MicroRNA-650 Was a Prognostic Factor in Human Lung Adenocarcinoma and Confers the Docetaxel Chemoresistance of Lung Adenocarcinoma Cells via Regulating Bcl-2/Bax Expression

Increasing evidence shows that dysregulation of microRNAs (miRNAs) is involved in malignant transformation. We investigated the clinical significance of miR-650 and its involvement in chemoresistance to docetaxel. Our results showed that the relative expression level of miR-650 was significantly higher in LAD tissues than in corresponding nontumor tissues and high level of miR-650 expression was found to be significantly associated with high incidence of lymph node metastasis, advanced clinical stage and poor prognosis of LAD patients. Univariate and multivariate analyses indicated that high miR-650 expression was an independent prognostic factor for survival. Also, we found that the level of miR-650 in LAD tissues was correlated with the response of patients to docetaxel-based chemotherapy. Silencing of miR-650 could increase the in vitro sensitivity of docetaxel-resistant LAD cells to docetaxel, while upregulation of miR-650 decreased the sensitivity of parental LAD cells to docetaxel both in vitro and in vivo. Additionally, silencing of miR-650 could enhance the caspase-3-dependent apoptosis, which might be correlated with the decreased ratio of Bcl-2/Bax. Further researches suggested that inhibitor of growth 4 (ING4) was a direct target of miR-650. Downregulated or upregulated ING4 expression could partially rescue the effects of miR-650 inhibitor or mimics in docetaxel-resistant or parental LAD cells. Furthermore, we found that ING4 was upregulated in docetaxel-responding LAD tissues, and its expression was inversely correlated with miR-650. Thus, miR-650 is a novel prognostic marker in LAD and its expression is a potential indicator of chemosensitivity to docetaxel-based chemotherapy regimen.


Introduction
Lung cancer is one of the commonest malignant malignancies all over the world [1]. The estimated incidence of new cases of lung cancer in the world is nearly 1,000,000 every year. In China, about 300,000 new lung cancer patients and more than 250,000 deaths from the disease are predicted each year. Adenocarcinoma of the lung (LAD) is the most common type of lung cancer and accounts for 30 to 35 percent of primary lung tumors [2]. Lung carcinogenesis is a complex, stepwise process that involves an accumulation of genetic and epigenetic changes that lead to alterations in normal lung epithelium, to in situ carcinoma, and finally to invasive and metastatic cancers [3]. Thus, elucidation of the molecular mechanisms involved in the pathogenesis of LAD will be helpful for the development of better prognostic markers and novel therapeutic targets to improve clinical treatment of LAD patients.
MicroRNAs (miRNAs) are a class of short (20-23 nucleotides in length), endogenous, single-stranded RNAs that regulate gene expression [4]. Mature miRNAs and Argonaute (Ago) proteins form the RNA-induced silencing complex (RISC), which mediates post-transcriptional gene silencing through induction of messenger RNA (mRNA) degradation or translational inhibition [5,6]. Although they account for only a minor fraction of the expressed genome, miRNAs have been found to play vital roles in various cellular processes such as growth, differentiation, apoptosis, motility and malignant transformation [7][8][9]. Increasing evidence shows that miRNAs are grossly dysregulated in human cancers including LAD, and may serve as oncogenes or tumour suppressors [10,11]. A single miRNA can have multiple targets involved in different oncogenic pathways, while a single gene can also be regulated by several miRNAs [12,13]. Thus, identification of dysregulated miRNAs and their target genes is critical to understand the characterization of the miRNA pathways and their underlying molecular mechanisms, which will help us to further understand the pathogenesis of LAD and to develop miRNA-direct therapeutics against this disease, including reversing tumor chemoresistance.
In this study, we determined the expression of miR-650 in LAD tissues, and found that miR-650 was upregulated in LAD tissues and correlated with advanced clinical stage and a higher incidence of lymph node metastasis in LAD patients. Also, high miR-650 expression was a poor prognostic factor for LAD patients. Furthermore, we have investigated the functional role for miR-650 in LAD cells, and showed that miR-650 could contribute to docetaxel chemoresistance of LAD cells by posttranscriptionally downregulating ING4.

Cell culture
Two human LAD cell lines (SPC-A1 and NCI-H1299) were purchased from the Tumor Cell Bank of Chinese Academy of Medical Science (Shanghai, China). The docetaxel-resistant SPC-A1 and NCI-H1299 cell lines (SPC-A1/DTX and H1299/ DTX) were established and preserved in our lab. All the cell lines were cultured in RPMI 1640 medium containing 10% fetal bovine serum and ampicillin and streptomycin at 37°C in a humidified atmosphere of 95% air and 5% CO 2 .

