Long non-coding RNA HOTAIR promotes tumorigenesis and forecasts a poor prognosis in cholangiocarcinoma

Cholangiocarcinoma (CCA) arising from the neoplastic transformation of cholangiocytes with increasing incidence in the worldwide. Unfortunately, a large amount of CCA patients lost their chance for surgery because it is hard to diagnose in the early stages. Long non-coding RNAs (lncRNAs) is closely associated with development and progression of various malignant tumors. Hox transcript antisense intergenic (HOTAIR), a negative prognostic factor for patients with gastric, liver and pancreatic carcinoma. Its transcription levels and functional roles in CCA is still unknown. Therefore, we aimed to explore the effect of HOTAIR in CCA including cell proliferation, apoptosis, migration, invasion and epithelial-to-mesenchymal transition (EMT). The results showed that HOTAIR was highly expressed both in CCA tissue samples and cell lines compared with corresponding normal bile duct tissues and Human intrahepatic biliary epithelial cells (HIBEC). Its overexpression was closely correlated with Tumor size, TNM stage and postoperative recurrence in CCA patients. Moreover, up-regulation of HOTAIR has correlation with prognosis in CCA patients. Knockdown of HOTAIR by siRNAs significantly decreased the migration and invasion but increased apoptosis of CCA cells in vitro. Overall, our study revealed that HOTAIR may play as a new potential therapeutic target and forecast poor prognosis for this fatal disease.

indicated that HOTAIR plays imperative roles in the modulation of cancer progression. Nevertheless, little is known about the impact of HOTAIR on CCA carcinogenesis or metastasis.
In order to understand the functional role of HOTAIR in CCA development and progression, in our study, we investigated the expression levels of HOTAIR in CCA tissues and analyzed its connection to clinical and pathological characteristics. In addition, we investigated the crucial roles of HOTAIR in CCA by analyzing proliferation, apoptosis, migration, invasion and epithelial-mesenchymal transition (EMT) after silencing of HOTAIR in vitro.

Results
HOTAIR is overexpressed in human CCA cell lines and tissues. In order to elucidate the expression levels of HOTAIR in CCA, 70 pairs of tissues and diverse CCA cell lines including HuCCT1, CCLP-1, Huh-28, QBC939 and RBE were investigated by qRT-PCR experiments. The experimental results showed that the relative expression of HOTAIR in CCA specimens was obviously superior than that of the corresponding normal tissues (Fig. 1A). Besides, the expression of HOTAIR was up-regulated in CCA cell lines compared to HIBEC cells (Fig. 1B). The fact above indicated that HOTAIR might play key roles in tumorigenesis and progression in CCA.
Up-regulated HOTAIR is negatively correlated with prognosis. 70 patients were divided into high HOTAIR expression group or low expression group. Clinicopathological factors were analyzed between the high and low HOTAIR expression groups. As a result, the high HOTAIR expression group showed higher postoperative recurrence (p = 0.020), larger tumor size (p = 0.028) and more advanced TNM stage (p = 0.021) than the low HOTAIR expression group. However, patients' gender, age, lymph node invasion and other essential information were not related to HOTAIR (Table 1). Moreover, for the overall survival rate after radical surgery, patients with high HOTAIR expression group had a significantly poorer prognosis than those with low HOTAIR expression patients (p = 0.004; Fig. 1C). The univariate Cox regression analyses of overall survival showed that the lymph node invasion (P = 0.034), TNM stage (P = 0.023), postoperative recurrence (P = 0.011) and HOTAIR expression (P = 0.006) were good prognostic indicators. Furthermore, multivariate Cox regression analyses confirmed the expression of HOTAIR as an independent predictor for overall survival in patients with CCA (P = 0.039, Table 2).
