Long noncoding RNA highly upregulated in liver cancer promotes the progression of hepatocellular carcinoma and attenuates the chemosensitivity of oxaliplatin by regulating miR‐383‐5p/vesicle‐associated membrane protein‐2 axis

Abstract We aimed to explore the function and underlying mechanism of highly upregulated in liver cancer (HULC; an long noncoding RNAs) in hepatocellular carcinoma (HCC) and chemosensitivity of oxaliplatin (Oxa). The expression of HULC, miR‐383‐5p, and vesicle‐associated membrane protein‐2 (VAMP2) was detected by quantitative real‐time polymerase chain reaction. Western blot assay was applied for measuring the protein expression of cyclinD1, cleaved‐caspase‐3, light Chain 3 I/II, p62, and VAMP2. Cell viability and Oxa IC50 value were determined by Cell Counting Kit‐8 assay. A colony formation assay was conducted to evaluate colony formation ability. Cell apoptosis was assessed by flow cytometry. The interaction between miR‐383‐5p and HULC or VAMP2 was predicted by bioinformatics analysis and verified by dual‐luciferase reporter assay and RNA immunoprecipitation assay. The mice xenograft model was established to investigate the roles of HULC in vivo. HULC and VAMP2 were overexpressed whereas miR‐383‐5p was lowly expressed in HCC tissues. HULC overexpression promoted the progression of HCC cells and inhibited chemosensitivity of Oxa by increasing cell proliferation and protective autophagy and inhibiting apoptosis, whereas HULC silence presented opposite effects. Moreover, miR‐383‐5p was a direct target of HULC and miR‐383‐5p reversed the effects of HULC on the progression of HCC cells and chemosensitivity of Oxa. Besides, HULC acted as a molecular sponge of miR‐383‐5p to regulate VAMP2 expression. HULC promoted the progression of HCC and inhibited Oxa sensitivity by regulating miR‐383‐5p/VAMP2 axis, elucidating a novel regulatory mechanism for chemosensitivity of Oxa and providing a potential lncRNA‐targeted therapy for HCC.


| INTRODUC TI ON
Hepatocellular carcinoma (HCC) is one of the most frequent human malignancies in the world and it accounts for 85%-90% of primary liver cancer. 1,2 Over the past few decades, despite advances in the treatment of HCC, such as operation technique, chemotherapeutic, and radiotherapeutic approaches, the prognosis remains very poor because most of the HCC patients are usually diagnosed at advanced stages. 3 Autophagy, a conserved catabolic process, is responsible for providing nutrition (ATP, amino acids, etc.) for cell survival by disposing and recycling cellular proteins and damaged/ excessive organelles during nutritional deficiency. 4 Previous studies have demonstrated that multiple chemotherapeutic drugs, such as cisplatin, 5-fluorouracil, and sorafenib can induce autophagy in HCC cells, decreasing cell apoptosis and chemosensitivity, resulting in the survival of HCC cells. 5,6 Oxaliplatin (Oxa), a third-generation platinum anti-cancer drug, is extensively used in chemotherapeutic regimes to treat advanced HCC. 7 However, Oxa resistance is one of the most vital barriers for patients with HCC. Hence, it is essential to better understand the molecular mechanisms of HCC and improve the chemosensitivity of Oxa.
Long noncoding RNAs (lncRNAs), one type of endogenous cellular RNAs (>200 nucleotides), lack protein-coding potential and have pivotal roles in a variety of biological processes. 8 LncRNAs have been suggested to participate in multiple cellular processes, such as cell differentiation, cell cycle, apoptosis, and autophagy. 9 Besides, some reports have shown that lncRNAs also could regulate chemosensitivity in cancer cells. 10,11 LncRNA highly upregulated in liver cancer (HULC), an important lncRNA with a wide variety functions, is specifically overexpressed and serves as an oncogenic lncRNA in human HCC. 12 Although HULC-triggered autophagy has been shown to reduce the chemosensitivity of Oxa, 13 more functions of HULC and its underlying mechanism are still largely unknown in HCC progression and chemosensitivity of Oxa.
A large number of evidence has shown that lncRNAs can function as competing endogenous RNAs (ceRNAs) through sponging microRNAs (miRNAs) to modulate gene expression. 14,15 MiRNAs are a kind of short noncoding RNAs and modulate gene levels via binding with the 3′-untranslated region (3′UTR) of target mRNAs, resulting in mRNAs degradation or suppression of their translation. 16,17 Emerging evidence has revealed that aberrant expression of miR-NAs is tightly related to the occurrence and development of HCC as well as chemosensitivity. 18,19 MiR-383-5p has been suggested to play an anti-cancer role in many tumors, including HCC. 20 However, the interaction between HULC and miR-383-5p and the biological functions of miR-383-5p in chemosensitivity of Oxa remain unclear.
Vesicle-associated membrane protein-2 (VAMP2) has been demonstrated to be expressed at a high level in HCC tissues and cells. In view of these findings, we are interested in whether miR-383-5p exerts its effect by regulating VAMP2 expression.
In this research, we determined the abundance of HULC, miR-383-5p, and VAMP2 in HCC tissues and HCC cells treated with Oxa.
Moreover, the roles of HULC, miR-383-5p, and VAMP2 in cell growth, apoptosis, and autophagy were examined in HCC cells treated with Oxa. Besides, we also probed the regulatory network of HULC/miR-383-5p/VAMP2 in HCC cells. The aim of this study was to provide a new insight and treatment strategy for Oxa-based chemotherapeutics.