Tissue samples
A total of 96 primary LADs were collected from the Department of Cardiothoracic Surgery, Jinling Hospital between 2003 and 2006. The study was approved by the Ethic Committee of Jinling Hospital and it was performed in compliance with the Helsinki Declaration. All patients did not receive chemotherapy or radiotherapy prior to surgery. Patient characteristics are shown in Table 1. Disease histology was determined in accordance to the criteria of the World Health Organization. Pathologic staging was performed in accordance to the current International Union Against Cancer tumor-lymph node-metastasis classification. 26 randomly selected LAD tumors and their matched histologically normal lung parenchyma adjacent to the tumors (within 1 cm of the discrete tumor margin) were immediately frozen in liquid nitrogen and stored at -70°C until use. Also, 44 tissues were collected from patients with advanced LAD who received docetaxel-based chemotherapy. Patients met all of the following criteria: patients who suffered from primary LAD; a histological diagnosis of LAD with at least one measurable lesion; a clinical stage of IIIB~IV; first-line chemotherapy either with docetaxel 75 mg/m 2 and cisplatin 100 mg/m 2 or docetaxel 75 mg/m 2 and carboplatin AUC 6.0 mg/ml min administered every 3 weeks for a maximum of 6 cycles; and availability of sufficient tumor tissue in paraffin blocks for assessment by immunohistochemistry. Tumor response was examined by computed tomography and evaluated according to the Response Evaluation Criteria in Solid tumors (RECIST) as complete response (CR), partial response (PR), stable disease (SD), or progressive disease (PD). All tissue samples were snap-frozen in liquid nitrogen, which were transferred to 500 µl TRIzol solution (Invitrogen, CA, USA) immediately after harvesting in order to avoid mRNA degradation. Written informed consent was obtained from all patients according to the guidelines approved by the institutional research board.

Ethics statement
The study was approved by the Ethic Committee of Nanjing University and it was performed in compliance with the Helsinki Declaration. Written informed consent was obtained for all patient samples. All experimental animals were housed under specific pathogen-free conditions. All experimental procedures were approved by the Institutional Review Board of the Nanjing University. All procedures were performed in accordance with

Transfection of oligonucleotides or plasmids and stable selection
An miR-650 mimic (100 nM) was transfected into SPC-A1 or H1299 cells, and miR-650 inhibitor (anti-miR-650, 200nM) was transfected into SPC-A1/DTX or H1299/DTX cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instruction. All DNA plasmids (pLMP-miR-650 or pLMP-miR-NC) for transfection were extracted by DNA Midiprep or Midiprep kit (Qiagen). The SPC-A1 cells were transfected with those recombinant DNA vectors containing a puromycin selection marker and were selected with 5.0 µg/mL of puromycin (Sigma, USA) for 14 days. Then, single clones were isolated and expanded in media containing 2.5 µg/mL of puromycin.

RNA extraction and real-time quantitative RT-PCR (qRT-PCR) assay
Total RNA was extracted from the tissues or cultured cells using Trizol (Invitrogen, CA, USA) according to the manufacturer's protocol, reverse transcribed using Taqman™ microRNA reverse transcription kit and subjected to real-time PCR using TaqMan™ MicroRNA Assay kit (Applied Biosystems, USA) according to the manufacturer's instructions. Reactions were performed using Stratagene Mx3000 instrument in triplicate. MiRNA expression was normalized to U6 snRNA.

Western blot assay
Cell protein lysates were separated in 10% SDS polyacrylamide gels, electrophoretically transferred to polyvinylidene difluoride membranes (Roche), then detected with anti-cleaved caspase-3 or pro-caspase-3, ING4, Bcl-2 and Bax proteins. Protein loading was estimated using mouse anti-GAPDH monoclonal antibody. Lab Works TM Image Acquisition and Analysis Software (UVP, Upland, CA, USA) were used to quantify band intensities. Antibodies were purchased from Univ-bio Inc (Shanghai, China).

MTT assay
The cells were seeded into 96-well culture plates. After overnight incubation, cells treated with various concentrations of chemotherapeutic drugs. Following incubation for 24h, cell growth was measured following addition of 0.5 mg/ml MTT solution (Sigma). About 4 h later, the medium was replaced with 100µL dimethylsulfoxide (DMSO, Sigma) and vortexed for 10 min. Absorbance (A) was then recorded at 490 nm using a microplate reader (Bio-Rad, USA).

