Long non‐coding RNA DLX6‐AS1 mediates proliferation, invasion and apoptosis of endometrial cancer cells by recruiting p300/E2F1 in DLX6 promoter region

Abstract Endometrial cancer features abnormal growth of cells of the inner lining of the uterus with the potential to invade to other organs. Accumulating evidence suggests that aberrant expression of long non‐coding RNA (lncRNA) may facilitate cancer progression. The aim of the present study was to identify the molecular mechanisms of the lncRNA known as DLX6 antisense RNA 1 (DLX6‐AS1) in endometrial cancer. Microarray‐based analysis was utilized to predict expression profile and possible function pattern of DLX6‐AS1 and DLX6 in endometrial cancer, and their expression was quantified in 78 clinically obtained endometrial cancer tissues and also in cell lines. We next assessed the effects of DLX6‐AS1 and DLX6 on proliferation, invasion and apoptosis of endometrial cancer cells. A mouse xenograft model was established to confirm DLX6‐AS1 functions and explore its underlying regulatory mechanisms in vivo. DLX6‐AS1 and DLX6 were highly expressed in endometrial cancer tissues and cells, and their silencing weakened the proliferative and invasive abilities of endometrial cancer cells and tumours, while promoting apoptosis. Mechanistic investigations indicated that DLX6‐AS1 formed a triplex structure with DLX6 via interaction with p300/E2F1 acetyltransferase. Thus, we find that functional up‐regulation of DLX6‐AS1 can promote endometrial cancer progression via a novel triplex mechanism that may prove to be great clinical significance for future treatments of endometrial cancer.


| Ethics statement
The study was approved by the Institutional Review Board of Linyi People's Hospital. All participants provided written informed consent prior to participation. The animal protocols and experiment procedures were conducted in accordance with the International Guide for the Care and Use of Laboratory Animals, following a protocol approved by the institution animal research committee.

| Microarray-based analysis
Analysis of the JASPAR database (http://jaspar.gener eg.net/) revealed the existence of binding sites of transcription factor E2F1 on the DLX6-AS1 promoter and the DLX6 promoter. GEPIA database (http://gepia.cance r-pku.cn/index.html) showed that DLX6-AS1 and DLX6 were both highly expressed in endometrial cancer and had a strong positive correlation. Furthermore, DLX6-AS1 was predicted to have a mainly nuclear localization according to analy-

| Study subjects
A total of 78 female patients (mean age: 56.81 ± 12.47 years, ranging from 35 to 74 years) with endometrial cancer diagnosed in Linyi People's Hospital from June 2016 to June 2017, who had not received any anticancer therapies prior to their operation in our study, with diagnosis according to World Health Organization criteria.
Patients were excluded if they also suffered from a serious infection, other solid tumours or diseases of the immune system. Endometrium collected by hysterectomy from patients for other reasons but malignancy was used as controls.

| RNA isolation and quantitation
Total RNA was extracted using Trizol (15596026, Invitrogen Inc, Carlsbad, CA, USA). RNA was reversely transcribed to cDNA according to instructions in the PrimeScript RT reagent kit (RR047A, Takara, Tokyo, Japan). The synthesized cDNA was subjected to RT-qPCR using the Fast SYBR Green PCR kit (Applied Biosystems, Oyster Bay, NY, USA) and the ABI PRISM 7300 RT-qPCR system (Applied Biosystems, Oyster Bay, NY, USA). Triplicates were set for each well.
With glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as the internal control, the fold changes of DLX6-AS1 expression were calculated by means of the relative quantification (2 −ΔΔCt method). 10 Primer sequences are shown in Table 1.

