This article is mentioned in:

Introduction

Chemotherapy is usually given systemically to patients with cancer following surgical intervention or radiotherapy (1). However, certain patients are unresponsive to chemotherapy whereby the cancer cells become resistance to chemotherapeutic drugs, a phenomenon known as multidrug resistance (MDR). MDR is resistance to multiple structurally and mechanistically unassociated classes of anticancer drugs by the cells (1), and has become a major impediment to cancer treatment due to cancer relapse which eventually leads to mortality.

Numerous reports have suggested a plausible association between MDR and Y-box binding protein-1 (YB-1), an evolutionary conserved DNA or RNA binding protein. YB-1 was initially discovered as a transcription factor that binds to the Y-box sequence (inverted CCAAT), present in the promoter region of the major histocompatibility complex class (MHC) II (2). It has pleotropic functions and is implicated in fundamental processes, including transcriptional regulation, DNA repair, mRNA splicing, translation and stability (3). The presence of nuclear YB-1 has been demonstrated as associated with enhanced transcription of the multidrug resistance 1 (MDR1) gene, eventually leading to the upregulation of P-glycoprotein, which is indicative of chemoresistance, cancer progression and poor prognosis (4). YB-1 was also suggested to be a promising prognostic biomarker for patients with breast cancer (5). A recent report has further suggested the involvement of YB-1 in cisplatin resistance in gastric cancer (6).

Janus kinase (JAK)/signal transducer and activator of transcription (STAT) signaling is an important pathway involved in fundamental cellular processes, including tumourigenesis, immune responses and sex determination (7-9). In particular, aberrant expression of STAT3 is associated with cancer progression through the facilitation of cell proliferation, metastasis and chemoresistance (10). Dysregulated STAT3 signaling has been implicated in gastric carcinogenesis (11,12). Immunohistochemical staining of STAT3 in gastric adenocarcinoma tissues demonstrated that STAT3 serves as a predictor of poor prognosis and is associated with lymph node metastasis (13,14). Disruption of STAT3 leads to a decrease in gastric cancer cell survival through a reduction in the expression of the anti-apoptotic molecule, survivin (15). Furthermore, inhibition of STAT3 has been reported to increase apoptosis in gastric cancer cells treated with cisplatin (16). Notably, the expression levels of STAT3 downstream targets, including B cell lymphoma-2 (Bcl-2) and c-Myc, were reported to be down-regulated, while Bax and p53 levels were increased, elevating the sensitivity of gastric cancer cells to apoptosis (16).

As YB-1 and JAK/STAT signaling are known to enhance chemoresistance in gastric cancer, the present study aimed to determine whether they interact with each other to promote chemoresistance. The results of the current study suggested the involvement of YB-1 and STAT3 in the chemoresistance of gastric cancer NUGC3 cells, whereby the expression of multi-drug resistance-associated proteins, ABBC3 and ABCC2, were transcriptionally regulated by YB-1 and STAT3, respectively. Notably, the inhibition of YB-1 and STAT3 activities were demonstrated to exhibit a synergistic effect on suppressing chemoresistance and cell invasion, suggesting that the combination of YB-1 antagonists with small molecule inhibitors of STAT3 may serve as an effective therapy for chemoresistance in gastric cancer.

Methods and materials

Cell culture

The poorly differentiated gastric adenocarcinoma cell line NUGC3 (American Type Culture Collection, Manassas, VA, USA) was routinely cultured in RPMI-1640 supplemented with 10% fetal bovine serum (FBS) (both form HyClone; GE Healthcare Life Sciences, Logan, UT, USA). When the cells were 80-90% confluent, subculturing of the cells were performed.

Short interfering RNA (siRNA/si) transfection

NUGC3 cells were plated in 6-well plates at a density of 2×105 cells/well. YBX1, ABCC2 and ABBC3 ON-TARGETplus SMARTpool siRNAs, as well as non-targeting pool siRNA were obtained from GE Healthcare Dharmacon, Inc. (Lafayette, CO, USA), while siRNA for STAT3 was obtained from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). Table I list the sequences of siRNAs used in the present study. NUGC3 cells were then transfected with 20 nM of siNegative, siYB-1, siSTAT3, siABCC2 or siABCC3 using DharmaFECT1 Transfection reagent (GE Healthcare Dharmacon, Inc.). Following an overnight incubation with siRNAs, culture medium was replaced prior to further incubation. The transfected cells were then collected at 48 h post transfection for RNA isolation and 72 h post transfection for protein isolation. For double knockdown of STAT3 and YB-1, each siRNA was used at a final concentration of 20 nM.

