Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Implication of epithelial-mesenchymal transition in IGF1R-induced resistance to EGFR-TKIs in advanced non-small cell lung cancer

Implication of epithelial-mesenchymal transition in IGF1R-induced resistance to EGFR-TKIs in... www.impactjournals.com/oncotarget/ Oncotarget, Vol. 6, No. 42 Implication of epithelial-mesenchymal transition in IGF1R- induced resistance to EGFR-TKIs in advanced non-small cell lung cancer 1,2,* 2,* 2,* 2 2 2 Juan Zhou , Jinjing Wang , Yunyun Zeng , Xi Zhang , Qiaoting Hu , Jihua Zheng , 2 2 2 Bei Chen , Bo Xie , Wei-Min Zhang Department of Oncology, Guangzhou Clinical College of the Second Military Medical University, Guangzhou, Guangdong 510010, China Department of Oncology, General Hospital of Guangzhou Military Command of PLA, Guangzhou, Guangdong 510010, China These authors have contributed equally to this work Correspondence to: Wei-Min Zhang, e-mail: [email protected] Keywords: epidermal growth factor receptor-tyrosine kinase inhibitors, epithelial-mesenchymal transition, type 1 insulin-like growth factor receptor, non-small cell lung cancer, drug resistance Received: May 21, 2015 Accepted: October 26, 2015 Published: November 05, 2015 ABSTRACT The underlying mechanisms for acquired resistance to epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs) in about 30%-40% of non-small cell lung cancer (NSCLC) patients remain elusive. Recent studies have suggested that activation of epithelial-mesenchymal transition (EMT) and type 1 insulin-like growth factor receptor (IGF1R) is associated with acquired EGFR-TKIs resistance in NSCLC. Our study aims to further explore the mechanism of EMT and IGF1R in acquired EGFR- TKIs resistance in NSCLC cell lines with mutant (PC-9) or wild-type EGFR (H460). Compared to parental cells, EGFR-TKIs-resistant PC-9/GR and H460/ER cells displayed an EMT phenotype and showed overexpression of IGF1R. SiIGF1R in PC-9/GR and H460/ER cells reversed EMT-related morphologies and reversed their resistance to EGFR-TKIs. Exogenous IGF-1 alone induced EMT in EGFR-TKIs-naïve PC-9 and H460 cells and increased their resistance against EGFR-TKIs. Inducing EMT by TGF-β1 in PC-9 and H460 cells decreased their sensitivity to EGFR-TKIs, whereas reversing EMT by E-cadherin overexpression in PC-9/GR and H460/ER cells restored their sensitivity to EGFR-TKIs. These data suggest that IGF1R plays an important role in acquired drug resistance against EGFR-TKIs by inducing EMT. Targeting IGF1R and EMT may be a potential therapeutic strategy for advanced NSCLC with acquired EGFR-TKIs resistance. The insulin-like growth factor (IGF) system, INTRODUCTION including IGF ligands, their receptors and binding proteins, Acquired drug resistance has become a bottleneck is important in promoting tumor development. Previous in the treatment of advanced non-small cell lung cancer studies showed that activation of IGF1R is involved in (NSCLC) using epidermal growth factor receptor-tyrosine EGFR-TKIs resistance in NSCLC cell lines [9] and in lung kinase inhibitors (EGFR-TKIs) [1–3]. Currently, the known cancer patients [10]. IGF1R tyrosine kinase inhibitors have mechanisms underlying this acquired drug resistance been reported to reverse the drug resistance of NSCLC to include T790M mutation [4, 5] and amplification of MET EGFR-TKIs in vitro and in vivo [7]. IGF1R activates the genes [6]. These mechanisms account for about 60–70% downstream pathways of EGFR signaling, such as the of acquired drug resistance. However, the underlying phosphatidylinositol 3 kinase/protein kinase B (PI3K/AKT) mechanisms for approximately 30%-40% of cases are still pathway and the extracellular signal-regulated kinases/ unclear. Recent studies show that the activation of epithelial- mitogen-activated protein kinase (ERK/MAPK) pathway, mesenchymal transition (EMT) and type 1 insulin-like leading to secondary drug resistance to EGFR-TKIs [11– growth factor receptor (IGF1R) is associated with acquired 13]. However, the exact mechanisms of IGF1R-induced drug resistance against EGFR-TKIs in NSCLC [7, 8]. acquired EGFR-TKIs resistance remain to be elucidated. www.impactjournals.com/oncotarget 44332 Oncotarget Interestingly, IGF1R has been shown to play an of EGFR and the phosphorylation of EGFR (pEGFR) important role in EMT [7] and IGF1R activation can showed no significant changes (Figure 1B). induce EMT in breast epithelial cells [14] and prostate To further evaluate whether activation of cancer cells [15]. EMT is a biological process of losing the IGF1R pathway was the major mechanism epithelial features and acquiring mesenchymal properties, underlying acquired drug resistance, we targeted characterized by E-cadherin reduction and Vimentin silencing of IGF1R using siRNA technology in PC-9/ induction. It has been reported that a subgroup of NSCLC GR and H460/ER cells. The expression of IGF1R with pronounced EMT was EGFR-TKIs resistant [3, 8, 16, in PC-9/GR-siIGF1R and H460/ER-siIGF1R cells 17], suggesting that EMT may render NSCLC insensitive was downregulated, indicating that the interference to EGFR inhibition. Furthermore, decreased expression was successful, while the expression of EGFR and of E-cadherin [8, 16, 17] was associated with reduced pEGFR did not show significant changes (Figure sensitivity to EGFR-TKIs, and restoration of E-cadherin 1C). PC-9/GR-siIGF1R and H460/ER-siIGF1R cells expression improved cells’ sensitivity to EGFR-TKIs [18]. had the same EGFR and KRAS gene state as their Consistently, clinical studies have suggested a prognostic parent cells (Supplementary Figure S4). Furthermore, value of E-cadherin in NSCLC patients treated with EGFR- knockdown of IGFIR significantly decreased the IC50 TKIs [19–21]. of gefitinib and erlotinib in PC-9/GR and H460/ER Previously, we reported the association between cells, respectively, suggesting that IGFIR may play an EMT, IGF1R expression and drug response in advanced important role in restoring their sensitivity to gefitinib NSCLC patients treated with gefitinib [22]. NSCLC or erlotinib (Figure 1D). patients with negative EMT or lower IGF1R expression have a significantly higher objective response rate. Both, EMT is involved in the acquirement of the IGF1R expression and EMT occurrence correlated with EGFR-TKIs-resistance phenotype the development of acquired drug resistance to EGFR- TKIs in NSCLC patients. In the present study, we further After the acquisition of drug resistance, PC-9/ examined the relationship between EMT and IGF1R GR and H460/ER cells showed EMT phenotypes, with expression with sensitivity to EGFR-TKIs in NSCLC cell loose cell junctions and long, spindle-type morphology lines with wild-type or mutant EGFR. Furthermore, using (Figure 2A). Expression of the epithelial marker in vitro assays, we provided evidence that IGF1R induced E-cadherin decreased in PC-9/GR cells, and was EGFR-TKIs resistance by inducing EMT and explored undetectable in H460 and H460/ER cells. In contrast, the possible cellular mechanism. Our data highlight the the mesenchymal marker Vimentin, transcription factor importance of EMT in IGF1R-induced resistance to Snail and nuclear β-catenin all increased in PC-9/GR EGFR-TKIs in NSCLC and implicate both EMT and and H460/ER cells, compared to their parental cells IGF1R as potential therapeutic targets for advanced (Figure 2B). Specifically, Vimentin expression gradually NSCLC. increased in H460/ER cells in a time-dependent manner upon erlotinib incubation (Figure 2C). Cell migration RESULTS and invasion abilities increased significantly in PC-9/ GR and H460/ER cells compared to their parental cells as shown by scratch test and transwell experiment IGF1R activation is involved in the acquirement (Figure 2D, 2E, 2F, 2G). of the EGFR-TKIs-resistance phenotype Treatment with TGF-β1 has also been shown in As expected, the resistant cells PC-9/GR and H460/ many cancer cells to promote a shift from epithelial ER exhibited significantly decreased sensitivity to EGFR- to mesenchymal phenotype [25, 26]. To further TKIs, compared to the parental PC-9 and H460 cells, evaluate whether EMT constituted a major mechanism respectively (Figure 1A). The delE746-A750 deletion underlying acquired drug resistance, parental PC-9 mutation in exon 19 of EGFR was detected in PC-9 and and H460 cells were treated with exogenous TGF-β1 PC-9/GR cells by qPCR-HRM, but not in H460 or H460/ (10 ng/mL) for 72 h. After TGF-β1 treatment, ER; however, the T790M mutation was not detected in any both PC-9 and H460 cells acquired a spindle-like of the cell lines. FISH analysis showed no amplification morphology (Figure 3A) and expressed notably of c-MET in PC-9/GR or H460/ER cells (Supplementary reduced E-cadherin and increased Vimentin proteins Figure S2). No EGFR mutation in H460/ER cells was (Figure 3B). Interestingly, following TGF-β1-induced found, and all cell lines harbored wild-type KRAS before transient EMT, the sensitivity of PC-9 and H460 to and after the induction of drug resistance (Supplementary EGFR-TKIs was dramatically reduced (Figure 3C), Figure S3). Additionally, the expression of IGF1R and which phenocopied the PC-9/GR and H460/ER cells, the phosphorylation of IGF1R (pIGF1R) increased respectively. These results indicated that transient significantly in PC-9/GR and H460/ER cells after the induction of mesenchymal-like phenotypes is sufficient acquisition of drug resistance, while the expression to induce resistance to EGFR-TKIs in NSCLC cells. www.impactjournals.com/oncotarget 44333 Oncotarget Figure 1: Role of IGF1R on the sensitivity to gefitinib and erlotinib in EGFR-TKIs-resistant cells. A. The sensitivity to gefitinib and erlotinib of PC-9/GR, H460/ER, and their parental cells was assessed by MTT assays. Cells were treated with the indicated doses of gefitinib or erlotinib for 72 h. IC50 values for different conditions are provided in the table within individual figures. B. Expression of IGF1R, phosphor-IGF1R, EGFR, and phosphor-EGFR in EGFR-TKIs-resistant cells by immunoblotting analysis. C. Effect of IGF1R siRNA on expression of IGF1R, phosphor-IGF1R, EGFR, and phosphor-EGFR in EGFR-TKIs-resistant cells. β-actin was used as an internal control. D. IC50 of gefitinib/erlotinib in PC-9/GR and H460/ER cells increased significantly following IGF1R knock-down when compared with the control cells. Data represent means ± S.D. of three independent experiments. www.impactjournals.com/oncotarget 44334 Oncotarget Figure 2: EMT in EGFR-TKIs-resistant cells. A. Morphology of PC-9/GR, H460/ER, and their parental cells grown for 3 days until 90% confluence. In contrast to the parental cells, the PC-9/GR and H460/ER cells displayed long spindle-like shape with loose cell junctions. Photographs were taken at × 200 magnification. B–C. Loss of E-cadherin was seen in PC-9/GR cells, H460 cells did not express E-cadherin, and there was increased expression of Vimentin, transcription factor Snail and nuclear β-catenin in PC-9/GR and H460/ER cells shown by immunoblotting analysis. In addition, Vimentin expression at the protein level increased in a time dependent manner after the induction of drug resistance. β-actin was used as an internal control. D, E. Enhanced migratory capacity of EGFR-TKIs-resistant cells according to Scratch assay. Confluent cells were scraped by a pipette tip to generate wounds and then were cultured in serum-free culture medium for 48 h. Representative images of wounds were taken at 0 and 48 h. Cell motility was examined with a light microscope (×40) and the width of the wound was quantified. F, G. Enhanced invasiveness of EGFR-TKIs-resistant cells according to transwell assay. The cells were incubated for 24 h in