| Western blot analysis
Cells were harvested and lysed by enhanced radioimmunoprecipitation assay (RIPA) lysis (Boster Biological Technology Co., Ltd., Wuhan, Hubei, China) containing protease inhibitor. Protein concentration was determined with a bicinchoninic acid (BCA) kit (Boster Biological Technology Co., Ltd., Wuhan, Hubei, China). The protein was separated by 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis and transferred onto a polyvinylidene fluoride membrane. The membrane was treated with 5% bovine serum albumin (BSA) at room temperature for 2 hours to block non-specific binding and then incubated at 4ºC overnight with diluted primary rabbit antibody to DLX6 (ab137079, 1:2000, Abcam Inc, Cambridge, MA, USA). After washing with phosphate-buffered saline Tween-20 (PBST) three times, the horseradish peroxidase (HRP)-conjugated secondary goat anti-rabbit antibody (ab205719, 1:2000, Abcam Inc, Cambridge, MA, USA) was applied for a 1-h incubation at room temperature, followed by three PBST washes. The membrane was then visualized by enhanced chemiluminescence (Millipore, Temecula, CA, USA). The grey value was quantified and analysed using ImageJ software. GAPDH expression was used as the loading control. Each reaction was run in triplicate. 10

| 5-ethynyl-2'-deoxyuridine (EdU) staining
Cells in the logarithmic growth phase were plated in a 24-well plate, with triplicate wells set for each group. EdU (C10341-1, Ribo Biotechnology Co., Ltd., Guangzhou, Guangdong, China) was added into the medium to a final concentration of 10 µmol/L, followed by 2-h incubation. After removal of the medium, cells were fixed in phosphate buffer saline (PBS) containing 4% paraformaldehyde at room temperature for 15 minutes, washed twice with PBS containing 3% BSA and incubated with PBS containing 0.5% Triton-100 at room temperature for 20 minutes, followed by another two washes in PBS containing 3% BSA. Then, 100 µL Apollo ® 567 (Ribo Biotechnology Co., Ltd., Guangzhou, Guangdong, China) was added into each well for 30-min incubation at room temperature in the dark, followed by another two washes with PBS containing 3% BSA.
Then, the sample was stained with 1 × Hoechst 33342 for 30 minutes and washed three times in PBS. The number of EdU positive cells (red) in randomly selected fields was counted under a fluorescence microscope (FM-600, Shanghai Pudan Optical Instrument Co., Ltd., Shanghai, China). Each experiment was run in triplicate.

| Transwell assay
The basement membrane of apical chambers in Transwell chamber was coated with Matrigel (Becton, Dickinson and Company, Franklin L., New Jersey, USA) and allowed to stand at 37ºC for 30 minutes to allow polymerization. The basement membrane was hydrated before use. Cells were cultured in serum-free medium for 12 hours, collected, and re-suspended in serum-free medium (1 × 10 5 cells/mL).
The basolateral chamber was added with medium containing 10% FBS, and 100 μL cell suspension was added into the Transwell chamber for incubation at 37ºC for 24 hours. Cells that did not invade the Matrigel surface was removed gently with cotton swabs. Following fixation in 100% methanol, cells were stained in 1% toluidine blue Abbreviations: DLX6-AS1, DLX6 antisense RNA 1; F, forward; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; R, reverse.

| Flow cytometry
The endometrial cancer cells were seeded in 6-well plates containing penicillin-streptomycin-free medium at a density of 1 × 10 6 cells/

| Subcellular fraction
Nucleus and cytoplasm were separated following the manufacturer's instructions in the PARIS kit (Life Technologies, Carlsbad, CA, USA). 13

| RNA immunoprecipitation (RIP)
The binding between DLX6-AS1 and p300 protein was detected by the RIP kit (Millipore, Temecula, CA, USA and E2F1-DLX6 promoter was then obtained and de-crosslinked.
The purified fragments enriched with DLX6-AS1 and DLX6 promoter were detected by PCR with Input as the internal reference.
Each experiment was run in triplicate. 17

| In vitro triplex pull-down assay
Primers were designed to amplify the 5'untranslated region (

| In vivo triplex capture assay
Nuclei (2 × 10 6 ) were isolated from cells that had been transfected with DLX6 TSS fragments and then incubated with 8 pmoL bioti-   The cells with stable transfection were dispersed into a cell suspension (5 × 10 7 cells/mL). Then, for tumorigenesis, 0.2 mL cell suspension was subcutaneously inoculated into the left armpit of BALB/c nude mice (n = 8/each group) using a 1 mL syringe. After 5 weeks, mice were euthanized with the size and weight of tumour recorded.