In vitro chemosensitivity assay
The single-cell suspensions were prepared and dispersed in 96-well plates. After incubation for 72 h with the docetaxel compounds (Sigma, Saint Louis, MO, USA), the MTT (Sigma, USA) solution (0.5 mg/ml) was added. Following incubation for 4 h, 100 µL of extraction buffer were added to each well. After an overnight incubation, absorbance at 490 nm was measured using using a microplate reader (Bio-Rad, Model 680).

Colony formation assay
The transfected LAD cells were placed in a fresh six-well plate and maintained in DMEM containing 10% FBS. 24h later, the medium was replaced with new medium. After 14 days, cells were fixed with methanol and stained with 0.1% crystal violet. Visible colonies were manually counted.

Detection of caspase-3 activity
The activity of caspase-3 was determined using the colorimetric CaspACE Assay System (Promega Corp., Madison, WI) following the manufacturer's instructions. Each determination was performed in triplicate.

Flow cytometric detection of apoptosis
An annexin V-fluorescein isothiocyanate (FITC) apoptosis detection kit (Oncogene Research Products, Boston, MA) was used to detect apoptosis according to the manufacturer's instructions. All of the samples were assayed in triplicate.

Hoechst staining assay
The cells were cultured on 6-well tissue culture plates to confluence, and Hoechst 33342 (Sigma, USA) was added to the culture medium of living cells; changes in nuclear morphology were detected by fluorescence microscopy using a filter for Hoechst 33342 (365 nm).

Detection of luciferase activity
Human docetaxel-resistant LAD cells (SPC-A1/DTX or H1299/DTX) grown in a 48-well plate were cotransfected with miRNA mimics/inhibitors, and pLUC firefly luciferase vectors containing empty, wild-type or mutant ING4 3'-UTR sequence using lipofectamine 2000 (Invitrogen, Carlsbad, CA). Cotransfected with two days post-transfection, the cells were collected and lysed for measurement of luciferase activities using the Dual-Luciferase Assay kit (Promega). Relative luciferase activity was calculated by normalizing the ratio of Firefly/Renilla luciferase to that of negative control-transfected cells.

In vivo chemosensitivity assay
Animal studies were performed according to institutional guidelines. The mock or stably transfected SPC-A1 cells (SPC-A1/pLMP-miR-650 and SPC-A1//pLMP-miR-NC) were suspended in 100 µL PBS and injected subcutaneously into the right side of the posterior flank of female BALB/c athymic nude mice (Department of comparative medicine, Jinling Hospital, Nanjing, China) at 5 to 6 weeks of age. When the average tumor size reached about 50 mm 3 , docetaxel was given through intraperitoneal injection with a concentration of 1.0 mg/kg, one dose every other day with 3 doses totally. Tumor growth was examined weekly for at least 5 wk. After 35 days, the mice were killed, necropsies were performed, and tumors were weighted. Tumor volumes were calculated by using the equation V (in mm 3 ) = A×B 2 /2, where A is the largest diameter, and B is the perpendicular diameter.

Bioinformatic and statistical analysis
The online miRNA databases (TargetScan, miRBase, and PicTarget) were used for prediction of miR-650 target genes. The SPSS10.0 program was used for general statistical and survival analysis. Experimental data were expressed as the mean±SEM. For comparison of means between two groups, a two-tailed t-test was used, and for comparison of means among three groups, one-way ANOVA was used The Spearman correlation test was used for analyses of primary tumors. Survival probabilities were determined using Kaplan-Meier analysis and the significance of difference was analyzed by a log-rank test. Significance was accepted at P<0.05.

MiR-650 was significantly upregulated in LAD tissues
To gain insight into the biological role of miR-650 in human LAD progression, we first detected the expression of miR-650 in 26 human LAD tissues and corresponding non-tumor tissues by qRT-PCR ( Figure 1A). Compared with corresponding nontumor tissues, 23 of 26 human LAD tissues (88.5%) showed that miR-650 was significantly upregulated 11.6-fold to 2.5-fold, but only 3 cases (11.5%) were downregulated in different degree. In general, the mean expression level of miR-650 in LAD tissues was significantly higher than that in corresponding non-tumor tissues (P<0.01; Figure 1B). Thus, these data indicated that upregulation of miR-650 might be, at least in part, involved in human LAD development.

