Phosphorylation of paxillin confers cisplatin resistance in non-small cell lung cancer via activating ERK-mediated Bcl-2 expression AbstractPaxillin (PXN) is required for receptor tyrosine kinase-mediated ERK activation, and the activation of the Raf/MEK/ERK cascade has been linked with Bcl-2 expression. We hypothesized that phosphorylation of PXN by the EGFR/Src pathway might contribute to cisplatin resistance via increased Bcl-2 expression. We show that cisplatin resistance was dependent on PXN expression, as evidenced by PXN overexpression in TL-13 and TL-10 cells and PXN knockdown in H23 and CL1-5 cells. Specific inhibitors of signaling pathways indicated that the phosphorylation of PXN at Y118 and Y31 via the Src pathway was responsible for cisplatin resistance. We further demonstrated that ERK activation was also dependent on this PXN phosphorylation. Bcl-2 transcription was upregulated by phosphorylated PXN-mediated ERK activation via increased binding of phosphorylated CREB to the Bcl-2 promoter. A subsequent increase in Bcl-2 levels by a PXN/ERK axis was responsible for the resistance to cisplatin. Animal models further confirmed the findings of in vitro cells indicating that xenograft tumors induced by TL-13-overexpressing cells were successfully suppressed by cisplatin combined with Src or ERK inhibitor compared with treatment of cisplatin, Src inhibitor or ERK inhibitor alone. A positive correlation of phosphorylated PXN with phosphorylated ERK and Bcl-2 was observed in lung tumors from NSCLC patients. Patients with tumors positive for PXN, phosphorylated PXN, phosphorylated ERK and Bcl-2 more commonly showed a poorer response to cisplatin-based chemotherapy than did patients with negative tumors. Collectively, PXN phosphorylation might contribute to cisplatin resistance via activating ERK-mediated Bcl-2 transcription. Therefore, we suggest that Src or ERK inhibitor might be helpful to improve the sensitivity for cisplatin-based chemotherapy in NSCLC patients with PXN-positive tumors. IntroductionPaxillin (PXN) is a 68-kDa focal adhesion protein that functions as an adapter protein. It is phosphorylated by FAK and Src after integrin engagement and in turn binds to other downstream proteins, facilitating their recruitment into the signal cascade.1, 2, 3 Phosphorylation of PXN seems essential for maintaining the labile adhesions required for cell migration.4, 5 Our previous report indicated that PXN overexpression via the reduction of microRNA-218 promotes tumor progression and may predict poor overall survival (OS) in non-small cell lung cancer (NSCLC) patients.6 Recent reports indicated that aggressive tumor growth as a consequence of the epithelial鈥搈esenchymal transition was associated with anti-apoptosis for drug resistance, which resulted in poor tumor response in patients who received cisplatin-based chemotherapy.7, 8, 9 Therefore, PXN overexpression and its phosphorylation might be associated with tumor progression and drug resistance.PXN is required for receptor tyrosine kinase-mediated activation of ERK.3, 5, 10, 11, 12 EGFR activation leads to the phosphorylation of tyrosine residues on PXN, most likely by EGFR and Src.11, 12 This tyrosine phosphorylation of PXN is required for downstream activation of Raf, MEK and ERK,3, 11, 12 and the activation of the Raf/MEK/ERK cascade is linked with Bcl-2 expression.13, 14, 15, 16, 17, 18 Therefore, we hypothesized that phosphorylation of PXN by the EGFR and Src pathway might increase Bcl-2 expression via ERK activation, and consequently result in resistance to cisplatin.Our preliminary immunohistochemistry data showed that PXN expression was positively correlated with phosphorylation of PXN at Y118 and Y31. In addition, a prognostic significance of phosphorylation at Y118 and Y31 on survival and relapse was observed in a subset of lung cancer patients. Therefore, we performed mechanistic studies on a cell model to investigate whether PXN and its phosphorylation could modulate the resistance to cisplatin. We further verified which signaling pathway could be responsible for the phosphorylation of PXN, and whether PXN phosphorylation could promote resistance to cisplatin via ERK-mediated Bcl-2 expression, as Bcl-2 interacts with the LD4 motif of PXN to promote cell survival.19, 20, 21 Cisplatin resistance induced by PXN-mediated Bcl-2 expression via ERK activation was further verified in xenograft tumors of nude mice using Src or ERK inhibitor. The tumor response in patients who received cisplatin-based chemotherapy was investigated to understand whether PXN and its phosphorylation could be associated with tumor response and patient survival.ResultsPXN may confer resistance to cisplatin in lung cancer cellsA panel of lung cancer cells was enrolled to examine the possibility that PXN could confer resistance to cisplatin, determined by MTT assays. Ten lung cancer cell lines were treated with four doses of cisplatin and the dose response curve was used to calculate the concentration that yielded 50% inhibition (IC50). PXN mRNA expression levels in each lung cancer cell line were positively correlated with the IC50 for cisplatin (Figure 1a). This result suggests that PXN levels might be associated with the resistance to cisplatin in lung cancer cells.Figure 1PXN expression may confer cisplatin resistance via anti-apoptotic mechanisms. (a) Ten lung cancer cell types were treated with four doses of cisplatin and the dose-response curves were used to calculate the 50% inhibition concentration (IC50). PXN mRNA expression levels in each lung cancer cell type were examined by real-time PCR. (b) Two different PXN knockdown plasmids were transfected into high-PXN-expressing (H23 and CL1-5) cell lines. Alternatively, increasing amounts of expression plasmid were transfected into low PXN expressing (TL-13 and TL-10) cell lines. The total amount of transfected DNA was kept constant by adding the control vector. After 48鈥塰, the lysates were harvested and evaluated for levels of PXN and 尾-actin protein by western blotting. 