| Xenograft tumour in nude mice
The DLX6 expression was determined. Each experiment was run in triplicate. SPSS 21.0 software (IBM Corp., Armonk, NY, USA) was applied for data analysis. The measurement data were expressed as mean ± standard deviation. Comparison of paired data between two groups following normal distribution and homogeneity of variance was conducted by paired t test while unpaired data were conducted by unpaired t test.

| Statistical analysis
Comparison among multiple groups was conducted by one-way analysis of variance, followed by Tukey's post hoc test. Comparison at different time points within group was conducted by repeated measures analysis of variance, followed by Bonferroni post hoc test. Values of P < 0.05 were considered statistically significant.

| DLX6-AS1 is highly expressed in endometrial cancer
Initially, analysis from GEPIA database (http://gepia.cance r-pku. cn/index.html) showed that DLX6-AS1 was highly expressed in endometrial cancer ( Figure 1A), which was further confirmed by RT-qPCR of the patient samples ( Figure 1B). The correlation analysis between DLX6-AS1 and clinicopathological features of patients with endometrial cancer is depicted in Table 2, which revealed no significant correlation between DLX6-AS1 and age or degree promoter, as shown in Figure 1D. ChIP assay for binding between E2F1 and DLX6-AS1 promoter region showed that enrichment of E2F1 in DLX6-AS1 promoter region was higher in the RL-952 cell line than in PA cells (P < 0.05; Figure 1E). The aforemen-

| Silencing DLX6-AS1 represses endometrial cancer cell proliferation and invasion and contributes to apoptosis
We next focus on the potential role of DLX6-AS1 in regulating biological behaviours of endometrial cancer cells. After DLX6-AS1 was silenced or overexpressed in RL-952 and HHUA cells, RT-qPCR was conducted to detect the silence efficiency of DLX6-AS1 (Figure 2A

| DLX6 is highly expressed in endometrial cancer
According to GEPIA database (http://gepia.cance r-pku.cn/index. html), DLX6 is also highly expressed in endometrial cancer and positively correlated to DLX6-AS1 ( Figure 3A-B). RT-qPCR and Western blot analysis for DLX6 expression revealed higher DLX6 expression in endometrial cancer tissues than in normal tissues (P < 0.05; Figure 3C-D). Western blot analysis also showed that endometrial The results were measurement data and expressed as mean ± standard deviation. Comparison among multiples groups was conducted by one-way analysis of variance, followed by Tukey's post hoc test. Cell experiments were repeated three times independently had significantly higher DLX6 expression than did PA cells (P < 0.05; Figure 3E). Western blot analysis after silencing DLX6-AS1 revealed significantly decreased DLX6 expression when compared with sh-NC treatment (P < 0.05; Figure 3F). Taken together, these findings suggest high expression of DLX6 in endometrial cancer tissue and cells.

| Silencing DLX6 weakens endometrial cancer cell proliferation and invasion while inducing apoptosis
After confirming the aberrantly expressed DLX6 in endometrial cancer, we explored its effects on biological functions of endometrial cancer cells. Western blot analysis showed that sh-DLX6 led to significantly lower DLX6 expression in RL-952 and HHUA cells, while oe-DLX6 treatment led to significantly higher DLX6 expression (P < 0.05; Figure 4A). Then, EdU assay ( Figure 4B