Table I List of siRNA sequences used.

Table I

List of siRNA sequences used.

siRNA	Target sequence (5′-3′)
NT siRNA
D-001810-10	UGGUUUACAUGUCGACUAA
D-001810-11	UGGUUUACAUGUUGUGUGA
D-001810-12	UGGUUUACAUGUUUUCUGA
D-001810-13	UGGUUUACAUGUUUUCCUA
YB-1 siRNA
J-010213-06	CUGAGUAAAUGCCGGCUUA
J-010213-07	CGACGCAGACGCCCAGAAA
J-010213-08	GUAAGGAACGGAUAUGGUU
J-010213-09	GCGGAGGCAGCAAAUGUUA
ABCC2 siRNA
J-004225-05	GGAUGAAUCUCGACCCUUU
J-004225-06	GUAUCAGGUUUGCCAGUUA
J-004225-07	GGUCAAGUGUUCUACAGAU
J-004225-08	GAACCUGACUGUCUUCUUU
ABCC3 siRNA
J-007312-05	GGACAAAGGAGUAGUAGCU
J-007312-06	GCACACCGGCUUAACACUA
J-007312-07	CCACCCUGCUGAUACAGUA
J-007312-08	ACAAUGAUCCAGUCACCUA

[i] YB-1, Y-box binding protein-1; siRNA, small interfering RNA; ABC, ATP-binding cassette; NT, negative control.

RT-qPCR

For RT-qPCR, 48 h following siRNA transfection, total RNA was extracted using the RNeasy Mini kit (Qiagen GmbH, Hilden, Germany). A total of 1 µg of total RNA was then used to synthesize first strand cDNA with the SuperScript III First-Strand Synthesis system (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) as described previously (17). The HT7900 Fast Real-Time PCR system was used to examine alterations in gene expression using Fast SYBR-Green Master mix (both from Applied Biosystems; Thermo Fisher Scientific, Inc.). For the housekeeping gene, GAPDH was used for normalization. The thermocycling conditions used were as follows: initial activation at 95°C for 40 sec; and then 40 cycles consisting of melting at 95°C for 1 sec and annealing/extension at 60°C for 20 sec. The 2−∆∆Cq method (18) was utilized to express the changes in gene expression and each sample was run in triplicate. Table II contains the list of primers used.

Table II List of primers used for RT-qPCR.

Table II

List of primers used for RT-qPCR.

Gene	Forward primer sequence (5′→3′)	Reverse primer sequence (5′→3′)
YB-1	AAGTGATGGAGGGTGCTGAC	TTCTTCATTGCCGTCCTCTC
STAT3	ACCAGCAGTATAGCCGCTTC	GCCACAATCCGGGCAATCT
GAPDH	GAAGGTGAAGGTCGGAGTCAACG	TGCCATGGGTGGAATCATATTGG
ABCC1	ATGCTCACTTTCTGGCTGGTA	AGCGATCTGAGAAACAGGACA
ABCC2	TATCCCACGAAGTGACAGAGG	ATGGTCGTCTGAATGAGGTTG
ABCC3	GGACTTCCAGTGCTCAGAGG	TGGATGAGGTTGTCAGTCTCC
ABCC4	TGTTCTTCTGGTGGCTCAATC	AGAACCCTTGCAACTCCTCTC
ABCC5	CATCCGGACTACTTCCAAACA	CAGAGACCACACGTCTTCCAT
ABCG2	AACTTCTGCCCAGGACTCAAT	CAGGTAGGCAATTGTGAGGAA

[i] RT-qPCR, reverse transcription-quantitative polymerase chain reaction; YB-1, Y-box binding protein-1; ABC, ATP-binding cassette; STAT, signal transducer and activator of transcription.