| Cell culture and transfection
Human HCC cells (Hep3B and Huh7) and human embryonic kidney cells (293T) were bought from COBIOER. Cells were grown in RPMI 1640 medium (Invitrogen) containing 10% (v/v) fetal bovine serum (Invitrogen) in a humidified atmosphere with 5% CO 2 at 37°C. Oxa (Sigma-Aldrich) was applied to treat Hep3B and Huh7 cells.

| Quantitative real-time polymerase chain reaction
Trizol reagent (Invitrogen) WAS applied for isolating total RNA.
After that, the RNA samples were reversely transcribed to complementary DNA using the TIANScript RT Kit (Tiangen Biotech).

| Western blot assay
Tissue samples and cultured cells were lysed using the RNA immunoprecipitation assay (RIPA) buffer mixed with protease inhibitors

| Cell viability assay
Cell Counting Kit-8 (CCK-8; Beyotime) was applied for measuring cell viability. In brief, Hep3B and Huh7 cells (5 × 10 3 per well) were plated in a 96-well plate. CCK-8 (10 μl) reagent was added to the wells after treatment, followed by incubation for 3 h. The optical density value at 450 nm was recorded under a microplate reader (Bio-Rad). Oxa concentration causing 50% inhibition of growth (IC50) was measured using the relative survival curve.

| Colony formation assay
Transfected Hep3B and Huh7 cells were plated in 6-well plates replacing the medium every 3 days. Following incubation for 2 weeks, these cells were fixed using paraformaldehyde (4%) and stained with Giemsa solution (Sigma-Aldrich). After that, the number of colonies (a colony was defined as >50 cells) was calculated. Lastly, the cells were subjected to flow cytometry (Partec AG) for detecting the number of apoptotic cells.

| RIP assay
RNA immunoprecipitation assay was performed using the EZ-Magna RIP Kit (Millipore). First, Hep3B and Huh7 cells were harvested, lysed by complete RIP lysis buffer, followed by incubation with RIP buffer supplemented with magnetic beads conjugated to a human anti-Ago2 anti-Ago or anti-IgG (as control). After that, these samples were incubated using the proteinase K to digest proteins, followed by isolating the immunoprecipitated RNA. The enrichment of miR-383-5p or HULC was measured using qRT-PCR.

| Statistical analysis
The data were presented as mean ± standard deviation and the experiments were repeated at least 3 times in this study.
Correlation between miR-383-5p and HULC or VAMP2 was analyzed using the Spearman rank correlation. Statistical analyses were performed by GraphPad Prism 6.0 (GraphPad Software Inc.). The Student's t-test (for 2 groups) or a one-way analysis of variance (ANOVA; for multiple groups) was used to compare values of test and control samples. p < .05 was regarded as statistically significant.