Association of miR-650 expression with clinicopathological factors of LAD patients
To further analyze the association of miR-650 expression with clinicopathological factors of LAD patients, qRT-PCR assay was performed to detect miR-650 expression in a total of 96 LAD cases. Then, according to the median miR-650 expression level (median=8.45, normalized to U6 snRNA), LAD cases were classified into two expression groups: high-miR-650 expression group (n=52) and low-miR-650 expression group (n=44). By statistical analysis (Table 1), it was shown that the level of miR-650 expression was significantly correlated with lymph node metastasis and clinical stage of LAD patients (P=0.023 and 0.019, respectively), suggesting that high miR-650 expression group showed a higher incidence of lymph node metastasis (70.7%) and more advanced clinical stage (75.6%) than the low miR-60 expression group (51.2% and 40.0%, respectively). However, there were no significant differences between miR-650 expression and other clinicopathological facors, including gender, age, smoking condition, tumor differentiation and T-primary tumor (P=0.439, 0.861, 0.484, 0.502 and 0.851, respectively).

MiR-650 expression was an independent prognostic factor for LAD patients
To explore the association of miR-650 expression with the clinical outcome of LAD patients, the overall survival of LAD patients relating to miR-650 expression, status of tumor differentiation, lymph node metastasis and clinical stage was determined ( Figure 1C). LAD patients with high-miR-650 expression (n=52) showed a significantly worse prognosis than those patients with low-miR-650 expression (n=44). The median survival time for the LAD patients with high-miR-650 was 12.5 months (95% CI: 9.5-15.7 months), and for LAD patients with low-miR-650 was 28 To further investigate whether miR-650 expression might be a prognostic factor in LAD patients, univariate and multivariate data analyses were performed using the Cox proportional hazards regression model (Table 2). By univariate analysis, lymph node metastasis (N 1 +N 2 +N 3 /N 0 ), clinical stage (III+IV/I +II) and miR-650 expression (High/Low) were the prognostic factors to predict a poor prognosis (P=0.026, 0.012 and 0.008, MiR-650 in Lung Adenocarcinoma PLOS ONE | www.plosone.org respectively). By multivariate analysis of the prognostic factors, we confirmed that high miR-650 expression was an independent prognostic factor (unfavorable; risk ratio: 4.65; 95% confidence interval: 1.58-6.21; P=0.011), while lymph node metastasis was also an independent prognostic factor (P=0.028). These data strongly suggested that the miR-650 expression in LAD patients was closely related to a poor prognosis.

Expression of miR-650 was negatively correlated with responses of LAD patients to docetaxel
To explore the correlation between miR-650 expression and the efficiency of chemotherapy, we analyzed the survival data in patients who received docetaxel-based adjuvant chemotherapy (n=44). Tumors were divided into two groups: responding (CR+PR) and non-responding (SD+PD). Our data showed that the mean level of miR-650 expression in nonresponding tumor tissues (n=19) was significantly higher than that in responding tumor tissues (n=25) (P<0.001; Figure 2A). Then, ROC curve analysis was performed to establish the optimal cutoff value for the HSCORE of miR-650 expression level, which yielded a value of 143.5 (data not shown). Patients with a low level of miR-650 expression (HSCORE < 143.5) had a significantly longer progression-free survival (PFS) than did those with a high level of miR-650 expression (HSCORE ≥ 143.5) (P=0.0024; Figure 2B). Thus, it was concluded that the expression of miR-650 in advanced LAD tissues might be negatively correlated with the response of patients to docetaxel-based chemotherapy.