Overexpression of HOTAIR stimulates CCA cell proliferation. In order to uncover the functional roles of HOTAIR in CCA in vitro, we designed two siRNAs with specifically targeting HOTAIR and an isotype control (si-NC) for transfection. We use the inverted fluorescence microscopy and flow cytometry to detect transfection efficiency. As for the two selected cell lines (RBE and QBC939), transfection efficiency reached 80-90% after transfection ( Fig. 2A,B). Next, qRT-PCR was used to analyze the levels of HOTAIR expression in the two cell lines and as shown in Fig. 2C, the HOTAIR expression was significantly reduced after transfected with si-HOTAIRs. Cell Counting Kit-8 (CCK-8) was used to detect cell proliferation to assess the functional character of HOTAIR in CCA according to the instruction. The experimental data showed that the growth of QBC939 and RBE cells were both inhibited after knockdown of HOTAIR (Fig. 2D). And is similar with this should of is, colony formation assays showed that the clonogenic ability of two cell lines were significantly decreased after transfected with si-HOTAIR-1 or si-HOTAIR-2 (Fig. 2E). Western blot assays showed that the expression levels of proliferating cell nuclear antigen (PCNA) decreased significantly after knockdown of HOTAIR (Fig. 2F).
The impact of si-HOTAIR in promoting CCA cell apoptosis. In this study, we introduced acridine orange/ethidium bromide (AO/EB) double fluorescence staining, flow cytometry analysis, TdT-mediated dUTP Nick-End Labeling (TUNEL), caspase-3 and caspase-9 to explore the potential effect of HOTAIR on apoptosis of CCA cells. After transfected with si-HOTAIRs, the early and late apoptotic cells were both increased, nevertheless, most of cells were alive in the si-NC groups ( Fig. 3A-C). Previous studies suggested that Bcl-2 family proteins could play an important role in cell growth and death. In our study, the expression of Bax protein were activated. However, the protein level of Bcl-2 were suppressed proved by Western blot analysis (Fig. 3D). Previous research had revealed that Bcl-2 family proteins act could regulators of cell life and death. But beyond that caspase-3, caspase-9, Bcl-2 and Bax are closely correlated with mitochondrial pathway mediated apoptosis. In our study, the expression of caspase-3 and caspase-9 were both activated after HOTAIR silenced (Fig. 3E).

Knockdown of HOTAIR decreased the migration and invasion capacities of CCA cells. In order
to verify whether HOTAIR affects CCA cell migration and invasion, Transwell and wound healing assays were used after transfected with si-NC or si-HOTAIRs in QBC939 and RBE. Interestingly, as shown in Fig. 4A-D, after knockdown of HOTAIR, the migratory and invasion potentials were impeded compared with the si-NC groups. The results above indicated that down-regulation of HOTAIR distinctly suppressed RBE and QBC939 cell metastasis in vitro.
Silenced HOTAIR affects EMT in CCA cells. EMT is an important mechanism related to cell metastasis, its mark including E-cadherin, N-cadherin and Vimentin are closely related to the epithelial cell invasion capacity. Therefore, Western blot were performed to investigate the relationship between dysregulation of HOTAIR and EMT. According to the Western blot results, the expression levels of N-cadherin and Vimentin were lower in QBC939 and RBE cells after HOTAIR was silenced, however, E-cadherin was significantly up-regulated (Fig. 5A).

HOTAIR promoted CCA tumor formation and growth in vivo.
To confirm whether the HOTAIR can effect tumorigenesis in vivo, shHOTAIR QBC939 cells and control cells were subcutaneously injected into BALB/c separately. As is shown in Fig. 6A suppressed the growth of CCA xenografts. The average tumor weight in the shCtrl group was significantly higher than that in the shHOTAIR group (Fig. 6C). what's more, immunohistochemical staining indicated a decreased proliferative index Ki67 expression in shHOTAIR group (Fig. 6D). The qRT-PCR results showed lower expression levels of the HOTAIR transcription in shHOTAIR group (Fig. 6E).

Discussion
CCA is one of the most malignant tumors till now. Unfortunately, the incidence of CCA has increased significantly over the past few decades 1,21 . Though medical technology may improve the outcome of CCA patients, CCA still threatens people's health due to its late presentation and lack of clinical treatments 22 . Given the loss of life, it is imperative to find new preventive strategies and reliable methods to treat the patients with CCA. Recently, the virtue of the lncRNAs has been well recognized, but the significant function also unwitnessed in relatively long term especially for variety of malignant tumors. Several lncRNAs have been confirmed to play an important role in various types of malignant tumors, including PANDAR, ZFAS1, HOTAIR, TUG1 and so on [23][24][25][26] . Emerging evidence indicated that in the process of tumor progression, lncRNA HOTAIR plays an imperative role 27,28 . However, its clinical significance and expression levels in CCA have not been studied yet. HOTAIR, which originates from the HOXC cluster onchromosome, was the first lncRNA ever described to interact with polycomb proteins and suppress gene transcription. The high-expression of HOTAIR has been proved to play an imperative role in tumorigenesis and progression. Studies pointed out that lncRNA HOTAIR is overexpressed in breast cancer tissues compared with adjacent normal counterparts. Up-regulation of lncRNA HOTAIR was found in many other cancer tissues such as gallbladder cancer, pancreatic cancer, hepatocellular carcinoma, and nasopharyngeal carcinomas compared with their normal tissues 12,29-31 . These findings imply a potential role of HOTAIR in the development and progression of human cancers.
