The lncRNA HULC functions as an oncogene by targeting ATG7 and ITGB1 in epithelial ovarian carcinoma

Highly upregulated in liver cancer (HULC) is a long noncoding RNA (lncRNA), which has recently been identified as a key regulator in the progression of hepatocellular carcinoma, gliomas and gastric cancer. However, its role in epithelial ovarian carcinoma (EOC) remains unknown. In this study, HULC expression was examined in EOC, borderline and benign ovarian tumors, and normal ovarian tissues by RT-PCR. Ovarian cancer cell phenotypes, as well as autophagy-associated proteins were examined after HULC overexpression or downregulation by plasmid or small interfering RNA (siRNA) transfection, respectively. LncRNA–protein interactions were examined by ribonucleoprotein immunoprecipitation (RIP) assays. We found that HULC expression levels were higher in EOC tissues than normal samples. HULC overexpression induced cell proliferation, migration, invasion, whereas reduced cell apoptosis in vitro and induced tumor growth in vivo. In contrast, downregulation of HULC by siRNA transfection reduced cell proliferation, migration and invasion, and induced cell apoptosis and autophagy. Our results showed that HULC overexpression reduced ATG7, LC3-II and LAMP1 expression, while inducing SQSTM1 (P62) and ITGB1 expression. HULC downregulation had the opposite effects. Furthermore, RIP indicated that ATG7 interacted with HULC; ATG7 downregulation also induced cell proliferation, reduced apoptosis and inhibited autophagy in vitro by reducing LC3-II and LAMP1 expression, while inducing SQSTM1 expression. Furthermore, ATG7 co-transfection with HULC reversed the oncogenic effects of HULC both in vitro and in vivo; however, downregulating ATG7 did not affect cell migration and invasive ability. We found that ITGB1 siRNA co-transfection with HULC reversed the function of HULC in inducing ovarian cancer cell migration and invasive ability. Taken together, our results show that HULC may promote ovarian carcinoma tumorigenesis by inhibiting ATG7 and inducing progression by regulating ITGB1.

examined in EOC, borderline and benign ovarian tumors and in normal ovarian tissues by RT-PCR (Figure 1a). HULC expression levels were higher in EOC tissues than in the normal samples and benign ovarian tumors (Po0.05; Figure 1b).
HULC downregulation reduces ovarian carcinoma cell migration and invasion. Wound-healing assay showed that HULC downregulation reduced cell migration compared with the control and mock-transfected cells (Po0.05; Figure 3h). Transwell assays showed that cells transfected with si-HULC had reduced invasion ability compared with control and mock-transfected cells (Po0.05; Figure 3i).
HULC overexpression regulates ATG7, LC3, SQSTM1 (P62), LAMP1 and ITGB1 mRNA and protein expression. We observed induced mitochondria formation in cells transfected with HULC by transmission electron microscopy ( Figure 4a). Results from immunofluorescence assays indicated that HULC overexpression induced expression of SQSTM1 (Figure 4b), while inhibiting ATG7 (Figure 4c), LC3 Figure 1 Correlation of lncRNA HULC expression with pathogenesis of ovarian carcinoma. HULC expression in ovarian cancer tissues was higher than borderline ovarian tumors, benign ovarian tumors and normal ovarian tissues (a). HULC expression was significantly higher in EOCs than normal ovarian tissues and benign ovarian tumors (b). *versus normal ovarian tissues; # versus benign tumors; Po0.05 The role of HULC in ovarian carcinoma S Chen et al ( Figure 4d) and LAMP1 (Figure 4e) expression. We performed qRT-PCR and western blot analysis to measure ATG7, LC3, SQSTM1, LAMP1 and ITGB1 mRNA or protein expression levels after HULC overexpression in OVCAR3 cells. Both mRNA and protein expression of ATG7, LC3 and LAMP1 were significantly lower than in the control, whereas expression of SQSTM1 and ITGB1 was significantly increased compared with the negative control (Po0.05; Figures 4f and g). However, there were no significant differences in ATG5, Beclin-1, ATG12, Vps34, P150 and UVRAG expression (Figure 4h). and LAMP1 mRNA and protein expression both were significantly higher than in the control, whereas expression of SQSTM1 and ITGB1 was significantly decreased by HULC siRNA transfection compared with the negative control (Po0.05; Figures 5f and g).
