Y-box binding protein 1 (YB-1) promotes gefitinib resistance in lung adenocarcinoma cells by activating AKT signaling and epithelial–mesenchymal transition through targeting major vault protein (MVP)

Abstract

Purpose

Gefitinib is a first-line treatment option for epidermal growth factor receptor (EGFR)-mutated lung adenocarcinoma. However, most patients inevitably develop gefitinib resistance. The mechanism underlying this resistance is not fully understood. Y-box binding protein 1 (YB-1) has been reported to play a role in modulating drug sensitivity, but its role in gefitinib resistance is currently unknown. Here, we investigated the role of YB-1 in gefitinib resistance of lung adenocarcinoma.

Methods

We determined the expression of YB-1, epithelial–mesenchymal transition (EMT) and AKT signaling markers, as well as the viability of lung adenocarcinoma cell lines bearing mutant (HCC827, PC-9) or wild-type (H1299) EGFR. We also evaluated PC-9 cell migration and invasion using transwell assays. The clinical importance of YB-1 and major vault protein (MVP) was evaluated using primary lung adenocarcinoma patient samples.

Results

We found that YB-1 was significantly upregulated in gefitinib-resistant lung adenocarcinoma cells compared to gefitinib-sensitive cells. YB-1 augmented gefitinib resistance by activating the AKT pathway and promoting EMT. Decreased migration and invasion was observed upon MVP silencing in YB-1-overexpressing PC-9 cells, as well as restored gefitinib sensitivity. A retrospective analysis of 85 patients with lung adenocarcinoma revealed that YB-1 levels were significantly increased in tyrosine kinase inhibitor (TKI)-resistant patients compared to those in TKI-sensitive patients, indicating that YB-1 may serve as a biomarker to clinically predict acquired gefitinib resistance.

Conclusion

YB-1 activates AKT signaling and promotes EMT at least in part by directly activating MVP. Hence, targeting the YB-1/MVP axis may help to overcome gefitinib resistance in lung adenocarcinoma patients.

Introduction

Lung cancer is the leading cause of cancer-related deaths worldwide. Non-small cell lung cancer (NSCLC) represents 80% of these tumors, with lung adenocarcinoma being the major pathologic type of NSCLC [1]. Gefitinib is a specific and effective epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI) that has been approved for patients harboring EGFR exon 19 deletions or exon 21 substitutions [2]. Despite initial measurable efficacies during early stages of treatment in patients harboring these mutations, most patients eventually become resistant to these drugs, which ultimately leads to treatment failure [3]. Acquired or innate resistance to targeted therapies typically results from gene mutations, such as EGFR T790M, or mutations in EGFR downstream pathways [4–6]. To accurately select personalized therapeutics for patients with lung adenocarcinoma, detailed knowledge of the molecular mechanisms underlying acquired resistance to EGFR-TKIs is of paramount importance.

Drug resistance may arise from a broad range of mechanisms and has been found to be associated with ATP transporters, DNA repair mutations, enzymes, and activated signaling pathways, including the Notch, PI3K/AKT and NF-κB pathways. Additionally, several recent studies have suggested that for multiple types of cancer a close relationship exists between chemoresistance and enhanced epithelial-to-mesenchymal transition (EMT) [7, 8], which is characterized by loss of intercellular adhesion and gain of migratory and invasive properties [9, 10]. In fact, the acquisition of an EMT phenotype has been detected in both NSCLC tumor types and found to be associated with acquired resistance to EGFR-TKI in EGFR-mutant cell lines and primary tumor specimens [11–15]. Alternatively, restoring E-cadherin expression has been found to increase the sensitivity of cancer cells EGFR-TKIs [16].

Y-box binding protein 1 (YB-1, YBX1) is a multifunctional member of the cold-shock domain protein family, which supports tumorigenesis through transcriptional regulation of target genes that control various cellular processes including tumor cell proliferation, progression and multidrug resistance [17, 18]. YB-1 overexpression is associated with an aggressive tumor phenotype, relapse and a poor overall survival in various malignancies [19], including lung cancer [20, 21]. Several studies have reported that YB-1 is overexpressed and involved in treatment resistance in various cancers [17, 18]. Specifically, YB-1 has been found to deregulate drug resistance-related genes, including ABCB1, major vault protein (MVP)/lung resistance protein (LRP), PCNA, MYC, TOP2A, CD44, CD49f, p53, BCL2 and androgen receptor (AR), thereby affecting the sensitivity of tumor cells to a wide range of chemotherapeutic agents [22]. Additionally, YB-1 has been found to regulate the translation of several mRNAs whose protein products are required for EMT, including SNAIL [23]. Together, these studies suggest that YB-1 may contribute to EMT and foster chemoresistance in tumor cells. YB-1 has also been reported to promote basal and 5-fluorouracil-induced expression of the LRP/MVP gene, the promoter of which contains a Y-box, in human colon cancer [24]. MVP is the primary component of vaults, also known as LRP, and was initially reported to be involved in chemotherapy resistance of various tumor types [25] and to be associated with chemotherapy resistance-related pathways, such as the phosphoinositide-3-kinase/AKT and EGFR-induced MAPK pathways [26]. However, the specific role of YB-1 in establishing drug resistance has not yet been elucidated. In addition, YB-1 still has to be fully characterized in terms of clinical relevance.

In the present study, we show that YB-1 and MVP are pivotal regulators that contribute to chemoresistance in lung adenocarcinoma. YB-1 transcriptionally promotes the expression of MVP and activates the AKT pathway as well as the EMT process, and decreases the sensitivity of lung adenocarcinoma cells to gefitinib.