Recent studies have indicated that, in colon cancer, lncRNA loc285194 is a tumor-suppressive lncRNA through downregulating miRNA-211 32 . In gallbladder cancer, Ma et al. demonstrated that HOTAIR and miRNA-130a have a negative correlation. They found that the expression level of miRNA-130a is upregulated after HOTAIR silenced. In addition, the study found that knockdown of HOTAIR inhibited the invasiveness and proliferation of gallbladder cancer cells, miRNA-130a inhibitor reversed the effects of knockdown of HOTAIR exerted 12 20,33,34 . Consistent with the above research, our study confirmed that HOTAIR was overexpressed in CCA tissue samples compared with corresponding normal tissue samples. HOTAIR may be an oncogene and a negative prognostic factor in patients with CCA because high expression of HOTAIR was closely associated with the advanced TNM stage, larger tumor size, postoperative relapse and dismal 5-year overall survival.
CCK-8 and colony formation assays were used to determine the proliferation of CCA cells after suppressing HOTAIR expression. The results suggested that CCA cell multiplication capacities were affected by HOTAIR. PCNA is an essential factor in DNA replication, DNA repair, chromatin assembly and RNA transcription. Lots of researches have shown that amounts of PCNA were detected in normal proliferating cells as well as in transformed cells and tumors 35 . In our study, the results of Western blot experiments revealed that the expression of PCNA significantly decreased after knockdown of HOTAIR in CCA. Previous studies suggested that Bcl-2 family proteins could regulate cell growth and death. Bcl-2 and Bax could regulate tumor cell growth by regulating cell apoptosis rather than proliferation. In fact, it has been observed that overexpression of Bcl-2 or down-regulation of Bax can inhibit many tumor cell apoptosis, while down regulation of Bcl-2 or up-regulation of Bax can increase the apoptosis rate of tumor cells. What's more, caspase-3 and caspase-9 play pivotal roles in the regulation of cell apoptosis among the 14 mammalian caspases identified to date. In addition, Receptor mediated pathway and mitochondrial pathway are related to the regulation of Bax, Bcl-2, caspase-3 and caspase-9. Moreover, in our study Bax was upregulated after knockdown of HOTAIR, on the other hand, Bcl-2 with low expression levels, this is probably an indirect argument that HOTAIR promotes tumor development and progression in CCA [36][37][38][39][40] .
Other experiments including flow cytometric analysis and AO/EB staining assays were also determined that as the HOTAIR expression decreased, the rate of apoptosis increased. Transwell and scratch wound assays showed that knockdown of HOTAIR with siRNAs significantly reduced the migration and invasion of QBC939 and RBE cells. EMT has been proved to play an important role in tumor invasiveness, metastasis and prognosis 41,42 . Recent researches indicated that N-cadherin, Vimentin and Fibronectin are upregulated, however, E-cadherin and Claudins are down-regulated. It has been reported that it is an essential step in the transformation between E-cadherin and N-cadherin to manipulate cell metastasis in CCA 43 . Knockdown of HOTAIR could reverse EMT progress and then impair migratory and invasive potential of QBC939 and RBE cells proved by Western blot analysis. However, there are still some limitations to come to an incontrovertible conclusion in our research. A larger cohort of patients should be recruited to further validate the clinical value of HOTAIR in CCA. The mechanisms underlying the regulatory actions of HOTAIR should be further investigated. In conclusion, these studies indicated that HOTAIR was remarkedly upregulated in CCA tissues and cell lines compared with normal tissues and HIBEC. Our findings identified an oncogenetic role of HOTAIR in the tumorigenesis and development of CCA both in vitro and in vivo. Upregulated HOTAIR may be a negative prognostic factor for CCA patients. Besides, HOTAIR may become a novel therapeutic target for the treatment of CCA.  