| DLX6-AS1 increases DLX6 expression by recruiting p300/E2F1 to DLX6 promoter
Subsequently, we explored the regulatory mechanism underlying DLX6-AS1 and DLX6 effects. The analysis available on the lncATLAS website (http://lncat las.crg.eu/) revealed that DLX6-AS1 was mainly localized in the nucleus ( Figure 5A), which was in line with results of our FISH assay ( Figure 5B) and subcellular fraction study ( Figure 5C).
To explore the relation between E2F1 and DLX6, we turned to the JASPAR website (http://jaspar.gener eg.net/) for prediction on the binding sequences between E2F1 and DLX6 promoter, which promoter region showed that sh-DLX6-AS1 treatment resulted in less E2F1 enrichment while oe-DLX6-AS1 increased E2F1 enrichment (P < 0.05; Figure 5H). These findings highlight that DLX6-AS1 recruits p300/E2F1 to the DLX6 promoter to activate DLX6 expression. In vitro triplex, pull-down assay showed that DLX6 obtained from streptomycin-coated magnetic beads increased with oe-DLX6-AS TFO treatment than with oe-control oligo treatment. Besides, in studies using 7-deaza-dGTP to amplify the DLX6 double-strand and maturation of TTS, there was decreased DLX6-AS enrichment, which lowered the binding affinity of DNA double-strand to DLX6-AS1 (P < 0.05; Figure 6A). The EMSA assay demonstrated that triplex formation was sensitive to RNase A but resistant to RNase H. Next, DLX6-AS TFO WT and MUT was incubated with 32 P-labelled WT DLX6 TTS, respectively, results of which showed that enrichment of MUT DLX6-AS1 reduced significantly ( Figure 6B). To further verify the in vivo binding between DLX6-AS1 and DLX6, the in vivo triplex capture assay was performed using biotin-labelled DLX6-AS1 TFO and endometrial cancer cell lysate.

| DLX6-AS1 binds to DLX6 promoter in the form of DNA-RNA triplex
This test revealed significant enrichment of DLX6 on DLX6-AS1 TFO rather than the oligonucleotide sequence (P < 0.05; Figure 6C). To conclude, these findings demonstrate that DLX6-AS1 binds to the DLX6 promoter in the form of a DNA-RNA triplex.