尾-actin was used as a protein-loading control. P: parental control. NC: non-specific RNAi control. VC: vector control. PXN-knocked down or PXN-overexpressing lung cancer cells were treated with four doses of cisplatin and the dose-response curves were used to calculate the IC50 value. (c) Flow cytometric analysis of apoptosis after annexin V and PI staining. Indicated cells were treated with 0.1% DMSO or 25鈥壩?span >M cisplatin for 48鈥塰. The cells were then subjected to annexin V and PI staining, followed by flow cytometry. (d) Percentage of apoptotic cells including with the Annexin V+/PI鈭?population (early apoptosis) plus Annexin V+/PI+ (late apoptosis/secondary necrosis) was summarized by flow cytometric analysis. Data are expressed as means卤s.d., n=3.Full size imageWe investigated this possibility further using H23 and CL1-5 cells, which had the highest cisplatin IC50 values and TL-13 and TL-10 cells, which had the lowest IC50 values. The H23 and CL1-5 cells were transfected with two PXN RNAi while the TL-13 and TL-10 cells were transfected with two doses of the GFP-PXN expression vector. As expected, PXN expression, evaluated by western blotting, was decreased in PXN-knockdown cells and increased in PXN-overexpressing cells (Figure 1b). The IC50 value was decreased by PXN knockdown in the H23 and CL1-5 cells when compared with their parental and non-specific RNAi transfection control (NC) cells (9.5鈥壩?span >M and 10.9鈥壩?span >M vs 26.9鈥壩?span >M and 26.2鈥壩?span >M for H23; 8.7鈥壩?span >M and 10.2鈥壩?span >M vs 21.2鈥壩?span >M and 20.5鈥壩?span >M for CL1-5; Figure 1b). Conversely, the IC50 value was increased by PXN overexpression in TL-13 and TL-10 cells as compared with their parental and vector control (VC) cells (26.0鈥壩?span >M and 19.8鈥壩?span >M vs 7.5鈥壩?span >M and 6.7鈥壩?span >M for TL-13; 24.5鈥壩?span >M and 15.4鈥壩?span >M vs 7.5鈥壩?span >M and 8.7鈥壩?span >M for TL-10; Figure 1b). Annexin-V/PI assay was performed to test whether cell apoptosis could be responsible for the cell viability reduced by cisplatin. The representative cisplatin-induced apoptotic cells in PXN-knocked down H23 and PXN-overexpressing TL-13 cells are shown in Figure 1c. The cell viability reduced by cisplatin in PXN-knocked down H23 and CL1-5 cells and PXN-overexpressing TL-13 and TL-10 cells was consistent with apoptotic cells (Figure 1d). These results clearly indicate that PXN expression may confer resistance to cisplatin in lung cancer cells.The EGFR, Src and ERK pathways might be responsible for the resistance to cisplatin in phosphorylated PXN-overexpressing lung cancer cellsPXN phosphorylation is expected to be associated with the resistance to cisplatin and its phosphorylation may be mediated through the JNK, NF-魏B, PI3K/AKT, EGFR, Src, and MEK/ERK signaling pathways.1, 5, 10, 11, 12, 22, 23, 24, 25, 26 We examined which signal pathway(s) could be responsible for the resistance to cisplatin in lung cancer cells by combining the specific inhibitor of each signaling pathway with cisplatin to treat high-PXN-expressing H23 cells. The percentage of apoptotic cells induced by cisplatin in H23 cells was increased by 鈭?/span>18% by c-Met inhibitor (SU11274), 鈭?/span>11% by JNK inhibitor (SP600125), 鈭?/span>20% by EGFR inhibitors (PD153035 and Gefitinib) and 鈭?/span>40% by Src (PP2 and Dasatinib) and MEK/ERK inhibitors (U0126 and AZD6244) (Figure 2a, upper panel). However, the percentage of apoptotic cells induced by cisplatin in H23 cells was not changed by inhibitors of NF-魏B (BAY11-7082), PI3K/AKT (LY294002 and wortmannin) or p38/MAPK (SB203580) (Figure 2a, upper panel). Similarly, cisplatin-induced apoptosis in PXN-overexpressing TL-13 cells was increased by EGFR inhibitors (PD153035 and Gefitinib), Src (PP2 and Dasatinib) and MEK/ERK inhibitors (U0126 and AZD6244) (Figure 2a, down panel). Cisplatin resistance due to PXN overexpression via the EGFR, Src, and ERK signaling pathway was further confirmed by direct knockdown of EGFR, Src and ERK, respectively (Figure 2b). These results suggest that EGFR, Src and MEK/ERK pathway might contribute to the resistance to cisplatin in lung cancer cells with high PXN expression.Figure 2PXN expression may confer resistance to cisplatin via EGFR/Src/ERK pathway in lung cancer cells. (a) H23 and PXN-overexpressing TL-13 cells were treated with inhibitors of PI3K/AKT (25鈥壩糾ol/l LY294002; 10鈥壩糾ol/l wortmannin), NF-魏B (20鈥壩糾ol/l BAY11-7082), p38/MAPK (5鈥壩糾ol/l SB203580), c-Met (5鈥壩糾ol/l SU11274), JNK (10鈥壩糾ol/l SP600125), EGFR (0.5鈥壩糾ol/l PD153035; 5鈥壩糾ol/l Gefitinib), Src (10鈥壩糾ol/l PP2; 0.2鈥壩糾ol/l Dasatinib) and ERK (10鈥壩糾ol/l U0126; 10鈥壩糾ol/ AZD6244) for 5鈥塰 and then these inhibitors were removed to treat with 25鈥壩糾ol/l cisplatin for the additional 48鈥塰. (b) H23 and PXN-overexpressing TL-13 cells were transfected with two different types of AKT RNAi, NF-魏B (p65) RNAi, P38 RNAi, JNK RNAi, c-Met RNAi, EGFR RNAi, Src RNAi and ERK RNAi for 24鈥塰 and then treated with cisplatin (25鈥壩糾ol/l) for the additional 48鈥塰. Percentage of apoptosis was summarized by flow cytometric analysis. (c) H23 and PXN-overexpressing TL-13 cells were treated with EGFR (0.5鈥壩糾ol/l PD153035; 5鈥壩糾ol/l Gefitinib), Src (10鈥壩糾ol/l PD153035; 0.2鈥壩糾ol/l Dasatinib) and ERK (10鈥壩糾ol/l U0126; 10鈥壩糾ol/ AZD6244) inhibitors for 5鈥塰, and then the cells lysates were separated by SDS鈥揚AGE for the evaluation of pY118-PXN, pY31-PXN, PXN, p-Src, total Src, p-ERK, total ERK, EGFR and 尾-actin expression by specific antibodies using western blotting. (d) H23 and PXN-overexpressing TL-13 cells were transfected with two different type AKT RNAi, NF-魏B (p65) RNAi, P38 RNAi, JNK RNAi, c-Met RNAi, EGFR RNAi, Src RNAi and ERK RNAi for 48鈥塰 and then the cells lysates were separated by SDS鈥揚AGE for the evaluation of pY118-PXN,pY31-PXN, PXN, p-Src, total Src, p-ERK, total ERK, EGFR and 尾-actin expression by specific antibodies using western blotting.Full size imageWe next examined the possibility that phosphorylation of PXN could be mediated through EGFR, Src and MEK/ERK signaling pathway, giving rise to cisplatin resistance in lung cancer cells with high PXN expression. The expression levels of PXN phosphorylated at Y118 and Y31 was not changed by ERK inhibitors (U0126 and AZD6244) in H23 cells or PXN-overexpressing TL-13 cells (Figure 2c). Conversely, phosphorylated PXN levels were reduced to a greater extent by Src inhibitors (PP2 and Dasatinib) than by EGFR inhibitors (PD153035 and Gefitinib) in H23 and PXN-overexpressing TL-13 cells (Figure 2c). We also observed that phosphorylated ERK levels were markedly decreased by Src and EGFR inhibitors in both cell types. As expected, phosphorylated Src and ERK expression levels were suppressed by Src and ERK inhibitors in both cell types (Figure 2c). These results were further confirmed by direct knockdown of EGFR, Src and ERK, respectively (Figure 2d). We thus suggest that the phosphorylation of PXN at Y118 and Y31 is predominately regulated by the EGFR-mediated Src pathway, not by the ERK pathway.Phosphorylation of PXN at Y118 and Y31 is required for ERK-mediated Bcl-2 and Mcl-1 expressions, but Bcl-2 has a more important role than Mcl-1 in cisplatin resistancePhosphorylation of PXN by HGF may activate ERK phosphorylation via promotion of the interaction between PXN and Raf in mouse epithelial cells,10 and ERK activation is associated with elevated Bcl-2 expression.13, 14, 15, 16, 17, 18 We therefore explored the possibility that PXN phosphorylation at Y118 and Y31 could activate ERK phosphorylation to promote Bcl-2 transcription induced via increased CREB phosphorylation. The phosphorylation of PXN and ERK were concomitantly decreased by PXN knockdown in H23 cells, but the phosphorylation of both molecules was increased by PXN overexpression in TL-13 cells in a dose-dependent manner (Figure 3a). This result suggests that PXN expression could be associated with the phosphorylation of ERK.Figure 3Phosphorylation of PXN at Y118 and Y31 confer cisplatin resistance via increasing ERK-mediated Bcl-2 expression. (a) The PXN in H23 cells was knocked down by PXN RNAi and