Long non‐coding RNA highly up‐regulated in liver cancer promotes epithelial‐to‐mesenchymal transition process in oral squamous cell carcinoma

Abstract Oral squamous cell carcinoma (OSCC) is an oral and maxillofacial malignancy that exhibits high incidence worldwide. In diverse human cancers, the long non‐coding RNA (lncRNA) highly up‐regulated in liver cancer (HULC) is aberrantly expressed, but how HULC affects OSCC development and progression has remained mostly unknown. We report that HULC was abnormally up‐regulated in oral cancer tissues and OSCC cell lines, and that suppression of HULC expression in OSCC cells not only inhibited the proliferation, drug tolerance, migration and invasion of the cancer cells, but also increased their apoptosis rate. Notably, in a mouse xenograft model, HULC depletion reduced tumorigenicity and inhibited the epithelial‐to‐mesenchymal transition process. Collectively, our findings reveal a crucial role of the lncRNA HULC in regulating oral cancer carcinogenesis and tumour progression, and thus suggest that HULC could serve as a novel therapeutic target for OSCC.

that are not translated into proteins. 14 Transcription of HULC yields ~500 nt long, spliced and polyadenylated ncRNA that localizes to the cytoplasm, where it has been reported to be associated with ribosomes. 15 HULC has been shown to perform critical functions in diverse tumours, including gastric cancer, pancreatic cancer, osteosarcoma and liver metastasis of colorectal cancer. 14,[16][17][18][19] However, no study to date has reported a regulatory role of HULC in OSCC.
To investigate HULC function in OSCC development, we quantified HULC expression levels in oral cancer tissues and adjacent normal tissues by using qRT-PCR. HULC expression was higher in the cancer tissues than in the normal tissues, and, similarly, was higher in OSCC cell lines than in normal keratinocytes, and HULC down-regulation in the OSCC cell lines SCC15 and SCC25 affected the proliferation, migration and invasion abilities of these cells. Moreover, in a nude mouse xenograft model that we constructed, HULC depletion reduced tumorigenicity and inhibited the epithelial-to-mesenchymal transition (EMT) process. Our results not only reveal a previously unreported regulatory role of HULC in OSCC, but also suggest that HULC represents a potential therapeutic target for OSCC.

| Patients and tissue samples
Oral cancer tissues and their adjacent normal tissues were ob-

| Cell culture and transfection
The human OSCC cell lines SCC9, SCC15, SCC25 and CAL27 were obtained from the College of Stomatology, Wuhan University (Wuhan, China). HOK cells were obtained from the cell bank of the Chinese Academy of Sciences (Shanghai, China). All cells were cultured at 37°C and 5% CO 2 in a humidified atmosphere. SCC15, and control siRNA-2:5′-UAAAACGAAUGGAAUUCACUU-3′. The results shown here were obtained using control siRNA-1. For HULC shRNA construction, a lentiviral vector was obtained from GenePharma (Shanghai, China), and the primers for the HULC shRNA were the following: forward, 5′-GGAGAACACTTAAATAAGTTT-3′; reverse, 5′-ACTTATTTAAGTGTTCTCCTA-3′.

| RNA extraction and qRT-PCR
RNA extraction and qRT-PCR were performed as previously described. 20 Total RNA from frozen tissues or cultured cells was extracted using TRIzol reagent (Invitrogen), according to the manufacturer's protocol. A PrimeScript RT Reagent Kit (Takara Bio, Nojihigashi, Kusatsu, Japan) was used for reverse-transcribing the RNA into cDNA, as per the manufacturer's instructions. qRT-PCR was performed with SYBR-Green Premix Ex Taq (Takara Bio) and was monitored using an ABI PRISM 7500 Sequence Detection System

| CCK-8 assay
We seeded 100 μl of transfected OSCC (SCC15 and SCC25) cell suspensions in 96-well plates at a density of 2 × 10 3 cells/well, and then added CCK-8 solution (Beyotime, Shanghai, China) at 10 μl/well at 24, 48, 72 and 96 hours. The plates were incubated for 1 hour and the 450-nm absorbance was measured, and the OD values at various time-points were compared.