| D ISCUSS I ON
Endometrial cancer is one of the most common gynaecological cancers all over the world. In low-grade stages, prognosis of patients F I G U R E 5 DLX6-AS1 up-regulates DLX6 expression by recruiting p300/E2F1 to the DLX6 promoter. A, subcellular localization of DLX6-AS1 predicted on lncATLAS. B, Subcellular localization of DLX6-AS1 detected by FISH assay (400 ×). C, subcellular localization of DLX6-AS1 following subcellular fraction. D, Binding ability between sequences of DLX6-AS1 and p300 predicted by the RPIseq database. E, Binding between DLX6-AS1 and p300 detected by RIP. F, DLX6-AS1 pulling down of p300 detected by RNA pull-down assay. G, Enrichment of p300 in the DLX6 promoter region detected by ChIP. H, Enrichment of E2F1 in the DLX6 promoter region detected by ChIP. *P < 0.05 vs sh-NC treatment or oe-NC treatment or DMSO treatment. The results were measurement data and expressed as mean ± standard deviation. Comparison between two groups was conducted by non-paired t test. Comparison among multiples groups was conducted by one-way analysis of variance (ANOVA), followed by Tukey's post hoc test. Cell experiments were repeated three times independently    20 Therefore, in this study, we scrutinized the regulation between lncRNA DLX6-AS1 and the DLX6 coding, with the result that lncRNA DLX6-AS1 was proved to recruit p300 and E2F1 to form an RNP complex at the promoter region of F I G U R E 6 DLX6-AS1 and DLX6 promoter initiates the DNA-RNA triplex formation. A, Binding between DLX6-AS1 and DLX6 promoter detected by in vitro triplex pull-down assay. B, DNA-RNA triplex formation between DLX6-AS1 and DLX6 detected by electrophoretic mobility shift assay. C, Binding between DLX6-AS1 and DLX6 promoter detected by in vivo triplex capture assay. *P < 0.05 vs the oe-NC, oe-control oligo or dGTP group. The results were measurement data and expressed as mean ± standard deviation. Comparison between two groups was conducted by non-paired t test. Comparison among multiples groups was conducted by one-way analysis of variance (ANOVA), followed by Tukey's post hoc test. Cell experiments were repeated three times independently  Besides, the knockdown of lncRNA DLX6-AS1 could suppress proliferation, migration and invasion of HCC in vitro as well as inhibiting tumour growth in vivo. 22 In another gynaecological cancer, cervical cancer, lncRNA DLX6-AS1 similarly functioned as an oncogene by sponging miR-199a and promoting cell proliferation. 8 Furthermore, Yang et al have shown that lncRNA DLX6-AS1 could bind to and inhibit miR-199a, which promoted the proliferation, migration and invasion of nasopharyngeal carcinoma cells. 23 These various findings drew our attention to the possible mechanism between lncRNA DLX6-AS1 and miR-199a, which we shall explore in future studies. Much less is known about DLX6 protein in cancer.
A single published report suggested that overexpression of DLX6 was associated with metastasis of the human breast cancer cell line MDA-MB-231. 24 Bioinformatics analysis predicted that DLX6 protein could play important roles in embryonic development and tissue differentiation. Additionally, in the IntAct protein interaction database, DLX6 was identified to interacted with HSP 90beta and DNA damage-inducible transcript 4-like protein, which suggested an association between DLX6 and cell ROS status or DNA restoration. However, the molecular mechanism by which DLX6 protein could mediate effects of lncRNA DLX6-AS1 to enhance tumour growth is still unknown. Hence, more investigations should be carried out to explore the function of DLX6.
Second, we found lncRNA DLX6-AS1 could specifically bind at the promoter region of DLX6 and then recruit p300 protein to promote transcription of DLX6. Epigenetic regulation includes genome methylation and acetylation of histones or other DNA binding proteins. p300 is a histone acetyltransferase that can increase acetylation of histones and loose chromatin structure to facilitate the initiation of transcription. 25 p300 protein shows diverse roles in various human diseases including cancers. Usually, loss of p300 protein function by mutation or knockout leads to increased occurrence of leukaemia, which suggests a tumour suppressive role of p300/CBP. 26 On one hand, p300/CBP is indispensable to transcribe certain essential tumour suppressor genes such as p53, BRCA1 and FOXO3a. On the other hand, p300 can also serve as an enhancer of c-Myc, c-Myb, and AR to promote tumour growth. 27 These dual functions suggest contradictory or opposing roles of p300/CBP in the regulation of tumorigenesis via diverse signalling pathways. For instance, p300/CBP can promote the development of prostate cancer in an AR-dependent transcription. 28 In this study, we proved the tumour-activator role of p300 in an E2F1-dependent mechanism.
Interestingly, a previous report showed that lncRNA DLX6-AS1 could relieve E2F1 by sponging miR-197-5p and thus promote the glioma carcinogenesis. 29 These studies suggested that lncRNA DLX6-AS1 could elevate gene expression through improving E2F1mediated transcription.
In general, the regulation of a gene by its antisense transcript is a common mechanism in lncRNA pathways. We found that DLX6 and its antisense transcript lncRNA DLX6-AS1 were highly expressed in endometrial cancer ( Figure 8). Meanwhile, silencing lncRNA DLX6-AS1 or DLX6 significantly suppressed tumour growth and increased cell apoptosis. Furthermore, the effects on endometrial cancer cells were dependent upon the recruitment of p300 and E2F1 via F I G U R E 8 A schematic diagram depicting the molecular basis of the DLX6-AS1/p300/E2F1/DLX6 axis in regulating the progression of endometrial cancer. To be specific, DLX6-AS1 up-regulates DLX6 expression by recruiting transcription factor p300/E2F1 to the DLX6 promoter region, with the end results of promoting the progression of endometrial cancer E2F1 DLX6-AS1 P300 P300 E2F1

DXL6
Promote the progression of endometrial cancer Nucleus lncRNA DLX6-AS1. Therefore, it seems that lncRNA DLX6-AS1 and DLX6 both have oncogenic properties, which could justify efforts to develop an antagonist as an experimental treatment for endometrial cancer.

ACK N OWLED G EM ENT
We acknowledge and appreciate our colleagues for their valuable efforts and comments on this paper.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.