overexpressed using wild-type (WT). After 48鈥塰, the cells lysates were separated by SDS鈥揚AGE for the evaluation of pY118-PXN, pY31-PXN, PXN, p-ERK, total ERK, EGFR and 尾-actin expression by specific antibodies using western blotting. (b) PXN wild-type (WT) or -mutated (Y31F, Y118F, Y31/118F) were overexpressed in TL-13 cells. Cell viability was then evaluated by MTT assay and the percentage of apoptosis was summarized by flow cytometric analysis. (c) The PXN in H23 cells was knocked down by PXN RNAi and overexpressed using wild-type (WT) or mutated (Y31F, Y118F, Y31/118F) PXN in TL-13 cells. After 48鈥塰, the cells lysates were separated by SDS鈥揚AGE for the evaluation of Mcl-1, Bcl-XL, Bcl-w, A1, Bcl-2 and 尾-actin expression by specific antibodies using western blotting. (d) H23 and PXN-overexpressing TL-13 cells were knocked down by two different Mcl-1 and Bcl-2 RNAi. The indicated cells were treated with 0.1% DMSO or 25鈥壩?span >M cisplatin for 48鈥塰 for analyzing cell viability and apoptosis. Indicated cells were treated with two doses of ABT-199 for 5鈥塰, and then these inhibitors were removed. The indicated cells were treated with 0.1% DMSO or 25鈥壩?span >M cisplatin for 48鈥塰 for analyzing cell apoptosis.Full size imageExpression vectors for PXN mutations at Y118 and/or Y31 were constructed by site-directed mutagenesis. Mutated PXN expression vectors were transfected into TL-13 cells to explore whether phosphorylation of ERK could be abrogated by the mutation of PXN phosphorylation at Y118 or Y31. As expected, PXN phosphorylation at Y118 or Y31 was only observed in TL-13 cells with mutated PXN-Y31 or PXN-Y118 expression vectors (Figure 3b, upper panel). Interestingly, the phosphorylated ERK expression levels were markedly decreased in TL-13 cells after transfection with the mutated PXN expression vectors when compared with TL-13 cells transfected with a wild-type PXN expression vector. The decrease in phosphorylated ERK expression was almost abolished in TL-13 cells transfected with a PXN-Y118 and -Y31 double-mutated expression vector (Figure 3b, upper panel). The percentage of apoptotic cells increased by cisplatin modulated by wild-type PXN or mutated PXN expression vector transfections in TL-13 cells were inversely correlated with phosphorylated ERK expression levels (Figure 3b). These results clearly indicate that phosphorylation of PXN at Y118 and Y31 is required for ERK activation and may contribute to cisplatin resistance in lung cancer cells with high PXN expression.We next explored the possibility that anti-apoptosis genes such as Mcl-1, Bcl-XL, Bcl-w, A1, and Bcl-2 could be upregulated by PXN-activated ERK pathway to contribute to cisplatin resistance. Western blot showed that Mcl-1 and Bcl-2 expression levels were decreased by PXN knockdown in H23 cells and increased by PXN overexpression in TL-13 cells; however, Bcl-XL, Bcl-w, and A1 were not changed by PXN knockdown and -overexpression in both cell types (Figure 3c). In addition, Mcl-1 and Bcl-2 expressions were decreased by mutated PXN expression vectors in TL-13 cells when compared with those with wild-type PXN expression vector transfection (Figure 3c, right panel). Bcl-2 and Mcl-1 expressions in both cell types were expectedly reduced by Bcl-2- and Mcl-1-knockdown (Figure 3d, upper panel). The percentage of apoptotic cells in H23 cells and wild-type PXN-overexpressing TL-13 cells was markedly increased by Bcl-2-knockdown, but slightly increased by Mcl-1-knockdown in both cell types (Figure 3d, lower panel). As expected, the percentage of apoptotic cells induced by cisplatin in PXN-knocked down H23 cells was dependent on ABT-199 (Bcl-2 inhibitors) treatment (Figure 3d, left panel). Consistent finding was also observed in PXN-overexpressing TL-13 cells with ABT-199 treatment (Figure 3d, right panel). These results suggest that Bcl-2 elevated by PXN-mediated ERK activation may have a more important role than Mcl-1 in cisplatin resistance of lung cancer cells with high PXN expression.Upregulation of Bcl-2 transcription by PXN-mediated ERK activation is through increased phosphorylation of CREB binding to Bcl-2 promoterWe next examined the possibility that ERK activation by PXN phosphorylation may upregulate Bcl-2 transcription via phosphorylation of CREB (p-CREB).13, 15, 27 Western blotting data showed that p-CREB and Bcl-2 expression were concomitantly decreased by PXN knockdown in H23 cells (Figure 4a, left panel). More interestingly, the p-CREB and Bcl-2 expressions were markedly reduced in PXN-overexpressing TL-13 cells following transfection with PXN-Y118-mutated, Y31-mutated or double-mutated expression vectors when compared with TL-13 cells transfected with a wild-type PXN expression vector (Figure 4a, right panel). In addition, Luciferase reporter assay and real-time PCR analysis showed that the reporter activity of Bcl-2 promoter and its mRNA expression were significantly decreased by PXN knockdown in H23 cells and were increased by PXN overexpression in TL-13 cells (Figure 4a, lower panel). P-ERK, p-CREB and Bcl-2 expressions were decreased by ERK inhibitors, and ERK knockdown in H23- and PXN-overexpressing TL-13 cells (Figure 4b, upper panel). The p-CREB to the Bcl-2 promoter, evaluated by chromatin immunoprecipitation (ChIP), was significantly decreased by ERK inhibitors, and ERK knockdown in H23- and PXN-overexpressing TL-13 cells (Figure 4b, middle panel). In addition, Luciferase reporter assay and real-time PCR analysis showed that the reporter activity of Bcl-2 promoter and its mRNA expression were significantly decreased by ERK inhibitors and ERK knockdown in H23- and PXN-overexpressing TL-13 cells (Figure 4b, lower panel). Collectively, we suggested that ERK activation by PXN phosphorylation may upregulate Bcl-2 transcription via phosphorylation of CREB (p-CREB).Figure 4ERK-mediated Bcl-2 expression was regulated by via phosphorylation of CREB. (a) H23 cells were transfected with two different types of PXN RNAi (1鈥壩糶) and 0.5鈥壩糶 of Bcl-2 P1-Luc reporter. TL-13 cells were transfected with 1鈥壩糶 of wild-type (WT) or mutated (Y31F, Y118F and Y31/118F) PXN and 0.5鈥壩糶 of Bcl-2 P1-Luc reporter. Luciferase activity was measured at 48鈥塰 post-transfection. In all experiments, the relative luciferase activity was shown as fold-activation relative to that of the control cells. (b) H23 and wild-type (WT) PXN-overexpressing TL-13 cells were treated by U0126, AZD6244, ERK RNAi #1 (p42#1 and p44#1) and #2 (p42#2 and p44#2). After 48鈥塰, the cells lysates were separated by SDS鈥揚AGE for the evaluation of p-CREB, CREB, BCL-2, p-ERK, total ERK and 尾-actin expressions by western blotting and real-time PCR. For ChIP assay, the lysates were immunoprecipitated by p-CREB, and PCR amplification of immunoprecipitated DNA was carried out with diluted aliquots, using the primers consisting of the oligonucleotides that encompass the promoter region of Bcl-2. Bcl-2 transcriptional activity was evaluated by real-time RT鈥揚CR and luciferase assay. The total amount of transfected DNA was kept constant by adding the control