HULC overexpression induces ovarian carcinoma cell tumorigenesis in vivo. In our study, mice injected with A2780-HULC overexpression cells showed a significantly higher rate of tumorigenicity after inoculation compared with the control group (Po0.05; Figure 6c), and exhibited larger tumor volumes during the same observation period (Figures 6a and b). Immunohistochemistry (IHC) analysis showed significantly lower ATG7 and LC3 expression, while SQSTM1 and ITGB1 expression was upregulated in the HULC overexpression group compared with the control group ( Figure 6d).

Discussion
Studies have demonstrated that the lncRNA HULC is associated with the proliferation, invasion, metastasis and survival of tumor cells in certain cancers. [17][18][19][20][21][22] Furthermore, overexpression of HULC serves as an independent indicator of patient prognosis, by predicting the rates of recurrence and disease-free survival. 20 Our results showed that HULC The mRNA (f) and protein (g) expression of ATG7, LC3 and LAMP1 were significantly lower than in the control, whereas expression of SQSTM1 and ITGB1 was significantly increased compared with the negative control. No significant differences were with ATG5, Beclin-1, ATG12, Vps34, P150 and UVRAG expression (h). *Po0.05 The role of HULC in ovarian carcinoma S Chen et al The mRNA (f) and protein (g) expression of ATG7, LC3 and LAMP1 were significantly higher than in the control, whereas expression of SQSTM1 and ITGB1 was significantly decreased compared with the negative control. *Po0.05 The role of HULC in ovarian carcinoma S Chen et al expression was significantly higher in EOC than in benign tumor, and normal ovarian tissues. Our results are consistent with Peng et al. 23 , who showed that the lncRNA HULC is a novel biomarker in patients with pancreatic cancer and diffuse large B-cell lymphoma. 24 This indicates that HULC may have potential as a biomarker in EOC diagnosis.
HULC is known to induce proliferation, migration and invasion in cell culture. 21 We confirmed this observation by showing that the overexpression of HULC in vitro induced cell proliferation, migration, invasion and reduced cell apoptosis, whereas HULC downregulation by siRNA transfection reduced cell proliferation, migration, invasion and induced cell The role of HULC in ovarian carcinoma S Chen et al apoptosis, which suggests that HULC may functions as an oncogene in EOC. Furthermore, through electron microscopy, we found that HULC increased mitochondria formation, while si-HULC transfection induced autophagosome; we also found that adding the autophagy inhibitor 3-MA reversed the effects of si-HULC in inducing apoptosis and inhibiting cell proliferation, suggesting than HULC may promote tumorigenesis by inhibiting autophagy.
