PDK1 results in the activation of the PDK1 downstream signaling pathways, e.g. Akt/PKB activation. Moreover, recent studies have shown that Akt/PKB activation is involved in p53 degradation by enhancing the Mdm2-mediated ubiquitination and degradation of p53 (26). Overall, these observations suggest that the activation of Akt/PKB by the RET/PTC3-PDK1 pathway may explain the down-regulation of p53 by RET/PTC. However, the indispensable roles of Akt/PKB in the down-regulation of p53 by the RET/PTC3-PDK1 activation process require more cell biological investigation.

In summary, colocalization of RET/PTC3 and PDK1, and RET/PTC3-induced Y9 phosphorylation, triggers activation of signaling pathways downstream of PDK1, leading to suppression of p53 function. These data may elucidate the oncogenic potential of RET/PTC in the development of papillary thyroid cancer.

MATERIALS AND METHODS

Materials

Media, cell culture reagents, and materials were purchased from Life Technologies, Inc. (Gaithersburg, MD), Sigma (St. Louis, MO), Fisher Scientific (Fairlawn, NJ), Corning, Inc. (Corning, NY), and Hyclone Laboratories, Inc. (Logan, UT). Wortmannin and PP1 were from Calbiochem (La Jolla, CA). LY294002 was obtained from New England Biolabs (Beverly, MA). Antibodies for RET and p53 were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA), anti-PDK1 antibody and antiphosphotyrosine antibody (4G10) from Upstate Biotechnology, Inc., and monoclonal anti-Myc and anti-HA antibodies from New England Biolabs, Inc. Anti-phospho-specific PDK1 (PDK1-pY9 and pY373/6) antibodies were gifts from Dr. Hemmings (25). Pervanadate was prepared with 0.2 mm H₂O₂ (25). All other materials, including N-α-tosyl-l-phenylalanyl chloromethyl ketone and rapamycin, were purchased from Sigma (St. Louis, MO).

Plasmids

Myc-PDK1-Y9F and Myc-PDK1-Y373/6F were supplied by Dr. Hemmings (25). pcDNA3-RET/PTC1 (iso 9) and pcDNA3-RET/PTC3 (iso 9) were generously provided by Dr. Jhiang (49). The pcDNA3-RET/PTC3-Y284M and pcDNA3-Y588F mutants were constructed by site-directed mutagenesis using PCR with TaKaRa Ex-Taq polymerase (TaKaRa) in a Perkin-Elmer 9700 thermocycler (Perkin-Elmer Corp.). The following PCR conditions were used: predenaturation at 95 C for 5 min, followed by 30 cycles of denaturation at 94 C for 1 min, annealing at 55 C for 45 sec, and elongation at 72 C for 1 min. The primers used were: RET/PTC3 forward, 5′-cccaagcttatgaataccttccaa-3′; RET/PTC3 K284M forward, 5′-gtggccgtgatgatgctgaa-3′; RET/PTC3 K284M reverse, 5′-ttcagcatcatcacggccac-3′; and RET/PTC3 reverse, 5′-ccctctagactagaatctagtaaatg-3′. The PTC3-Y588F mutant was constructed using a similar PCR protocol with the following primer pairs: RET/PTC3 forward, 5′-cccaagcttatgaataccttccaa-3′; RET/PTC3 Y588F forward, 5′-caaactctttggtagaatt-3′; RET/PTC3 Y588F reverse, 5′-aattctaccaaagagtttg-3′ and RET/PTC3 reverse, 5′-ccctctagactagaatctagtaaatg-3′. Right and left fragments were employed in a further PCR with the RET/PTC3 forward and RET/PTC3 reverse primers. The amplified full-length fragment was digested with HindIII and XbaI and cloned into a pcDNA3 expression vector. pcDNA3-HA-RET/PTC3 (iso 51) was generated from HA-RET (iso 51) supplied by Dr. Worby (50). Wild-type human p53 cDNA (a 1.8-kb XbaI fragment) was ligated into pCMV-NEO-BAM. HA-Akt/PKB and Myc-PDK1 were inserted into the pcDNA3 vector. All plasmid constructs were confirmed by automated DNA sequencing.