vector. Luciferase activity was measured at 48鈥塰 post-transfection.Full size imagePXN-mediated xenograft tumors were more effectively suppressed by cisplatin combined with Dasatinib or AZD6244As previously described, we expected that PXN-mediated cisplatin resistance can be dissolved by Src or ERK inhibitor via decreased Bcl-2 expression. The representative xenograft tumors in each group of nude mice were shown in Figure 5a. The tumor volume in nude mice with TL-13 VC cell injection was grown slowly during 7鈥?7 days, but the tumor volume of the nude mice with VC cells injection was almost suppressed by cisplatin treatment (VC vs VC+Cisplatin; Figure 5b, left panel). However, the tumor volume of nude mice with PXN-overexpressing TL-13 cells injection was increased markedly during the time intervals; surprisingly, the tumor volume of nude mice with PXN-overexpressing TL-13 cells injection was almost not changed by cisplatin treatment (PXN vs PXN+Cisplatin; Figure 5b, left panel). More interestingly, the tumor volume of nude mice injected with PXN-overexpressing TL-13 cells was decreased markedly by cisplatin plus ERK inhibitor AZD6244 or Src inhibitor Dasatinib when compared with PXN-overexpressing TL-13 cells-injected nude mice that were treated with cisplatin, AZD6244 or Dasatinib alone (Figure 5b, right panel). These results from nude mice model strongly support the findings from the cell experiments.Figure 5The antitumor effect of cisplatin and Dasatinib or AZD6244 was performed in the mouse xenograft model. TL-13 VC xenografts were treated with vehicle and cispatin (5鈥塵g/kg). TL-13 PXN overexpression xenografts were treated with vehicle, cisplatin (5鈥塵g/kg), Dasatinib (5鈥塵g/kg), AZD6244 (5鈥塵g/kg) or a combination of both. (a) The representative tumor burdens in the eight groups. (b) Tumor volumes in the nude mice of the eight groups were measured at 3-day intervals from days 9 to 27. Mean卤s.e.m. values (mm3) were calculated from the tumor volume of 3 nude mice in each group.Full size imageThe correlation of PXN with phosphorylated PXN, phosphorylated ERK and Bcl-2 in lung tumors from NSCLC patientsWe collected 183 tumors from NSCLC patients to verify whether PXN expression could be associated with the expressions of phosphorylated PXN, phosphorylated ERK and Bcl-2. The representative immunohistochemical results for PXN, PXN phosphorylated at Y118 and Y31, phosphorylated ERK and Bcl-2 are shown in Figure 6a. Immunohistochemistry analysis showed that PXN expression was positively correlated with PXN phosphorylation (phosphorylated Y118 and/or Y31), phosphorylated ERK and Bcl-2 expression in lung tumors (P 0.001 for phosphorylated PXN, P=0.027 for phosphorylated ERK, P=0.005 for Bcl-2; Table 1). In addition, the correlation of phosphorylated PXN with phosphorylated ERK and Bcl-2 expression (P=0.003 for phosphorylated ERK, P=0.015 for Bcl-2; Table 1), and the association between phosphorylated ERK and Bcl-2 were also observed in this study population (P=0.002; Table 1). These results were consistent with the findings of the cell model indicating that PXN phosphorylation may elevate Bcl-2 expression via ERK activation.Figure 6PXN, p-PXN, p-ERK and Bcl-2 expressions are associated with OS and RFS in HPV-infected NSCLC. (a) A representative figure of positive- or negative PXN, pY31-PXN, pY118-PXN, p-ERK and Bcl-2 expression in lung cancer patients. (b) Lung cancer patients with tumors positive for PXN, p-PXN, p-ERK and Bcl-2 had poor outcomes.Full size imageTable 1 The correlation between Paxillin (PXN), phosphorylated Paxillin (p-PXN), phosphorylated ERK (p-ERK), and Bcl-2 expressionsFull size tableThe prognostic value of PXN and its phosphorylation at Y118 and/or Y31, phosphorylated ERK and Bcl-2 on OS and relapse-free survival in NSCLC patientsPXN overexpression is associated with poor OS in NSCLC patients. We further hypothesized that PXN and its phosphorylated Y118 and/or Y31 forms, phosphorylated ERK and Bcl-2 might be associated with OS and also with relapse-free survival (RFS). Kaplan鈥揗eier analysis showed that patients with tumors positive for PXN and its phosphorylated forms had shorter OS and RFS periods than did those with tumors negative for PXN and its phosphorylated forms (PXN: P=0.001 for OS, P 0.001 for RFS; phosphorylated PXN: P=0.005 for OS, P 0.001 for RFS; Figure 6b, left panel). Phosphorylated ERK and Bcl-2 also showed prognostic value for determining OS and RFS in resected NSCLC patients (phosphorylated ERK: P=0.001 for OS, P 0.001 for RFS; Bcl-2: P=0.034 for OS, P=0.038 for RFS; Figure 6b, right panel). Cox regression analysis further indicated that an independent prognostic value of PXN, phosphorylated PXN and phosphorylated ERK on OS and RFS in resected NSCLC patients (OS: HR, 1.80, 95% CI, 1.21鈥?.67, P=0.004 for PXN; HR, 1.55, 95% CI, 1.05鈥?.26, P=0.026 for phosphorylated PXN; HR, 1.79, 95% CI, 1.20鈥?.67, P=0.004 for phosphorylated ERK; RFS: HR, 1.80, 95% CI, 1.23鈥?.64, P=0.003 for PXN; HR, 1.66, 95% CI, 1.12鈥?.41, P=0.008 for phosphorylated PXN; HR, 1.86, 95% CI, 1.27鈥?.72, P=0.001; Table 2); however, no independent prognostic value was observed for Bcl-2 in this study population. These results support the findings of the cell models and indicate that PXN phosphorylation may increase ERK-mediated Bcl-2 expression, resulting in relapse and poor survival in resected NSCLC patients.Table 2 Mutivariate Cox regression analysis for the combined effects of PXN, p-PXN, p-ERK and BCL-2 on OS and RFS in NSCLCFull size tableAssociation of PXN, phosphorylated PXN, phosphorylated ERK, and Bcl-2 with the tumor response in patients who received cisplatin-based chemotherapyA total of 100 of 183 patients who received cisplatin-based chemotherapy after surgical resection were enrolled in this cohort study. The tumor response to cisplatin-based chemotherapy in 100 patients was collected from chart reviews, and the information was further confirmed by at least two medical doctors. We classified patients with stable disease and progressive disease as the unfavorable tumor response group and the partial response group; the complete response was classified as the favorable tumor response group. As shown in Table 3, patients with tumors positive for PXN, phosphorylated PXN, phosphorylated ERK and Bcl-2 had more unfavorable tumor responses than did those with tumors negative for PXN, phosphorylated PXN, phosphorylated ERK, and Bcl-2 (70 vs 41%, P=0.004 for PXN; 75 vs 39%, P 0.001 for phosphorylated PXN; 71 vs 26%, P 0.001 for phosphorylated ERK; 70 vs 38%, P=0.001 for Bcl-2; Table 3). These results strongly support the findings of the cell model, which showed that the resistance to cisplatin due to PXN phosphorylation may be mediated through ERK-mediated Bcl-2 expression.Table 3 Association between PXN, p-PXN, p-ERK and Bcl-2 expressions in lung tumors and tumor response to cisplatin-based chemotherapy in these patientsFull size tableDiscussionIn the present study, we provide the evidence that phosphorylation of PXN at Y118 and Y31 is required for cisplatin resistance in lung cancer cells. The possible mechanism would involve activation of the ERK-CREB signaling pathway by PXN phosphorylation, leading to the upregulation of Bcl-2 expression in transcriptional levels. Previous studies have indicated that phosphorylation of PXN is dependent on Src activation via the phosphorylation of EGFR and the c-Met pathway.2, 3, 10, 12, 22, 24, 25, 26, 28, 29, 30 Consistent results were also observed in this study, where an involvement of the EGFR, Src and MEK/ERK pathway was implicated in the phosphorylation of PXN; however, no evidence was obtained for involvement of the PI3K/AKT and MAPK pathways (Figure 2a). Interestingly, the percentage of apoptotic cells induced by cisplatin in PXN-overexpressing TL-13 cells increased by 鈭?