Autophagy involves numerous steps: initiation, nucleation, elongation, closure of the membranes that form the autophagosome, fusion with the lysosome and the recycling of macromolecular precursors. Specific autophagy-related proteins regulate each step. Two ubiquitin-like conjugation systems elongate the autophagosome membrane. The ubiquitin-like protein ATG12 is conjugated to ATG5 in a process requiring the E1-like enzyme ATG7. A similar lipid conjugation system (also using ATG7) attaches phosphatidylethanolamine (PE) to the microtubule-associated protein 1 light chain 3 (MAP1LC3) and GABA type A receptorassociated protein (GABARAP) protein families. Furthermore, Beclin-1 (BCL-2-interacting moesin-like coiled-coil protein 1), its signaling complex P150, VPS34 (class III phosphoinositide-3-kinase) and ultraviolet irradiation resistant-associated gene (UVRAG) are all responsible for vesicle nucleation of the Si-ATG7 transfection had no effect in migration and invasion ability (e and f). Si-ATG7 transfection induced SQSTM1 expression (g), inhibited ATG7 (h), LC3 (i) and LAMP1 expression (j) by Immunofluorescence. The mRNA (k) and protein (l) expression of ATG7, LC3 and LAMP1 were significantly lower than in the control, whereas expression of SQSTM1 was significantly increased compared with the negative control. *Po0.05 The role of HULC in ovarian carcinoma S Chen et al phagophore membrane. 25,26 LC3 was originally identified as a subunit of microtubule-associated proteins 1A and 1B and was subsequently found to be similar to the yeast protein Atg8/ Aut7/Cvt5, which is critical for autophagy. 24,27 The conversion of LC3 to the lower migrating form, LC3-II, has been used as an indicator of autophagy. 28-30 SQSTM1 (P62), has been implicated as a potential oncogene in other settings, including human hepatocellular carcinomas, 31 lung carcinomas, 32 pancreatic carcinomas, breast carcinomas, 33,34 prostate cancer 35 and in immortalized baby mouse kidney cells. 36 Recently, studies have shown that its accumulation represents a block to autophagosome clearance, and it has been well The role of HULC in ovarian carcinoma S Chen et al studied as a negative regulator of autophagy. [37][38][39][40][41][42][43][44] It can also be conjugated to LC3, participating in autophagy. Studies have shown that when LC3-II is upregulated, SQSTM1 (P62) is reduced, indicating that the autophagy is progressing, otherwise the autophagy flow is blocked. 45,46 Following ATG7 knockdown, inhibition of autophagy was verified by LC3-II downregulation and overexpression of the autophagy substrate SQSTM1/P62. 47 Researchers have also shown that LAMP1, a lysosome surface marker, could be used for detecting the combination of autophagy and lysosomes. Our results showed that HULC overexpression reduced ATG7, LC3-II and LAMP1 expression, while inducing SQSTM1 expression. In contrast, HULC downregulation led to ATG7, LC3, LAMP1 overexpression and SQSTM1 inhibition. Furthermore, HULC overexpression in vivo induced tumor formation, and reduced ATG7, LC3 expression, while inducing SQSTM1 expression; however, we found that HULC overexpression did not influence ATG5, ATG12, Beclin-1, VPS34, P150 or UVRAG expression. Therefore, we suggest that the lncRNA HULC may cause changes in cell homeostasis by inhibiting autophagy and promoting ovarian cancer by regulating ATG7, LC3, LAMP1 and SQSTM1 expression.
LncRNAs frequently function both in cis (at the site of their transcription), as well as in trans (at sites on other chromosomes), which highlights potential functions as interfaces with the epigenetic machinery, roles in chromatin organization and regulation of gene expression. The biogenesis of many lncRNAs is similar to mRNAs. 48 It is currently believed that lncRNAs conduct their regulatory functions in the form of RNA-protein complexes through interactions with chromatinmodifying complexes and regulation of gene expression. Our RIP results showed that ATG7, but not LC3, SQSTM1 or LAMP1 could interact with HULC. Our results showed that ATG7 downregulation could also induce ovarian cancer cell proliferation, reduce apoptosis by reducing LC3-II and LAMP1 expression, and induce SQSTM1 expression; furthermore, ATG7 co-transfection with HULC partly reversed the function of HULC in tumorigenesis both in vitro and in vivo. This Figure 9 si-ITGB1 co-transfection with HULC reverses HULC's effect in inducing migration and invasion. Compared with HULC overexpression group, si-ITGB1 cotransfection with HULC reduced ITGB1 expression (a), inhibited migration (b) and invasion ability (c). Results are representative of three separate experiments; data are expressed as the mean ± S.D., *Po0.05 The role of HULC in ovarian carcinoma S Chen et al suggests that HULC may combine with ATG7 and inhibit the ATG7 pathway. Therefore, we suggest that HULC may function as an oncogene and autophagy inhibitor through inhibiting ATG7 in EOC. However, dysfunction of ATG7 does not make sense in ovarian cancer migration and invasion.