Western blot analysis

Total protein extraction and protein measurement were performed as previously described (17). Radioimmunoprecipitation lysis buffer (Pierce; Thermo Fisher Scientific, Inc.) was used to lyse the cells. Following Bradford protein assay quantification of the isolated proteins, 30 µg of proteins were separated using a 8 or 10% SDS-PAGE gel, then transferred onto a polyvinyl difluoride membrane with a semi-dry system (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Next, membranes were incubated with 5% bovine serum albumin (BSA) (Santa Cruz Biotechnology, Inc.), at room temperature for 1 h, followed by incubation at 4°C overnight with the following primary antibodies: rabbit poly-clonal anti-YB-1 [1:1,000; RIKEN, Wako-shi, Japan; (17,19), the antibody was obtained from Dr Ken Matsumoto, who is stationed at RIKEN, Wakoshi.]; rabbit polyclonal anti-STAT3 (1:2,000; cat. no. 12640); rabbit polyclonal anti-phospho-STAT3 (Ser727) (1:1,000; cat. no. 9134); rabbit polyclonal anti-phospho-STAT3 (Tyr705) (1:1,000; cat. no. 9145) (all from Cell Signaling Technology, Inc., Danvers, MA, USA); and mouse monoclonal anti-β-actin (1:6,000) (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany; cat. no. A2228). Subsequent incubation of the membranes was performed with horseradish peroxidase-conjugated goat anti-rabbit IgG (1:5,000; Sigma-Aldrich, Merck KgaA; cat. no. A0545) or goat anti-mouse IgG (1:6,000; Sigma-Aldrich, Merck KgaA; cat. no. A9044) secondary antibodies at room temperature for 1 h. Proteins were identified with SuperSignal West Pico Chemiluminescent ECL substrate (Pierce; Thermo Fisher Scientific, Inc.) and quantified by GS-800 optical densitometry with Quantity-One Image Analysis software version 4.62 (both from Bio-Rad Laboratories, Inc.).

Cell migration and invasion assays

Migration assays were performed with 6.5-mm polycarbonate membrane Transwell inserts (Corning Incorporated, Corning, NY, USA), while BD BioCoat™ Matrigel™ Invasion chambers (BD Biosciences, Franklin Lakes, NJ, USA) were used for the invasion assays. A total of 8×104 NUGC3 cells were seeded in 200 µl of RPM1-1640 serum-free media in the upper part of the Transwell migration or invasion inserts, 48 h post siRNA transfection. Subsequently, 600 µl of RPMI-1640 with 20% FBS was placed in the lower chamber of each well, which acted as a chemoattractant for the cells. The cells were then allowed to migrate or invade for 20 or 48 h, respectively, prior to fixation with absolute methanol for 3 min and stained with 0.5% (w/v) crystal violet in 20% methanol for 30 min at room temperature. Inserts were then washed with distilled water. Next, the upper membrane of the inserts were wiped with wet cotton swabs to remove cells that did not migrate or invade. To visualize migrated or invaded NUGC3 cells, images of five different fields were captured with a Nikon SMZ 1500 stereo microscope (Nikon Corporation, Tokyo, Japan) at ×100 magnification and stained cells were counted. The assay was performed in triplicates.

Determination of inhibitory concentration (IC50) of chemotherapy drugs

NUGC3 cells were grown at 4×104 cells/well in a 24-well plate, and transfected with siNegative or siYB-1 after 24 h post seeding using the aforementioned protocol. Then, following transfection for 48 h, treatment with varying concentrations of doxorubicin hydrochloride (0, 0.3125, 0.625, 1.25, 2.5 and 5 µM) or epirubicin hydrochloride (0, 0.3125, 0.625, 1.25, 2.5, 10 and 20 µM) (both from Sigma-Aldrich; Merck KGaA) was performed. The corresponding percentages of cell viability at different concentrations of drugs were determined by MTS assay, after 24 h incubation. Subsequently, 600 µl of MTS mixture (diluted 1:5 in RPMI-1640 with 10% FBS; Promega Corporation, Madison, WI, USA) were added to each of the wells. A negative control set of blank wells containing only MTS solution with no cells was also prepared. Incubation of cells was performed in the dark at 37°C for 4 h. CellTiter 96® AQueous One Solution Cell Proliferation assay reagent (Promega Corporation) was added to cells to measure cell viability. A wavelength of 490 nm was used for measuring of the absorbance with an ELISA plate reader. The IC50 value of each drug was then determined by plotting the survival rate after 24 h of drug treatment against varying drug concentrations.

Transmission electron microscopy (TEM)

The effect of doxorubicin treatment in NUGC3 cells was evaluated using TEM. NUGC3 cells were treated with IC50 of doxorubicin for 48 h. Cells in the supernatant were obtained by centrifugation at 1,000 × g for 5 min at room temperature and adherent cells were detached and then collected by centrifuging at 300 × g for 5 min at room temperature. Then, 2.5% glutaraldehyde was used to fix the cell pellets for 1 h at room temperature, prior to washing with 1X PBS thrice and post-fixation with 1% osmium tetroxide (Agar Scientific Ltd., Stansted, UK) for 1 h at room temperature. Following dehydration with increasing concentrations of ethanol, the pellet was embedded in araldite. The fixation in araldite was performed at 40°C for 45 min, then 50°C for 30 min and finally 55°C or 30 min. The samples were sectioned and mounted on a copper grid coated with formvar. Sections of the samples were stained with uranyl acetate for 10 min and lead citrate for 8 min, both at room temperature and finally viewed under a Philips CM120 BioTwin TEM (FEI; Thermo Fisher Scientific, Inc.) at ×2,800 or ×4,400 magnifications.