| EdU cell-proliferation assay
Transfected OSCC cells were seeded in 96-well plates at a density of 4 × 10 4 cells/well and cultured to logarithmic growth phase.
The cells were incubated with diluted EdU solution for 2 hours, fixed with 4% paraformaldehyde, and then stained with Apollo staining-reaction solution and Hoechst 33342 reaction solution in the dark. Subsequently, images were acquired and analysed.

| Flow cytometry
Cells were routinely transfected and cultured for 48 hours and then digested with trypsin without EDTA. An Annexin V-FITC apoptosis assay kit (Biyuntian Biotechnology Co., Ltd.) was used to estimate the apoptosis rate, according to the manufacturer's instructions.
Cells were suspended in 1× annexin-binding buffer, and then 5 μl of Annexin V and 1 μl of PI reagents were added to 100 μl of the cell suspension and mixed. The mixture was incubated in the dark for 15 minutes at room temperature, and then 400 μl of the 1× annexinbinding buffer was added to each sample to terminate the staining.
The apoptosis rate was determined using a FACSCalibur flow cytometer (BD).

| Hoechst staining assay
We plated 4 × 10 5 cells/well in 24-well plates containing sterile glass coverslips, and following overnight incubation, fixed the cells with 4% paraformaldehyde for 15 minutes. Hoechst 33258 staining solution (Beyotime) was added and the cells were incubated for 10-15 minutes, and after air-drying, images were acquired and examined for typical apoptosis-related changes (chromatin concentration, aggregation, destruction, etc.).

| Wound-healing assay
Transfected cells were spread on 6-well plates and cultured until confluence. Before wounding, the cells were cultured in DMEM without FBS for 1 day. A sterile 200-μl pipette tip was used to scratch the cell monolayers, and after wounding, the cells were washed thrice with phosphate-buffered saline (PBS) and incubated with fresh medium containing 10% FBS. Images were acquired at 0 and 48 hours or 72 hours.

| Migration and invasion assays
Migration and invasion assays were performed with , respectively, Transwell chambers and Matrigel pre-coated Transwell chambers (Corning, NY). Cells were resuspended in DMEM without FBS and added to the upper chamber, and medium containing 10% FBS was added to the lower chamber; after incubation for 24 or 48 hours, the cells in upper chamber were wiped off, and the cells in the lower chamber were fixed in 4% paraformaldehyde, stained with 0.1% crystal violet, washed with PBS, and dried. Lastly, images were acquired and analysed.

| Cell-viability assay
The cell-viability assay was designed according to the method of

| Western blotting analysis
Cells extracts were prepared at 4°C in RIPA buffer (

| Image processing and statistical analysis
All images shown are wide-field microscopy images that were acquired at sufficient resolution. Results in graphs are shown as means ± SEM from three independent experiments. All statistical data were analysed using SPSS 17.0 software (SPSS, Chicago, IL).
Two-tailed Student's t test was used to determine P values; P < 0.05 was considered significant.

| HULC is highly expressed in OSCC
Highly up-regulated in liver cancer (HULC) mRNA levels were determined using qRT-PCR. Analysis of 30 pairs of clinical oral cancer tissues and their adjacent normal tissues revealed that HULC expression was higher in the cancer tissues than in the normal tissues ( Figure 1A). The clinicopathological features of the 30 OSCC patients are shown in Table 1. We also measured HULC levels in four OSCC cell lines (SCC15, SCC25, SCC9 and CAL27), which revealed that HULC expression was markedly up-regulated in the cancer cell lines relative to that in a normal oral keratinocyte cell line (human oral keratinocyte (HOK) cells) ( Figure 1B). Because similar results were obtained with both the oral cancer tissues and the OSCC cell lines, we used the OSCC cell lines SCC15 and SCC25 for subsequent studies.