RNA extraction and quantitative real-time PCR. Trizol reagent (Thermo Fisher Scientific, Waltham, MA, USA) was used to extract total RNA from tissues or cultured cells. qRT-PCR assays were performed by using FastStart Universal SYBR Green Master (Roche, Germany) in a BIO-RAD C1000 Thermal Cycler, and total RNA was subjected to cDNA by Transcriptor First Strand cDNA. GAPDH was selected as the negative control. The primers used for GAPDH and HOTAIR were as follows: HOTAIR Forward, 5′-GGGAGCCAAAAGGGTCAT-3′ and Reverse, 5′-GAGTCCTTCCACGATACCAA-3′; GAPDH Forward, 5′-GGGAGCCAAAAGGGTCAT -3′ and Reverse, 5′-GAGTCCTTCCACGATACCAA -3′. siRNAs and transfection. RBE and QBC939 were selected for the knockdown study, on the basis of the expression of HOTAIR in CCA cell lines. We cultured cells in serum-free medium and allowed them to grow to half confluence prior to transfected with si-HOTAIR using Lipofectamine 3000 transfection reagent (Thermo Fisher Scientific) for 48 h. The target sequences for si-HOTAIR are as follows: si-HOTAIR-1 sense 5′-UUCUAAAUCCGUUCCAUUCCACUGCGA-3′; antisense 5′-GCAGUGGAAUGGAACGGAUUUAGAA-3′; si-HOTAIR-2 sense 5′-AGCGAACCACGCAGAGAAAUGCAGG-3′; antisense 5′-CCUGCAUUUCUCUGCGUGGUUCGCUUU-3′. The sequence for negative control FAM is: sense 5′-UCUCCGAACGUGUCACGUTT-3′; antisense 5′-GUGACACGUUCGGAGAATT-3′.
Proliferation assays. For CCK-8 analysis, in 96-well plates, seeding 1500 transfected cells/well after QBC939 and RBE cells were transfected with si-HOTAIR or si-NC. Using CCK-8 (Dojindo, Tokyo, Japan) to detect viability at following time (0, 24, 48, 72 and 96 h), added 10 uL of CCK-8 into the corresponding wells. A microplate reader (Tecan, Männedorf, Switzerland) was used to analyze the absorbance at 450 nm after incubated at 37 °C for 2 h. For the clonogenic assay, QBC939 and RBE cells were trypsinized into a single-cell suspension and plated in a 3.5-cm dish at a total of 500 cells per well. The cells were maintained in an incubator for approximately 2 weeks until there were visible colonies.
Flow cytometry for cell apoptosis. Collecting QBC939 and RBE cells after transfected with si-NC or si-HOTAIR and washed twice with cold PBS. Binding buffer was used to re-suspend the cells. After staining with 5 μL FITC-Annexin V and 5 μl Propidium iodide (PI) using FITC Annexin V Apoptosis Detection Kit (BD, Biosciences, USA), the stained cells were measured by flow cytometry (FACScan; BD Biosciences, USA).
Acridine orange/ethidium bromide (AO/EB) double fluorescence staining. The exponential growth phase cells were cultured in an incubator of 5% CO 2 at 37 °C and transfected with si-HOTAIR or si-NC, and then, stained with prepared AO/EB mixing solution for 5 min (Solarbio, Beijing, China). Because of different abilities to penetrate the cell membrane, AO/EB could tell live cells from apoptotic cells. Apoptotic cells DNA were dyed red or orange while the normal cells with green fluorescence. At last, the fluorescence microscope (Leica, Germany) was used to take photographs and determined the apoptotic cells.
TdT-mediated dUTP Nick-End Labeling (TUNEL) assay. To explore the apoptotic in QBC939 and RBE cells, apoptosis was examined using One Step TUNEL Apoptosis Assay Kit (Beyotime, Beijing, China). Transfected cells were washed by PBS after fixed with paraformaldehyde for 30 min. Then, the cells were incubated by Immunostaining Permeabilization Buffer with Triton X-100 (Beyotime, Beijing, China). Afterwards, QBC939 and RBE cells transfected with si-NC or si-HOTAIR were incubated at 37 °C for 60 min after treated with 45 μl fluorescent labeled reagent and 5 μl terminal deoxynucleotidyl transferase (TdT). A fluorescent microscope (Leica, Germany) was used to detect the apoptotic cells.