| HULC was overexpressed and Oxa-activated autophagy in HCC
To explore the potential role of HULC in HCC, we investigated HULC expression in HCC tissues and the corresponding adjacent normal tissues. The results of qRT-PCR showed that HULC expression was obviously higher in HCC tissues than that in normal tissues ( Figure 1A).
Next, we explored whether autophagy was associated with HCC.
Western blot was used to measure the levels of autophagy-related proteins (LC3 and p62). LC3 is a special protein in the early stage of autophagy, and LC3I is converted to LC3II during autophagy, and p62 degradation is another marker of autophagy. 21,22 The data presented that LC3II/LC3I ratio was enhanced, suggesting that LC3I was converted to LC3II, as well as p62 expression was decreased in HCC tissues relative to normal tissues ( Figure 1B). To investigate the effect of Oxa on cell viability, CCK-8 assay was conducted in Hep3B and Huh7 cells treated with different concentrations of Oxa (0, 0.5, 1, 2, 5, 10, and 20 μM). The results presented that Oxa IC50 value was 4.48 ± 0.65 (μM) in Hep3B and 7.69 ± 0.54 (μM) in Huh7 cells ( Figure 1C). Moreover, we found that HULC had a pivotal role in the sensitivity of HCC cells in response to Oxa. As presented in Figure 1D, Oxa dramatically enhanced the expression of HULC in Hep3B and Huh7 cells in a dose-dependent manner. Furthermore, we uncovered that Oxa treatment dose-dependently increased LC3II/LC3I ratio and decreased p62 expression in Hep3B and Huh7 cells ( Figure 1E), indicating that Oxa treatment induced protective autophagy. These results suggested that HULC might act as an oncogene in HCC.

| HULC promoted proliferation and autophagy whilst inhibited apoptosis and chemosensitivity of Oxa
First, qRT-PCR was used to determine the transfection efficiency of oe-HULC and si-HULC. As displayed in Figure 2A

| MiR-383-5p was a direct target of HULC
To explore the expression of miR-383-5p in HCC, we performed qRT-PCR assay. As illustrated in Figure 3A, the level of miR-383-5p was decreased in HCC tissues in comparison with normal tissues.
Moreover, Oxa treatment decreased the expression of miR-383-5p in a dose-dependent manner in Hep3B and Huh7 cells ( Figure 3B). In addition, the correlation between miR-383-5p and HULC expression was analyzed in HCC tissues. As shown in Figure 3C, miR-383-5p abundance was negatively correlated with HULC level in HCC tissues (r = −.6463, p < .0001). Next, we further explored the interaction between miR-383-5p and HULC in HCC. Bioinformatics analysis (miRcode) provided the putative binding sites of miR-383-5p and HULC ( Figure 3D), implying the potential relationship between miR-383-5p and HULC. Subsequently, the prediction was validated by dual-luciferase reporter assay and RIP assay. The data presented that the luciferase activity of HULC-wt was significantly decreased in 293T cells transfected with miR-383-5p mimic, but the luciferase activity of HULC-mut was not evidently affected after transfection with miR-383-5p mimic ( Figure 3E). Moreover, RIP assay showed that enrichment of miR-383-5p and HULC was obviously enhanced in anti-Ago2 group compared with that in anti-IgG group ( Figure 3F).
Taken together, these results indicated that miR-383-5p could bind to HULC.

| HULC exerted its functions by regulating miR-383-5p in HCC cells
QRT-PCR analysis was used to examine the transfection efficiency.
The results showed that transfection of miR-383-5p mimic in-

383-5p
To investigate the expression of VAMP2 in HCC tissues, VAMP2 mRNA and protein levels were determined by qRT-PCR and western blot assays, respectively. The results displayed that mRNA and protein expression of VAMP2 were elevated in HCC tissues with respect to normal tissues ( Figure 5A,B). Moreover, Oxa treatment dose-dependently increased the mRNA and protein levels of VAMP2 in Hep3B and Huh7 cells ( Figure 5C,D). Furthermore, VAMP2 mRNA level was inversely correlated with miR-383-5p expression in HCC tissues ( Figure 5E) (r = −.4349, p = .009). Interestingly, we found that VAMP2 contained putative binding sites of miR-383-5p in the 3′UTR ( Figure 5F). To verify whether VAMP2 was a direct target of miR-383-5p, the dual-luciferase reporter assay was carried out. Results indicated that overexpression of miR-383-5p greatly reduced the luciferase activity of VAMP2 3′UTR-wt in 293T cells, whereas miR-383-5p upregulation had no impact on the luciferase activity of VAMP2 3′UTR-mut ( Figure 5G). In a word, these data indicated that miR-383-5p directly targeted VAMP2.