| Suppression of HULC reduces proliferation and promotes apoptosis in OSCC cells
To investigate the role of HULC in regulating cell-proliferation activity, we performed the CCK-8 assay on SCC15 and SCC25 cells in which HULC was knocked down. Transfection of HULC siRNA into SCC15 and SCC25 cells led to HULC knockdown with an efficiency of roughly 90% and 74%, respectively ( Figure S1A Another assay that involved EdU staining was also performed to confirm the proliferation results; here, nuclei were stained red when the cells were in S phase. Determination of the proliferation ratio in SCC15 and SCC25 cells revealed that after HULC depletion, the ratio was decreased by approximately 12% relative to that in the control group ( Figure 2B,C).
Next, the apoptosis rate in HULC-depleted cells was estimated by performing Annexin V-FITC/PI dual-label flow cytometry experiments. In the case of SCC15 cells, the early apoptosis and late apoptosis proportions were 0.85% and 0.97% in the control group, respectively, which were lower than those in the HULC-depletion group (early apoptosis: 4.35%; late apoptosis: 3.78%; Figure 3A).
For SCC25 cells, the early and late apoptosis proportions measured were the following (respectively): HULC-depletion group, 1.90% and 4.47%; control group, 0.30% and 1.02% ( Figure 3B). These results indicate that the suppression of HULC expression strongly promoted apoptosis in SCC15 and SCC25 cells. Here, we also performed Hoechst staining on the SCC15 and SCC25 cells transfected with HULC siRNA and then counted the apoptotic cells in each group: the numbers of apoptotic cells in the HULC-depletion groups were 5.6-fold (SCC15) and 7-fold (SCC25) higher than those in the corresponding control groups, respectively ( Figure 3C). Collectively, these findings indicate that HULC depletion reduces the proliferation of OSCC cells and promotes their apoptosis.

| HULC down-regulation inhibits OSCC cell migration and invasion abilities
To determine whether HULC influences OSCC cell migration, we performed wound-healing assays on control and HULC-depleted SCC15 and SCC25 cells. Measurement of the scratch area at 0 and 48 hours after wounding revealed that the wound-closure rate in HULC-depleted cells was significantly lower than that in control  cells ( Figure 4A, B). The closure percentages at 48 hours in HULCdepleted cells were roughly 21% lower (SCC15; Figure 4A) and 25% lower (SCC25; Figure 4B) than those in the control cells.
In addition, we also tested the migration ability in HOK cells with high expression of HULC. Up-regulation of HULC caused higher wound-closure rate in HOK cells ( Figure 4C). The closure percentages at 72 hours in HULC-overexpressed cells were roughly 16% higher than those in the control cells ( Figure 4C). To quantify the migration ability of HULC-depleted OSCC cells, Transwell assays were used. After 48-hours incubation, the numbers of HULC-depleted SCC15 and SCC25 cells that had passed through to the lower chamber were approximately 600 and 580, respectively, which were considerably lower than that in the case of control cells (roughly 820; Figure 4D). Overall, the results suggested that the OSCC cell migration ability was impaired following the depletion of HULC. To investigate the OSCC cell invasion ability, we again used the Transwell invasion assay and counted the cells that had crossed through the chamber coated with Matrigel.
F I G U R E 2 Suppression of HULC expression inhibits OSCC cell proliferation. A, SCC15 and SCC25 cells were transfected with control or HULC siRNA, and the CCK-8 assay was used to measure cell proliferation after different transfection durations. HOK cells were transfected with vector control or HULC, respectively. The cell proliferation were measured using CCK-8 assay. (B, C) EdU incorporation assay was used to measure the proliferation ratio of control and HULC-depleted cells. Data are presented as means ± SEM of three independent experiments. Student's t test, *P < 0.05, **P < 0.01, ***P < 0.001; scale bar = 20 μm HULC depletion strongly suppressed the invasiveness of SCC15 and SCC25 cells, with the invading HULC-depleted cells in both cases being only around half as many as the invading control cells ( Figure 4E). Taken together, these results suggest that HULC regulates OSCC cell migration and invasion.