Downregulation of miR-650 reverses the resistance of docetaxel-resistant LAD cells to docetaxel
Previously, two docetaxel-resistant LAD cell lines (SPC-A1/DTX and H1299/DTX) were developed from the docetaxelnonresistant LAD cell lines (SPC-A1 and H1299) in our lab, respectively [14]. To further investigate whether miR-650 is an important regulator of acquired docetaxel resistance in LAD cells, qRT-PCR was performed to detect the expression of miR-650 in docetaxel-resistant and parental LAD cells. As shown in Figure 2C, the level of miR-650 expression in SPC-A1/DTX or H1299/DTX cells was significantly higher than that in parental SPC-A1 or H1299 cells (P<0.05). Next, we testify whether docetaxel treatment could affect the expression of miR-650 in LAD cells. The miR-650 levels after treatment with docetaxel (5.0 µg/L) were increased in the parental SPC-A1 or H1299 cells, but not in the SPC-A1/DTX or H1299/DTX cells ( Figure 2D).
In order to examine the correlation between the upregulation of miR-650 and the resistance of LAD cells to docetaxel, miR-650 inhibitor was transfected into docetaxel-resistant LAD cells. 48h after transfection, qRT-PCR assay showed that the expression of miR-650 in anti-miR-650-transfected SPC-A1/DTX or H1299/DTX cells was significantly decreased compared with that in anti-miR-NC-transected cells (P<0.05; Figure 3A). MTT assay was performed to detect the effect of miR-650 expression on the growth of docetaxel-resistant LAD cells. Compared with anti-miR-NC, anti-miR-650 could induce the decreased growth of SPC-A1/DTX or H1299/DTX cells in a time-dependent manner ( Figure 3B). Meanwhile, anti-miR-650 could lead to the decreased colony formation capacity of SPC-A1/DTX or H1299/DTX cells (P<0.05; Figure 3C). Next, we analyzed the changes of docetaxel sensitivity of chemoresistant LAD cells induced by anti-miR-650. Compared with anti-miR-NC-transfected cells, the IC 50 -value of docetaxel in anti-miR-650-transfected SPC-A1/DTX or H1299/DTX cells was significantly decreased by approximately 40.2% or 43.2%, respectively (P<0.05; Figure 3D). Thus, downregulation of miR-650 could reverse the resistance of docetaxel-resistant LAD cells to docetaxel by inhibiting cell growth.

Upregulation of miR-650 reduces the in vitro sensitivity of parental LAD cells to docetaxel
Next, we investigated whether upregulation of miR-650 could affect the sensitivity of LAD cells to docetaxel. Then, miR-650 mimics was transfected into parental SPC-A1 or H1299 cells. 48h after transfection, qRT-PCR assay showed that the expression of miR-650 in miR-650 mimics-transfected SPC-A1/DTX or H1299/DTX cells could be significantly increased (P<0.01; Figure 4A). Results of MTT assay indicated that miR-650 mimics could induce the increased growth ability of parental SPC-A1 or H1299 cells compared with miR-NC mimics ( Figure 4B). Likewise, the ability of colony formation in SPC-A1 or H1299 cells transfected with miR-650 mimics was increased compared with cells transfected with miR-NC mimics (P<0.05; Figure 4C). More importantly, compared with miR-NC mimics-transfected cellsh, the IC 50 -value of docetaxel in miR-650 mimics-transfected SPC-A1 or H1299 cells could be increased by approximately 72.3% or 41.1%, respectively ( Figure 4D). These data suggested that upregulation of miR-650 might be an important inducer of docetaxel resistance of LAD cells.

ING4 was a functional target of miR-650 in docetaxelresistant LAD cells
Previously, we have reported that downregulation of inhibitor of growth 4 (ING4) was involved in docetaxel resistance of LAD cells [14]. However, the molecular mechanisms involved in downregulation of ING4 remain unclear. As miRNAs function mainly through the inhibition of target genes, the targets of miR-650 that function in docetaxel-resistant LAD cells were further investigated. The targets of miR-650 were predicted through at least three databases (Pictarget, miRBas and TargetScan), and ING4 was selected as a putative target, which was also recently reported in gastric cancer  downregulated compared with that in parental SPC-A1 or H1299 cells ( Figure 5A). Then, to further confirm target specificity between miR-650 and ING4 in LAD cells, we performed luciferase reporter assay with a vector containing the putative ING4 3'-UTR target site downstream of the luciferase reporter gene, which was transfected into LAD cells. Base pairing between miR-650 and wild-type (wt) or mutant (mut) target site in the 3'-UTR of ING4 mRNA was shown in Figure 5B. Luciferase activities of LAD cells transfected with ING4-wt construct were significantly lower after transfection of miR-650 mimics and were significantly higher after transfection of anti-miR-650, whereas those with ING4-mut construct showed no significant difference ( Figure 5C). Next, we analyzed the effect of miR-650 on ING4 protein expression, and found that miR-650 mimics could lead to the decreased expression of ING4 protein in SPC-A1/DTX or H1299/DTX cells, while anti-miR-650 could induce the increased expression of ING4 protein ( Figure 5D).
In our previous study, we have reported that upregulation of ING4 could reverse the docetaxel resistance of docetaxelresistant LAD cells, suggesting that upregulation of ING4 could mimic the effects of anti-miR-650 on the chemosensitivity of docetaxel-resistant LAD cells to docetaxel [15]. To further testify the roles of ING4 in miR-650-induced chemoresistance of LAD cells, siRNA/ING4 was transfected into anti-miR-650transfected SPC-A1/DTX cells. It was found that siRNA/ING4 could reverse the increased ING4 protein expression in SPC-A1/DTX cells induced by anti-miR-650 ( Figure 6A). Also, siRNA/ING4 could partially rescue the growth inhibition of SPC-A1/DTX cells induced by anti-miR-650 ( Figure 6B). Meanwhile, siRNA/ING4 could also partially rescue the decreased IC 50value of docetaxel in SPC-A1/DTX cells induced by anti-miR-650 ( Figure 6C). Next, pcDNA/ING4 vector was transfected into miR-650 mimics-transfected SPC-A1 cells. Results indicated that pcDNA/ING4 could reverse the decreased ING4 protein expression in SPC-A1 cells induced by miR-650 mimics ( Figure 6D). Likewise, pcDNA/ING4 could partially rescue the growth promotion and increased IC 50 -value of docetaxel in SPC-A1 cells induced by miR-650 mimics ( Figure 6E-F). Collectively, these findings further suggest that ING4 is a functional and direct target gene for miR-650 in LAD cells.