Materials and methods

Cell lines and drugs

Eight cell lines were used: H3255, HCC827 and PC-9 (Cell Bank of Chinese Academy of Sciences, Shanghai, China), which carry EGFR mutations in exon 19 or 21 and are sensitive to TKI inhibitors. The H1975 cell line carrying an EGFR T790M mutation in exon 20 associated with TKI resistance, and the EGFR wild-type cell lines H358, H1299, H520 and A549 were all purchased from the American Type Culture Collection (ATCC; Rockville, MD, USA). They were maintained in RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum serum (FBS, GEMINI), and100 μg/ml penicillin/streptomycin. The cells were incubated at 37 °C with 5% CO2.

Gefitinib was purchased from MCE (MedChemExpress, China). Aliquots of gefitinib were dissolved in dimethyl sulfoxide (DMSO) and stored at −20 °C until use. The respective lung adenocarcinoma cell lines were treated with 0.01, 0.1, 1, 10 and 20 μM geifitinib for 48 h.

RNA interference and lentiviral transduction assays

All small interfering RNAs (siRNAs) used in this study were synthesized by GenePharma (Shanghai, China). The RNA oligonucleotides were transfected using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), and the medium was replaced 6–8 h after transfection. YB-1 gene-specific siRNAs (YB-1siRNA-1, sense: 5’GGUCCUCCACGCAAUUACCAGCAAA-3’,YB-1siRNA-2, sense 5’-GACCCUAUGGGCGUCGACCACAGUA-3′), MVP gene-specific siRNA (MVP-siRNA,sense: 5’-ATCATTCGCACTGCTGTCTT-3′), Snail gene-specific siRNA (Snail-siRNA, sense: 5’-AGACCCACUCAGAUGUCAAGAAGUA-3′) and nonspecific siRNA (siRNA-NC, sense: 5’-GTACCTGACTAGTCGCAGA-3′) were used for gene knockdown experiments.

For the establishment of stable cells overexpressing YB-1, lentiviral vectors with YB-1 and control sequences were designed by FulenGene (FulenGene Co., Guangzhou, China) and transductions were conducted according to the manufacturer’s instructions followed by selection with puromycin (2 μg/ml) for 2–3 days.

RNA isolation and quantitative real-time PCR (qRT-PCR)

Total mRNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA), and reverse transcription was performed using a GoScriptTM Reverse Transcription System (Promega, USA). Real-time PCR was performed using a reaction system containing cDNA, SYBR Green Mix (GoTaq® qPCR Master Mix, Promega, CA, USA) and primers (Table 1) following the manufacturer’s instructions. ß-actin was used as an internal control. The amplification cycling parameters (40 cycles) were set as follows: denaturation for 15 s at 95 °C, annealing for 15 s at 60 °C and extension for 45 s at 72 °C. The relative expression of the target genes was expressed as 2 -△△Ct.

Table 1.

Primers of target genes used in this study

gene	Forward	Reverse
YB-1	CGAAACCCAGAAAGCTATGC	TGTGGAGGACCTTCAACTCC
MVP	TTTGATGTCACAGGGCAAGTTCGGC	CACCAAATCCAGAACCTCCTCAAAC
SNAIL	TCGGAAGCCTAACTACAGCGA	AGATGAGCATTGGCAGCGAG
Vimentin	AGTCCACTGAGTACCGGAGAC	CATTTCACGCATCTGGCGTTC
ZEB-1	GATGATGAATGCGAGTCAGATGC	ACAGCAGTGTCTTGTTGTTGT
E-cadherin	ATTTTTCCCTCGACACCCGAT	TCCCAGGCGTAGACCAAGA
MDR1	GAGCCCATCCTGTTTGACTGC	TGCCATGCTCCTTGACTCTGC
BCRP	GGCCTATAATAACCCTGCAGACTTC	GCCCAAAGTAAATGGCACCT
MPR2	CCTGGTTCCTGTCCCTATTCT	GGCATCTTGGCTTTGACTCTG
β-actin	GGACTTCGAGCAAGAGATGG	AGCACTGTGTTGGCGTACAG

Western blot analysis and antibodies

Cells were harvested and homogenized with RIPA lysis buffer. A BCA protein assay kit (Pierce, Thermo Fisher Scientific, USA) was used to determine the concentration of the proteins. Total protein (50 μg) was separated on 10%–12% SDS-PAGE gels and transferred to PVDF membranes (Millipore, Bedford, MA, USA). The membranes were subsequently blocked with 5% nonfat dry milk in TTBS for two hours at 37 °C, followed by overnight incubation with primary antibodies at 4 °C. After washing, secondary antibodies were incubated for 1 h at room temperature. The primary antibodies and secondary antibodies used are listed in Table 2. Protein signals were captured using a chemiluminesence (ECL) detection system.

Table 2.

Primary and secondary antibodies used in this study

Antibodies	Purchase information	Host	Application and dilution
YB-1	Abcam	Rabbit	1:1000 for WB and 1:400 for IHC
AKT	Cell Signaling Technology	Rabbit	1:1000 for WB
p-AKT (Ser473)	Cell Signaling Technology	Rabbit	1:1000 for WB
mToR	Abcam	Rabbit	1:1000 for WB
p-mTOR (Ser2448)	Cell Signaling Technology	Rabbit	1:1000 for WB
MVP	HuaAn Biotechnology Co.	Rabbit	1:1000 for WB and 1:200 for IHC
SNAIL	GeneTex	Rabbit	1:1000 for WB
Twist	Affinity Biosciences	Rabbit	1:1000 for WB
E-cad	Cell Signaling Technology	Rabbit	1:1000 for WB
Vimentin	Cell Signaling Technology	Rabbit	1:1000 for WB
ZEB-1	Cell Signaling Technology	Rabbit	1:1000 for WB
ß-actin	Abclonal	Rabbit	1:1000 for WB
Goat anti-Rabbit IgG	KPL	Goat	1:5000 for WB

WB: Western blotting, IHC:Immunohistochemistry

Clinical specimens and immunohistochemistry

Paraffin embedded samples of patients with lung adenocarcinoma (n = 85), who received EGFR mutation detection tests, were collected from the Pathology department of the Second Hospital of Hebei Medical University, Shijiazhuang (China), with written informed consent of the patients. All samples were histologically classified and graded according to TNM by a blinded clinical pathologist. All experimental protocols were approved by the Institutional Review Committee of the the Second Hospital of HEBMU, Shijizhuang (China).