/span>40% in response to an EGFR inhibitor, but the percentage of apoptotic cells induced by cisplatin in H23 cells only increased by 鈭?/span>20% in response to the same inhibitor. The different impact of EGFR inhibitors on cisplatin cytotoxicity in both cell types was due to the higher expression of c-Met in H23 than in TL-13 cells (Supplementary Figure 1). The percentage of apoptotic cells induced by cisplatin in H23 cells in response to a c-Met inhibitor SU11274 was similar with that of H23 cells induced by cisplatin in response to EGFR inhibitors (Figure 2a). We also used EGF and HGF to activate the EGFR and c-Met pathways in H23 cells. Phosphorylated PXN and Src levels were concomitantly increased by EGF and HGF in a dose-dependent manner (Supplementary Figure 1b). Levels of phosphorylated PXN and Src in H23 cells were further elevated by the combined treatment with EGF and HGF (Supplementary Figure 1c, left panel). Conversely, the expression levels of phosphorylated PXN and Src were diminished by the combined treatment with SU11274 and Gefitinib (Supplementary Figure 1c, right panel). These results are consistent with previous reports indicating that the Src pathway can be activated independently by the EGFR and c-Met pathways to phosphorylate PXN.2, 3, 10, 12, 22, 24, 25, 26, 28, 29, 30 In addition, p-ERK and Bcl-2 expression were synergistically reduced in these lung cancer cells, supporting the idea that phosphorylation of PXN by the Src pathway has a crucial role in cisplatin resistance in lung cancer cells (Supplementary Figure 1c).In the present study, PXN-mediated Bcl-2 transcription via ERK activation may be responsible for cisplatin resistance (Figures 2, 3, 4). Notably, PXN phosphorylation at Y118 and Y31 may affect the interaction between PXN LD4 and Bcl-2 BH4 (Supplementary Figures 2 and 3a). In addition, cycloheximide and ubiquitination pattern assay showed that the interaction between PXN and Bcl-2 may increase Bcl-2 protein stability. (Supplementary Figure 3a and b). These results suggest that interaction of PXN with Bcl-2 may increase Bcl-2 expression via protecting Bcl-2 protein degraded by ubiquitin proteasomal pathway. PXN LD4 motif is a conserved binding domain to form a complex via recruiting GIT1/b-PIX/Pak or TACE. PXN at S272 has been shown to be phosphorylated by Pak kinase(s).31, 32 PXN phosphorylation at S272 expression levels in TL-13 cells was almost diminished by Src inhibitors and EGFR inhibitors when compared with those cells without inhibitor treatments (Supplementary Figure 4a). Cisplatin-mediated cell apoptosis was increased markedly in TL-13 cells with PXN-mutated S272A transfection when compared with those cells with PXN wild-type transfection (Supplementary Figure 4b). However, the percentage of apoptotic cells were not changed by mutated PXN S85A, S178A and T538A transfection (Supplementary Figure 4b). Mutated PXN S272A (located at PXN鈥?LD4 motif) also abolished the interaction of PXN with Bcl-2 to decrease Bcl-2 expression levels (Supplementary Figure 4c and d). Interestingly, PXN phosphorylation at Y118 and Y31 seemed to contribute to PXN phosphorylation at S272 via increased p21 activated kinase (PAK1) phosphorylation due to PAK1 activation (Supplementary Figure 4). Therefore, Bcl-2 protein stability increased by PXN was not only indirectly via phosphorylation at Y118 and Y31 but also directly via PXN phosphorylation at S272. However, the in vivo xenograft tumors showed that the Src or ERK inhibitor was able to markedly suppress tumor formation in nude mice which were injected with PXN-overexpressing cells and combined with cisplatin treatment. This observation was consistent with the results from the in vitro cell model. Therefore, we suggest that phosphorylation of PXN may contribute more to cisplatin resistance via activating ERK-mediated Bcl-2 transcription than Bcl-2 protein stability due to PXN interacting with Bcl-2.Previous reports indicated that the increase in Bcl-2 expression due to ERK phosphorylation was associated with the survival of estrogen-induced human macrophages, valproate-treated neuron-like cells and pancreatic cancer cells.13, 14, 15, 16 Bcl-2 has a crucial role in chemoresistance in various human cancers including ovarian, bladder, colorectal, prostate, nasopharyngeal and lung cancers.33, 34, 35, 36, 37, 38, 39 Therefore, the signaling pathways that increase Bcl-2 expression are important in the reduction of chemoresistance. We showed that PXN phosphorylation is required for ERK activation, and consequently for the upregulation of Bcl-2 transcription via phosphorylation of CREB (Figure 4). ChIP analysis further indicated that the phosphorylated CREB upregulated Bcl-2 transcription via an increased binding of phosphorylated CREB to the Bcl-2 promoter (Figure 4). To the best of our knowledge, this is the first study to indicate that PXN phosphorylation can activate the ERK pathway and, in turn, upregulate Bcl-2 expression via promotion of binding of phosphorylated CREB to the Bcl-2 promoter. We further evidenced that PXN phsphorylation at Y31 and Y118 has a crucial role in the interaction between PXN and Bcl-2 to maintain Bcl-2 protein stability via blocking ubiquitin proteasomal pathway (Supplementary Figures 2 and 3). The involvement of two mechanisms in PXN-mediated cisplatin resistance due to increased Bcl-2 expression was further evidenced in TL-13 cells with PXN E268R plus CREB RNAi transfection (Supplementary Figure 5). Therefore, Bcl-2 expression, elevated by PXN phosphorylation at Y31 and Y118, was responsible for the resistance to cisplatin in lung cancer cells.Previous reports indicated that tumors with high PXN expression were more common in late-stage patients than in early-stage patients.6, 19 PXN protein expression was consistent with its mRNA expression in lung tumors from lung cancer patients.6, 19 In present study, PXN expression in lung tumors may be used to predict tumor response in patients who received cisplatin-based chemotherapy. Therefore, we expected that PXN mRNA expression in blood plasma could be associated with its expression in lung tumors, and it could be useful as an indicator of the potential effectiveness of cisplatin-based chemotherapy in advanced NSCLC patients. The blood plasma and tumor tissues were obtained