Immunofluorescence staining

NUGC3 cells were grown on a Lab-Tek 4-Chambered coverglass (Nalge Nunc International; Thermo Fisher Scientific, Inc.) until 60-70% confluency was achieved. For cell fixation, cells were incubated with 4% para-formaldehyde for 20 min at room temperature. Subsequently, 0.05% Tween-20/PBS was used for washing twice, followed by permeabilisation with 0.2% Triton-X/PBS for 5 min and blocking with 1% (w/v) BSA at room temperature for 30 min. The YB-1 primary antibody (1:250) (17) was diluted in 1% BSA (w/v) 1X PBS and incubated at 4°C overnight. Washing was performed thrice with 0.05% Tween-20/PBS. Next, cells were incubated with goat anti-rabbit IgG Alexa Fluor 594 secondary antibody (1:400; cat. no. A11012; Thermo Fisher Scientific, Inc.) at room temperature in the dark for 1 h. The cover glass was removed following washing thrice again with 0.05% Tween-20/PBS and mounted on a coverslip with VECTASHIELD fluorescent mounting medium containing DAPI (Vector Laboratories., Burlingame, CA, USA), which counterstains the nucleus. Subsequently, an Olympus Fluoview FV1000 Laser Scanning confocal microscope (Olympus Corporation, Tokyo, Japan) was used to visualize the stained cells at ×400 or ×600 magnifications.

For siYB-1, siSTAT3 or siNegative transfected-NUGC3 cells, immunofluorescence staining was performed on the cover slips with seeded cells at 48 h post transfection. The primary antibodies used were anti-phospho-STAT3 (Tyr705) (1:50), anti-total STAT3 (1:1,000) (both from Cell Signaling Technology, Inc.) and anti-YB-1 (1:250) (17), followed by the use of goat anti-rabbit IgG Alexa Fluor 594 (1:400) (cat. no. A11012) or goat anti-rabbit IgG Alexa Fluor 488 (1:400) (both from Thermo Fisher Scientific, Inc.; cat. no. A11008) secondary antibodies. The procedures were the same as aforementioned with the exception that the cells were fixed with 100% methanol for 20 min for immunofluorescence staining of phospho-STAT3 (Tyr705) and total STAT3 at room temperature.

Cotreatment of the JAK inhibitor AG490 and doxorubicin hydrochloride, following YB-1 knockdown in NUGC3 cells

Cell were transfected as aforementioned for 48 h and 0 or 50 µM AG490 was used for the 48 h treatment, prior to exposure to medium with or without IC50 of doxorubicin hydrochloride for 24 h. Subsequently, an MTS assay was performed to determine cell viability.

Doxorubicin hydrochloride treatment in siABCC2 and siABCC3 NUGC3 cells

NUGC3 cells were grown in a 24-well plate at a density of 4×104 cells/well, followed by treatment with siABCC2, siABBC3 or siNegative, to a final concentration of 20 nM, with the aforementioned transfection procedures. At 48 h post transfection, treatment with doxorubicin hydrochloride for 24 h was performed, then cell viability was evaluated using an MTS assay.

Statistical analysis

The GraphPad Prism 5.0 (GraphPad Software, Inc., La Jolla, CA, USA) statistical package was used. For the means between two groups, a two tailed Student’s t-test was used; whereas for comparison between ≥3 groups, one-way analysis of variance was used followed by Tukey’s post-hoc test. Data are presented as mean ± standard error of the mean. P<0.05 was considered to indicate a statistically significant difference. All experiments were repeated in triplicates.

Results

YB-1 protein translocalizes to the nucleus upon chemotherapy treatment in NUGC3 cells

Several reports have suggested that chemotherapeutic drugs kill cancer cells susceptible to the treatment through promoting apoptosis, and have implicated nuclear YB-1 in chemoresistance (19-21). Hence, whether the treatment of NUGC3 cells with chemotherapy drugs, including doxorubicin hydrochloride were able to induce cell death was examined. NUGC3 cells treated with doxorubicin hydrochloride at a concentration of 4.65 µM (IC50) (data not shown) for 48 h exhibited morphological changes, including cytoplasmic blebbing (Fig. 1A; black arrows), indicative of apoptosis. Subsequently, whether nuclear translocation of YB-1 increased upon doxorubicin hydrochloride treatment was investigated. Notably, a marked increase in the levels of nuclear YB-1 was detected upon doxorubicin hydrochloride compared with the control, suggesting the potential role of YB-1 in chemoresistance (Fig. 1B).