Histologic sections (4 μm thick) were prepared from formalin-fixed paraffin-embedded (FFPE) tissue blocks. After deparaffinization, antigen retrieval was performed in citrate buffer for 5 min. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide for 30 min. Subsequent immunohistochemistry was performed using an IHC assay kit (ZSGB-bio, China) following the manufacturer’s instructions. The samples were stained with primary antibodies directed against YB-1 and MVP diluted 1:400 and 1:200 in PBS, respectively, and incubated overnight at 4 °C. Next, the samples were incubated with a diluted secondary antibody for 30 min at 37 °C. Visualization was achieved using peroxidase-labeled streptavidin-biotin and diaminobenzidine (DAB) staining. The immunostaining results were evaluated by two pulmonary pathologists using a blinded protocol design. For each specimen, the total score of expression intensity (negative staining: 0 point; weak staining: 1point; moderate staining: 2 points; and strong staining: 3 points) multiplied by cell numbers (positive cells as ≤ 25% of the cells: 1 point; 26–50% of the cells: 2 points; 51–75% of the cells: 3 points; > 75% of the cells: 4 points) of YB-1 or MVP was determined. When the sample was scored ≥ 6 points, it was defined it as high expression, otherwise as low expression.

Cell viability assay

The cytotoxic effects of drugs on lung adenocarcinoma cells were assessed using a CCK8 assay. To this end, cells were seeded in 96-well plates at a density of 2000 cells per well overnight in medium with 10% FBS. Next, the indicated concentrations of relevant reagents were added to the experimental and control groups, respectively. After incubation for 48 h, the number of viable cells was determined using a CCK-8 assay kit (MedChemExpress, China) according to the manufacturer’s protocol, with incubation at 37 °C for 2 h. The resulting absorbance of each well was measured at 450 nm using a Microplate Reader. Each assay comprised triplicates and each assay was independently repeated at least three times. The data are presented as the percentage of control cells calculated from the absorbance corrected for background.

Scratch wound healing migration assay

Cells were seeded in six-well plates, cultured overnight and subsequently scraped with a 200 ul pipette tip to create sterile wound gaps. The wounds were evaluated and photographed by inverted microscopy at 0 h, 24 h and 48 h as indicated in the figure legends.Wound sizes (perpendicular to the wound) were measured at five randomly selected sites. Three independent experiments were performed. The cell migration capacity in each group was calculated using image J software.

Transwell migration and invasion assays

Transwell migration assays were performed using polycarbonate membrane transwell inserts (Corning, NY,USA), with a membrane diameter of 8.0 μm. A BD Bio-Coat™ Matrigel™ Invasion Chamber (BD Biosciences, San Jose, CA, USA) with an 8 μm pore size Polyethylene Terephthalate membrane, was used for the cell invasion assays. 48 h post-transfection with siRNA, 15,000 H1299 cells/well or 20,000 HCC827 cells/well were seeded in triplicates in the upper transwell chambers to assess their migration and invasion capacities, respectively. For PC-9 cells, 10,000 cells/well and 20,000 cells/well were seeded and incubated for 48 h for the migration and invasion assays, respectively. The migratory or invasive cells were fixed with absolute methanol for 30 min, washed twice with PBS and stained with 0.5% crystal violet in water for 8 min. Removal of excess crystal violet stain was carried out by dipping the upper chamber inserts in distilled water and wiping the upper membranes of the inserts with cotton swabs. Migrated and invaded cells were counted (10 random 200× fields per well) and calculated using image J software. Cell counts were expressed as the mean number of cells per field of view. Three independent experiments were performed. Data are presented as mean ± standard deviation (SD).

Dataset collection

Microarray-based gene expression data were downloaded from the Gene Expression Ominibus (GEO) public database under accession number GSE30219, and were used for detecting differential YB-1 expression in cancer and normal tissues.

Gene set enrichment analysis

We used gene set enrichment analysis (GSEA v2.0, available online: http://www.broad.mit.edu/gsea/) to assess associations between YB-1 expression and HALLMARK_PI3K_AKT_mTOR_SIGNALING. Pre-defined gene sets were obtained from the Molecular Signatures Database, MsigDB (http://software.broadinstitute.org/gsea/msigdb). Samples from the TCGA datasets were divided into high- or low-YB-1 expression groups using the median as cutoff. Default settings were used and thresholds for significance were determined by permutation analysis (1000 permutations). False discovery rates (FDRs) were calculated. A gene set was considered significantly enriched when the FDR score was < 0.25.

Statistical analysis

Statistical analyses were performed using SPSS 21.0 and GraphPad Prism software (La Jolla, CA, USA). Statistical significance of differences between experimental groups was determined using Student's t tests. The x² test was used to assess associations between YB-1 and MVP expression and clinicopathologic parameters. Kaplan-Meier curves were used to assess survival. Data are presented as mean ± standard deviation (S.D.), and each experiment was performed at least three times. P < 0.05 was considered statistically significant.