modified Boyden chambers. Those cells that migrated through the filters were stained and counted under a light microscope. Quantification was done in 10 randomly chosen fields. The data are reported as means ± S.D. The photographs were taken at × 200 magnification. www.impactjournals.com/oncotarget 44335 Oncotarget E-cadherin was stably transfected into PC-9/GR and H460/ Inhibition of EMT increased sensitivity to ER cells. The EGFR and KRAS gene status in the resulting EGFR-TKIs in NSCLC cells cells, PC9/GR-CDH1 and H460/ER-CDH1, remained the To further evaluate the role of EMT in the resistance same as their parent cells (Supplementary Figure S5). PC9/ to EGFR-TKIs in NSCLC cells, CDH1 which encodes GR-CDH1 and H460/ER-CDH1 presented an epithelial- Figure 3: Effect of TGF-β1 on EMT and the sensitivity to gefitinib and erlotinib in EGFR-TKIs-resistant cells. A. Morphology of PC-9 and H460 cells grown with 10 ng/mL TGF-β1 for 3 days until 90% confluence. The photographs were taken at × 200 magnification. B. TGF-β1-induced downregulation of E-cadherin and upregulation of Vimentin in PC-9 and H460 cells according to immunoblotting analysis. β-actin was used as an internal control. C. The effects of sequential treatment with the TGF-β1 on cell viability of PC-9 and H460 cells exposed to gefitinib and erlotinib by MTT uptake assays, respectively. The data represent the means ± S.D. of three independent experiments. www.impactjournals.com/oncotarget 44336 Oncotarget like morphology (Figure 4A). The expression of PC9/GR and H460/ER cells (Figure 6C, Supplementary E-cadherin in PC-9/GR-CDH1 and H460/ER-CDH1 cells Figure S6B). Furthermore, activation of IGF1R by IGF-1 was significantly higher than that in the empty vector induced activation of pERK, but not of pAKT in PC9 and controls, indicating that the lentiviral transfection of H460 cells (Figure 6D and 6E). Finally, knockdown of E-cadherin was successful. Additionally, compared to the IGF1R dramatically attenuated the increase in total ERK empty vector control, the expression of Vimentin, Snail and pERK in PC-9/GR, and resulted in decreased pERK and nuclear β-catenin decreased in PC-9/GR-CDH1 and in H460/ER (Figure 6D and 6E). H460/ER-CDH1 cells (Figure 4B). Overexpression of CDH1 correlated with a reduction of the nuclear β-catenin DISCUSSION in PC-9/GR and H460/ER by immunofluorescence staining experiment (Supplementary Figure S6A). Most In the present study, both PC-9 cells, which harbor importantly, the sensitivity to gefitinib and erlotinib the mutant EGFR, and H460 cells, with wild-type EGFR, dramatically increased in E-cadherin-overexpressing were used to establish EGFR-TKIs-resistant sublines PC9/GR and H460/ER cells, respectively (Figure 4C), (PC-9/GR and H460/ER) by continuously culturing in suggesting that E-cadherin repressed EMT in NSCLC gefitinib or erlotinib, respectively. We did not detect the cells, and subsequently enhanced the cytotoxic effect classical T790M mutation or c-MET gene amplification of EGFR-TKIs. Finally, overexpression of E-cadherin in either PC-9/GR or H460/ER cells. However, inhibited the motility and invasiveness of EGFR-TKIs- expression of IGF1R and pIGF1R, but not that of EGFR, resistant NSCLC cells shown by scratch test and transwell were enhanced in these two resistant cell lines. These assays (Figure 4D, 4E, 4F, 4G). results suggested that IGF1R has a major role in acquired drug resistance to EGFR-TKIs, which is consistent with previous studies. For example, Morgillo et al. [7] found Exogenous IGF-1 induced EMT in EGFR-TKIs- decreased EGFR and increased pIGF1R in erlotinib- naïve NSCLC cells and increased their resistance resistant H460 cells. They also found that the combined to EGFR-TKIs treatment of an IGF1R tyrosine kinase inhibitor with erlotinib inhibits cell proliferation significantly. Vazquez- To test whether IGF1R overexpression alone can Martin et al. [27] and Cortot et al. [28] also suggested a trigger EMT and EGFR-TKIs resistance, exogenous IGF- role of IGF1R in acquired drug resistance to EGFR-TKIs 1 was applied to EGFR-TKIs-naïve PC-9 and H460 cells. in PC-9 cells. Twenty-four hours after IGF-1 induction, both IGF1R and To further evaluate the role of IGF1R in acquired pIGF1R were induced, and no significant change of EGFR EGFR-TKIs resistance, we knocked down IGF1R by was observed (Figure 5A). The IGF-1 induced cells also siRNA in PC-9/GR and H460/ER cells and found that acquired a mesenchymal phenotype and a spindle-like both cells restored their sensitivity to gefitinib or erlotinib. morphology (Figure 5B). EMT occurrence was further Furthermore, our data showed that exogenous IGF-1 confirmed by decreased expression of E-cadherin, and application in EGFR-TKIs-naïve PC-9 and H460 cells increased expression of Vimentin, nuclear β-catenin and triggered IGF1R activation, and increased cells’ resistance Snail (Figure 5A). In addition, β-catenin was shown to to EGFR-TKIs. Guix et al. [29] also discovered that IGF- translocate from cell membrane into the nuclear after IGF- 1 activated IGF1R in lung cancer cells, and showed that 1 induction by Immunofluorescence experiment (Figure only cells, which depended on IGF1R and its downstream 5C). IGF-1 alone does not promote lung cancer cell signaling pathways to promote cell growth, acquired the proliferation (data not shown), however IGF-1 induced resistance ability against geftinib. Taken together, our PC-9 and H460 cells showed decreased sensitivity to results provided direct evidence that activation of IGF1R EGFR-TKIs compared to parental cells (Figure 5D). These is one of the mechanisms underlying acquired drug results indicated that activation of IGF1R is associated resistance to EGFR-TKIs in lung cancer cells. with EMT process. There are still few studies addressing the mechanism of IGF1R activation in EGFR-TKIs resistance in lung IGF1R induced EMT of NSCLC cells by cancer cells. One possible mechanism is that after EGFR enhancing ERK/MAPK signaling pathway is inhibited, IGF1R pathway is activated to To clarify the association among IGF1R activation, promote cell survival and proliferation [7]. Morgillo et al. EMT, and resistance to EGFR-TKIs, we explored whether [7] found that erlotinib treatment of H460 cells increase the induction of EMT might proceed through the activation the levels of EGFR/IGFIR heterodimer on cell membrane, of IGF1R to promote resistance to EGFR-TKIs in NSCLC subsequently activate IGFIR and its downstream signaling cells. Knockdown of IGF1R by siRNA repressed the EMT- mediators. Data in the present study, which were consistent related morphological features (Figure 6A and 6B). We with Peled et al’s [9] report, showed that exogenous IGF-1 also demonstrated that knockdown of IGF1R attenuated alone does not promote lung cancer cell proliferation, but the decrease of E-cadherin expression and the increase instead increases cell survival against erlotinib. Further of Vimentin, Snail and nuclear β-catenin expression in study is needed in this regard. www.impactjournals.com/oncotarget 44337 Oncotarget Figure 4: Effect of E-cadherin (CDH1) overexpression on EMT and the sensitivity to gefitinib and erlotinib in EGFR- TKIs-resistant cells. A. E-cadherin-overexpressing cell lines PC-9/GR and H460/ER after CDH1 transfection showed an epithelial- like morphology and a remarkably increased expression of E-cadherin according to immunofluorescence assay. The nuclei were stained with DAPI (blue fluorescence), and E-cadherin was stained with Cy3-conjugated antibodies (red fluorescence). B. Effect of E-cadherin overexpression on expression levels of EMT markers in EGFR-TKIs resistant cells. β-actin was used as an internal control. C. The IC50 of gefitinib/erlotinib in E-cadherin-overexpressing PC-9/GR-CDH1 and H460/ER-CDH1 cells was significantly greater than that in PC-9/GR- control and H460/ER-control cells. The data represent the means ± S.D. of three independent experiments. D, E. E-cadherin overexpression repressed migration of EGFR-TKIs-resistant cells according to Scratch assay. F, G. E-cadherin overexpression suppressed invasion of EGFR-TKIs-resistant cells according to transwell assay. www.impactjournals.com/oncotarget 44338 Oncotarget Figure 5: IGF1R activation led to EMT and decreased sensitivity against EGFR-TKIs in PC-9 and H460 cells upon IGF-1 induction. EGFR-TKIs-naïve lung cancer cells PC-9 and H460 were serum-starved overnight and then treated with fresh RPMI 1640 containing 0.5% FBS and 200 ng/ml IGF-I for 24 h. A. After IGF-1 induction, IGF1R and pIGF1R were activated. EMT phenotype, decreased expression of E-cadherin, increased Vimentin, nuclear β-catenin and Snail were observed. β-actin was used as an internal control. B. Mesenchymal phenotype of PC-9 and H460 cells after IGF-1 induction. C. β-catenin relocated from cell membrane to nucleus after IGF-1 induction as shown by immunofluorescence experiment. The photographs were taken at × 200 magnification. D. Exogenous IGF-1 application increased resistance to EGFR-TKIs in PC-9 and H460 cells. Data represent means ± S.D.of three independent experiments. www.impactjournals.com/oncotarget 44339 Oncotarget It is well established that EMT is involved in role of EMT in acquired resistance to EGFR-TKIs, we acquired drug resistance to EGFR-TKIs in NSCLC cells firstly induced EMT with TGF-β1 and then evaluated carrying wild-type EGFR, for instance, in A549 [8, 30] cells’ sensitivity to EGFR-TKIs before and after EMT in and in H1229 cells [30]. However, there is still paucity PC-9 and H460 cells, which are sensitive to EGFR-TKIs of evidence to support a role of EMT in acquired drug at baseline. We observed a mesenchymal phenotype and resistance to EGFR-TKIs in NSCLC cells with EGFR decreased sensitivity to EGFR-TKIs after EMT induction. mutations. By downregulating N-cadherin expression We then reversed EMT of PC-9/GR and H460/ER cells via siRNA, Zhang et al. [31] found that the loss of EMT by overexpressing E-cadherin and studied their sensitivity can reduce the proliferation and invasion of erlotinib- to EGFR-TKIs. Reversal of EMT-related morphological resistant H1650ER cells harboring a mutation in exon 19 and transcriptional features by E-cadherin overexpression of the EGFR gene. Vazquez-Martin et al. [27] showed in PC-9/GR and H460/ER cells was sufficient to recover that EMT also occurred after PC-9 cells acquired drug cells’ sensitivity to gefitinib or erlotinib. These data resistance to erlotinib. To obtain direct evidence for the suggested that EMT, independent of EGFR mutation and Figure 6: Effects of IGF1R on EMT and ERK/MAPK signaling of EGFR-TKIs-resistant cells. A, B. Morphologic change of PC-9/GR and H460/ER after IGF1R knockdown. Green fluorescence represented the successful transfection of IGF1R siRNA. C. Upregulation of E-cadherin. and downregulation of Vimentin. nuclear β-catenin and Snail by IGF1R silencing. D, E. Attenuated ERK/ MAPK signaling in PC-9 and H460 after IGF1R activation by IGF-1 and IGF1R knockdown. β-actin was used as an internal control.