from 11 stage-I and stage-II patients and from 67 stage-III and stage-IV patients from another medical center in Central Taiwan (Chung Shan Medical University Hospital, Taichung, Taiwan). PXN mRNA expression was more frequently detected in blood plasma from late-stage patients than in early-stage patients (57 vs 9%, P=0.007; Supplementary Table 1). More interestingly, patients with high PXN mRNA tumors were more commonly associated with patients鈥?blood plasma with high PXN mRNA expression levels (64 vs 36%, P=0.013). Therefore, we suggest that evaluation of PXN mRNA expression in blood plasma may be useful for predicting the tumor response in advanced lung cancer patients.In summary, we have provided in vitro evidence demonstrating that phosphorylation of PXN at Y118 and Y31 promoted ERK-mediated Bcl-2 expression to consequently result in cisplatin resistance. Consistent observations were also seen in in vivo animal models and lung tumors from lung cancer patients. Therefore, we suggest that Src or ERK may be potential therapeutic targets in patients whose tumors show high PXN expression and are likely to be cisplatin-resistant.Materials and methodsStudy subjectsThe study included 183 patients who underwent resection at the Department of Chest Surgery, Taichung Veteran General Hospital, Taichung, Taiwan, between June 1994 and December 2006. The tumor type and stage of each specimen was histologically determined according to the World Health Organization鈥檚 (WHO) classification system. Seventy eight female (42.6%), 105 male (57.4%), 110 nonsmokers (60.1%), 73 smokers (39.9%), 105 adenocarcinoma (57.4%), 78 squamous cell carcinoma (42.6%), 67 stage-I (36.6%), 29 stage-II (15.8%), 82 stage-III (44.8%) and 5 stage-IV patients (2.7%) were enrolled in this study (Supplementary Table 2). The median follow-up time was 26.3 months (range from 1 to 165.3 months) and the end of the follow-up period was December 2007. One hundred and one patients had tumor relapse, whereas 82 patients did not. One hundred out of 183 patients received adjuvant cisplatin-based chemotherapy. Among these, 85 patients had tumor relapse, whereas 15 patients did not, but were in a late stage (II, III or IV). The chemotherapeutic drugs used for these patients are listed in Supplementary Table 3. The IRB protocol CS07159 was approved by Chung Shang Medical University Hospital.Tumor responseAmong these patients, 100 patients were treated with cisplatin-based chemotherapy. Responses were categorized as follows: complete response: a complete disappearance of all the tumors; partial response: a decrease in size or number of the tumor lesions by 猢?/span>50%; progressive disease: at least 25% increase in size or number of the tumor lesions, and stable disease: neither sufficient shrinkage to qualify for partial response nor sufficient increase to qualify for progressive disease. Therefore, the favorable response (complete response and partial response) is a decrease in tumor size of least 猢?/span>50%.Cell linesA549, Ch27, H23, H358, H1355, H157 and H1299 cells were obtained from the American Type Culture Collection (ATCC) and cultured as described. CL1-5 cells were kindly provided by Dr P-C Yang (Department of Internal Medicine, National Taiwan University Hospital, Taiwan). TL-10 and TL-13 cells were culturally established from pleural effusions of three patients by the Ficoll鈥揚aque method. The detailed methods for culture and identification were shown in pervious study.40 These cells were cultured and stored according to the suppliers鈥?instructions and used at passages 5鈥?0.Chemicals and antibodiesPD153035 was obtained from Calbiochem (La Jolla, CA, USA). Dasatinib was obtained from LC Laboratories (Woburn, MA, USA). Gefitinib and AZD6244 were obtained from Selleckchem.com (Houston, TX, USA). ABT-199 was obtained from ActiveBiochem (Maplewood, NJ, USA). All other chemicals were acquired from Sigma Chemical (St Louis, MO, USA) unless otherwise indicated. Anti-EGFR (epidermal growth factor receptor), anti-total Src, anti鈥損hospho-Src (p-Src), anti-total ERK and anti鈥損hospho-ERK (p-CREB) antibodies were obtained from Cell Signaling (Danvers, MA, USA). Anti-PXN antibody was obtained from NeoMarker (Fremont, CA, USA). Anti-phosphoY31-PXN (pY31-PXN) and anti-Bcl-2 were obtained from Genetex (Irvine, CA, USA). Anti-phosphoS85-PXN (pS85-PXN), Anti-phosphoS178-PXN (pS178-PXN), Anti-phosphoS272-PXN (pS85-PXN), Anti-phosphoT538-PXN (pT538-PXN) were obtained from ECM bioscience (Versailles, KY, USA). All other antibodies were purchased from Santa Cruz Biotechnology (Dallas, TX, USA).Plasmid constructs and transfectionThe PXN-overexpressing plasmid was kindly provided by Dr Salgia (The University of Chicago, USA). Mutated PXN expression constructs containing multiple-point mutations (Y31F, Y118F, Y31/118F, E286R, S272A, S85A, S178A and T538A) and 螖BH4 Bcl-2 overexpression constructs were constructed by the QuickChange site-directed mutagenesis system (Stratagene, La Jolla, CA, USA). The Bcl-2 P1 promoter-Luciferase plasmid and Bcl-2 overexpression plasmid was purchased from Addgene (Addgene Company, Cambridge, MA, USA). AKT (AKT1), NF-魏B (p65), p38, JNK, c-Met, EGFR, Src, ERK, CREB1, Bcl-2 and PXN RNAi were purchased from National RNAi Core Facility, Academia Sinica, Taiwan (Supplementary Table 4). Different concentrations of expression plasmids were transiently transfected into lung cancer cells (1 脳 106) using the Turbofect reagent (Formentas, Hanover, MD, USA).41 After 48鈥塰, cells were harvested and whole-cell extracts were assayed in subsequent experiments.Real-time quantitative RT鈥揚CR analysis, western blotting and luciferase reporter assayReal-time quantitative RT鈥揚CR analysis, western blotting and luciferase reporter assay were performed to assess gene expression as previously described.42 Primers used for real-time PCR analysis are listed in Supplementary Table 5.In vivo immunoprecipitation and ChIP assaysThe immunoprecipitation and ChIP procedures and quantification methods were performed as described previously.43 The primers consisting of the oligonucleotides that encompass the promoter region of CREB are shown in Supplementary Table 5.Immunohistochemical analysisAnti-mouse PXN antibodies were purchased from Neomarkers (Fremont, CA, USA). Anti-rabbit phosphorylated-PXN (Y118 and Y31) and Bcl-2 antibodies were purchased from Santa Cruz. Anti-rabbit phosphorylated ERK antibodies were purchased from Cell Signaling. The immunohistochemical procedures and quantification methods were performed as described previously.6MTT cytotoxicity assayThe cell lines were cultured in a humidified incubator containing 95% air and 5% CO2 at 37鈥壜癈 in 96-well flat-bottomed microtiter plates containing RPMI and DMEM supplemented with 10% heat-inactivated fetal bovine serum, 100鈥塙/ml penicillin and 100鈥塙/ml streptomycin. Before cisplatin treatment, the cells in the exponential growth phase were pretreated with overexpression and knockdown plasmids for 24鈥塰 or procaine for 2鈥塰. After 48鈥塰 of incubation, the in vitro cytotoxic effects of these treatments were determined by MTT assay (at 570鈥塶m) and the cell viability was expressed as a percentage of the control (untreated) cells (% of control).Annexin-V stainingThe cells were collected by trypsinization and centrifugation at 1000鈥?i>g for 5鈥塵in. Following resuspension in binding buffer (10鈥塵M HEPES-NaOH, 140鈥塵M NaCl, 2.5鈥塵M CaCl2) at a final cell density of 1鈥? 