Figure 1 YB-1 translocates to the nucleus upon chemotherapy drug treatment. (A) Transmission electron microscopy micrographs of NUGC3 cells treated with IC50 of doxorubicin hydrochloride for 48 h revealed that the cells underwent apoptosis-like changes, as evidenced by cytoplasmic blebbing (right, black arrows), as compared with untreated cells (left). Scale bar, 2 µm. (B) Immunofluorescence staining for YB-1 in NUGC3 cells treated with doxorubicin hydrochloride. In untreated cells, YB-1 was primarily located in the cytoplasm (left), but YB-1 translocated to the nucleus upon doxorubicin hydrochloride treatment (middle). Scale bar, 100 µm. YB-1, Y-box binding protein-1.

Knockdown of YB-1 decreases chemoresistance of NUGC3 cells

To investigate whether YB-1 functions in chemoresistance of gastric cancer NUGC3 cells, the inhibitory effects of YB-1 on cell viability, following exposure to chemotherapy drugs were evaluated. The siRNA-mediated silencing efficiency of YB-1 at the mRNA and protein levels was ~99 and 56%, respectively (Fig. 2A and B). Notably, inhibition of YB-1 induced a decrease in chemoresistance of the cells to doxorubicin hydrochloride and epirubicin hydrochloride, two of the commonly used chemotherapy drugs for cancer treatment. While the IC50 of doxorubicin hydrochloride in NUGC3 cells transfected with siYB-1 was 2.8 µM, the IC50 of doxorubicin hydrochloride in control cells was 7.2 µM (Fig. 2C). Similarly, the IC50 of epirubicin hydrochloride in cells transfected with siYB-1 was decreased compared with control cells (2.1 vs. 6.3 µM); however, this was not statistically significant (Fig. 2D).

Figure 2 YB-1 knockdown suppresses chemoresistance of NUGC3 cells. YB-1 knockdown caused a significant decrease in YB-1 mRNA (A) and protein (B) levels, as compared with the controls at 72 h post siRNA transfection. Representative dose response curves illustrate that the viability of NUGC3 cells treated with doxorubicin hydrochloride (C) or epirubicin hydrochloride (D) further decreased upon YB-1 knockdown compared with the controls. Data are presented as the mean ± standard error of the mean from three independent experiments, **P<0.01 and ***P<0.001. YB-1, Y-box binding protein-1; si, small interfering RNA.

Synergistic inhibitory effects of YB-1 and JAK/STAT signaling on chemoresistance

The JAK/STAT signaling pathway has also been implicated in the chemoresistance of cancer cells (10,22,23). For example, STAT3-dependent myeloid cell leukemia-1 (Mcl-1) expression was reported to affect chemo-sensitivity in melanoma, whereby Mcl-1 confers survival advantage against B-Raf proto-oncogene serine/threonine kinase inhibitors (24). Therefore, the present study examined the potential interaction between JAK/STAT signaling and YB-1 in the chemoresistance of NUGC3 cells. The viability of siNegative-transfected control cells co-treated with the JAK2 inhibitor AG490 and doxorubicin hydrochloride compared with that of the control cells treated with AG490 or doxorubicin hydrochloride alone was examined. As shown in Fig. 3, AG490 or doxorubicin hydrochloride treatment alone reduced the viability of cells compared with the untreated control group. Notably, co-treatment of AG490 and doxorubicin hydrochloride further enhanced the decreased viability induced by AG490 or doxorubicin hydrochloride. The effects of co-treatment with AG490 and doxorubicin hydrochloride on cell viability were examined in YB-1 knockdown cells. Cell death in YB-1 knockdown cells treated with AG490, doxorubicin hydrochloride, or AG490 and doxorubicin hydrochloride, was further enhanced compared with that in siNegative-transfected control cells treated with AG490, doxorubicin hydrochloride, or AG490 and doxorubicin hydrochloride, respectively. These suggest that JAK/STAT signaling and YB-1 function synergistically in promoting chemoresistance of gastric cancer cells to doxorubicin hydrochloride.

Figure 3 Synergistic effect of YB-1 and JAK/STAT signaling on chemore-sistance. Note that co-treatment of the JAK inhibitor AG490 and doxorubicin hydrochloride enhanced the decreased viability caused by AG490 at 50 µM or IC50 of doxorubicin hydrochloride alone. Notably, cell death in siNegative-transfected control cells treated with AG490, doxorubicin hydrochloride or AG490 and doxorubicin hydrochloride, was further enhanced upon YB-1 knockdown, suggesting the key role of YB-1 in chemoresistance. Data are presented as the mean ± standard error of the mean from three independent experiments, ***P<0.001. YB-1, Y-box binding protein-1; si, small interfering RNA; JAK, Janus kinase; STAT, signal transducer and activator of transcription.