Integrins are a large family of cell surface adhesion proteins that are involved in epithelial cell-matrix interactions. The upregulation of integrins is associated with malignancy, particularly during invasion, metastasis and angiogenesis. Increasing evidence suggests that ITGB1 is frequently upregulated in ovarian cancer, and promotes ovarian tumorigenesis and cancer progression. 49 We found that HULC overexpression induces ITGB1 expression, while si-HULC overexpression had the opposite effect. Moreover, si-ITGB1 co-transfection with HULC inhibited the tumor-promoting effect of HULC by inhibiting tumor metastasis and invasion, suggesting that HULC may promote ovarian cancer progression by regulating ITGB1.
The objective of this study demonstrates that HULC may promote ovarian carcinoma tumorigenesis by inhibiting ATG7 and induce progression by regulating ITGB1. We suggest that HULC may be an important diagnostic marker and potential therapeutic target in EOC. The inhibition of HULC expression may prove to be an effective genetic therapeutic strategy for EOC. However, further investigation will be required to elucidate the specific molecular mechanisms involved and to identify potential clinical applications of HULC in the treatment of epithelial ovarian carcinoma (EOC).
Materials and Methods EOC specimens. EOC tissues, borderline tumor tissues, benign tumor tissues and normal ovarian tissue specimens were collected from patients who had undergone surgical resection at the Department of Gynaecology of the First Affiliated Hospital of China Medical University (Shenyang, Liaoning, China). The tumor specimens were independently confirmed by two pathologists. None of the patients had preoperative chemotherapy or radiotherapy. Informed written consent was obtained from all participants and the research protocol was approved by the China Medical University Ethics Committee (no: 2014-27).
Real-time cell analyzer. For real-time cell proliferation assays, 50 μl medium was added to each well of a 96-well E-plate for establishment of background levels. Subsequently, 5 × 10 3 cells in 100 μl medium were seeded per well into the E-plate. After incubation at room temperature for 30 min, the E-plates containing cells were placed on the RTCA SP/MP station positioned in a cell culture incubator. The CI values were measured automatically every 15 min (up to 99 h) to obtain a continuous proliferation curve.
Cell cycle analysis. Cells were fixed with 70% ice-cold ethanol in − 20°C overnight, washed with phosphate-buffered saline (PBS), and then stained with PI (contain RNAase) following the manufacturer's protocol (BD Biosciences, San Jose, CA, USA). The PI signal was examined by a flow cytometry; a total of 10 000 cells were assessed for each sample.
Apoptosis assay. Apoptosis was quantified using 7-AAD/PI staining and flow cytometry with PE/FITC-labeled annexin V (BD Pharmingen, San Diego, CA, USA) following the manufacturer's protocol and flow cytometry. Cells were collected 48 h after transfection, washed twice with cold PBS, and resuspended at 1 × 10 6 cells/ml and mixed with 100 μl of 1 × buffer and 5 μl Annexin V-PE/FITC and 7-AAD/PI, incubated for 15 min in the dark, 400 μl 1 × buffer was added, and the cells were subjected to cytometry flow within 1 h.
Wound-healing assay. Cells were cultured to 80% confluence in a six-well culture dish. After scratching with a 200 μl pipette tip, cells were washed with PBS and cultured in FBS-free medium. Wounds were observed by microscopy and photographed at 0 and 24 h. The wound areas were measured using Image J software (National Institutes of Health, Bethesda, MD, USA). The rate of wound healing was calculated as: area of original wound − area of actual wound at different times)/area of original wound × 100%. Autophagolysosome detection by transmission electron microscopy. Cells were fixed in 0.2 % glutaraldehyde in PBS (pH 7.4) for 2 h at room temperature, postfixed in 1 % osmium tetroxide in water for 1 h and then stained in 2 % uranyl acetate in water for 1 h in the dark. After dehydration in an ascending series of ethanol, the samples were embedded in Durcupan ACM for 6 h, and cut into 80 nm sections. These sections were stained with uranyl acetate and lead citrate, and examined with a transmission electron microscope (Philips CM, Amsterdam, The Netherlands).