* p < 0.05 www.impactjournals.com/oncotarget 44340 Oncotarget MET amplification, is a novel mechanism of EGFR-TKIs- Both of the lung cancer cells were cultured in RPMI-1640 resistance. with 10% fetal bovine serum (FBS) in a 37°C incubator Previous studies have shown that IGF-1/IGF1R under 5% CO . Drug resistant PC-9 and H460 cells was activation can promote EMT in prostate cancer cells [15] obtained by continuous culturing the cells in gefitinib (2.5 and human mammary epithelial cells [32]. Nurwidya et μmol/L) or erlotinib (10 μmol/L), each for six months. The al. [33] found that inhibition of IGF1R reversed hypoxia- resistant PC-9 and H460 cells were designated as PC-9/ induced EMT. We further investigated whether one of GR and H460/ER cells, accordingly. the underlying mechanisms of IGF1R on acquired drug resistance is through EMT. We found that along with IGF1R DNA extraction and detection of EGFR gene upregulation, both PC-9/GR and H460/ER cells acquired mutations EMT phenotype. Exogenous IGF-1 administration in EGFR- TKIs-naïve PC-9 and H460 cells also triggered IGF1R Lung cancer cells in the logarithmic growth activation, and subsequently EMT. Furthermore, we showed phase were trypsinized and collected. DNA extraction that siIGF1R in PC-9/GR and H460/ER cells reversed EMT- was performed using QIAamp DNA Mini Kit (Qiagen, related morphological and transcriptional features, restored Germany), according to the instruction manual. The cells’ sensitivity to gefitinib or erlotinib. These data suggest integrity of the isolated DNA was assessed by 1.2% IGF1R plays an important role in acquired drug resistance agarose gel electrophoresis. EGFR mutations were against EGFR-TKIs by inducing EMT. detected using quantitative PCR-high-resolution melting The mechanism of IGF1R-induced EMT remained (qPCR-HRM) curve analysis technology [23]. unclear in current literatures. Loss of E-cadherin is a critical step in EMT and corresponds with the morphological and C-Met Fluorescence in situ hybridization cellular alterations [34, 35]. Factors that inhibit E-cadherin MET fluorescence in situ hybridization (FISH) was expression, such as nuclear transcription factor Snail, ZEB, performed on unstained formalin-fixed lung cancer cell Slug etc [36] are involved in EMT process. Internalization of suspension, using a MET/CEP7 probe cocktail (Kreatech the protein complex leads to the degradation of E-cadherin and Diagnostics, Amsterdam, Netherlands) according to the relocalization of β-catenin into the nucleus also contributes manufacturer’s instructions [24]. A MET/CEP ratio was to this process [37–39]. Data of the present study showed that established on the basis of a count of at least 200 cells. the expression of Snail was significantly increased after cells Specifically, two independent observers scored at least acquired drug resistance, and was significantly decreased 100 non-overlapping interphase nuclei to determine the when IGF1R was silenced, indicating that Snail may be the number of MET gene-specific (red) and CEP7-specific major transcription factors involved in EMT. Our results (green) signals. Samples with a ratio of MET/CEP7 also showed that β-catenin relocated from cell membrane greater than 2 were considered to have MET amplification into nucleus after IGF-1 induction in NSCLS cells, whereas (low amplification < 2, high amplification ≥ 2). β-catenin decreased when IGF1R was silenced. Graham et al. [15] showed that IGF-1 up-regulates ZEB1 expression through Cell proliferation assay with MTT the ERK/MAPK pathway in prostate cancer cells. We herein showed that activation of IGF1R by IGF-1 induced activation Cells in the logarithmic growth phase were of the pERK, but not of pAKT in both PC-9 and H460 cells, inoculated into 96-well culture plates at 1,500 cells whereas pERK decreased dramatically when IGF1R was per well. After adherent growth was observed, 100 μL silenced. Our results indicated that IGF1R may induce EMT in of culture medium (0.5% FBS) containing different NSCLC cells by upregulating the expression of Snail through concentrations of gefitinib and erlotinib were added to the activating ERK/AKT pathway, and promoting β-catenin cells. After 72 h incubation, 20 μL of MTT (5g/L) were translocation away from cytoplasmic membrane into the added to each well, and incubated for an additional 4 h. nucleus, which directly, repress the expression of E-cadherin. The supernatant was then discarded, and 150 μL of DMSO Taken together, our findings highlight the importance were added. The plates were then agitated for 10 min and of EMT in IGF1R-induced resistance to EGFR-TKIs the absorbance value (D) were measured at 492 nm with a in NSCLC. Interference of EMT or IGF1R might be a microplate reader. Cell survival rates after drug treatment promising therapeutic strategy for advanced NSCLC. and the half maximal inhibitory concentration (IC50) were calculated. Each treatment included six replicates, and MATERIALS AND METHODS three independent experiments were performed. Western blotting Cell preparation Cells from each group were detached with trypsin, Human lung cancer PC-9 cells (mutant EGFR) and centrifuged, and washed 3 times with pre-chilled PBS. H460 cells (wild-type EGFR) were provided by Shanghai Cell lysis buffer (Cell Signaling), containing protease Institute Cell Bank of the Chinese Academy of Sciences. www.impactjournals.com/oncotarget 44341 Oncotarget and phosphatase inhibitors, was subsequently added and and non-invasive cells remaining in the top chamber were incubated on ice for protein extraction. Nuclear β-catenin wiped using a cotton swab, followed by crystal violet was extracted using NE-PER Nuclear and Cytoplasmic staining. The cells were placed under an Olympus inverted Extraction Reagents (Thermo Scientific). Protein fluorescence microscope (200 ×) and photographed. Three concentration was determined using the BCA Bradford fields were chosen for counting, and the values were protein assay (Bio-Rad). Equal amounts of proteins were averaged. separated via 10% SDS-PAGE and then transferred to a membrane (Amersham). The membrane was blocked in SiIGF1R 5% bovine serum albumin (BSA) for 2 h and incubated According to the sequence information of IGF1R with an appropriate amount of primary antibody (working gene (GenBank No., NM-000875), three shRNA was dilutions of antibodies: Vimentin 1:100, E-cadherin designed using the RNAi design software provided 1:1,000, Snail 1:2,000, EGFR 1:400, p-EGFR 1:1,000, by Ambion. The most efficient interference target IGF1R 1:2,000, p-IGF-1R 1:1,000, ERK 1:2,000, p-ERK sequence “CGA AGA TTT CAC AGT CAA A” was 1:2,000, AKT1 1:2,000, p-AKT1 1:200, c-MET 1:1000 chosen (Supplementary Figure S1). Both strands of the and β-actin 1:400) (Amersham) in a shaker at 4°C interference sequences were synthesized by Shanghai overnight. Detection was by horseradish peroxidase- Genechem. DNA oligo: IGF1-siRNA-a: 5'-CCG GGC conjugated secondary antibodies and chemiluminescence. CGA AGA TTT CAC AGT CAA ACT CGA GTT TGA An integrated optical density analysis was performed CTG TGA AAT CTT CGG CTT TTT G-3' and IGF1- using Quantity One imaging analysis software. siRNA-b: 5'-AAT TCA AAA AGC CGA AGA TTT CAC AGT CAA ACT CGA GTT TGA CTG TGA AAT Immunofluorescence CTT CGG C-3'. Through BLAST homology analysis, Cells were plated into 12-well plates at 2.0 × 10 no homology with any other genes was found for the cells per well. After 48 h, the cells were fixed in 4% selected sequences. In addition, a nonsense sequence paraformaldehyde for 30 min. After washing with PBS was synthesized as the negative control sequence. for 3 times, the cells were blocked in blocking solution Plasmid DNA with the siRNA and control sequence (3% BSA, 0.3% Triton X-100) at room temperature for were purified and used for lentiviral packaging and 60 min. The cells were subsequently incubated with the titer determination. Then, 293T cells in the logarithmic primary antibodies (rabbit anti-E-cadherin at 1:50; rabbit phase were co-transfected with the recombinant pGC- anti-β-catenin at 1:50; CST) at 4°C overnight. The primary LV plasmid, pHelper 1.0 plasmid, and pHelper 2.0 antibodies were then removed, and secondary antibody plasmid using Lipofectamine 2000 (Invitrogen). After 8 was added (Cy3-conjugated secondary antibody at 1:500; h of transfection, the culture medium was replaced with Life Technologies), and incubated for 1 h. Finally, the complete culture medium. Following culture for another nuclei were stained with 0.5 μg/mL DAPI (Sigma), and 48 h, the supernatant with enriched lentiviral particles was the fixed cells were observed and photographed using collected. The supernatant was concentrated to obtain a an Olympus IX71 inverted fluorescence microscope high-titer lentiviral concentrate. PC-9/GR and H460/ER (Olympus Optical). cells in the logarithmic phase were trypsinized to obtain cell suspension and were inoculated into 6-well plates. When cell confluence reached approximately 30%, an Scratch test and invasion assay appropriate amount of virus was added at a multiplicity For Scratch test, cells from each group were of infection (MOI) of 100. Puromycin (5 μg/mL) was inoculated into 6-well plates at 5.0 × 10 cells per well. added 48 h later. The cells were photographed under a When cell confluence reached 90%, the cells were starved fluorescence microscope to observe the infection rate overnight in serum-free medium. Three parallel lines were 72 hours later. A negative control group transfected with scratched on the bottom of the culture plate using a sterile lentivirus expressing the empty IGF1R vector (siControl) 20 μL micropipette tip. The cells were washed twice using and a gene silencing group transfected with effective serum-free medium and cultured for another 48 h. The lentivirus expressing IGF1R (siIGF1R) were included in changes in cell motility were observed and photographed the experiment. under an Olympus inverted fluorescence phase-contrast microscope (40 ×). Three independent experiments were Stable LV-CDH1 (E-cadherin) gene performed. overexpression Cell invasion was determined using Boyden Chamber assay. The top transwell chamber was inoculated The CDH1(5386–1)-P1 and CDH1(5386–1)-P2 with 5 × 10 cells in 200 μL of serum-free culture medium. primers, containing exchange pairing bases, enzymatic The bottom chamber was filled with 500 μL of RPMI cutting sites, and a partial 5′-terminal target gene sequence, culture medium containing 10% FBS. After the cells were were designed, synthesized, and then used to amplify cultured at 37°C for 24 h, the top chamber was removed, the target gene. The sequences of the primers were as www.impactjournals.com/oncotarget 44342 Oncotarget follows: CDH1(5386–1)-P1: GAG GAT CCC CGG Bonomi P. Determinants of tumor response and survival GTA CCG GTC GCC ACC ATG GGC CCT TGG AGC with erlotinib in patients with non--small-cell lung cancer. CGC AG and CDH1(5386–1)-P2: TCA CCA TGG TGG J Clin Oncol. 2004; 22:3238-3247. CGA CCG GGT CGT CCT CGC CGC CTC CGT ACA. 2. Riely GJ, Pao W, Pham D, Li AR, Rizvi N, Venkatraman The amplification products were digested with enzymes ES, Zakowski MF, Kris MG, Ladanyi M, Miller VA. and ligated into the lentiviral expression vector GV218. Clinical course of patients with non-small cell lung cancer Following transformation, positive clones were picked, and and epidermal growth factor receptor exon 19 and exon 21 plasmids were extracted after the CDH1 (5386–1) sequence mutations treated with gefitinib or erlotinib. Clin Cancer was confirmed via sequencing. GV218-CDH1 (5386–1) Res. 2006; 12:839-844. and the empty control GV218 plasmid were obtained and 3. Sequist LV, Waltman BA, Dias-Santagata D, Digumarthy transfected into 293T cells. The cells were then cultured in S, Turke AB, Fidias P, Bergethon K, Shaw AT, Gettinger Opti-MEM culture medium containing 1% FBS for 48–72 S, Cosper AK, Akhavanfard S, Heist RS, Temel J, et al. h. H460/ER and PC-9/GR cells were infected at an MOI Genotypic and histological evolution of lung cancers of 100. Lentivirus production and cell infection procedures acquiring resistance to EGFR inhibitors. Sci Transl Med. were the same as described above. Cells with more than 80% 2011; 3:26r-75r. fluorescence rate were used for subsequent experiments. A 4. Kobayashi S, Boggon TJ, Dayaram T, Janne PA, Kocher negative control transfected with an empty vector lentivirus O, Meyerson M, Johnson BE, Eck MJ, Tenen DG, Halmos (Control) and an overexpression control transfected with B. EGFR mutation and resistance of non-small-cell lung effective lentivirus expressing CDH1 (CDH1) were included cancer to gefitinib. N Engl J Med. 2005; 352:786-792. in the experiment. 