MiR-650 and ING4 affects the apoptosis of docetaxelresistant LAD cells by way of regulating Bcl-2/Bax expression
As miR-650 could reverse the resistance of docetaxelresistant LAD cells to docetaxel and we confirmed that ING4 was a direct and functional target, we investigated the molecular mechanisms by which miR-650 and ING4 mediated the regulation of chemosensitivity of LAD cells. First, we analyzed the effect of miR-650 expression on apoptosis of SPC-A1/DTX cells, and showed that anti-miR-650 could induce the increased apoptosis of SPC-A1/DTX ( Figure 7A). Also, using the Hoechst staining assay, condensed and fragmented nuclei in anti-miR-650-transfected SPC-A1/DTX cells were obviously observed ( Figure 7B). In previous study, we have showed that upregulation of ING4 could significantly increase the apoptosis of docetaxel-resistant LAD cells, suggesting that downregulation of miR-650 could lead to the same apoptosisinducing effect as upregulation of ING4. Also, we found that both anti-miR-650 and pcDNA/ING4 could increase the expression of cleaved caspase-3 in SPC-A1/DTX cells but had no effects on the expression of pro-caspase-3 protein ( Figure  7C). Meanwhile, compared with anti-miR-NC or pcDNA/controltransfected cells, the activity of caspase-3 in anti-miR-650 or pcDNA/ING4-transfected SPC-A1/DTX cells was increased by 152.3% or 185.8%, respectively ( Figure 7D). Further, Western blot assay confirmed the decreased expression of Bcl-2 protein and the increased expression of Bax protein in anti-miR-650 or  Figure 7E), which led to a decrease in the cell antiapoptotic ability (Bcl-2/Bax ratio).
To further investigate whether ING4 was involved in the effect of miR-650 on apoptosis in SPC-A1/DTX cells, we then performed rescue experiments. After transfected with anti-miR-650, SPC-A1/DTX cells were co-transfected with siRNA/ ING4. Flow cytometric analysis showed that the cotramsfection could partially rescue the increased apoptosis in SPC-A1/DTX cells induced by anti-miR-650 ( Figure S1A). Also, the co-tramsfection could partially rescue the increased expression of cleaved caspase-3 protein and the enhanced activity of caspase-3 in SPC-A1/DTX cells induced by anti-miR-650 ( Figure S1B-C). Furthermore, the co-tramsfection could partially rescue the decreased expression of Bcl-2 protein and the increased expression of Bax protein in SPC-A1/DTX cells induced by anti-miR-650 ( Figure S1D). These data indicated that upregulation of ING4 was potentially involved in the apoptosis-promoting function of anti-miR-650, which could induce the caspase-3-dependent apoptosis by way of regulating the expression of Bcl-2/Bax.

Upregulation of miR-650 reduces the in vivo sensitivity of parental LAD cells to docetaxel
To further investigate the possible effect of miR-650 expression on LAD cell sensitivity to docetaxel in vivo, SPC-A1 cells were stably transfected with pLMP-miR-650 or control pLMP-miR-NC vector. As shown in Figure 8A, the expression  Figure  8B). Also, the IC 50 -value of docetaxel in SPC-A1/pLMP-miR-650 cells was modestly increased by approximately 53.4% ( Figure 8C). Next, s.c. tumors were induced in nude mice followed by treatment with docetaxel. Following docetaxel treatments, the tumors formed from SPC-A1/pLMP-miR-650 cells grew faster than those formed from SPC-A1/pLMP-miR-NC cells (P<0.05; Figure 8D). These data support the notion that miR-650 could induced the increased docetaxel resistance of LAD cells in vivo.