Cell Culture and Transfection

The CHO cell line was maintained in Ham’s F-12 medium (Sigma) containing 10% fetal bovine serum (Hyclone Laboratories, Inc.) in an atmosphere of 5% CO₂ at 37 C. Colon carcinoma HCT116 and src(−/−) MEF cells (51) were cultured in DMEM with 10% fetal bovine serum. All media contained 100 U/ml penicillin and 100 μg/ml streptomycin.

The cells were transiently transfected with the RET/PTC and Myc-PDK1 expression plasmids using lipofectAMINE PLUS (Invitrogen) according to the manufacturer’s instructions. Briefly, the plasmids were mixed with the Plus reagent and then incubated with lipofectAMINE. The lipofectAMINE Plus DNA complex was added to the cells and further incubated at 37 C for 3 h. The control cells received the lipofectAMINE PLUS alone. The cell viability was detected using a trypan blue dye exclusion test. After incubation, the medium was removed and replaced with fresh medium and cells were maintained for an additional 24 h. The transfection efficiency was evaluated with pCMV-SPORT-βgal (500 ng) and pEGFP-PDK1 (500 ng) in a six-well culture plate.

Immunoblotting

Cells were lysed by adding sodium dodecyl sulfate sample buffer [62.5 mm Tris-HCl (pH 6.8), 6% (wt/vol) sodium dodecyl sulfate, 30% glycerol, 125 mm dithiothreitol (DTT), and 0.03% (wt/vol) bromophenol blue]. Total cell lysates were denatured by boiling for 5 min, resolved on sodium dodecyl sulfate-polyacrylamide gels, and transferred to nitrocellulose membranes. Membranes were blocked in Tris-buffered saline containing 5% (wt/vol) milk and 0.1% Tween for 1 h, and incubated for 2 h with primary antibody. The blot was developed using horseradish peroxidase-conjugated secondary antibody (Phototope-Horseradish Peroxidase Western Blot Detection Kit, New England Biolabs).

Confocal Microscopy

CHO cells were grown on coverslips and transfected with pEGFP-PDK1 and pCDNA3-RET/PTC3 by the LipofectAmine method (Life Technologies, Inc.). At 24 h after transfection, cells were washed three times with cold PBS and fixed in 3.7% formaldehyde for 40 min. Fixed cells were mounted on glass slides with PBS and observed with a laser-scanning confocal microscope (Olympus Corp., Lake Success, NY). For detection of pCDNA3-RET/PTC3, cells mounted on glass slides were permealized with 2 ml PBS containing 0.1% Triton X-100 and 0.1 m glycine at room temperature, incubated for 15 min, washed three times with 1× PBS, and blocked with 3% (wt/vol) BSA in PBS for 10 min at RT. Cells were incubated with primary anti-RET antibody for 1 h at 37 C, washed three times with 1× PBS, and incubated for 1 h with rhodamine-conjugated antirabbit secondary antibody (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) at 37 C.

Protein Kinase Assays

CHO cells were transiently transfected with pcDNA3-RET/PTC3 and Myc-PDK1. Lysates were extracted with lysis Buffer A [containing 50 mm Tris-HCl (pH 7.5), 1% Nonidet P-40, 120 mm NaCl, 25 mm sodium fluoride, 40 mm β-glycerol phosphate, 0.1 mm sodium orthovanadate, 1 mm phenylmethylsulfonyl fluoride, 1 mm benazmidine and 2 μm microcystin-LR]. Lysates were centrifuged for 15 min at 12,000 × g and immunoprecipitated with anti-Myc monoclonal antibody coupled to protein G-Sepharose (Amersham Pharmacia Biotech, Arlington Heights, IL). Immune complexes were washed with lysis buffer containing 500 mm NaCl and, finally, kinase assay buffer [containing 50 mm Tris-HCl (pH 7.5), 0.1% (vol/vol) 2-mercaptoethanol]. PDK1 activities were assayed in a reaction mixture comprising kinase buffer, immunoprecipitated with Myc-PDK1, 10 mm MgCl₂, 1 μm protein kinase A inhibitor peptide, 100 μm [r-³²P]ATP (specific activity, 3000 Ci/mmol; NEN Life Science Products, Boston, MA) with 100 μm Suntide (RRKDGATMKTFCGTPE) as substrate at 30 C for 30 min. Samples were subjected to liquid scintillation counting (Hewlett-Packard Co., Palo Alto, CA).