5. Shih JY, Gow CH and Yang PC. EGFR mutation conferring primary resistance to gefitinib in non-small-cell lung cancer. Exogenous IGF-1 and TGF-β1 induction N Engl J Med. 2005; 353:207-208. 6. Engelman JA, Zejnullahu K, Mitsudomi T, Song Y, Hyland EGFR-TKIs-naïve lung cancer cells PC-9 and H460 C, Park JO, Lindeman N, Gale CM, Zhao X, Christensen J, were serum-starved overnight and then treated with fresh Kosaka T, Holmes AJ, Rogers AM, et al. MET amplifica - RPMI 1640 containing 0.5% FBS and 200 ng/ml IGF-I tion leads to gefitinib resistance in lung cancer by activating (Sigma-aldrich) for 24 h or TGF-β1 (Sigma-aldrich, 10 ng/ ERBB3 signaling. Science. 2007; 316:1039-1043. mL) for 72 h before proceeding to downstream experiments. 7. Morgillo F, Woo JK, Kim ES, Hong WK, Lee HY. Statistical analysis Heterodimerization of insulin-like growth factor receptor/ epidermal growth factor receptor and induction of survivin Statistical analyses were performed using the SPSS expression counteract the antitumor action of erlotinib. 17.0 software. The data are presented as means ± standard Cancer Res. 2006; 66:10100-10111. deviation (SD). Comparisons between two groups were 8. Rho JK, Choi YJ, Lee JK, Ryoo BY, Na II, Yang SH, performed using the student t-test, while comparisons Kim CH, Lee JC. Epithelial to mesenchymal transi- among multiple groups were performed using one- tion derived from repeated exposure to gefitinib deter - way analysis of variance (ANOVA) with post hoc least mines the sensitivity to EGFR inhibitors in A549, a significant difference (LSD) test. P < 0.05 was considered non-small cell lung cancer cell line. Lung Cancer. 2009; to indicate a statistically significant difference. 63:219-226. 9. Peled N, Wynes MW, Ikeda N, Ohira T, Yoshida K, Qian ACKNOWLEDGMENTS J, Ilouze M, Brenner R, Kato Y, Mascaux C, Hirsch FR. Insulin-like growth factor-1 receptor (IGF-1R) as a bio- This work was supported by grants from the marker for resistance to the tyrosine kinase inhibitor gefi - National Natural Science Foundation of China (NO. tinib in non-small cell lung cancer. Cell Oncol (Dordr). 81472172, 81172225), and Science and technology 2013; 36:277-288. planning project of Guangdong Province (NO. 10. Yeo CD, Park KH, Park CK, Lee SH, Kim SJ, Yoon HK, 2014A020212254). The authors declare that they have Lee YS, Lee EJ, Lee KY, Kim T. Expression of insulin-like no competing interests. growth factor 1 receptor (IGF-1R) predicts poor responses to epidermal growth factor receptor (EGFR) tyrosine CONFLICTS OF INTEREST kinase inhibitors in non-small cell lung cancer patients harboring activating EGFR mutations. Lung Cancer. 2015; The authors declare no conflicts of interest. 87:311-317. 11. Desbois-Mouthon C, Baron A, Blivet-Van EM, Fartoux L, REFERENCES Venot C, Bladt F, Housset C, Rosmorduc O. Insulin-like growth factor-1 receptor inhibition induces a resistance 1. Perez-Soler R, Chachoua A, Hammond LA, Rowinsky EK, mechanism via the epidermal growth factor receptor/HER3/ Huberman M, Karp D, Rigas J, Clark GM, Santabarbara P, www.impactjournals.com/oncotarget 44343 Oncotarget AKT signaling pathway: rational basis for cotargeting insu- 22. Chen B, Xiao F, Li B, Xie B, Zhou J, Zheng J, Zhang W. lin-like growth factor-1 receptor and epidermal growth fac- The role of epithelial-mesenchymal transition and IGF-1R tor receptor in hepatocellular carcinoma. Clin Cancer Res. expression in prediction of gefitinib activity as the second- 2009; 15:5445-5456. line treatment for advanced nonsmall-cell lung cancer. Cancer Invest. 2013; 31:454-460. 12. Gallagher EJ, LeRoith D. The proliferating role of insulin and insulin-like growth factors in cancer. Trends Endocrinol 23. Chen B, Zheng J, Zeng Y, Li B, Xie B, Zheng J, Zhou J, Metab. 2010; 21:610-618. Zhang W. Sequence-dependent antiproliferative effects of gefitinib and docetaxel on non-small cell lung cancer 13. van der Veeken J, Oliveira S, Schiffelers RM, Storm G, (NSCLC) cells and the possible mechanism. Plos One. van Bergen EHP, Roovers RC. Crosstalk between epider- 2014; 9:e114074. mal growth factor receptor- and insulin-like growth factor-1 receptor signaling: implications for cancer therapy. Curr 24. Casadevall D, Gimeno J, Clave S, Taus A, Pijuan L, Arumi Cancer Drug Targets. 2009; 9:748-760. M, Lorenzo M, Menendez S, Canadas I, Albanell J, Serrano S, Espinet B, Salido M, et al. MET expression and copy 14. Irie HY, Pearline RV, Grueneberg D, Hsia M, Ravichandran number heterogeneity in nonsquamous non-small cell lung P, Kothari N, Natesan S, Brugge JS. Distinct roles of Akt1 cancer (nsNSCLC). Oncotarget. 2015; 6:16215-16226. doi: and Akt2 in regulating cell migration and epithelial-mesen- 10.18632/oncotarget.3976. chymal transition. J Cell Biol. 2005; 171:1023-1034. 25. Argast GM, Krueger JS, Thomson S, Sujka-Kwok I, Carey 15. Graham TR, Zhau HE, Odero-Marah VA, Osunkoya AO, K, Silva S, O'Connor M, Mercado P, Mulford IJ, Young GD, Kimbro KS, Tighiouart M, Liu T, Simons JW, O'Regan Sennello R, Wild R, Pachter JA, et al. Inducible expression of RM. Insulin-like growth factor-I-dependent up-regulation of TGFbeta, snail and Zeb1 recapitulates EMT in vitro and in vivo ZEB1 drives epithelial-to-mesenchymal transition in human in a NSCLC model. Clin Exp Metastasis. 2011; 28:593-614. prostate cancer cells. Cancer Res. 2008; 68:2479-2488. 26. Pirozzi G, Tirino V, Camerlingo R, Franco R, La Rocca A, 16. Thomson S, Buck E, Petti F, Griffin G, Brown E, Liguori E, Martucci N, Paino F, Normanno N, Rocco G. Ramnarine N, Iwata KK, Gibson N, Haley JD. Epithelial Epithelial to mesenchymal transition by TGFbeta-1 induc- to mesenchymal transition is a determinant of sensitivity tion increases stemness characteristics in primary non small of non-small-cell lung carcinoma cell lines and xenografts cell lung cancer cell line. Plos One. 2011; 6:e21548. to epidermal growth factor receptor inhibition. Cancer Res. 2005; 65:9455-9462. 27. Vazquez-Martin A, Cufi S, Oliveras-Ferraros C, Torres- Garcia VZ, Corominas-Faja B, Cuyas E, Bonavia R, Visa 17. Yauch RL, Januario T, Eberhard DA, Cavet G, Zhu W, Fu J, Martin-Castillo B, Barrajon-Catalan E, Micol V, Bosch- L, Pham TQ, Soriano R, Stinson J, Seshagiri S, Modrusan Barrera J, Menendez JA. IGF-1R/epithelial-to-mesenchy- Z, Lin CY, O'Neill V, et al. Epithelial versus mesenchymal mal transition (EMT) crosstalk suppresses the erlotinib- phenotype determines in vitro sensitivity and predicts clini- sensitizing effect of EGFR exon 19 deletion mutations. Sci cal activity of erlotinib in lung cancer patients. Clin Cancer Rep. 2013; 3:2560. Res. 2005; 11:8686-8698. 28. Cortot AB, Repellin CE, Shimamura T, Capelletti M, 18. Witta SE, Gemmill RM, Hirsch FR, Coldren CD, Hedman Zejnullahu K, Ercan D, Christensen JG, Wong KK, Gray K, Ravdel L, Helfrich B, Dziadziuszko R, Chan DC, NS, Janne PA. Resistance to irreversible EGF recep- Sugita M, Chan Z, Baron A, Franklin W, et al. Restoring tor tyrosine kinase inhibitors through a multistep mecha- E-cadherin expression increases sensitivity to epidermal nism involving the IGF1R pathway. Cancer Res. 2013; growth factor receptor inhibitors in lung cancer cell lines. 73:834-843. Cancer Res. 2006; 66:944-950. 29. Guix M, Faber AC, Wang SE, Olivares MG, Song Y, Qu 19. Miyanaga A, Gemma A, Ando M, Kosaihira S, Noro R, S, Rinehart C, Seidel B, Yee D, Arteaga CL, Engelman JA. Minegishi Y, Kataoka K, Nara M, Okano T, Miyazawa H, Acquired resistance to EGFR tyrosine kinase inhibitors in Tanaka T, Yoshimura A, Kobayashi K, et al. E-cadherin cancer cells is mediated by loss of IGF-binding proteins. J expression and epidermal growth factor receptor mutation Clin Invest. 2008; 118:2609-2619. status predict outcome in non-small cell lung cancer patients treated with gefitinib. Oncol Rep. 2008; 19:377-383. 30. Ahmad A, Maitah MY, Ginnebaugh KR, Li Y, Bao B, Gadgeel SM, Sarkar FH. Inhibition of Hedgehog signal- 20. Ren S, Su C, Wang Z, Li J, Fan L, Li B, Li X, Zhao C, Wu ing sensitizes NSCLC cells to standard therapies through C, Hou L, He Y, Gao G, Chen X, et al. Epithelial phenotype modulation of EMT-regulating miRNAs. J Hematol Oncol. as a predictive marker for response to EGFR-TKIs in non- 2013; 6:77. small cell lung cancer patients with wild-type EGFR. Int J Cancer. 2014; 135:2962-2971. 31. Zhang X, Liu G, Kang Y, Dong Z, Qian Q, Ma X. N-cadherin expression is associated with acquisition of 21. Frederick BA, Helfrich BA, Coldren CD, Zheng D, Chan EMT phenotype and with enhanced invasion in erlotinib- D, Bunn PJ, Raben D. Epithelial to mesenchymal transi- resistant lung cancer cell lines. Plos One. 2013; 8:e57692. tion predicts gefitinib resistance in cell lines of head and neck squamous cell carcinoma and non-small cell lung car- 32. Kim HJ, Litzenburger BC, Cui X, Delgado DA, Grabiner cinoma. Mol Cancer Ther. 2007; 6:1683-1691. BC, Lin X, Lewis MT, Gottardis MM, Wong TW, Attar www.impactjournals.com/oncotarget 44344 Oncotarget RM, Carboni JM, Lee AV. Constitutively active type I loss of adhesion and of responsiveness to the motogenic insulin-like growth factor receptor causes transformation trefoil factor 2 in colon carcinoma cells. Proc Natl Acad Sci and xenograft growth of immortalized mammary epithelial U S A. 1999; 96:2316-2321. cells and is accompanied by an epithelial-to-mesenchymal 36. Thiery JP, Acloque H, Huang RY, Nieto MA. Epithelial- transition mediated by NF-kappaB and snail. Mol Cell Biol. mesenchymal transitions in development and disease. Cell. 2007; 27:3165-3175. 2009; 139:871-890. 33. Nurwidya F, Takahashi F, Kobayashi I, Murakami A, 37. Liu Y, Ye X, Zhang JB, Ouyang H, Shen Z, Wu Y, Wang Kato M, Minakata K, Nara T, Hashimoto M, Yagishita W, Wu J, Tao S, Yang X, Qiao K, Zhang J, Liu J, Fu Q and S, Baskoro H, Hidayat M, Shimada N, Takahashi K. Xie Y. PROX1 promotes hepatocellular carcinoma prolif- Treatment with insulin-like growth factor 1 receptor inhibi- eration and sorafenib resistance by enhancing beta-catenin tor reverses hypoxia-induced epithelial-mesenchymal tran- expression and nuclear translocation. Oncogene. 2015.Epub sition in non-small cell lung cancer. Biochem Biophys Res ahead of print Commun. 2014; 455:332-338. 38. He W, He S, Wang Z, Shen H, Fang W, Zhang Y, Qian 34. Wheeler JM, Kim HC, Efstathiou JA, Ilyas M, Mortensen W, Lin M, Yuan J, Wang J, Huang W, Wang L, Ke Z. NJ, Bodmer WF. Hypermethylation of the promoter region Astrocyte elevated gene-1(AEG-1) induces epithelial-mes- of the E-cadherin gene (CDH1) in sporadic and ulcerative enchymal transition in lung cancer through activating Wnt/ colitis associated colorectal cancer. Gut. 2001; 48:367-371. beta-catenin signaling. BMC Cancer. 2015; 15:107. 35. Efstathiou JA, Liu D, Wheeler JM, Kim HC, Beck NE, 39. Morali OG, Delmas V, Moore R, Jeanney C, Thiery JP, Ilyas M, Karayiannakis AJ, Mortensen NJ, Kmiot W, Larue L. IGF-II induces rapid beta-catenin relocation to Playford RJ, Pignatelli M, Bodmer WF. Mutated epithelial the nucleus during epithelium to mesenchyme transition. cadherin is associated with increased tumorigenicity and Oncogene. 2001; 20:4942-4950. www.impactjournals.com/oncotarget 44345 Oncotarget http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Oncotarget Unpaywall

Implication of epithelial-mesenchymal transition in IGF1R-induced resistance to EGFR-TKIs in advanced non-small cell lung cancer

Download PDF
14 pages

Loading next page...