Results

YB-1 expression is related to gefitinib resistance in lung adenocarcinoma cells

A panel of lung adenocarcinoma cell lines was selected to encompass three groups as follows: (1) no EGFR mutation (H358, H1299, H520 and A549), (2) EGFR mutation in exon 19 and 21 and sensitive to TK inhibitors (HCC27, PC9 and H3255) and (3) EGFR T790M mutation and an additional TKI resistance-inducing mutation (H1975). YB-1 expression levels were determined in all seven cell lines using qRT-PCR (Fig. 1a) and Western blotting (Fig. 1b-c), respectively. We found that H1299 cells expressed the highest YB-1 level, HCC827 cells expressed the next highest YB-1 level, and PC-9 cells expressed the lowest YB-1 level. To explore a possible correlation between gefitinib sensitivity and YB-1 expression, we assessed the effect of gefitinib on the viability of these cancer cell lines using a CCK-8 assay (Fig. 1d). Significant differences in gefitinib sensitivity were observed among the H1299, HCC827 and PC-9 cell lines, and these were consistent with trends in YB-1 levels. Therefore, these cell lines were chosen for subsequent analyses. The data suggest that YB-1 may be involved in the development of gefitinib resistance in lung adenocarcinoma.

Fig. 1.

YB-1 expression level affects sensitivity to geifitinib treatment. (a-b) Expression of YB-1 in cell lines determined by qRT-PCR and Western blotting. (c) Intensities of the immunoreactive bands quantified by densitometric scanning. * p < 0.05, ** p < 0.01, compared with PC-9 cells. (d) H1299, HCC827 and PC-9 cells were cultured in 96-well plates and treated with different doses of gefitinib for 48 h before measuring cell viabilities using CCK8 assays. (e) H1299 and HCC827 cells were transfected with either control-siRNA (siRNA-NC) or YB-1-siRNA (YB-1siRNA-1, YB-1siRNA-2). PC-9 cells were transduced with either control lentiviruses (vector) or YB-1 lentiviruses (YB-1). Transfection/transduction efficiencies were determined by qRT-PCR. (f) Expression of YB-1 determined by Western blotting (left panels). Band densities were normalized to β-actin (right panels). (g-h) H1299 and HCC827 cells were transfected with siRNA-NC or YB-1siRNA followed by treatment with various doses of gefitinib (0, 0.01, 0.1, 1, 10 and 20 μM) for 48 h, after which cell viabilities were determined using CCK-8 assays. (i) Vector or YB-1 overexpressing PC-9 cells were treated with various doses of gefitinib for 48 h (0, 0.001, 0.01, 0.1, 1 and 10 μM), after which cell viabilities were determined using CCK-8 assays. The values represent the mean ± SD from three independent experiments, * p < 0.05, ** p < 0.01, ^## p < 0.01 and ^††p < 0.01 versus corresponding control cells

To confirm that YB-1 contributes to gefitinib resistance in lung adenocarcinoma cells, we used both loss-of-function and gain-of-function experimental approaches. First, we knocked down YB-1 expression in H1299 and HCC827 cells using siRNAs (siRNA-YB1–1, siRNA-YB1–2). The respective transfection efficiencies were assessed by qRT-PCR and Western blot analyses. YB-1 siRNAs significantly reduced YB-1 mRNA and protein levels (Fig. 1e-1f). In addition, we found that YB-1 knockdown significantly increased the sensitivity of H1299 and HCC827 cells to gefitinib compared to the same cells transfected with nonspecific siRNAs (Fig. 1g–1h). Subsequently, we employed a gain-of-function approach. YB-1 was overexpressed in PC-9 cells by transduction with a lentiviral vector, followed by treatment with various doses of gefitinib. We found that exogenous YB-1 overexpression significantly increased the percentage of surviving cells, indicating that YB-1 promotes gefitinib resistance in PC-9 cells (Fig. 1i). These results indicate that YB-1 may regulate gefitinib resistance and suggest that YB-1 may serve as a molecular biomarker for gefitinib sensitivity in lung adenocarcinoma cells.

YB-1 regulates gefitinib resistance via EMT in lung adenocarcinoma cells

Previous studies have shown that EMT can induce cancer cell resistance to EGFR-TKI [14, 15]. Therefore, we first set out to investigate the involvement of YB-1 in EMT. To this end, we treated lung adenocarcinoma cells with gefitinib and next determined the effect of YB-1 on cell migration and EMT. Scratch wound healing and transwell assays revealed that YB-1 depletion significantly inhibited the migration and invasion of H1299 and HCC827 cells. Consistent with these data, exogenous YB-1 overexpression increased the ability of PC-9 cells to migrate and invade (Fig. 2a–2b). Next, to confirm the role of YB-1 in EMT, we assessed the expression of EMT-related markers. We found that depletion of YB-1 in H1299 and HCC827 cells led to significant decreases in the expression of ZEB-1, vimentin and SNAIL, whereas the expression of the epithelial marker E-cadherin was significantly upregulated in HCC827 cells. In addition, we found that exogenous YB-1 overexpression in PC-9 cells led to opposite molecular changes (Fig. 2c–2g). Collectively, these data suggest that increased YB-1 expression may contribute to lung adenocarcinoma cell metastasis by increasing cell migration and invasion, as well as by promoting EMT.

Fig. 2.

YB-1 contributes to EMT in lung adenocarcinoma cells. H1299 and HCC827 cells were transfected with YB-1 siRNA and PC-9 cells with YB-1 overexpression vector. (a) Scratch wound healing assays of the indicated cells; representative images (upper panels) and quantification (lower panels) of wound closure; images were taken at 0, 24 h and 48 h. (b) Representative images (left panels) and quantification (right panels) of invaded (without Matrigel) and migrated (with Matrigel) cells in Transwell assays. Representative fields of the migration and invasion assays at 100x magnification are shown. (c-f) EMT pathway markers detected by qRT-PCR and Western blotting. (g) Intensities of the immunoreactive bands quantified using densitometric scanning. (h) CCK-8 viability assay of H1299 and HCC827 cells following gefitinib treatment after transfection with si-Snail or si-Snail+si-YB-1. (i) qRT-PCR and (j) Western blot detection of E-cadherin, Vimentin and SNAIL expression after transfection of si-Snail or si-Snail+si-YB-1. (k) Intensities of the immunoreactive bands quantified using densitometric scanning. Values represent the mean ± SD from three independent experiments, * p < 0.05, ^# p < 0.05 ** p < 0.01 and ^## p < 0.01 versus corresponding control cells