Preparation of Crude Plasma Membrane Fraction

The CHO cells cultured in the 10-cm dishes were transfected with RET/PTC3 (4 μg/dish) and/or Myc-PDK1 (4 μg/dish). Twenty-four hours after transfection, the cells were treated with or without pervanadate (100 μm) and then placed on ice. After washing once in ice-cold PBS, the cells were scraped in 500 μl of an ice-cold fractionation buffer containing 20 mm HEPES-NaOH, pH 7.4; 250 mm sucrose; 25 mm sodium fluoride; 1 mm sodium pyrophosphate; 0.1 mm sodium orthovanadate; 2 μm microcystin LR; 1 mm phenylmethylsulfonyl fluoride; and 1 mm benzamidine; and then homogenized by passing them through a 26-gauge needle 10 times. The homogenates were centrifuged at 14,000 × g for 10 min to separate the cytosolic fraction (supernatant) from the organelles (pellet). The resulting pellet was resuspended in 1 ml of the fractionation buffer and layered onto 10 ml of a sucrose cushion (20 mm HEPES-NaOH, pH 7.4; and 1.15 m sucrose) and centrifuged at 77,000 × g for 60 min in a Beckman SW41 rotor (Beckman Coulter, Inc., Fullerton, CA). The diffuse band at the interface of the sucrose solutions was collected (1 ml), mixed with 1 ml of the fractionation buffer, and centrifuged at 100,000 × g for 20 min to pellet the crude plasma membrane fraction. The pellet was resuspended in the fractionation buffer, and the protein concentration was determined.

Luciferase Assay

p53-luc containing the firefly luciferase gene, p53 expression vector (pCMV-NEO-BAM-p53), and pcDNA3-RET/PTC3 were transfected with 50 ng pRL-SV40 plasmid encoding Renilla luciferase (Promega Corp., Madison, WI) into HCT116 colon carcinoma cells. After transfection, cells were allowed to recover for 24 h. The cells were washed with 1× PBS and lysed with 100 μl lysis buffer containing 40 mm Tricine (pH 7.8), 50 mm NaCl, 2 mm EDTA, 1 mm MgSO₄, 5 mm DTT, and 1% Triton X-100. Extracts were assayed in triplicate for luciferase activity in a total volume of 130 μl containing 30 μl cell extract, 20 mm Tricine, 0.1 mm EDTA, 1 mm magnesium carbonate, 2.67 mm MgSO₄, 33.3 mm DTT, 0.27 mm coenzyme A, 0.47 mm luciferin, and 0.53 mm ATP. Light intensity was measured using a luminometer (Berthold, Bad Wildbad, Germany). Firefly and Renilla luciferase activities were measured with a Dual-Luciferase Reporter assay system (Promega Corp., Madison, WI). Luciferase activity was integrated over a 10-sec period. Firefly luciferase values were standardized to Renilla values.

Other Assays

Protein concentration was determined by the Bradford method (Bio-Rad Laboratories, Inc., Hercules, CA), using recrystallized BSA as standard.

Acknowledgments

We are grateful to Dr. M. Billaud and Dr. C. A. Worby for providing the expression vectors for MEN2A and HA-RET. We further acknowledge Dr. S. M. Jhiang for supplying expression vectors for RET/PTC1 and 3 (iso 9) forms, and Dr. Y. Y. Kong for providing Src-deficient MEF cells.

This work was supported by the National Research Laboratory Program (M1-0104-00-0014) (Ministry of Science and Technology, Korea).

Abbreviations:

 
• CHO,

  Chinese hamster ovary;

 
• DTT,

  dithiothreitol;

 
• HA,

  hemagglutinin;

 
• MEF,

  mouse embryo fibroblast;

 
• PDK1,

  phosphoinositide-dependent kinase 1;

 
• PI3K,

  phosphatidylinositol 3-kinase;

 
• PKB,

  protein kinase B;

 
• RET/PTC,

  rearranged in transformation/papillary thyroid carcinomas;

 
• Shc,

  Src homology 2 domain containing.