 
/lp/unpaywall/implication-of-epithelial-mesenchymal-transition-in-igf1r-induced-GX3TFeB6MQ

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
1949-2553
DOI
10.18632/oncotarget.6293
Publisher site
See Article on Publisher Site

Abstract

www.impactjournals.com/oncotarget/ Oncotarget, Vol. 6, No. 42 Implication of epithelial-mesenchymal transition in IGF1R- induced resistance to EGFR-TKIs in advanced non-small cell lung cancer 1,2,* 2,* 2,* 2 2 2 Juan Zhou , Jinjing Wang , Yunyun Zeng , Xi Zhang , Qiaoting Hu , Jihua Zheng , 2 2 2 Bei Chen , Bo Xie , Wei-Min Zhang Department of Oncology, Guangzhou Clinical College of the Second Military Medical University, Guangzhou, Guangdong 510010, China Department of Oncology, General Hospital of Guangzhou Military Command of PLA, Guangzhou, Guangdong 510010, China These authors have contributed equally to this work Correspondence to: Wei-Min Zhang, e-mail: [email protected] Keywords: epidermal growth factor receptor-tyrosine kinase inhibitors, epithelial-mesenchymal transition, type 1 insulin-like growth factor receptor, non-small cell lung cancer, drug resistance Received: May 21, 2015 Accepted: October 26, 2015 Published: November 05, 2015 ABSTRACT The underlying mechanisms for acquired resistance to epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs) in about 30%-40% of non-small cell lung cancer (NSCLC) patients remain elusive. Recent studies have suggested that activation of epithelial-mesenchymal transition (EMT) and type 1 insulin-like growth factor receptor (IGF1R) is associated with acquired EGFR-TKIs resistance in NSCLC. Our study aims to further explore the mechanism of EMT and IGF1R in acquired EGFR- TKIs resistance in NSCLC cell lines with mutant (PC-9) or wild-type EGFR (H460). Compared to parental cells, EGFR-TKIs-resistant PC-9/GR and H460/ER cells displayed an EMT phenotype and showed overexpression of IGF1R. SiIGF1R in PC-9/GR and H460/ER cells reversed EMT-related morphologies and reversed their resistance to EGFR-TKIs. Exogenous IGF-1 alone induced EMT in EGFR-TKIs-naïve PC-9 and H460 cells and increased their resistance against EGFR-TKIs. Inducing EMT by TGF-β1 in PC-9 and H460 cells decreased their sensitivity to EGFR-TKIs, whereas reversing EMT by E-cadherin overexpression in PC-9/GR and H460/ER cells restored their sensitivity to EGFR-TKIs. These data suggest that IGF1R plays an important role in acquired drug resistance against EGFR-TKIs by inducing EMT. Targeting IGF1R and EMT may be a potential therapeutic strategy for advanced NSCLC with acquired EGFR-TKIs resistance. The insulin-like growth factor (IGF) system, INTRODUCTION including IGF ligands, their receptors and binding proteins, Acquired drug resistance has become a bottleneck is important in promoting tumor development. Previous in the treatment of advanced non-small cell lung cancer studies showed that activation of IGF1R is involved in (NSCLC) using epidermal growth factor receptor-tyrosine EGFR-TKIs resistance in NSCLC cell lines [9] and in lung kinase inhibitors (EGFR-TKIs) [1–3]. Currently, the known cancer patients [10]. IGF1R tyrosine kinase inhibitors have mechanisms underlying this acquired drug resistance been reported to reverse the drug resistance of NSCLC to include T790M mutation [4, 5] and amplification of MET EGFR-TKIs in vitro and in vivo [7]. IGF1R activates the genes [6]. These mechanisms account for about 60–70% downstream pathways of EGFR signaling, such as the of acquired drug resistance. However, the underlying phosphatidylinositol 3 kinase/protein kinase B (PI3K/AKT) mechanisms for approximately 30%-40% of cases are still pathway and the extracellular signal-regulated kinases/ unclear. Recent studies show that the activation of epithelial- mitogen-activated protein kinase (ERK/MAPK) pathway, mesenchymal transition (EMT) and type 1 insulin-like leading to secondary drug resistance to EGFR-TKIs [11– growth factor receptor (IGF1R) is associated with acquired 13]. However, the exact mechanisms of IGF1R-induced drug resistance against EGFR-TKIs in NSCLC [7, 8]. acquired EGFR-TKIs resistance remain to be elucidated. www.impactjournals.com/oncotarget 44332 Oncotarget Interestingly, IGF1R has been shown to play an of EGFR and the phosphorylation of EGFR (pEGFR) important role in EMT [7] and IGF1R activation can showed no significant changes (Figure 1B). induce EMT in breast epithelial cells [14] and prostate To further evaluate whether activation of cancer cells [15]. EMT is a biological process of losing the IGF1R pathway was the major mechanism epithelial features and acquiring mesenchymal properties, underlying acquired drug resistance, we targeted characterized by E-cadherin reduction and Vimentin silencing of IGF1R using siRNA technology in PC-9/ induction. It has been reported that a subgroup of NSCLC GR and H460/ER cells. The expression of IGF1R with pronounced EMT was EGFR-TKIs resistant [3, 8, 16, in PC-9/GR-siIGF1R and H460/ER-siIGF1R cells 17], suggesting that EMT may render NSCLC insensitive was downregulated, indicating that the interference to EGFR inhibition. Furthermore, decreased expression was successful, while the expression of EGFR and of E-cadherin [8, 16, 17] was associated with reduced pEGFR did not show significant changes (Figure sensitivity to EGFR-TKIs, and restoration of E-cadherin 1C). PC-9/GR-siIGF1R and H460/ER-siIGF1R cells expression improved cells’ sensitivity to EGFR-TKIs [18]. had the same EGFR and KRAS gene state as their Consistently, clinical studies have suggested a prognostic parent cells (Supplementary Figure S4). Furthermore, value of E-cadherin in NSCLC patients treated with EGFR- knockdown of IGFIR significantly decreased the IC50 TKIs [19–21]. of gefitinib and erlotinib in PC-9/GR and H460/ER Previously, we reported the association between cells, respectively, suggesting that IGFIR may play an EMT, IGF1R expression and drug response in advanced important role in restoring their sensitivity to gefitinib NSCLC patients treated with gefitinib [22]. NSCLC or erlotinib (Figure 1D). patients with negative EMT or lower IGF1R expression have a significantly higher objective response rate. Both, EMT is involved in the acquirement of the IGF1R expression and EMT occurrence correlated with EGFR-TKIs-resistance phenotype the development of acquired drug resistance to EGFR- TKIs in NSCLC patients. In the present study, we further After the acquisition of drug resistance, PC-9/ examined the relationship between EMT and IGF1R GR and H460/ER cells showed EMT phenotypes, with expression with sensitivity to EGFR-TKIs in NSCLC cell loose cell junctions and long, spindle-type morphology lines with wild-type or mutant EGFR. Furthermore, using (Figure 2A). Expression of the epithelial marker in vitro assays, we provided evidence that IGF1R induced E-cadherin decreased in PC-9/GR cells, and was EGFR-TKIs resistance by inducing EMT and explored undetectable in H460 and H460/ER cells. In contrast, the possible cellular mechanism. Our data highlight the the mesenchymal marker Vimentin, transcription factor importance of EMT in IGF1R-induced resistance to Snail and nuclear β-catenin all increased in PC-9/GR EGFR-TKIs in NSCLC and implicate both EMT and and H460/ER cells, compared to their parental cells IGF1R as potential therapeutic targets for advanced (Figure 2B). Specifically, Vimentin expression gradually NSCLC. increased in H460/ER cells in a time-dependent manner upon erlotinib incubation (Figure 2C). Cell migration RESULTS and invasion abilities increased significantly in PC-9/ GR and H460/ER cells compared to their parental cells as shown by scratch test and transwell experiment IGF1R activation is involved in the acquirement (Figure 2D, 2E, 2F, 2G). of the EGFR-TKIs-resistance phenotype Treatment with TGF-β1 has also been shown in As expected, the resistant cells PC-9/GR and H460/ many cancer cells to promote a shift from epithelial ER exhibited significantly decreased sensitivity to EGFR- to mesenchymal phenotype [25, 26]. To further TKIs, compared to the parental PC-9 and H460 cells, evaluate whether EMT constituted a major mechanism respectively (Figure 1A). The delE746-A750 deletion underlying acquired drug resistance, parental PC-9 mutation in exon 19 of EGFR was detected in PC-9 and and H460 cells were treated with exogenous TGF-β1 PC-9/GR cells by qPCR-HRM, but not in H460 or H460/ (10 ng/mL) for 72 h. After TGF-β1 treatment, ER; however, the T790M mutation was not detected in any both PC-9 and H460 cells acquired a spindle-like of the cell lines. FISH analysis showed no amplification morphology (Figure 3A) and expressed notably of c-MET in PC-9/GR or H460/ER cells (Supplementary reduced E-cadherin and increased Vimentin proteins Figure S2). No EGFR mutation in H460/ER cells was (Figure 3B). Interestingly, following TGF-β1-induced found, and all cell lines harbored wild-type KRAS before transient EMT, the sensitivity of PC-9 and H460 to and after the induction of drug resistance (Supplementary EGFR-TKIs was dramatically reduced (Figure 3C), Figure S3). Additionally, the expression of IGF1R and which phenocopied the PC-9/GR and H460/ER cells, the phosphorylation of IGF1R (pIGF1R) increased respectively. These results indicated that transient significantly in PC-9/GR and H460/ER cells after the induction of mesenchymal-like phenotypes is sufficient acquisition of drug resistance, while the expression to induce resistance to EGFR-TKIs in NSCLC cells. www.impactjournals.com/oncotarget 44333 Oncotarget Figure 1: Role of IGF1R on the sensitivity to gefitinib and erlotinib in EGFR-TKIs-resistant cells. A. The sensitivity to gefitinib and erlotinib of PC-9/GR, H460/ER, and their parental cells was assessed by MTT assays. Cells were treated with the indicated doses of gefitinib or erlotinib for 72 h. IC50 values for different conditions are provided in the table within individual figures. B. Expression of IGF1R, phosphor-IGF1R, EGFR, and phosphor-EGFR in EGFR-TKIs-resistant cells by immunoblotting analysis. C. Effect of IGF1R siRNA on expression of IGF1R, phosphor-IGF1R, EGFR, and phosphor-EGFR in EGFR-TKIs-resistant cells. β-actin was used as an internal control. D. IC50 of gefitinib/erlotinib in PC-9/GR and H460/ER cells increased significantly following IGF1R knock-down when compared with the control cells. Data represent means ± S.D. of three independent experiments. www.impactjournals.com/oncotarget 44334 Oncotarget Figure 2: EMT in EGFR-TKIs-resistant cells. A. Morphology of PC-9/GR, H460/ER, and their parental cells grown for 3 days until 90% confluence. In contrast to the parental cells, the PC-9/GR and H460/ER cells displayed long spindle-like shape with loose cell junctions. Photographs were taken at × 200 magnification. B–C. Loss of E-cadherin was seen in PC-9/GR cells, H460 cells did not express E-cadherin, and there was increased expression of Vimentin, transcription factor Snail and nuclear β-catenin in PC-9/GR and H460/ER cells shown by immunoblotting analysis. In addition, Vimentin expression at the protein level increased in a time dependent manner after the induction of drug resistance. β-actin was used as an internal control. D, E. Enhanced migratory capacity of EGFR-TKIs-resistant cells according to Scratch assay. Confluent cells were scraped by a pipette tip to generate wounds and then were cultured in serum-free culture medium for 48 h. Representative images of wounds were taken at 0 and 48 h. Cell motility was examined with a light microscope (×40) and the width of the wound was quantified. F, G. Enhanced invasiveness of EGFR-TKIs-resistant cells according to transwell assay. The cells were incubated