ING4 is often downregulated in docetaxel-resistant LAD tissues and negatively correlated with miR-650 expression
Because ING4 is the target of miR-650, immunohistochemistry was performed to detect the expression of ING4 in LAD tissues collected from patients who received docetaxel-based chemotherapy (n=44). As shown in Figure 9A, ING4 expression was higher in the responding than in the nonresponding LAD tissues. To further investigate the correlation of ING4 expression with the responses of LAD patients to docetaxel, Western blot assays were performed to detect the expression of ING4 protein in those tissues. In contrast to miR-650, the relative expression level of ING4 protein in responding tumor tissues was significantly higher than that in non-responding tumor tissues (P=0.022; Figure 9B). Further, the inverse correlation between miR-650 and ING4 protein expression was verified by linear regression analysis (Spearman rank test r=-0.645; P=0.018) ( Figure 9C).

Discussion
Increasing evidence suggests that in addition to proteinencoding genes, miRNAs have more than a cursory role in the pathogenesis of human cancer development [16,17]. Thus, miRNA expression signatures seem to hold great promise in tumor characterization and could be used as potential diagnostic and prognostic markers for tumor diagnosis and therapy [18]. The key finding of this study is that upregulation of miR-650 expression is correlated with LAD development and progression. Statistical analyses indicate that LAD patients with high expression of miR-650 have a poorer prognosis. Also, our results reveal that upregulation of miR-650 could induce docetaxel resistance of LAD cells via regulating Bcl-2/Bax expression by targeting tumor suppressor ING4. These data suggest that miR-650 will be a significant predictor of poor prognosis for LAD patients and be a potential target for chemosensitizing LADs.
Several recent studies have demonstrated that many miRNAs are dysregulated in human LADs. Jiang, et al conducted miRNA expression profiling in matched lung adenocarcinoma and uninvolved lung using 56 pairs of freshfrozen and 47 pairs of formalin-fixed, paraffin-embedded samples from never smokers and showed that increased miR-708 expression in NSCLC and its association with poor survival in LAD from never smokers [19]. Rothschild and colleagues showed that both miR-381 and miR-29b repressed ID1 and were dysregulated in LAD. Kaduthanam, et al identified microRNAs (miRNAs) in serum of patients associated with early relapse in pulmonary adenocarcinoma by using qRT-PCR based low-density arrays (664 miRNAs) and showed that serum miR-142-3p was associated with early relapse in operable LAD patients [20]. Also, Patnaik and colleagues identified microRNA expression profiles of whole blood in LAD, and showed that microRNAs miR-190b, miR-630, miR-942, and miR-1284 were the most frequent constituents of the classifiers [21]. In our previous study, we have reported that miR-451 could function as a tumor suppressor not only in squamous cell carcinoma but also in lung adenocarcinoma [22]. These findings imply that miRNAs are involved in LAD development and progression and could serve as potential diagnostic markers and therapeutic targets. In the present study, we found that the average level of miR-650 expression was significantly higher in LAD tissues as compared with paired adjacent nontumor tissues. Results from quantitative analysis of qRT-PCR indicated that high level of miR-650 expression was closely correlated with advanced clinical stage and a high incidence of lymph node metastasis in LAD patients. Also, LAD patients with high-level miR-650 expression had a shorter median overall survival (12.5 months) than those with low-level miR-650 expression (28.3 months). These data, together with further univariate and multivariate analyses, suggest that high expression of miR-650 might be a significant predictor of poor prognosis for LAD patients. More importantly, further analyses showed that the expression of miR-650 was correlated with the efficiency of chemotherapy. It was found that the mean level of miR-650 expression in non-responding tumor tissues was significantly higher than that in responding tumor tissues. Additionally, the PFS of LAD patients with highlevel miR-650 who received docetaxel-based chemotherapy was significantly lower than that of patients with low-level miR-650. Thus, it was concluded that overexpression of miR-650 might be involved in docetaxel-resistant phenotypes of human LADs.