Real-time RT-PCR. Total RNA was isolated from ovarian cancer cell lines or tissues using TRIzol Reagent (Takara, , Shiga, Japan) and was reverse transcribed to cDNA using an avian myeloblastosis virus reverse transcriptase and random primers (Supplementary Table 1) according to the manufacturer's protocol. The cDNA was then amplified by real-time quantitative PCR using a SYBR Premix Ex Taq II kit (Takara). Expression levels of each target gene was normalized to 18S mRNA. The data analysis was performed based on the sample threshold cycle (Ct) value from three independent experiments.
Nude mouse xenograft assay. All animal experiments were undertaken in accordance with the National Institutes of Health Guide for the Care and Use of The role of HULC in ovarian carcinoma S Chen et al Laboratory Animals, with the approval of the China Medical University Animal Care and Use Committee. Female BALB/c nude mice, 4 weeks old were obtained from Vital River Laboratories (Beijing, China) and were routinely housed in rooms that were temperature-and light-controlled (12-h dark/12-h light). Animals had free access to food and water. A total of 1 × 10 7 cells, resuspended in 200 μl FBS-free culture medium were injected subcutaneously into the right flanks of the mice. The tumor volume was directly measured following inoculation and weight calculated using the formula: (length × width 2 )/2.
RNA-binding protein immunoprecipitation assay. The RIP assay was performed using the Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore, Bedford, MA, USA) following the manufacturer's protocol. Briefly, cells at 80-90% confluency were collected and lysed using RIP lysis buffer. One hundred microliters of cell extract was then incubated with RIP buffer containing magnetic beads conjugated to human anti-ATG7 antibody or negative control normal mouse IgG. The samples were incubated with Proteinase K to digest the protein and then the immunoprecipitated RNA was isolated. The purified RNA was used for qRT-PCR analysis.
Immunohistochemistry. Paraffin-embedded tissue sections were deparaffinized in xylene and rehydrated in a graded series of ethanol solutions and then incubated for 20 min in 3% H 2 O 2 to quench the endogenous peroxidase activity. Next, the sections were heated in target retrieval solution (Dako, Carpinteria, CA, USA) for 15 min in a microwave oven (Oriental Rotor, Tokyo, Japan) to retrieve the antigen. Nonspecific binding was blocked by incubating with 10% goat serum for 2 h at room temperature. The slides were then incubated overnight at 4°C with anti-ATG7, LC3, ITGB1 or SQSTM1 (1:100) primary antibody, after which secondary antibody was added and incubated for 20 min at 37°C, and the binding was visualized with 3, 39-diaminobenzidine tetrahydrochloride. After each treatment, the slides were washed three times with TBST for 5 min.
IF staining. IF staining was performed according to the standard protocol (Santa Cruz Biotechnology). Briefly, a 5-μm frozen section from each sample was fixed in acetone at 4°C overnight. After washing with PBS three times, sections were blocked with 1% bovine serum albumin for 30 min and incubated overnight at 4°C with rabbit anti-human LC3, SQSTM1, ATG7 or LAMP1 primary antibody (1 : 50). Following PBS washes, the sections were incubated with anti-rabbit IgG-TRITC (1 : 100, Santa Cruz Biotechnology) for 2 h at room temperature in the dark, and then washed again with PBS. Nuclei were stained with diamino-phenyl-indole (DAPI, 1 μg/ml; Sigma-Aldrich, St Louis, MO, USA) for 5 min at room temperature. Coverslips were mounted with SlowFade Gold Antifade Reagent (Invitrogen, Carlsbad, CA, USA), and all sections were imaged using a laser confocal microscope (Olympus, Tokyo, Japan).
Statistical analyses. Statistical analyses were performed using SPSS 17.0 (SPSS, Chicago, IL, USA). Spearman's correlation test was used to analyze the rank data and a Mann-Whitney U-test used to differentiate the means of different groups. Kaplan-Meier survival plots were generated and comparisons were constructed with log-rank statistics. All data are shown as mean ± S.D. from at least three separate experiments. Po0.05 was considered statistically significant.

Conflict of Interest
The authors declare no conflict of interest.