for 24 h in modified Boyden chambers. Those cells that migrated through the filters were stained and counted under a light microscope. Quantification was done in 10 randomly chosen fields. The data are reported as means ± S.D. The photographs were taken at × 200 magnification. www.impactjournals.com/oncotarget 44335 Oncotarget E-cadherin was stably transfected into PC-9/GR and H460/ Inhibition of EMT increased sensitivity to ER cells. The EGFR and KRAS gene status in the resulting EGFR-TKIs in NSCLC cells cells, PC9/GR-CDH1 and H460/ER-CDH1, remained the To further evaluate the role of EMT in the resistance same as their parent cells (Supplementary Figure S5). PC9/ to EGFR-TKIs in NSCLC cells, CDH1 which encodes GR-CDH1 and H460/ER-CDH1 presented an epithelial- Figure 3: Effect of TGF-β1 on EMT and the sensitivity to gefitinib and erlotinib in EGFR-TKIs-resistant cells. A. Morphology of PC-9 and H460 cells grown with 10 ng/mL TGF-β1 for 3 days until 90% confluence. The photographs were taken at × 200 magnification. B. TGF-β1-induced downregulation of E-cadherin and upregulation of Vimentin in PC-9 and H460 cells according to immunoblotting analysis. β-actin was used as an internal control. C. The effects of sequential treatment with the TGF-β1 on cell viability of PC-9 and H460 cells exposed to gefitinib and erlotinib by MTT uptake assays, respectively. The data represent the means ± S.D. of three independent experiments. www.impactjournals.com/oncotarget 44336 Oncotarget like morphology (Figure 4A). The expression of PC9/GR and H460/ER cells (Figure 6C, Supplementary E-cadherin in PC-9/GR-CDH1 and H460/ER-CDH1 cells Figure S6B). Furthermore, activation of IGF1R by IGF-1 was significantly higher than that in the empty vector induced activation of pERK, but not of pAKT in PC9 and controls, indicating that the lentiviral transfection of H460 cells (Figure 6D and 6E). Finally, knockdown of E-cadherin was successful. Additionally, compared to the IGF1R dramatically attenuated the increase in total ERK empty vector control, the expression of Vimentin, Snail and pERK in PC-9/GR, and resulted in decreased pERK and nuclear β-catenin decreased in PC-9/GR-CDH1 and in H460/ER (Figure 6D and 6E). H460/ER-CDH1 cells (Figure 4B). Overexpression of CDH1 correlated with a reduction of the nuclear β-catenin DISCUSSION in PC-9/GR and H460/ER by immunofluorescence staining experiment (Supplementary Figure S6A). Most In the present study, both PC-9 cells, which harbor importantly, the sensitivity to gefitinib and erlotinib the mutant EGFR, and H460 cells, with wild-type EGFR, dramatically increased in E-cadherin-overexpressing were used to establish EGFR-TKIs-resistant sublines PC9/GR and H460/ER cells, respectively (Figure 4C), (PC-9/GR and H460/ER) by continuously culturing in suggesting that E-cadherin repressed EMT in NSCLC gefitinib or erlotinib, respectively. We did not detect the cells, and subsequently enhanced the cytotoxic effect classical T790M mutation or c-MET gene amplification of EGFR-TKIs. Finally, overexpression of E-cadherin in either PC-9/GR or H460/ER cells. However, inhibited the motility and invasiveness of EGFR-TKIs- expression of IGF1R and pIGF1R, but not that of EGFR, resistant NSCLC cells shown by scratch test and transwell were enhanced in these two resistant cell lines. These assays (Figure 4D, 4E, 4F, 4G). results suggested that IGF1R has a major role in acquired drug resistance to EGFR-TKIs, which is consistent with previous studies. For example, Morgillo et al. [7] found Exogenous IGF-1 induced EMT in EGFR-TKIs- decreased EGFR and increased pIGF1R in erlotinib- naïve NSCLC cells and increased their resistance resistant H460 cells. They also found that the combined to EGFR-TKIs treatment of an IGF1R tyrosine kinase inhibitor with erlotinib inhibits cell proliferation significantly. Vazquez- To test whether IGF1R overexpression alone can Martin et al. [27] and Cortot et al. [28] also suggested a trigger EMT and EGFR-TKIs resistance, exogenous IGF- role of IGF1R in acquired drug resistance to EGFR-TKIs 1 was applied to EGFR-TKIs-naïve PC-9 and H460 cells. in PC-9 cells. Twenty-four hours after IGF-1 induction, both IGF1R and To further evaluate the role of IGF1R in acquired pIGF1R were induced, and no significant change of EGFR EGFR-TKIs resistance, we knocked down IGF1R by was observed (Figure 5A). The IGF-1 induced cells also siRNA in PC-9/GR and H460/ER cells and found that acquired a mesenchymal phenotype and a spindle-like both cells restored their sensitivity to gefitinib or erlotinib. morphology (Figure 5B). EMT occurrence was further Furthermore, our data showed that exogenous IGF-1 confirmed by decreased expression of E-cadherin, and application in EGFR-TKIs-naïve PC-9 and H460 cells increased expression of Vimentin, nuclear β-catenin and triggered IGF1R activation, and increased cells’ resistance Snail (Figure 5A). In addition, β-catenin was shown to to EGFR-TKIs. Guix et al. [29] also discovered that IGF- translocate from cell membrane into the nuclear after IGF- 1 activated IGF1R in lung cancer cells, and showed that 1 induction by Immunofluorescence experiment (Figure only cells, which depended on IGF1R and its downstream 5C). IGF-1 alone does not promote lung cancer cell signaling pathways to promote cell growth, acquired the proliferation (data not shown), however IGF-1 induced resistance ability against geftinib. Taken together, our PC-9 and H460 cells showed decreased sensitivity to results provided direct evidence that activation of IGF1R EGFR-TKIs compared to parental cells (Figure 5D). These is one of the mechanisms underlying acquired drug results indicated that activation of IGF1R is associated resistance to EGFR-TKIs in lung cancer cells. with EMT process. There are still few studies addressing the mechanism of IGF1R activation in EGFR-TKIs resistance in lung IGF1R induced EMT of NSCLC cells by cancer cells. One possible mechanism is that after EGFR enhancing ERK/MAPK signaling pathway is inhibited, IGF1R pathway is activated to To clarify the association among IGF1R activation, promote cell survival and proliferation [7]. Morgillo et al. EMT, and resistance to EGFR-TKIs, we explored whether [7] found that erlotinib treatment of H460 cells increase the induction of EMT might proceed through the activation the levels of EGFR/IGFIR heterodimer on cell membrane, of IGF1R to promote resistance to EGFR-TKIs in NSCLC subsequently activate IGFIR and its downstream signaling cells. Knockdown of IGF1R by siRNA repressed the EMT- mediators. Data in the present study, which were consistent related morphological features (Figure 6A and 6B). We with Peled et al’s [9] report, showed that exogenous IGF-1 also demonstrated that knockdown of IGF1R attenuated alone does not promote lung cancer cell proliferation, but the decrease of E-cadherin expression and the increase instead increases cell survival against erlotinib. Further of Vimentin, Snail and nuclear β-catenin expression in study is needed in this regard. www.impactjournals.com/oncotarget 44337 Oncotarget Figure 4: Effect of E-cadherin (CDH1) overexpression on EMT and the sensitivity to gefitinib and erlotinib in EGFR- TKIs-resistant cells. A. E-cadherin-overexpressing cell lines PC-9/GR and H460/ER after CDH1 transfection showed an epithelial- like morphology and a remarkably increased expression of E-cadherin according to immunofluorescence assay. The nuclei were stained with DAPI (blue fluorescence), and E-cadherin was stained with Cy3-conjugated antibodies (red fluorescence). B. Effect of E-cadherin overexpression on expression levels of EMT markers in EGFR-TKIs resistant cells. β-actin was used as an internal control. C. The IC50 of gefitinib/erlotinib in E-cadherin-overexpressing PC-9/GR-CDH1 and H460/ER-CDH1 cells was significantly greater than that in PC-9/GR- control and H460/ER-control cells. The data represent the means ± S.D. of three independent experiments. D, E. E-cadherin overexpression repressed migration of EGFR-TKIs-resistant cells according to Scratch assay. F, G. E-cadherin overexpression suppressed invasion of EGFR-TKIs-resistant cells according to transwell assay. www.impactjournals.com/oncotarget 44338 Oncotarget Figure 5: IGF1R activation led to EMT and decreased sensitivity against EGFR-TKIs in PC-9 and H460 cells upon IGF-1 induction. EGFR-TKIs-naïve lung cancer cells PC-9 and H460 were serum-starved overnight and then treated with fresh RPMI 1640 containing 0.5% FBS and 200 ng/ml IGF-I for 24 h. A. After IGF-1 induction, IGF1R and pIGF1R were activated. EMT phenotype, decreased expression of E-cadherin, increased Vimentin, nuclear β-catenin and Snail were observed. β-actin was used as an internal control. B. Mesenchymal phenotype of PC-9 and H460 cells after IGF-1 induction. C. β-catenin relocated from cell membrane to nucleus after IGF-1 induction as shown by immunofluorescence experiment. The photographs were taken at × 200 magnification. D. Exogenous IGF-1 application increased resistance to EGFR-TKIs in PC-9 and H460 cells. Data represent means ± S.D.of three independent experiments. www.impactjournals.com/oncotarget 44339 Oncotarget It is well established that EMT is involved in role of EMT in acquired resistance to EGFR-TKIs, we acquired drug resistance to EGFR-TKIs in NSCLC cells firstly induced EMT with TGF-β1 and then evaluated carrying wild-type EGFR, for instance, in A549 [8, 30] cells’ sensitivity to EGFR-TKIs before and after EMT in and in H1229 cells [30]. However, there is still paucity PC-9 and H460 cells, which are sensitive to EGFR-TKIs of evidence to support a role of EMT in acquired drug at baseline. We observed a mesenchymal phenotype and resistance to EGFR-TKIs in NSCLC cells with EGFR decreased sensitivity to EGFR-TKIs after EMT induction. mutations. By downregulating N-cadherin expression We then reversed EMT of PC-9/GR and H460/ER cells via siRNA, Zhang et al. [31] found that the loss of EMT by overexpressing E-cadherin and studied their sensitivity can reduce the proliferation and invasion of erlotinib- to EGFR-TKIs. Reversal of EMT-related morphological resistant H1650ER cells harboring a mutation in exon 19 and transcriptional features by E-cadherin overexpression of the EGFR gene. Vazquez-Martin et al. [27] showed in PC-9/GR and H460/ER cells was sufficient to recover that EMT also occurred after PC-9 cells acquired drug cells’ sensitivity to gefitinib or erlotinib. These data resistance to erlotinib. To obtain direct evidence for the suggested that EMT, independent of EGFR mutation and Figure 6: Effects of IGF1R on EMT and ERK/MAPK signaling of EGFR-TKIs-resistant cells. A, B. Morphologic change of PC-9/GR and H460/ER after IGF1R knockdown. Green fluorescence represented the successful transfection of IGF1R siRNA. C. Upregulation of E-cadherin. and downregulation of Vimentin. nuclear β-catenin and Snail by IGF1R silencing. D, E. Attenuated ERK/ MAPK signaling in PC-9 and H460 after IGF1R activation by IGF-1 and IGF1R knockdown. β-actin was used as an internal control.