YB-1 and STAT3 regulate the expression of different ABC transporters

To obtain insight into the possible interaction between YB-1 and JAK/STAT signaling in chemoresistance, the effect of YB-1 knockdown on STAT3 expression and vice versa was evaluated. However, no significant changes were observed in total STAT3 and phosphorylated STAT3 (Tyr705 and Ser727) levels upon YB-1 knockdown in NUGC3 cells (Fig. 4A, upper panel). Furthermore, immunofluorescence staining of siYB-1-transfected NUGC3 cells revealed no significant changes in the subcellular localization of phosphorylated STAT3 (Tyr705) and total STAT3 in the nucleus (Fig. 4A, lower panel). Similarly, STAT3 knockdown did not significantly affect YB-1 expression (Fig. 4B, upper panel) or the subcellular localization of YB-1 in the nucleus (Fig. 4B, lower panel).

Figure 4 Expression of YB-1 and STAT3 is not mutually regulated. (A) YB-1 and STAT3 mRNA levels upon YB-1 knockdown (upper left and middle panels). No significant changes in STAT3 mRNA were detected when YB-1 was inhibited in NUGC3 cells. YB-1 knockdown also failed to affect total and phosphorylated STAT3 levels (upper right panel). Similarly, immunofluorescence staining of phospho-STAT3 (Tyr705) and total STAT3 demonstrated no changes in their subcellular localization upon YB-1 knockdown (lower panel). (B) STAT3 knockdown resulted in a significant decrease in STAT3 mRNA levels, but not YB-1 mRNA levels (upper left and middle panels). Consistently, STAT3 inhibition did not affect YB-1 expression (upper right panel). Immunofluorescence staining revealed no changes in the subcellular localization of YB-1 upon STAT3 knockdown (lower panel). Data are presented as the mean ± standard error of the mean from three independent experiments, ***P<0.001. Scale bar, 20 µm. YB-1, Y-box binding protein-1; si, small interfering RNA; JAK, Janus kinase; STAT, signal transducer and activator of transcription; p, phosphorylated.

As ABC transporters are implicated in multidrug resistance in various cancer types (25-28), the present study aimed to evaluate the regulatory effect of YB-1 and STAT3 on ABC transporters, as the two proteins act as transcriptional factors. Upon YB-1 knockdown in NUGC3 cells, a signifi-cant reduction in the ABBC3 transporter gene was observed out of the six different ABC transporters (ABCC1, ABCC2, ABCC3, ABCC4, ABCC5 and ABCG2) evaluated (Fig. 5A). However, STAT3 knockdown only led to a significant decrease in ABCC2 expression (Fig. 5A). These results suggest that YB-1 and STAT3 act independently in chemoresistance by regulating the expression of different ABC transporters, which serve essential roles in multidrug resistance in various cancer types. The possible synergistic effects of YB-1 and STAT3 on ABC transporter expression were also examined. However, no significant synergistic effects were observed, confirming that YB-1 and STAT3 functions independently in regulating the expression of ABC transporters.

Figure 5 YB-1 and STAT3 function to regulate the expression of ABC transporters. (A) ABC transporter gene expression in NUGC3 cells. YB-1 knockdown led to a significant downregulation of ABCC3 in NUGC3 cells, while STAT3 knockdown decreased the expression of ABCC2. (B) The knockdown efficiencies of siABCC3 (left panel) and siABCC2 (right panel) at mRNA levels. (C) Cell viability of NUGC3 cells upon knockdown of ABCC3 (left panel) or ABCC2 (right panel), in the presence of doxorubicin hydrochloride. Data are presented as the mean ± standard error of the mean from three independent experiments. *P<0.05, **P<0.01 and ***P<0.001, compared with the siNegative group. YB-1, Y-box binding protein-1; ABC, ATP-binding cassette; STAT, signal transducer and activator of transcription; si, small interfering RNA.

The role of ABCC2 and ABCC3 on the viability of NUGC3 cells treated with doxorubicin hydrochloride was then examined. The knockdown efficiencies of siABCC3 and siABCC2 were 59.9% (Fig. 5B, left panel) and 56.5% (Fig. 5B, right panel), respectively. ABCC3 knockdown in NUGC3 cells led to decreased cell viability upon doxorubicin hydrochloride treatment, as compared with the negative control (Fig. 5C, left panel). However, no significant reduction in the viability of ABCC2 knockdown cells was observed upon treatment with doxorubicin hydrochloride (Fig. 5C, right panel).