To next determine the effect of EMT on YB-1-mediated gefitinib resistance, we transfected H1299 and HCC827 cells with negative control siRNA, SNAIL siRNA, or SNAIL siRNA+YB-1 siRNA. Subsequent CCK-8 assays revealed that SNAIL knockdown enhanced gefitinib sensitivity in H1299 and HCC827 cells compared to that in controls (Fig. 2h). YB-1 knockdown did not reverse the effect of SNAIL siRNA. Moreover, qRT-PCR and Western blotting results revealed that there were no significant differences in EMT-related protein levels between the SNAIL siRNA and SNAIL siRNA+YB-1 groups (Fig. 2i–2k). These results indicate that YB-1 may regulate gefitinib resistance via EMT.

YB-1 contributes to EMT via MVP

To further determine how YB-1 affects gefitinib resistance, we evaluated the expression of various genes related to drug resistance, including MDR1, MRP2, BCRP1 and MVP. Western blot and qRT-PCR analyses revealed that YB-1 knockdown led to marked MVP expression suppression at both the mRNA and protein levels in H1299 and HCC827 cells. Conversely, exogenous YB-1 overexpression in PC-9 cells led to elevated MVP mRNA and protein levels (Fig. 3a–3b). We also assessed the expression of MVP in lung adenocarcinoma cell lines that expressed varying levels of YB-1. We found that lung adenocarcinoma cell lines with elevated YB-1 levels (H1299 and HCC827) expressed higher levels of MVP, and that PC-9, a cell line with a low YB-1 level, also expressed reduced levels of MVP (Fig. 3c-3d).

Fig. 3.

YB-1 contributes to EMT via MVP. (a) Relative mRNA levels of drug resistance-related genes quantified in H1299 and HCC827 cells after YB-1 knockdown and YB-1 overexpression in PC-9 cells. MDR1, MVP, BCRP1 and MPR2 expression levels were determined by qRT-PCR. (b) MVP expression levels determined by Western blotting (left panel). Band densities were normalized to β-actin (right panel). (c-d) Expression MVP at mRNA and protein levels in H1299, HCC827 and PC-9 cells determined by qRT-PCR and Western blot analyses. Data represent the mean ± SD of three independent experiments, * p < 0.01, ** p < 0.01, ^## p < 0.01, and ^††p < 0.01 versus corresponding control cells

In a recent study, Stein et al. showed that YB-1 can bind a Y-box motif within the MVP promoter based on electrophoretic mobility shift assays (EMSAs) using a single-stranded Y-box oligonucleotide derived from the MVP promoter in human colon carcinoma HCT116 cells [24]. Another study reported that depletion of MVP may reverse EMT in triple-negative breast cancer cells [27]. Based on these results, we hypothesized that YB-1 may contribute to EMT via MVP in lung adenocarcinoma cells. To test this hypothesis, we employed a “rescue” experiment by transfecting a siRNA targeting MVP (siMVP) in PC-9 cells exhibiting YB-1 overexpression. We found that YB-1 overexpression increased MVP expression, and that this increase was abrogated by siMVP. Furthermore, in cells with YB-1-induced MVP upregulation, the expression of EMT-related markers was also found to be enhanced, and these effects were reversed when the cells were co-transfected with siMVP (Fig. 4a–4c).

Fig. 4.

MVP silencing in YB-1 overexpressing PC-9 cells decreases migration and invasion. YB-1 overexpressing PC-9 cells were transfected with MVP siRNA (siMVP+YB-1) and corresponding control siRNA (siNC+YB-1). EMT pathway markers were detected by qRT-PCR (a) and Western blotting (b). (c) Intensities of the immunoreactive bands were quantified using densitometric scanning. (d-e) Migration ability assessed by scratch wound healing assay and invasion ability assessed by transwell assay (magnification: 100x). (f) Changes in gefitinib sensitivity. Data represent the mean ± SD of three independent experiments, * p < 0.05, ** p < 0.01, ^# p < 0.05, and ^## p < 0.01, compared with corresponding control cells

We next examined the effect of siMVP on cell migration and invasion in YB-1-overexpressing PC-9 cells. Using a scratch wound healing assay, we found that MVP depletion significantly inhibited the migration of YB-1-overexpressing PC-9 cells. Based on a transwell assay, we found that by co-transfecting PC-9 cells with siMVP and a YB-1 overexpression vector, their migration and invasion abilities could be partially reversed (Fig. 4d–4e). Subsequently, we treated PC-9 cells with increasing concentrations of gefitinib, followed by a CCK-8 cell viability assay to confirm drug-resistant properties associated with YB-1 expression. We found that YB-1 overexpression increased the tolerance of PC-9 cells to gefitinib and, as expected, that MVP depletion partially reversed the enhanced gefitinib resistance owing to YB-1 overexpression (Fig. 4f). Collectively, these results support the hypothesis that YB-1 contributes to EMT via MVP in lung adenocarcinoma cells.

YB-1 modulates gefitinib resistance through the AKT/mTOR pathway mediated by MVP in lung adenocarcinoma cells

The AKT/mTOR signaling pathway plays an important role in tumor cell drug resistance. To evaluate a potential association between YB-1 and AKT signaling in lung adenocarcinoma cells, we performed gene set enrichment analysis. We indeed found that YB-1 was related to the AKT/mTOR pathway, and that key genes involved in its signaling activation were also upregulated in these cells (Fig. 5a). To validate this observation, we found by Western blotting that knocking down YB-1 in H1299 and HCC827 cells led to suppressed AKT signaling (Fig. 5b-5c). Exogenous YB-1 overexpression significantly promoted the phosphorylation of AKT (Ser473) in PC-9 cells (Fig. 5d-5e). These findings suggest that in lung adenocarcinoma cells the expression of YB-1 is positively associated with AKT activation.