Relative caspase activity determination. After post-transfection 48 h by using Caspase-3 Activity
Kit and Caspase-9 Activity Kit (Solarbio, Beijing, China). The relative of caspase-3 and caspase-9 activity were detected. Add cell proteins into 96-well plates with reaction buffer and substrate and then maintained at 37 °C for 4 h in the dark before quantified at a wavelength of 405 nm by a microplate reader (Tecan, Männedorf, Switzerland). Scratch wound assay. RBE and QBC939 cells were transfected by si-HOTAIR or si-NC and separately planted in 3.5 cm dishes until reaching about 80% confluence. Using phosphate buffer solution (PBS) to clean up cellular debris after mark out a straight wounds by a sterile 200-μL pipette tip. Migration distance were taken photographs after 0 h and 36 h.
Migration and invasion assays. Using transwell chambers with 8-μm polycarbonate nucleopore filters (Corning, NY, USA) to determine the alteration of migration and invasion capacities. 5 × 10 4 transfected cells were seeded in 200 ul serum-free medium loaded in the top section of a 24-well transwell. The bottom was loaded 600 ul complete medium. After the transwell chambers were incubated for 24 h, crystal violet was used to stain the cells on the lower surface. For invasion, first, put the Transwell chamber unit in 37 °C for 4 h after coated with 40 μl Matrigel (BD Biosciences, San Jose, CA, USA) in the top of Transwell chamber unit to reconstruct basement membrane. And then according to the methods of the migration assay.
Western blot analysis. RIPA buffer with protease inhibitors (Beyotime, Beijing, China) were used to lysed QBC939 and RBE cells at 48 h post-transfection. Sodium dodecyl sulfate polyacrylamide gel (SDS-PAGE) vertical electrophoresis were used to fractionated the extracted proteins. The proteins were transfered to a 0.45 μm polyvinylidene fluoride (PVDF) membrane (GE Healthcare, Piscataway, NJ, USA). Afterwards, 5% defatted milk diluted in Tris-buffered saline containing 0.05% Tween-20. Primary antibodies include GAPDH, E-cadherin, N-cadherin and Vimentin were used to incubate the membranes overnight. Then, the membranes were incubated in secondary antibody (Cell Signaling Technology, Danvers, USA) for 2 h before using eyoECL Plus Kit (Beyotime, Beijing, China) to visualize the blots.
Tumor xenografts study. All animal studies were performed under the supervision and guidelines of the Animal Care and Use Committee of Harbin Medical University. We confirm that the experimental protocol was approved by Animal Care and Use Committee of Harbin Medical University. 3 × 10 6 QBC939 cells were transfected with shHOTAIR and subcutaneously injected into either side of the flank area of 6-week-old female BALB/c nude mice (n = 3 per group). Tumor size was measured every 3 days and tumor volumes were calculated using the equation: V = 0.5 × D × d 2 (V, volume; D, longitudinal diameter; d, latitudinal diameter). The nude mice were euthanized after injected 18 days, and the tumors weight were measured. Immunohistochemistry study. Immunohistochemistry study was carried out on the paraffin-embedded tumor tissues from nude mice. The avidin-biotin-peroxidase method were used to explore the location and relative expression level of Ki67 protein. Image-Pro Plus 6.0 software was used to quantify the number of positive cells in each image. The results were expressed as the average percentage of positive cells per image. The Ki67 primary antibody (Abcam, Cambridge, MA, USA) was used at a dilution of 1:200.

Statistical analysis.
All the experiments were totally repeated for 3 times. The results are reported as the means ± standard deviation (SD) based on at least three replicates. The methods of calculating power of study were performed by using GraphPad Prism 5.01 software (GraphPad Software, Inc., La Jolla, CA, USA) and SPSS 20.0 statistical software package (IBM, Armonk, NY, USA). A paired and two-tailed Student's t-tests were used to evaluate the significant differences between groups. Fisher's exact test was used to analyze the link between HOTAIR expression and clinicopathological features. Survival curves were drawn by Kaplan-Meier analyses and tested using log-rank tests. Survival data were analyzed by using univariate and multivariate Cox proportional hazards modeling. The data were considered significant only when p-values less than 0.05.