| Drug tolerance of OSCC cells is HULCdependent
The drug most commonly used for OSCC treatment is cis-diam-

| HULC controls OSCC cell EMT process and tumour growth
To investigate the molecular basis of HULC regulation in OSCC cells, we examined the expression of several key proteins. Previous studies have shown that EMT is a crucial factor for epithelial cancer metastasis. 23 Here, to evaluate the EMT process, we suppressed HULC expression and detected EMT markers by performing Western blotting. As compared to control OSCC cells, HULC-depleted cells showed decreased expression of vimentin and N-cadherin and increased expression of Ecadherin ( Figure 6A), which indicates that HULC functions in the EMT process in OSCC cells. We next immunoblotted for the following proteins: Bcl-2 and BAX, which are critical indicators used for detecting cell proliferation and apoptosis 24,25 ; MMP-9, which plays a crucial role in tumour invasion and metastasis 26 ; and cyclin D1, a key regulator of cell cycle transition from G1 to S phase. 27 Our results showed that after HULC knockdown, BAX was up-regulated, whereas Bcl-2, MMP-9, and cyclin D1 were down-regulated ( Figure 6B). These results indicate that HULC participates in the EMT process and affects the expression levels of proteins that are crucial for cell proliferation and invasion.
Lastly, to investigate the potential of HULC as a new OSCC therapeutic target, we established a xenograft tumour model by using the SCC15 cell line in nude mice. In SCC15 cells, HULC was knocked down by using a lentiviral vector carrying a GFP-tagged shRNA ( Figure S2A); in the stably transfected SCC15 cells, HULC expression was lowered to approximately 20% of the control level ( Figure S2B).
All mice developed tumours at the injection sites, but the tumours in the HULC-depletion group were considerably smaller than those in the empty-vector group ( Figure 7A). Measurement of the tumour growth curve and final weight in the nude mice further revealed that both were suppressed in the HULC-depletion group relative to control ( Figure 7B, C). Moreover, the results of haematoxylin and eosin  Figure 7D). We also did the immunohistochemical (IHC) staining of xenograft tumours ( Figure 7E,F), which showed that the EMT maker vimentin was down-regulated ( Figure 7E) in HULC-depleted tumours but E-cadherin was up-regulated ( Figure 7F), which agreed with the Western blotting results. Collectively, these results showed that HULC is crucial for tumour growth and promotes the EMT process. Our findings further indicate that HULC could potentially serve as a new therapeutic target in OSCC treatment.

| D ISCUSS I ON
Oral squamous cell carcinoma (OSCC) is the tenth most common malignancy and accounts for 90% of all head and neck malignancies Although HULC has been found to be crucial for several cancer types, this is the first report on HULC in human OSCC. We have demonstrated for the first time that HULC is highly up-regulated in OSCC and is crucial for OSCC cell proliferation, migration and invasion, and our data further suggest that HULC could function as a potential oncogene and promote the malignant progression of OSCC; this provides a basis for the use of HULC as a tumour marker specific for OSCC. However, to further understand HULC regulatory mechanisms in OSCC, the pathways both upstream and downstream of HULC must be investigated in future studies.
In this study, we found that HULC expression was considerably higher in oral cancers than in adjacent normal tissues in patients.
However, Kaplan-Meier survival analysis was not completed because the patient number was limited and because of the loss of follow-up. Moreover, the relationship between the TNM stage and HULC expression level currently remains unclear. To further enhance our understanding of HULC function in OSCC, a more precise and comprehensive analysis of HULC expression in OSCC patients is required.

ACK N OWLED G EM ENTS
This study was supported by the National Natural Science Hospital Peking University).

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.