Fig. 5.

YB-1 activates the PI3K/AKT/mTOR signaling pathway. (a) GSEA analysis showing that YB-1 expression is positively associated with the PI3K_AKT_ACTIVATION gene set in lung cancer. (b-e) Western blot analysis of changes in p-AKT expression after YB-1 silencing in H1299 and HCC827 cells and YB-1 overexpression in PC-9 cells. Intensities of the immunoreactive bands were quantified using densitometric scanning. (f-g) Cell viability (%) of PC-9 cells with YB-1 overexpression exposed to different concentrations of LY294002 and 30 μM LY294002 for different time periods. (h) PC-9 cells were incubated with 30 μM LY294002 after which protein expression was detected using Western blot analysis. (i) Intensities of the immunoreactive bands were quantified using densitometric scanning. (j) Viability of PC-9 cells exposed to various doses of gefitinib and/or 30 μM LY294002 for 48 h determined by CCK-8 assays. (k) p-AKT expression in YB-1 overexpressing PC-9 cells after MVP silencing. (l) Intensities of the immunoreactive bands were quantified using densitometric scanning. Data represent the mean ± SD of three independent experiments, * p < 0.05, ** p < 0.01, ^# p < 0.05, ^## p < 0.01, ^††p < 0.01, compared with corresponding control cells

A previous study reported that YB-1 mediates mitoxantrone resistance via an AKT-dependent pathway in diffuse large B cell lymphoma [28]. Therefore, in the current study, we further examined the role of the PI3K/AKT pathway in YB-1-mediated drug resistance to gefitinib. For this, the PI3K/AKT/mTOR inhibitor LY294002 was tested in PC-9 cells. We found that LY294002 reduced the viability of PC-9 cells in a dose- and time-dependent manner (Fig. 5f-5g). Next, we examined whether blocking AKT could reverse the YB-1 overexpression-induced gefitinib resistance in PC-9 cells. To this end, the cells were incubated with 30 μmol LY294002 for 48 h, and next subjected to Western blot analysis. We found that LY294002 inhibited the activation of p-AKT (Fig. 5h-5i). Next, cell survival was assessed using CCK-8 assays. LY294002 significantly increased the gefitinib-induced cell growth inhibition. PC-9 cells overexpressing YB-1 showed a significantly higher viability compared to control cells at the same concentrations of gefitinib and LY294002 (Fig. 5j). These results indicate that AKT inhibition reverses resistance to gefitinib via p-AKT downregulation.

To determine whether YB-1 regulates the PI3K/AKT/mTOR signaling pathway via MVP, we performed rescue experiments in which we co-transfected PC-9 cells, which express high levels of YB-1, with pcDNA-YB-1 (YB-1) and siMVP (YB-1 + siMVP) or with the corresponding negative control vectors (YB-1 + siNC). Subsequent Western blot analysis revealed that the p-AKT (Ser473) levels were significantly decreased in YB-1-overexpressing PC-9 cells transfected with siMVP (Fig. 5k-5l). These results suggest that YB-1 may modulate gefitinib resistance through the AKT/mTOR pathway mediated by MVP in lung adenocarcinoma cells.

Clinicopathologic features, including patient survival and gefitinib resistance, are associated with YB-1 and/or MVP expression in primary lung adenocarcinomas

To assess the expression of YB-1 in primary human lung adenocarcinomas, we first searched the Oncomine database and analyzed YB-1 expression data retrieved from the GSE20347 dataset, which encompasses 293 lung adenocarcinoma tissues and 14 normal tissues. We found that YB-1 was significantly upregulated in lung adenocarcinomas compared to normal lung tissues (Fig. 6a). Subsequent Kaplan-Meier survival analyses of the GSE30219 dataset revealed that for the lung adenocarcinoma patients, a high YB-1 expression was associated with a shorter overall survival compared to that in patients with low YB-1-expressing tumors (Fig. 6b). To further confirm this observation and to explore the clinical relevance of YB-1 expression, we analyzed 85 archived lung adenocarcinoma tissues and 10 normal lung tissues by immunohistochemistry. We again found that YB-1 expression was elevated in the lung adenocarcinoma tumor tissues compared to that in normal lung tissues (Fig. 6c). Moreover, high YB-1 expression was significantly associated with a worse survival outcome based on a log-rank test (Fig. 6d-6e).

Fig. 6.

YB-1 expression significantly correlates with prognostic and clinicopathological parameters. (a) YB-1 expression in lung tumor and normal lung tissues derived from the GSE30219 dataset. (b) Kaplan-Meier survival analysis of the prognostic value of YB-1 using the GSE30219 dataset. High levels of YB-1 mRNA predict a poor prognosis. Patients were subdivided into high- or low-YB-1 expression groups using the median as cutoff. YB-1 and MVP protein expression levels and prognostic values in 85 lung adenocarcinoma patients are shown. (c) Expression of YB-1 is higher in lung adenocarcinomas than in normal lung tissues. (d) A high YB-1 expression level is significantly associated with a shorter overall survival (OS) time in 62 lung adenocarcinoma patients. (e) YB-1 and MVP protein expression in lung adenocarcinomas. (f) Association of YB-1 expression with advanced clinical stage, N stage, lymph node metastasis and TKI-resistance in 85 lung adenocarcinoma patients. (g) Association of MVP expression with advanced clinical stage, N stage, lymph node metastasis and distant metastasis in 85 lung adenocarcinoma patients