* p < 0.05 www.impactjournals.com/oncotarget 44340 Oncotarget MET amplification, is a novel mechanism of EGFR-TKIs- Both of the lung cancer cells were cultured in RPMI-1640 resistance. with 10% fetal bovine serum (FBS) in a 37°C incubator Previous studies have shown that IGF-1/IGF1R under 5% CO . Drug resistant PC-9 and H460 cells was activation can promote EMT in prostate cancer cells [15] obtained by continuous culturing the cells in gefitinib (2.5 and human mammary epithelial cells [32]. Nurwidya et μmol/L) or erlotinib (10 μmol/L), each for six months. The al. [33] found that inhibition of IGF1R reversed hypoxia- resistant PC-9 and H460 cells were designated as PC-9/ induced EMT. We further investigated whether one of GR and H460/ER cells, accordingly. the underlying mechanisms of IGF1R on acquired drug resistance is through EMT. We found that along with IGF1R DNA extraction and detection of EGFR gene upregulation, both PC-9/GR and H460/ER cells acquired mutations EMT phenotype. Exogenous IGF-1 administration in EGFR- TKIs-naïve PC-9 and H460 cells also triggered IGF1R Lung cancer cells in the logarithmic growth activation, and subsequently EMT. Furthermore, we showed phase were trypsinized and collected. DNA extraction that siIGF1R in PC-9/GR and H460/ER cells reversed EMT- was performed using QIAamp DNA Mini Kit (Qiagen, related morphological and transcriptional features, restored Germany), according to the instruction manual. The cells’ sensitivity to gefitinib or erlotinib. These data suggest integrity of the isolated DNA was assessed by 1.2% IGF1R plays an important role in acquired drug resistance agarose gel electrophoresis. EGFR mutations were against EGFR-TKIs by inducing EMT. detected using quantitative PCR-high-resolution melting The mechanism of IGF1R-induced EMT remained (qPCR-HRM) curve analysis technology [23]. unclear in current literatures. Loss of E-cadherin is a critical step in EMT and corresponds with the morphological and C-Met Fluorescence in situ hybridization cellular alterations [34, 35]. Factors that inhibit E-cadherin MET fluorescence in situ hybridization (FISH) was expression, such as nuclear transcription factor Snail, ZEB, performed on unstained formalin-fixed lung cancer cell Slug etc [36] are involved in EMT process. Internalization of suspension, using a MET/CEP7 probe cocktail (Kreatech the protein complex leads to the degradation of E-cadherin and Diagnostics, Amsterdam, Netherlands) according to the relocalization of β-catenin into the nucleus also contributes manufacturer’s instructions [24]. A MET/CEP ratio was to this process [37–39]. Data of the present study showed that established on the basis of a count of at least 200 cells. the expression of Snail was significantly increased after cells Specifically, two independent observers scored at least acquired drug resistance, and was significantly decreased 100 non-overlapping interphase nuclei to determine the when IGF1R was silenced, indicating that Snail may be the number of MET gene-specific (red) and CEP7-specific major transcription factors involved in EMT. Our results (green) signals. Samples with a ratio of MET/CEP7 also showed that β-catenin relocated from cell membrane greater than 2 were considered to have MET amplification into nucleus after IGF-1 induction in NSCLS cells, whereas (low amplification < 2, high amplification ≥ 2). β-catenin decreased when IGF1R was silenced. Graham et al. [15] showed that IGF-1 up-regulates ZEB1 expression through Cell proliferation assay with MTT the ERK/MAPK pathway in prostate cancer cells. We herein showed that activation of IGF1R by IGF-1 induced activation Cells in the logarithmic growth phase were of the pERK, but not of pAKT in both PC-9 and H460 cells, inoculated into 96-well culture plates at 1,500 cells whereas pERK decreased dramatically when IGF1R was per well. After adherent growth was observed, 100 μL silenced. Our results indicated that IGF1R may induce EMT in of culture medium (0.5% FBS) containing different NSCLC cells by upregulating the expression of Snail through concentrations of gefitinib and erlotinib were added to the activating ERK/AKT pathway, and promoting β-catenin cells. After 72 h incubation, 20 μL of MTT (5g/L) were translocation away from cytoplasmic membrane into the added to each well, and incubated for an additional 4 h. nucleus, which directly, repress the expression of E-cadherin. The supernatant was then discarded, and 150 μL of DMSO Taken together, our findings highlight the importance were added. The plates were then agitated for 10 min and of EMT in IGF1R-induced resistance to EGFR-TKIs the absorbance value (D) were measured at 492 nm with a in NSCLC. Interference of EMT or IGF1R might be a microplate reader. Cell survival rates after drug treatment promising therapeutic strategy for advanced NSCLC. and the half maximal inhibitory concentration (IC50) were calculated. Each treatment included six replicates, and MATERIALS AND METHODS three independent experiments were performed. Western blotting Cell preparation Cells from each group were detached with trypsin, Human lung cancer PC-9 cells (mutant EGFR) and centrifuged, and washed 3 times with pre-chilled PBS. H460 cells (wild-type EGFR) were provided by Shanghai Cell lysis buffer (Cell Signaling), containing protease Institute Cell Bank of the Chinese Academy of Sciences. www.impactjournals.com/oncotarget 44341 Oncotarget and phosphatase inhibitors, was subsequently added and and non-invasive cells remaining in the top chamber were incubated on ice for protein extraction. Nuclear β-catenin wiped using a cotton swab, followed by crystal violet was extracted using NE-PER Nuclear and Cytoplasmic staining. The cells were placed under an Olympus inverted Extraction Reagents (Thermo Scientific). Protein fluorescence microscope (200 ×) and photographed. Three concentration was determined using the BCA Bradford fields were chosen for counting, and the values were protein assay (Bio-Rad). Equal amounts of proteins were averaged. separated via 10% SDS-PAGE and then transferred to a membrane (Amersham). The membrane was blocked in SiIGF1R 5% bovine serum albumin (BSA) for 2 h and incubated According to the sequence information of IGF1R with an appropriate amount of primary antibody (working gene (GenBank No., NM-000875), three shRNA was dilutions of antibodies: Vimentin 1:100, E-cadherin designed using the RNAi design software provided 1:1,000, Snail 1:2,000, EGFR 1:400, p-EGFR 1:1,000, by Ambion. The most efficient interference target IGF1R 1:2,000, p-IGF-1R 1:1,000, ERK 1:2,000, p-ERK sequence “CGA AGA TTT CAC AGT CAA A” was 1:2,000, AKT1 1:2,000, p-AKT1 1:200, c-MET 1:1000 chosen (Supplementary Figure S1). Both strands of the and β-actin 1:400) (Amersham) in a shaker at 4°C interference sequences were synthesized by Shanghai overnight. Detection was by horseradish peroxidase- Genechem. DNA oligo: IGF1-siRNA-a: 5'-CCG GGC conjugated secondary antibodies and chemiluminescence. CGA AGA TTT CAC AGT CAA ACT CGA GTT TGA An integrated optical density analysis was performed CTG TGA AAT CTT CGG CTT TTT G-3' and IGF1- using Quantity One imaging analysis software. siRNA-b: 5'-AAT TCA AAA AGC CGA AGA TTT CAC AGT CAA ACT CGA GTT TGA CTG TGA AAT Immunofluorescence CTT CGG C-3'. Through BLAST homology analysis, Cells were plated into 12-well plates at 2.0 × 10 no homology with any other genes was found for the cells per well. After 48 h, the cells were fixed in 4% selected sequences. In addition, a nonsense sequence paraformaldehyde for 30 min. After washing with PBS was synthesized as the negative control sequence. for 3 times, the cells were blocked in blocking solution Plasmid DNA with the siRNA and control sequence (3% BSA, 0.3% Triton X-100) at room temperature for were purified and used for lentiviral packaging and 60 min. The cells were subsequently incubated with the titer determination. Then, 293T cells in the logarithmic primary antibodies (rabbit anti-E-cadherin at 1:50; rabbit phase were co-transfected with the recombinant pGC- anti-β-catenin at 1:50; CST) at 4°C overnight. The primary LV plasmid, pHelper 1.0 plasmid, and pHelper 2.0 antibodies were then removed, and secondary antibody plasmid using Lipofectamine 2000 (Invitrogen). After 8 was added (Cy3-conjugated secondary antibody at 1:500; h of transfection, the culture medium was replaced with Life Technologies), and incubated for 1 h. Finally, the complete culture medium. Following culture for another nuclei were stained with 0.5 μg/mL DAPI (Sigma), and 48 h, the supernatant with enriched lentiviral particles was the fixed cells were observed and photographed using collected. The supernatant was concentrated to obtain a an Olympus IX71 inverted fluorescence microscope high-titer lentiviral concentrate. PC-9/GR and H460/ER (Olympus Optical). cells in the logarithmic phase were trypsinized to obtain cell suspension and were inoculated into 6-well plates. When cell confluence reached approximately 30%, an Scratch test and invasion assay appropriate amount of virus was added at a multiplicity For Scratch test, cells from each group were of infection (MOI) of 100. Puromycin (5 μg/mL) was inoculated into 6-well plates at 5.0 × 10 cells per well. added 48 h later. The cells were photographed under a When cell confluence reached 90%, the cells were starved fluorescence microscope to observe the infection rate overnight in serum-free medium. Three parallel lines were 72 hours later. A negative control group transfected with scratched on the bottom of the culture plate using a sterile lentivirus expressing the empty IGF1R vector (siControl) 20 μL micropipette tip. The cells were washed twice using and a gene silencing group transfected with effective serum-free medium and cultured for another 48 h. The lentivirus expressing IGF1R (siIGF1R) were included in changes in cell motility were observed and photographed the experiment. under an Olympus inverted fluorescence phase-contrast microscope (40 ×). Three independent experiments were Stable LV-CDH1 (E-cadherin) gene performed. overexpression Cell invasion was determined using Boyden Chamber assay. The top transwell chamber was inoculated The CDH1(5386–1)-P1 and CDH1(5386–1)-P2 with 5 × 10 cells in 200 μL of serum-free culture medium. primers, containing exchange pairing bases, enzymatic The bottom chamber was filled with 500 μL of RPMI cutting sites, and a partial 5′-terminal target gene sequence, culture medium containing 10% FBS. After the cells were were designed, synthesized, and then used to amplify cultured at 37°C for 24 h, the top chamber was removed, the target gene. The sequences of the primers were as www.impactjournals.com/oncotarget 44342 Oncotarget follows: CDH1(5386–1)-P1: GAG GAT CCC CGG Bonomi P. Determinants of tumor response and survival GTA CCG GTC GCC ACC ATG GGC CCT TGG AGC with erlotinib in patients with non--small-cell lung cancer. CGC AG and CDH1(5386–1)-P2: TCA CCA TGG TGG J Clin Oncol. 2004; 22:3238-3247. CGA CCG GGT CGT CCT CGC CGC CTC CGT ACA. 2. Riely GJ, Pao W, Pham D, Li AR, Rizvi N, Venkatraman The amplification products were digested with enzymes ES, Zakowski MF, Kris MG, Ladanyi M, Miller VA. and ligated into the lentiviral expression vector GV218. Clinical course of patients with non-small cell lung cancer Following transformation, positive clones were picked, and and epidermal growth factor receptor exon 19 and exon 21 plasmids were extracted after the CDH1 (5386–1) sequence mutations treated with gefitinib or erlotinib. Clin Cancer was confirmed via sequencing. GV218-CDH1 (5386–1) Res. 2006; 12:839-844. and the empty control GV218 plasmid were obtained and 3. Sequist LV, Waltman BA, Dias-Santagata D, Digumarthy transfected into 293T cells. The cells were then cultured in S, Turke AB, Fidias P, Bergethon K, Shaw AT, Gettinger Opti-MEM culture medium containing 1% FBS for 48–72 S, Cosper AK, Akhavanfard S, Heist RS, Temel J, et al. h. H460/ER and PC-9/GR cells were infected at an MOI Genotypic and histological evolution of lung cancers of 100. Lentivirus production and cell infection procedures acquiring resistance to EGFR inhibitors. Sci Transl Med. were the same as described above. Cells with more than 80% 2011; 3:26r-75r. fluorescence rate were used for subsequent experiments. A 4. Kobayashi S, Boggon TJ, Dayaram T, Janne PA, Kocher negative control transfected with an empty vector lentivirus O, Meyerson M, Johnson BE, Eck MJ, Tenen DG, Halmos (Control) and an overexpression control transfected with B. EGFR mutation and resistance of non-small-cell lung effective lentivirus expressing CDH1 (CDH1) were included cancer to gefitinib. N Engl J Med. 2005; 352:786-792. in the experiment. 