脳 106 cells/ml, 100鈥壩糽 of a single-cell suspension (1鈥? 脳 105 cells) was incubated with 5鈥壩糽 annexin V-FITC and 5鈥壩糽 PI for 15鈥塵in at room temperature in the dark. After addition of 400鈥壩糽 of binding buffer, the samples were analyzed by using a BD FACSCalibur flow cytometer (BD Biosciences, San Jose, CA, USA) within 1鈥塰. For each sample, 10鈥?00 events were counted.In vivo animal model experimentsFor therapeutic experiments in tumor growth, tumor cells were injected subcutaneously into the back of 4鈥?-week-old female Balb/c nude mice. Xenograft size was measured every 3 days and tumor volume was determined as (length 脳 width2)/2. When tumors grew to 100鈥塵m3, the mice were randomized to indicated groups: vehicle (DMSO), cisplatin (5鈥塵g/kg/week), Dasatinib (5鈥塵g/kg/week), AZD6244 (5鈥塵g/kg/week), combined with cisplatin and Dasatinib, and combined with cisplatin and AZD6244. Drugs were administrated by intraperitoneal injection.Statistical analysisStatistical analysis was performed using the SPSS statistical software program (Version 18.0; SPSS Inc., Chicago, IL, USA). The association between tumor response and PXN protein expression was analyzed by the 蠂2-test. Survival plots were generated using the Kaplan鈥揗eier method, and differences between patient groups were determined by the log-rank test. Multivariate Cox regression analysis was performed to determine OS and RFS. The analysis was stratified for all known variables (age, gender, smoking status, tumor type and tumor stage) and protein expression. References1Aponte M, Jiang W, Lakkis M, Li MJ, Edwards D, Albitar L et al. Activation of platelet-activating factor receptor and pleiotropic effects on tyrosine phospho-EGFR/Src/FAK/paxillin in ovarian cancer. Cancer Res 2008; 68: 5839鈥?848.CAS聽 Article聽Google Scholar聽 2Sachdev S, Bu Y, Gelman IH . Paxillin-Y118 phosphorylation contributes to the control of Src-induced anchorage-independent growth by FAK and adhesion. BMC Cancer 2009; 9: 12.Article聽Google Scholar聽 3Webb DJ, Donais K, Whitmore LA, Thomas SM, Turner CE, Parsons JT et al. FAK-Src signalling through paxillin, ERK and MLCK regulates adhesion disassembly. Nat Cell Biol 2004; 6: 154鈥?61.CAS聽 Article聽Google Scholar聽 4Deakin NO, Turner CE . Paxillin comes of age. J Cell Sci 2008; 121: 2435鈥?444.CAS聽 Article聽Google Scholar聽 5Huang C, Rajfur Z, Borchers C, Schaller MD, Jacobson K . JNK phosphorylates paxillin and regulates cell migration. Nature 2003; 424: 219鈥?23.CAS聽 Article聽Google Scholar聽 6Wu DW, Cheng YW, Wang J, Chen CY, Lee H . Paxillin predicts survival and relapse in non-small cell lung cancer by microRNA-218 targeting. Cancer Res 2010; 70: 10392鈥?0401.CAS聽 Article聽Google Scholar聽 7Shintani Y, Okimura A, Sato K, Nakagiri T, Kadota Y, Inoue M et al. Epithelial to mesenchymal transition is a determinant of sensitivity to chemoradiotherapy in non-small cell lung cancer. Ann Thorac Surg 2011; 92: 1794鈥?804 (discussion 1804).Article聽Google Scholar聽 8Zhuo WL, Wang Y, Zhuo XL, Zhang YS, Chen ZT . Short interfering RNA directed against TWIST, a novel zinc finger transcription factor, increases A549 cell sensitivity to cisplatin via MAPK/mitochondrial pathway. Biochem Biophys Res Commun 2008; 369: 1098鈥?102.CAS聽 Article聽Google Scholar聽 9Haslehurst AM, Koti M, Dharsee M, Nuin P, Evans K, Geraci J et al. EMT transcription factors snail and slug directly contribute to cisplatin resistance in ovarian cancer. BMC Cancer 2012; 12: 91.CAS聽 Article聽Google Scholar聽 10Ishibe S, Joly D, Zhu X, Cantley LG . Phosphorylation-dependent paxillin-ERK association mediates hepatocyte growth factor-stimulated epithelial morphogenesis. Mol Cell 2003; 12: 1275鈥?285.CAS聽 Article聽Google Scholar聽 11Sen A, De Castro I, Defranco DB, Deng FM, Melamed J, Kapur P et al. Paxillin mediates extranuclear and intranuclear signaling in prostate cancer proliferation. J Clin Invest 2012; 122: 2469鈥?481.CAS聽 Article聽Google Scholar聽 12Sen A, O鈥橫alley K, Wang Z, Raj GV, Defranco DB, Hammes SR . Paxillin regulates androgen- and epidermal growth factor-induced MAPK signaling and cell proliferation in prostate cancer cells. J Biol Chem 2010; 285: 28787鈥?8795.CAS聽 Article聽Google Scholar聽 13Arany I, Megyesi JK, Reusch JE, Safirstein RL . CREB mediates ERK-induced survival of mouse renal tubular cells after oxidant stress. Kidney Int 2005; 68: 1573鈥?582.Article聽Google Scholar聽 14Boucher MJ, Morisset J, Vachon PH, Reed JC, Laine J, Rivard N . MEK/ERK signaling pathway regulates the expression of Bcl-2, Bcl-X(L), and Mcl-1 and promotes survival of human pancreatic cancer cells. J Cell Biochem 2000; 79: 355鈥?69.CAS聽 Article聽Google Scholar聽 15Creson TK, Yuan P, Manji HK, Chen G . Evidence for involvement of ERK, PI3K, and RSK in induction of Bcl-2 by valproate. J Mol Neurosci 2009; 37: 123鈥?34.CAS聽 Article聽Google Scholar聽 16Subramanian M, Shaha C . Up-regulation of Bcl-2 through ERK phosphorylation is associated with human macrophage survival in an estrogen microenvironment. J Immunol 2007; 179: 2330鈥?338.CAS聽 Article聽Google Scholar聽 17Wang J, Zhou JY, Wu GS . ERK-dependent MKP-1-mediated cisplatin resistance in human ovarian cancer cells. Cancer Res 2007; 67: 11933鈥?1941.CAS聽 Article聽Google Scholar聽 18Yoon H, Min JK, Lee JW, Kim DG, Hong HJ . Acquisition of chemoresistance in intrahepatic cholangiocarcinoma cells by activation of AKT and extracellular signal-regulated kinase (ERK)1/2. Biochem Biophys Res Commun 2011; 405: 333鈥?37.CAS聽 Article聽Google Scholar聽 19Jagadeeswaran R, Surawska H, Krishnaswamy S, Janamanchi V, Mackinnon AC, Seiwert TY et al. Paxillin is a target for somatic mutations in lung cancer: implications for cell growth and invasion. Cancer Res 2008; 68: 132鈥?42.CAS聽 Article聽Google Scholar聽 20Sheibani N, Tang Y, Sorenson CM . Paxillin鈥檚 LD4 motif interacts with bcl-2. J Cell Physiol 2008; 214: 655鈥?61.CAS聽 Article聽Google Scholar聽 21Sorenson CM . Interaction of bcl-2 with Paxillin through its BH4 domain is important during ureteric bud branching. J Biol Chem 2004; 279: 11368鈥?1374.CAS聽 Article聽Google Scholar聽 22Tabassam FH, Graham DY, Yamaoka Y . Paxillin is a novel cellular target for converging Helicobacter pylori-induced cellular signaling. Am J Physiol Gastrointest Liver Physiol 2011; 301: G601鈥揋611.CAS聽 Article聽Google Scholar聽 23Zou W, Deselm CJ, Broekelmann TJ, Mecham RP, Vande Pol S, Choi K et al. Paxillin contracts the osteoclast cytoskeleton. J Bone Miner Res 2012; 27: 2490鈥?500.CAS聽 Article聽Google Scholar聽 24Turner CE . Paxillin interactions. J Cell Sci 2000; 113 (Pt 23): 4139鈥?140.CAS聽 PubMed聽Google Scholar聽 25Johnson FM, Saigal B, Talpaz M, Donato NJ . Dasatinib (BMS-354825) tyrosine kinase inhibitor suppresses invasion and induces cell cycle arrest and apoptosis