YB-1 and STAT3 exhibit synergistic effects on facilitating cell invasion in NUGC3 cells

As previously mentioned, YB-1 and STAT3 exhibited a synergistic effect in mediating chemoresistance of NUGC3 cells (Fig. 3). Hence, the effects of their combination on cell migration and invasion were investigated. A double knockdown of YB-1 and STAT3 was performed in NUGC3 cells. STAT3 knockdown alone significantly decreased cell migration as compared with the control. YB-1 knockdown alone also led to a decline in cell migration, albeit statistically insignificant. The double knockdown of YB-1 and STAT3 failed to further enhance the decrease caused by STAT3 knockdown alone, suggesting that there was no synergistic effect on cell migration (Fig. 6A). In contrast, the double knockdown of YB-1 and STAT3 in NUGC3 cells further inhibited cell invasion caused by YB-1 or STAT3 knockdown alone (Fig. 6B), suggesting the synergistic role of YB-1 and STAT3 in cell invasion.

Figure 6 YB-1 and STAT3 synergistically act to decrease cell invasion in NUGC3 cells. (A) STAT3 knockdown resulted in a decrease in cell migration as compared with the control. However, no synergistic inhibitory effect of YB-1 on STAT3 knockdown-mediated decrease in cell migration was observed. (B) Double knockdown of YB-1 and STAT3 further inhibited cell invasion caused by YB-1 or STAT3 knockdown alone, suggesting that YB-1 and STAT3 act together to promote cell invasion in NUGC3 cells. Data are presented as the mean ± standard error of the mean from three independent experiments, *P<0.05, **P<0.01 and ***P<0.001. Scale bar, 100 µm. YB-1, Y-box binding protein-1; ABC, ATP-binding cassette; STAT, signal transducer and activator of transcription; si, small interfering RNA.

Discussion

YB-1 is a pleotropic protein which is upregulated in numerous human malignancies, including gastric cancer (29), nasopharyrngeal carcinoma (19) and breast cancer (17). YB-1 has been reported to serve a vital role in MDR in patients with cancer (4,30). Under normal conditions, YB-1 is predominantly expressed in the cytoplasm (31), but translocates to the nucleus upon exposure to environmental stress, including UV light, drug treatment and hyperthemia (4,31). Furthermore, the YB-1 nuclear localization is associated with MDR1 gene expression, whose product mediates chemoresistance in various cancer types, including breast, synovial sarcoma and osteosarcoma cancer (3). In line with the potential role of YB-1 in chemo-resistance in various cancer types, including nasopharyngeal cancer (19), breast cancer (21), melanoma (32) and ovarian cancer (33), the present study demonstrated the cytoplasmic to nuclear translocation of YB-1 in the NUGC3 gastric cancer cell line upon doxorubicin hydrochloride treatment. In support of this, YB-1 reduction led to a significant decrease in the chemoresistance of NUGC3 cells to doxorubicin hydrochloride and epirubicin hydrochloride.

Notably, constitutively-active JAK/STAT signaling has also been demonstrated to be associated with gastric cancer development and progression (34). In particular, it has been reported that STAT3 is phosphorylated by JAK, which subsequently translocates to the nucleus, and acts as a transcription factor to modulate the expression of a wide variety of genes that participate in proliferation, migration as well as invasion in gastric cancer cells (35,36). STAT3 also promotes anti-apoptotic molecule expression, including Bcl-2 and Mcl-1, thereby conferring pro-survival signals in gastric cancer cells (16). Furthermore, it serves as a useful biomarker in gastric cancer prognosis (37), with a high expression of STAT3 identified in gastric cancer (15).