5. Shih JY, Gow CH and Yang PC. EGFR mutation conferring primary resistance to gefitinib in non-small-cell lung cancer. Exogenous IGF-1 and TGF-β1 induction N Engl J Med. 2005; 353:207-208. 6. Engelman JA, Zejnullahu K, Mitsudomi T, Song Y, Hyland EGFR-TKIs-naïve lung cancer cells PC-9 and H460 C, Park JO, Lindeman N, Gale CM, Zhao X, Christensen J, were serum-starved overnight and then treated with fresh Kosaka T, Holmes AJ, Rogers AM, et al. MET amplifica - RPMI 1640 containing 0.5% FBS and 200 ng/ml IGF-I tion leads to gefitinib resistance in lung cancer by activating (Sigma-aldrich) for 24 h or TGF-β1 (Sigma-aldrich, 10 ng/ ERBB3 signaling. Science. 2007; 316:1039-1043. mL) for 72 h before proceeding to downstream experiments. 7. Morgillo F, Woo JK, Kim ES, Hong WK, Lee HY. Statistical analysis Heterodimerization of insulin-like growth factor receptor/ epidermal growth factor receptor and induction of survivin Statistical analyses were performed using the SPSS expression counteract the antitumor action of erlotinib. 17.0 software. The data are presented as means ± standard Cancer Res. 2006; 66:10100-10111. deviation (SD). Comparisons between two groups were 8. Rho JK, Choi YJ, Lee JK, Ryoo BY, Na II, Yang SH, performed using the student t-test, while comparisons Kim CH, Lee JC. Epithelial to mesenchymal transi- among multiple groups were performed using one- tion derived from repeated exposure to gefitinib deter - way analysis of variance (ANOVA) with post hoc least mines the sensitivity to EGFR inhibitors in A549, a significant difference (LSD) test. P < 0.05 was considered non-small cell lung cancer cell line. Lung Cancer. 2009; to indicate a statistically significant difference. 63:219-226. 9. Peled N, Wynes MW, Ikeda N, Ohira T, Yoshida K, Qian ACKNOWLEDGMENTS J, Ilouze M, Brenner R, Kato Y, Mascaux C, Hirsch FR. Insulin-like growth factor-1 receptor (IGF-1R) as a bio- This work was supported by grants from the marker for resistance to the tyrosine kinase inhibitor gefi - National Natural Science Foundation of China (NO. tinib in non-small cell lung cancer. Cell Oncol (Dordr). 81472172, 81172225), and Science and technology 2013; 36:277-288. planning project of Guangdong Province (NO. 10. Yeo CD, Park KH, Park CK, Lee SH, Kim SJ, Yoon HK, 2014A020212254). The authors declare that they have Lee YS, Lee EJ, Lee KY, Kim T. Expression of insulin-like no competing interests. growth factor 1 receptor (IGF-1R) predicts poor responses to epidermal growth factor receptor (EGFR) tyrosine CONFLICTS OF INTEREST kinase inhibitors in non-small cell lung cancer patients harboring activating EGFR mutations. Lung Cancer. 2015; The authors declare no conflicts of interest. 87:311-317. 11. Desbois-Mouthon C, Baron A, Blivet-Van EM, Fartoux L, REFERENCES Venot C, Bladt F, Housset C, Rosmorduc O. Insulin-like growth factor-1 receptor inhibition induces a resistance 1. Perez-Soler R, Chachoua A, Hammond LA, Rowinsky EK, mechanism via the epidermal growth factor receptor/HER3/ Huberman M, Karp D, Rigas J, Clark GM, Santabarbara P, www.impactjournals.com/oncotarget 44343 Oncotarget AKT signaling pathway: rational basis for cotargeting insu- 22. Chen B, Xiao F, Li B, Xie B, Zhou J, Zheng J, Zhang W. lin-like growth factor-1 receptor and epidermal growth fac- The role of epithelial-mesenchymal transition and IGF-1R tor receptor in hepatocellular carcinoma. Clin Cancer Res. expression in prediction of gefitinib activity as the second- 2009; 15:5445-5456. line treatment for advanced nonsmall-cell lung cancer. Cancer Invest. 2013; 31:454-460. 12. Gallagher EJ, LeRoith D. The proliferating role of insulin and insulin-like growth factors in cancer. Trends Endocrinol 23. Chen B, Zheng J, Zeng Y, Li B, Xie B, Zheng J, Zhou J, Metab. 2010; 21:610-618. Zhang W. Sequence-dependent antiproliferative effects of gefitinib and docetaxel on non-small cell lung cancer 13. van der Veeken J, Oliveira S, Schiffelers RM, Storm G, (NSCLC) cells and the possible mechanism. Plos One. van Bergen EHP, Roovers RC. Crosstalk between epider- 2014; 9:e114074. mal growth factor receptor- and insulin-like growth factor-1 receptor signaling: implications for cancer therapy. Curr 24. Casadevall D, Gimeno J, Clave S, Taus A, Pijuan L, Arumi Cancer Drug Targets. 2009; 9:748-760. M, Lorenzo M, Menendez S, Canadas I, Albanell J, Serrano S, Espinet B, Salido M, et al. MET expression and copy 14. Irie HY, Pearline RV, Grueneberg D, Hsia M, Ravichandran number heterogeneity in nonsquamous non-small cell lung P, Kothari N, Natesan S, Brugge JS. Distinct roles of Akt1 cancer (nsNSCLC). Oncotarget. 2015; 6:16215-16226. doi: and Akt2 in regulating cell migration and epithelial-mesen- 10.18632/oncotarget.3976. chymal transition. J Cell Biol. 2005; 171:1023-1034. 25. Argast GM, Krueger JS, Thomson S, Sujka-Kwok I, Carey 15. Graham TR, Zhau HE, Odero-Marah VA, Osunkoya AO, K, Silva S, O'Connor M, Mercado P, Mulford IJ, Young GD, Kimbro KS, Tighiouart M, Liu T, Simons JW, O'Regan Sennello R, Wild R, Pachter JA, et al. Inducible expression of RM. Insulin-like growth factor-I-dependent up-regulation of TGFbeta, snail and Zeb1 recapitulates EMT in vitro and in vivo ZEB1 drives epithelial-to-mesenchymal transition in human in a NSCLC model. Clin Exp Metastasis. 2011; 28:593-614. prostate cancer cells. Cancer Res. 2008; 68:2479-2488. 26. Pirozzi G, Tirino V, Camerlingo R, Franco R, La Rocca A, 16. Thomson S, Buck E, Petti F, Griffin G, Brown E, Liguori E, Martucci N, Paino F, Normanno N, Rocco G. Ramnarine N, Iwata KK, Gibson N, Haley JD. Epithelial Epithelial to mesenchymal transition by TGFbeta-1 induc- to mesenchymal transition is a determinant of sensitivity tion increases stemness characteristics in primary non small of non-small-cell lung carcinoma cell lines and xenografts cell lung cancer cell line. Plos One. 2011; 6:e21548. to epidermal growth factor receptor inhibition. Cancer Res. 2005; 65:9455-9462. 27. Vazquez-Martin A, Cufi S, Oliveras-Ferraros C, Torres- Garcia VZ, Corominas-Faja B, Cuyas E, Bonavia R, Visa 17. Yauch RL, Januario T, Eberhard DA, Cavet G, Zhu W, Fu J, Martin-Castillo B, Barrajon-Catalan E, Micol V, Bosch- L, Pham TQ, Soriano R, Stinson J, Seshagiri S, Modrusan Barrera J, Menendez JA. IGF-1R/epithelial-to-mesenchy- Z, Lin CY, O'Neill V, et al. Epithelial versus mesenchymal mal transition (EMT) crosstalk suppresses the erlotinib- phenotype determines in vitro sensitivity and predicts clini- sensitizing effect of EGFR exon 19 deletion mutations. Sci cal activity of erlotinib in lung cancer patients. Clin Cancer Rep. 2013; 3:2560. Res. 2005; 11:8686-8698. 28. Cortot AB, Repellin CE, Shimamura T, Capelletti M, 18. Witta SE, Gemmill RM, Hirsch FR, Coldren CD, Hedman Zejnullahu K, Ercan D, Christensen JG, Wong KK, Gray K, Ravdel L, Helfrich B, Dziadziuszko R, Chan DC, NS, Janne PA. Resistance to irreversible EGF recep- Sugita M, Chan Z, Baron A, Franklin W, et al. Restoring tor tyrosine kinase inhibitors through a multistep mecha- E-cadherin expression increases sensitivity to epidermal nism involving the IGF1R pathway. Cancer Res. 2013; growth factor receptor inhibitors in lung cancer cell lines. 73:834-843. Cancer Res. 2006; 66:944-950. 29. Guix M, Faber AC, Wang SE, Olivares MG, Song Y, Qu 19. Miyanaga A, Gemma A, Ando M, Kosaihira S, Noro R, S, Rinehart C, Seidel B, Yee D, Arteaga CL, Engelman JA. Minegishi Y, Kataoka K, Nara M, Okano T, Miyazawa H, Acquired resistance to EGFR tyrosine kinase inhibitors in Tanaka T, Yoshimura A, Kobayashi K, et al. E-cadherin cancer cells is mediated by loss of IGF-binding proteins. J expression and epidermal growth factor receptor mutation Clin Invest. 2008; 118:2609-2619. status predict outcome in non-small cell lung cancer patients treated with gefitinib. Oncol Rep. 2008; 19:377-383. 30. Ahmad A, Maitah MY, Ginnebaugh KR, Li Y, Bao B, Gadgeel SM, Sarkar FH. Inhibition of Hedgehog signal- 20. Ren S, Su C, Wang Z, Li J, Fan L, Li B, Li X, Zhao C, Wu ing sensitizes NSCLC cells to standard therapies through C, Hou L, He Y, Gao G, Chen X, et al. Epithelial phenotype modulation of EMT-regulating miRNAs. J Hematol Oncol. as a predictive marker for response to EGFR-TKIs in non- 2013; 6:77. small cell lung cancer patients with wild-type EGFR. Int J Cancer. 2014; 135:2962-2971. 31. Zhang X, Liu G, Kang Y, Dong Z, Qian Q, Ma X. N-cadherin expression is associated with acquisition of 21. Frederick BA, Helfrich BA, Coldren CD, Zheng D, Chan EMT phenotype and with enhanced invasion in erlotinib- D, Bunn PJ, Raben D. Epithelial to mesenchymal transi- resistant lung cancer cell lines. Plos One. 2013; 8:e57692. tion predicts gefitinib resistance in cell lines of head and neck squamous cell carcinoma and non-small cell lung car- 32. Kim HJ, Litzenburger BC, Cui X, Delgado DA, Grabiner cinoma. Mol Cancer Ther. 2007; 6:1683-1691. BC, Lin X, Lewis MT, Gottardis MM, Wong TW, Attar www.impactjournals.com/oncotarget 44344 Oncotarget RM, Carboni JM, Lee AV. Constitutively active type I loss of adhesion and of responsiveness to the motogenic insulin-like growth factor receptor causes transformation trefoil factor 2 in colon carcinoma cells. Proc Natl Acad Sci and xenograft growth of immortalized mammary epithelial U S A. 1999; 96:2316-2321. cells and is accompanied by an epithelial-to-mesenchymal 36. Thiery JP, Acloque H, Huang RY, Nieto MA. Epithelial- transition mediated by NF-kappaB and snail. Mol Cell Biol. mesenchymal transitions in development and disease. Cell. 2007; 27:3165-3175. 2009; 139:871-890. 33. Nurwidya F, Takahashi F, Kobayashi I, Murakami A, 37. Liu Y, Ye X, Zhang JB, Ouyang H, Shen Z, Wu Y, Wang Kato M, Minakata K, Nara T, Hashimoto M, Yagishita W, Wu J, Tao S, Yang X, Qiao K, Zhang J, Liu J, Fu Q and S, Baskoro H, Hidayat M, Shimada N, Takahashi K. Xie Y. PROX1 promotes hepatocellular carcinoma prolif- Treatment with insulin-like growth factor 1 receptor inhibi- eration and sorafenib resistance by enhancing beta-catenin tor reverses hypoxia-induced epithelial-mesenchymal tran- expression and nuclear translocation. Oncogene. 2015.Epub sition in non-small cell lung cancer. Biochem Biophys Res ahead of print Commun. 2014; 455:332-338. 38. He W, He S, Wang Z, Shen H, Fang W, Zhang Y, Qian 34. Wheeler JM, Kim HC, Efstathiou JA, Ilyas M, Mortensen W, Lin M, Yuan J, Wang J, Huang W, Wang L, Ke Z. NJ, Bodmer WF. Hypermethylation of the promoter region Astrocyte elevated gene-1(AEG-1) induces epithelial-mes- of the E-cadherin gene (CDH1) in sporadic and ulcerative enchymal transition in lung cancer through activating Wnt/ colitis associated colorectal cancer. Gut. 2001; 48:367-371. beta-catenin signaling. BMC Cancer. 2015; 15:107. 35. Efstathiou JA, Liu D, Wheeler JM, Kim HC, Beck NE, 39. Morali OG, Delmas V, Moore R, Jeanney C, Thiery JP, Ilyas M, Karayiannakis AJ, Mortensen NJ, Kmiot W, Larue L. IGF-II induces rapid beta-catenin relocation to Playford RJ, Pignatelli M, Bodmer WF. Mutated epithelial the nucleus during epithelium to mesenchyme transition. cadherin is associated with increased tumorigenicity and Oncogene. 2001; 20:4942-4950. www.impactjournals.com/oncotarget 44345 Oncotarget

Journal

Oncotarget – Unpaywall

Published: Nov 5, 2015

There are no references for this article.