Recently, the abberant expression of miR-650 has been reported to be correlated with the aggressiveness of colorectal cancer, gastric cancer, and chronic lymphocytic leukemia cells [23,24]. Feng and colleagues reported that downregulation of N-myc downstream-regulated gene 2 (NDRG2) expression involving promoter methylation and microRNA-650 promoted colorectal cancer progression [23]. In gastric cancer, Zhang' et al reported that microRNA-650 could target ING4 to promote gastric cancer tumorigenicity [25]. Also, Mraz and colleagues showed that microRNA-650 expression is influenced by immunoglobulin gene rearrangement and affects the biology of chronic lymphocytic leukemia (CLL) [24]. In their studies, higher expression of miR-650 was found to be associated with a favorable CLL prognosis and influenced the proliferation capacity of B cells. However, there were no reports about the association of miR-650 expression with the chemoresistance of tumor cells. Previously, the multidrug-resistant LAD cell lines (SPC-A1/DTX and H1299/DTX) were established from human LAD cell lines (SPC-A1 and H1299) in our lab by stepwise selection using docetaxel as inducing reagent. Furthermore, a subset of miRNAs was found to be differentially expressed in SPC-A1/DTX cells and its parental SPC-A1 cells [26]. Our further research data showed that miR-200b or miR-100 could modulate the resistance of LAD cells to docetaxel by targeting E2F3 or Plk1 [27,28]. In the present study, we firstly showed that miR-650 was significantly upregulated in docetaxelresistant LAD cells (SPC-A1/DTX and H1299/DTX) compared with in parental LAD cells (SPC-A1 and H1299). By modulating the miR-650 level in LAD cells, we revealed that miR-650 could mediate the docetaxel resistance of LAD cells both in vitro and in vivo. Further, we conducted luciferase reporter assay and Western blotting to confirm that ING4 was a direct and functional target of miR-650 in LAD cells. ING4, a novel member of ING family comprising six members characterized by a highly conserved C-terminal plant homedomain (PHD)-like zinc-finger domain, has been reported to be involved in a variety of human biological and pathological processes including oncogenesis, angiogenesis, and DNA repair [29,30]. The downregulation of ING4 has been found in many human cancers including gastric cancer, ovarian carcinoma, head and neck squamous cell carcinoma and lung cancer [31][32][33][34]. In our previous study, a cDNA microarray analysis was performed to compare differential gene expression profile between docetaxel-resistant and parental LAD cell lines, and further functional data suggest that ING4 could modulate the sensitivity of LAD cells to taxanes by regulating apoptosis-related proteins. It was observed that the effects of ING4 overexpression on the chemoresistance of docetaxl-resistant LAD cells were similar to the effects of miR-650 inhibitor. Furthermore, silencing of ING4 could partially rescue the effects of miR-650 inhibitor on the chemosensitivity of docetaxel-resistant LAD cells. We then further elucidated the possible mechanisms of chemosensitivity enhancement in LAD cells induced by miR-650 inhibitor or ING4 overexpression. Results from flow cytometry and Hoechst staining assays indicated that miR-650 inhibitor could induce the increased capase-3-dependent apoptosis. Previous researches have shown that ING4 could function as tumor suppressor in human cancers by regulating the Bcl-2 family proteins. In this study, it was found that docetaxel-resistant LAD cells transfected with miR-650 inhibitor presented with decreased expression of Bcl-2 protein and increased expression of Bax protein, and the decreased Bcl-2/Bax ratio led to the execution of apoptotic progression in docetaxel-resistant LAD cells. Likewise, silencing of ING4 could partially rescue the increased caspase-3-dependent apoptosis and decreased Bcl-2/Bax ratio induced by miR-650 inhibitor. At the same time, the expression of ING4 protein was negatively correlated with miR-650 expression in responding or non-responding LAD tissues. Our findings demonstrated that miR-650 confers the docetaxel chemoresistance of lung adenocarcinoma cells via regulating Bcl-2/Bax expression by targeting ING4. Of course, many studies have indicated that one miRNA could regulate multiple target genes, whereas one gene could also be regulated by multiple miRNAs. Therefore, miR-650 might regulate other target genes which are involved in formation of chemoresistant types in LAD cells, which needs to be further elucidated in future. This study showed that high level of miR-650 expression was correlated with enhanced malignant potential and poor prognosis of LAD patients. Furthermore, miR-650 could affect the chemosensitivity of LAD cells to docetaxel via regulating Bcl-2/Bax expression by directly targeting ING4. Since the size of tissue sample in the present study is small, further investigation of a larger patient population will be necessary to confirm the association of miR-650 and its target ING4 with the responses of LAD patients to docetaxel-based chemotherapy.