YB-1 and JAK/STAT signaling have been implicated in chemoresistance (7,20). Hence, a possible synergistic effect of YB-1 and JAK/STAT signaling on chemoresistance was examined using siYB-1 and the JAK2 inhibitor AG490 in gastric cancer NUGC3 cells. Notably, an enhanced reduction in the viability of NUGC3 cells co-treated with siYB-1 and AG490 in the presence of doxorubicin hydrochloride was observed, as compared with the viability in cells treated with siYB-1 or AG490 alone upon doxorubicin hydrochloride treatment. Thus, these results suggest that YB-1 and JAK/STAT signaling are involved in chemoresistance, and may act synergistically to promote chemoresistance of gastric cancer. To obtain more insight into the mechanism underlying their synergistic effects on chemoresistance, the inhibitory effect of YB-1 on STAT3 levels was ascertained. However, no alterations in total STAT3 and phosphorylated STAT3 levels upon YB-1 knockdown were detected. Similarly, STAT3 knockdown did not lead to any striking changes in YB-1 expression. YB-1 inhibition was shown to decrease phosphorylated STAT3 (Ser727) levels, but not phosphorylated STAT3 (Tyr705) or total STAT3 levels in breast cancer cells (38). This may suggest the cell type-dependent effect of YB-1 on STAT3 activity. Nonetheless, gene expression analysis of MDR-associated genes and ABC transporters upon siYB-1 or siSTAT3 revealed differential gene regulations, whereby YB-1 regulates ABCC3 expression while STAT3 regulates ABCC2 expression. In addition, no synergistic effects were observed for YB-1 and STAT3 on the ABC transporters, further suggesting that YB-1 and STAT3 regulate the expression of ABC transporters independently.

Regarding the role of ABCC2 and ABCC3 in chemo-resistance, siRNA mediated knockdown of the two ABC transporters was then performed. Cell viability was reduced to an extent in doxorubicin hydrochloride treated siABCC3 NUGC3 cells, but not in siABCC2 cells. The possible explanations for these observations may be i) the low knockdown efficiency of ABCC2; or ii) ABCC2 is not the sole target of STAT3 in chemoresistance. The ABCC2 siRNAs used in the present experiment consisted of four individual siRNAs that were pooled together, which would minimise the off-target effects of transfection. However, the knockdown efficiency was relatively low and therefore no significant difference in cell viability was observed when compared with the control cells. Alternatively, STAT3 may regulate chemoresistance via other targets, besides ABCC2. For example, STAT3 is also involved in the regulation of anti-apoptotic Bcl-2 and Mcl-1 proteins, which may confer chemoresistance in gastric cancer (16). Future experiments should investigate the role of ABCC2 and ABCC3 in mediating chemoresistance by determining whether ABCC2 and ABCC3 overexpression reverse the effects of YB-1 and STAT3 knockdown on NUGC3 cells. Furthermore, determining how YB-1 and STAT3 regulate the expression of ABCC3 and ABCC2, respectively, may be of importance. Using the more advanced clustered regularly inter-spaced short palindromic repeat/CRISPR-associated protein 9 system to knockout ABCC2 or ABCC3 gene expression may be performed to determine the effects on chemoresistance, as the knockdown efficiencies using siABCC2 and siABCC3 are relatively low.

The results demonstrated that YB-1 and STAT3 do not mutually regulate the expression of one another, but the two proteins act synergistically in the chemoresistance of gastric cancer cells, suggesting that they may also act together in other biological processes. A previous study reported that STAT3 expression is associated with spread to the lymph nodes in gastric cancer tissues, thus enhancing gastric cancer progression (39). Consistently, the knockdown of STAT3 led to a significant reduction in cell migratory and invasive properties. However, YB-1 has been implicated in the migratory property of NUGC3 cells, but not cell invasion and epithelial-to-mesenchymal transition, complex processes essential to the initial metastatic cascade (40). No significant inhibitory effects of YB-1 were detected with regards to the migratory and invasive properties of NUGC3 cells. Notably, STAT3 knockdown-mediated decrease in cell invasion was further enhanced upon YB-1 knockdown. It was reported that YB-1 prevents apoptosis in Her2-overexpressing breast cancer cells in vitro and promotes tumour growth in mice via the mammalian target of rapamycin/STAT3 signaling pathway (41), indicating a possible cross-talk between YB-1 and STAT3 signaling underlying the pathogenesis of breast cancer. Future experiments to determine the effects of YB-1 and STAT3 downregulation on gastric cancer in vivo may aid in determining the associated signaling pathways of these two molecules.

The present results indicate that YB-1 and STAT3 act synergistically in chemoresistance and cell invasion in NUGC3 gastric cancer cells, thus providing the potential for the development of novel therapeutic interventions targeting these two molecules in gastric cancer.

Funding

The present study was supported by the Ministry of Education grants (grant no. MOE2013-T2-1-129 to BHB) and (grant no. MOE T1-2016 Sep-13 to QH). JPL was a recipient of the Ong Hin Tiang Scholarship in Cancer Research.

Availability of data and materials

All data generated or analysed during the present study are included in this published article.

Authors’ contributions

PJC and TTG designed and performed the experiments. JPL wrote the manuscript and performed the experiments. PK performed the experiments. QH, BHB and GHB conceived the experiment designs and revised the manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Acknowledgments

Not applicable.

References