of head and neck squamous cell carcinoma and non-small cell lung cancer cells. Clin Cancer Res 2005; 11: 6924鈥?932.CAS聽 Article聽Google Scholar聽 26Bai L, Yang JC, Ok JH, Mack PC, Kung HJ, Evans CP . Simultaneous targeting of Src kinase and receptor tyrosine kinase results in synergistic inhibition of renal cell carcinoma proliferation and migration. Int J Cancer 2012; 130: 2693鈥?702.CAS聽 Article聽Google Scholar聽 27Meller R, Minami M, Cameron JA, Impey S, Chen D, Lan JQ et al. CREB-mediated Bcl-2 protein expression after ischemic preconditioning. J Cereb Blood Flow Metab 2005; 25: 234鈥?46.CAS聽 Article聽Google Scholar聽 28Hiscox S, Morgan L, Green TP, Barrow D, Gee J, Nicholson RI . Elevated Src activity promotes cellular invasion and motility in tamoxifen resistant breast cancer cells. Breast Cancer Res Treat 2006; 97: 263鈥?74.CAS聽 Article聽Google Scholar聽 29Sakata H, Takayama H, Sharp R, Rubin JS, Merlino G, LaRochelle WJ . Hepatocyte growth factor/scatter factor overexpression induces growth, abnormal development, and tumor formation in transgenic mouse livers. Cell Growth Differ 1996; 7: 1513鈥?523.CAS聽 PubMed聽Google Scholar聽 30Bertotti A, Bracco C, Girolami F, Torti D, Gastaldi S, Galimi F et al. Inhibition of Src impairs the growth of met-addicted gastric tumors. Clin Cancer Res 2010; 16: 3933鈥?943.CAS聽 Article聽Google Scholar聽 31Nayal A, Webb DJ, Brown CM, Schaefer EM, Vicente-Manzanares M, Horwitz AR . Paxillin phosphorylation at Ser273 localizes a GIT1-PIX-PAK complex and regulates adhesion and protrusion dynamics. J Cell Biol 2006; 173: 587鈥?89.CAS聽 Article聽Google Scholar聽 32Lee JH, Wittki S, Brau T, Dreyer FS, Kratzel K, Dindorf J et al. HIV Nef, paxillin, and Pak1/2 regulate activation and secretion of TACE/ADAM10 proteases. Mol Cell 2013; 49: 668鈥?79.CAS聽 Article聽Google Scholar聽 33Yang TM, Barbone D, Fennell DA, Broaddus VC . Bcl-2 family proteins contribute to apoptotic resistance in lung cancer multicellular spheroids. Am J Respir Cell Mol Biol 2009; 41: 14鈥?3.CAS聽 Article聽Google Scholar聽 34Li J, Viallet J, Haura EB . A small molecule pan-Bcl-2 family inhibitor, GX15-070, induces apoptosis and enhances cisplatin-induced apoptosis in non-small cell lung cancer cells. Cancer Chemother Pharmacol 2008; 61: 525鈥?34.CAS聽 Article聽Google Scholar聽 35Mano Y, Kikuchi Y, Yamamoto K, Kita T, Hirata J, Tode T et al. Bcl-2 as a predictor of chemosensitivity and prognosis in primary epithelial ovarian cancer. Eur J Cancer 1999; 35: 1214鈥?219.CAS聽 Article聽Google Scholar聽 36Miyake H, Hanada N, Nakamura H, Kagawa S, Fujiwara T, Hara I et al. Overexpression of Bcl-2 in bladder cancer cells inhibits apoptosis induced by cisplatin and adenoviral-mediated p53 gene transfer. Oncogene 1998; 16: 933鈥?43.CAS聽 Article聽Google Scholar聽 37Bendardaf R, Lamlum H, Ristamaki R, Syrjanen K, Pyrhonen S . Oncoprotein Bcl-2 and microsatellite instability are associated with disease-free survival and treatment response in colorectal cancer. Oncol Rep 2008; 20: 999鈥?004.CAS聽 PubMed聽Google Scholar聽 38Bray K, Chen HY, Karp CM, May M, Ganesan S, Karantza-Wadsworth V et al. Bcl-2 modulation to activate apoptosis in prostate cancer. Mol Cancer Res 2009; 7: 1487鈥?496.CAS聽 Article聽Google Scholar聽 39Low SY, Tan BS, Choo HL, Tiong KH, Khoo AS, Leong CO . Suppression of BCL-2 synergizes cisplatin sensitivity in nasopharyngeal carcinoma cells. Cancer Lett 2012; 314: 166鈥?75.CAS聽 Article聽Google Scholar聽 40Sung WW, Wang YC, Cheng YW, Lee MC, Yeh KT, Wang L et al. A polymorphic -844T/C in FasL promoter predicts survival and relapse in non-small cell lung cancer. Clin Cancer Res 2011; 17: 5991鈥?999.CAS聽 Article聽Google Scholar聽 41Wu DW, Tsai LH, Chen PM, Lee MC, Wang L, Chen CY et al. Loss of TIMP-3 promotes tumor invasion via elevated IL-6 production and predicts poor survival and relapse in HPV-infected non-small cell lung cancer. Am J Pathol 2012; 181: 1796鈥?806.CAS聽 Article聽Google Scholar聽 42Wu D-W, Lee M-C, Wang J, Chen C-Y, Cheng Y-W, Lee H . DDX3 loss by p53 inactivation promotes tumor malignancy via the MDM2/Slug/E-cadherin pathway and poor patient outcome in non-small-cell lung cancer. Oncogene 2014; 33: 1515鈥?526.CAS聽 Article聽Google Scholar聽 43Wu DW, Liu WS, Wang J, Chen CY, Cheng YW, Lee H . Reduced p21(WAF1/CIP1) via alteration of p53-DDX3 pathway is associated with poor relapse-free survival in early-stage human papillomavirus-associated lung cancer. Clin Cancer Res 2011; 17: 1895鈥?905.CAS聽 Article聽Google Scholar聽 Download referencesAcknowledgementsThis work was jointly supported by grants from the National Health Research Institute (NHRI96-TD-G-111-006; NHRI97-TD-G-111-006) and the National Science Council (NSC-100-2314-B-038-043-MY3) of Taiwan, ROC.Author informationAffiliationsGraduate Institute of Cancer Biology and Drug Discovery, Taipei Medical University, Taipei, Taiwan, ROCD-W Wu,聽Y-W Cheng聽 聽H LeeDepartment of Internal Medicine, Chung Shan Medical University, Taichung, Taiwan, ROCT-C WuDivision of Thoracic Surgery, Buddhist Tzu Chi Taichung General Hospital, Taichung, Taiwan, ROCJ-Y WuCollege of Medicine, Tzu Chi University, Hualien, Taiwan, ROCJ-Y WuGraduate Institute of Medical Sciences, College of Medicine, Taipei Medical University, Taichung, Taiwan, ROCY-C ChenDepartment of Thoracic Surgery, Taichung Veteran General Hospital, Taichung, Taiwan, ROCM-C LeeDepartment of Surgery, China Medical University Hospital, Taichung, Taiwan, ROCC-Y ChenAuthorsD-W WuView author publicationsYou can also search for this author in PubMed聽Google ScholarT-C WuView author publicationsYou can also search for this author in PubMed聽Google ScholarJ-Y WuView author publicationsYou can also search for this author in PubMed聽Google ScholarY-W ChengView author publicationsYou can also search for this author in PubMed聽Google ScholarY-C ChenView author publicationsYou can also search for this author in PubMed聽Google ScholarM-C LeeView author publicationsYou can also search for this author in PubMed聽Google ScholarC-Y ChenView author publicationsYou can also search for this author in PubMed聽Google ScholarH LeeView author publicationsYou can also search for this author in PubMed聽Google ScholarCorresponding authorCorrespondence to H Lee.Ethics declarations Competing interests The authors declare no conflict of interest. Additional informationSupplementary Information accompanies this paper on the Oncogene websiteSupplementary information Supplementary Information (DOC 2796 kb)Rights and permissionsReprints and PermissionsAbout this articleCite this articleWu, DW., Wu, TC., Wu, JY. et al. Phosphorylation of paxillin confers cisplatin resistance in non-small cell lung cancer via activating ERK-mediated Bcl-2 expression. Oncogene 33, 4385鈥?395 (2014). https://doi.org/10.1038/onc.2013.389Download citationReceived: 29 January 2013Revised: 05 July 2013Accepted: 02 August 2013Published: 07 October 2013Issue Date: 28 August 2014DOI: https://doi.org/10.1038/onc.2013.389KeywordsPXNBcl-2cisplatin-based chemotherapyNSCLC Tsung-Ying He, De-Wei Wu, Po-Lin Lin, Lee Wang, Chi-Chou Huang, Ming-Chih Chou Huei Lee Scientific Reports (2016)