Our results thus far indicate that YB-1 may regulate MVP in lung adenocarcinoma cells. To examine their clinical relevance, we next analyzed relationships between YB-1 and MVP expression and clinicopathological features of the lung adenocarcinoma patients. The clinicopathological features of the patients are listed in Table 3. We found that YB-1 expression was significantly associated with clinical stage (p = 0.009), N stage (p = 0.005) and lymph node metastasis (p = 0.006). In total 51 patients with EGFR mutations were recruited to our study, and YB-1 expression was found to be significantly increased in patients with tumors with an acquired TKI resistance (EGFR T790M mutation) compared to those with TKI-sensitive tumors (EGFR exon 19 or 21 deletion; p = 0.043) (Fig. 6f). MVP expression was also found to be significantly associated with clinical stage (p = 0.005), N stage (p = 0.002), lymph node metastasis p = 0.008) and distant metastasis (p = 0.001) (Fig. 6g). These results indicate that the YB-1 and MVP expression levels are associated with a worse prognosis. Additionally, the YB-1 and MVP expression levels were found to be significantly, and positively, correlated in all 85 lung adenocarcinoma tissue specimens tested (p = 0.001, r = 0.359) (Table 4). Overall, our results indicate that YB-1 may play a role in chemoresistance and serve as an effective prognostic factor for lung adenocarcinoma patients.

Table 3.

Relationships between YB-1 or MVP expression levels and clinicopathologic characteristics in lung adenocarcinoma

	YB-1			MVP
	High (%)	Low	P-value	High (%)	Low	P-value
overall	39 (45.9)	46		38 (44.7)	47
Age
≥ 60	21 (43.8)	27	0.653	25 (52.1)	23	0.119
< 60	18 (48.6)	19		13 (35.1)	24
Gender
Male	16 (40.0)	24	0.305	16 (40.0)	24	0.411
Female	23 (51.1)	22		22 (48.9)	23
Clinical Stage
I	0 (0.0)	2	0.009*	0 (0.0)	2	0.005*
II	1 (10.0)	9		1 (10.0)	9
III	5 (31.3)	11		4 (25.0)	12
IV	33 (57.9)	24		33 (57.9)	24
T classification
T1	9 (60.0)	6	0.097	9 (60.0)	6	0.141
T2	21 (42.9)	28		22 (44.9)	27
T3	9 (56.3)	7		9 (43.8)	7
T4	0 (0.0)	5		0 (0.0)	5
N classification
N0	2 (20.0)	8	0.005*	1 (10.0)	9	0.002*
N1	4 (33.3)	8		2 (16.7)	10
N2	9 (32.1)	19		12 (42.9)	16
N3	24 (68.6)	11		23 (65.7)	12
Lymph Note Metastasis
No	0 (0.0)	8	0.006*	0 (0.0)	8	0.008*
Yes	39 (50.6)	38		38 (49.4)	39
Distance Metastasis
No	11 (34.4)	21	0.098	7 (21.9)	25	0.001*
Yes	28 (52.8)	25		31 (59.6)	22
EGFR mutation
Wild-type	16 (47.1)	18	0.208	16 (47.1)	18	0.153
TKI-sensitive	16 (39.0)	25		15 (36.6)	26
TKI-resistant	7 (70.0)	3		7 (70.0)	3

*p < 0.05 indicates statistically significant difference

Table 4.

Positive correlation between YB-1 and MVP protein expression in lung adenocarcinoma specimens

	MVP		p-value	R-value
	Low expression	High expression
YB-1			0.001*	0.359
low expression	33	13
high expression	14	25

*p < 0.05 indicates statistically significant difference

Discussion

TKI resistance can significantly compromise the efficacy of chemotherapy, leading to recurrent or metastatic disease and overall poor clinical outcomes, as well as reduced lung adenocarcinoma patient survival [29, 30]. Therefore, an improved and more in-depth understanding of the mechanisms underlying acquired EGFR-TKI resistance is critical for the development of effective therapeutic strategies. Here, we show that increased YB-1 expression induces gefitinib resistance and EMT by activating the PI3K/AKT/mTOR axis, at least in part via MVP regulation. Importantly, we also found that YB-1 overexpression and MVP knockdown reversed the mesenchymal phenotype and sensitized lung adenocarcinoma cells to gefitinib. Furthermore, we observed strong co-expression of the MVP and YB-1 proteins in primary human lung adenocarcinoma specimens. Patients presenting TKI resistance showed a higher YB-1 expression compared to those designated as TKI-sensitive. These results provide new insights into the molecular mechanisms underlying gefitinib resistance elicited by increased YB-1 expression in lung adenocarcinoma cells and suggest that YB-1 may serve as a novel therapeutic target to reverse gefitinib resistance.

YB-1 is an oncogenic transcriptional and translational regulator that has been associated with various diseases, including breast cancer [31], ovarian cancer [32], lung cancer [20, 21], sarcoma [33], prostate cancer [34], glioblastoma [35] and non-Hodgkin’s lymphoma [36]. Interestingly, YB-1 has also been defined as a promising predictive marker for radio-resistance and chemo-radio-resistance in certain cancers [37]. YB-1 has also been reported to be overexpressed in cisplatin-resistant cells [38], and that antisense YB-1 RNA can augment cisplatin sensitivity [39]. Stratford et al. found that YB-1 can upregulate EGFR gene transcription and that YB-1 inhibition can enhance the sensitivity to gefitinib in basal-like breast cancer [40]. However, prior to our study no relationship had been defined between YB-1 expression and gefitinib resistance/EGFR status in lung adenocarcinoma. Here, we report that in human lung adenocarcinoma patients YB-1 is more highly expressed in TKI-resistant patients than in TKI-sensitive patients, thereby supporting the notion that high YB-1 expression causes gefitinib resistance. In addition, we found that increased YB-1 expression in lung adenocarcinoma cells is at least one of the mechanisms underlying intrinsic gefitinib resistance. We also found that up- and downregulation of YB-1 increased and decreased, respectively, the sensitivity of lung adenocarcinoma cells to gefitinib. This observation implies that YB-1 may be a vital regulator of the response to gefitinib in lung adenocarcinoma cells.

Why exactly YB-1 sensitizes lung adenocarcinoma cells to gefitinib remains an interesting question. YB-1 has been shown to regulate the MVP gene, which closely reflects the chemoresistance profile of many tumor cell lines [25, 41, 42]. The MVP promoter is known to contain an inverted CCAAT box, termed Y-box, which is a recognition site for the transcription factor YB-1 [43]. It has been reported that specific interactions between YB-1 and the Y-box binding motif of the MVP promoter enhance YB-1 binding in stably transfected colon cancer cells [24], indicating that YB-1 may be directly linked to MVP expression. In addition, it has been found that siRNA-mediated downregulation of YB-1 reduces the expression of EGFR and MVP, and that being positive for YB-1 and LRP expression confers a significantly worse prognosis to NSCLC patients [44]. Growing evidence suggests that YB-1 may serve as an effective prognostic predictors for various malignancies, including lung cancer [26, 33–35]. In the present study, in addition to YB-1-mediated regulation of MVP in lung adenocarcinoma cells, we found that inhibition of YB-1 through MVP silencing partially reversed gefitinib sensitivity. Therefore, YB-1 may directly be linked to MVP-mediated drug resistance. Additional analysis of pathologically confirmed tissue samples revealed that YB-1 is highly expressed in human lung adenocarcinomas with a strong association between YB-1 and MVP expression. The overall survival of patients with lung adenocarcinoma expressing high levels of YB-1 was significantly reduced compared to that of patients with tumors expressing lower levels. These results consistently suggest that YB-1 serves as a strong predictor of poor chemotherapy responses, promotes drug resistance at least in part through MVP and is indicative of a poor prognosis in patients with lung adenocarcinoma.

Also, EMT has been found to be closely related with treatment resistance and metastasis in many tumor types, including breast cancer, lung cancer, colon cancer and pancreatic cancer. Hence, targeting EMT has been considered a promising strategy for overcoming drug resistance [45]. Several recent studies reported that YB-1 not only facilitates lung adenocarcinoma metastasis by activating EMT [23], but also promotes drug resistance [46]. In conformity with this notion, we first found that YB-1 silencing decreased EMT, migration and invasion in H1299 and HCC827 cells, contributing to increased gefitinib sensitivity. Alternatively, when YB-1 was overexpressed in PC-9 cells, we observed a reverse phenotype, including a significant decrease in sensitivity to gefitinib and an increasing trend in cell migration and invasion, properties associated with EMT reversal. Additionally, no significant changes were observed in gefitinib sensitivity and EMT-related marker expression following simultaneous SNAIL and YB-1 suppression, suggesting that YB-1 silencing increases drug resistance by inducing EMT. This implies that EMT may play an important role in YB-1-mediated gefitinib resistance.

We also found that YB-1 downregulation inhibits gefitinib resistance, at least in part, by blocking AKT signaling through MVP modulation. The PI3K/AKT/mTOR axis represents a crucial downstream MVP pathway that regulates many biological processes, including tumor cell proliferation and survival, motility, migration and invasion [44, 45]. Moreover, abnormal activation of the PI3K/AKT signaling pathway is a key factor underlying tolerance to gefitinib in NSCLC cells [47, 48]. She et al. have previously shown that inhibition of the PI3-kinase pathway with LY294002 sensitizes cells to gefitinib [49] and that suppression of YB-1 expression increases the effect of gefitinib [40]. These reports are consistent with our finding that YB-1-dependent upregulation of AKT activity is responsible for resistance to gefitinib, and that specific inhibition of AKT reverses resistance to gefitinib. AKT is a key component of processes that have been shown to regulate EMT through suppression of E-cadherin expression via transcription factors such as SNAIL, ZEB and TWIST [50]. To determine whether crosstalk between the AKT pathway and the MVP and EMT pathways plays a role in mediating gefitinib resistance in lung adenocarcinoma, EMT was evaluated after MVP silencing in YB-1-overexpressing PC-9 cells. A decreased migration and invasion was observed following MVP knockdown in YB-1-overexpressing PC-9 cells, which was accompanied by a restored sensitivity to gefitinib and a reduced EMT. These results indicate that YB-1/MVP/AKT signaling acts as a pivotal regulator of EMT and gefitinib resistance of lung adenocarcinoma cells, and that this pathway may serve as a potential therapeutic target.

Certain limitations of this study need to be mentioned. Firstly, the precise mechanisms underlying gefitinib resistance in vivo remain to be investigated. In addition, the clinical feasibility of YB-1 as a therapeutic target to prevent drug resistance requires further analysis.

In summary, our data indicate that YB-1 can significantly reduce the efficacy of gefitinib in lung adenocarcinoma cells, in part by promoting MVP expression, which in turn can induce chemoresistance by activating AKT signaling and promoting EMT. In addition, we found that high levels of YB-1 and MVP in lung adenocarcinomas can predict poor patient outcomes and limited overall survival rates and may, therefore, serve as a unique set of predictive biomarkers for chemotherapy response in this subgroup of lung cancer patients. Our findings provide preliminary data that warrant further evaluation of YB-1-targeted chemotherapeutics to prevent acquired drug resistance.

Author’s contributions

LL was responsible for the study concept and design. FZ, GW and HS acquired the data for the study. JW and YL were responsible for data analysis and interpretation. LL prepared the manuscript and LX reviewed the manuscript. XZ is corresponding author. All other authors read and approved the final version of the manuscript.

Funding

This project was supported by the National Natural Science Foundation of China (Grant No.81502634).

Compliance with ethical standards

Conflict of interest

None declared.

Ethics approval

This study was approved by the Ethics Committee of The Second Hospital of Hebei Medical University, Shijiazhuang, China.

Consent for publication

All subjects participating in image acquisition signed consent forms.