| MiR-383-5p exerted its effect by modulating YWHA in HCC cells
Next, we further validated the association between VAMP2 and miR-383-5p in HCC cells. The results from qRT-PCR and whereas these effects were abated by downregulating VAMP2 or upregulating VAMP2 ( Figure 6C,D). Subsequently, Hep3B and Huh7 cells were exposed to Oxa (6 μM) for subsequent assays.

| VAMP2 was regulated by HULC and miR-383-5p in HCC cells
Next, we tried to investigate whether HULC served as a ceRNA of miR-383-5p to modulate VAMP2 expression. The results from qRT-PCR and western blot showed that overexpression of HULC increased the mRNA and protein levels of VAMP2 in Hep3B cells, which were weakened by upregulating miR-383-5p ( Figure 7A).

| Overexpression of HULC accelerated tumor growth by regulating miR-383-5p and VAMP2 in vivo
To explore the impact of HULC on tumor growth and chemosensitivity in vivo, we established a xenograft model in which the Hep3B cells stably transfected with Vector or oe-HULC were subcutaneously injected into BALB/c nude mice and treated with Oxa (5 mg/kg) twice a week. In agreement with in vitro results, overexpression of HULC increased tumor volume and weight in the xenograft model compared with the control group ( Figure 8A,B).

| DISCUSS ION
Hepatocellular carcinoma is an aggressive malignant tumor with a high recurrence rate. 23 Although Oxa-based chemotherapy is especially effective in a variety of cancers, chemoresistance development has become a major obstacle in cancer treatment. 24,25 Numerous lncRNAs have been suggested to participate in modulating tumor processes and chemosensitivity. 26,27 In this research, we focused on the biological role and underlying mechanism of HULC in HCC progression and chemosensitivity of Oxa.  29 Protective autophagy has been shown to be a possible mechanism for chemoresistance in various types of cancer cells, and the promotion of apoptosis is often beneficial to improve the sensitivity of chemotherapy drugs. 19,30,31 Notably, Xiong and his colleagues found that HULC-induced protective autophagy weakened the chemosensitivity of antitumor reagents in HCC cells. 13 In line with these findings, our results also proved that HULC was overexpressed in HCC tissues and its overexpression decreased the chemosensitivity of Oxa by triggering autophagy and inhibiting cell apoptosis. Meanwhile, HULC knockdown repressed cell growth and autophagy as well as accelerated apoptosis in HCC cells treated with Oxa, suggesting that HULC knockdown increased the chemosensitivity of Oxa.
Previous reports proved that miR-383-5p acted as an antioncogene in diverse tumors, such as ovarian cancer, 32 breast cancer, 33 and gastric cancer. 34 Besides, Chen et al. stated that miR-383 abundance was declined in HCC tissues and cells and its overexpression limited HCC cell proliferation by targeting APRIL. 20 Wang et al. pointed out that miR-383 limited HCC cell growth and facilitated cell apoptosis via regulating IL-17 and STAT3 signaling pathway. 35 Consistent with these previous findings, we also observed that miR-383-5p abundance was declined in HCC tissues. Besides, miR-383-5p expression was inversely associated with HULC level in HCC tissues. Interestingly, a large number of evidence has suggested that lncRNAs may function as ceRNAs or miRNA sponges to affect miRNAs, resulting in a change in the expression of miRNA F I G U R E 7 HULC positively regulated VAMP2 expression by sponging miR-383-5p in HCC cells. (A) The mRNA expression and protein expression of VAMP2 were determined by qRT-PCR and western blot analyses, respectively, in Hep3B cells transfected with Vector, oe-HULC, oe-HULC + mimic NC, or oe-HULC + miR-383-5p mimic (ANOVA). (B) QRT-PCR and western blot analyses were conducted to analyze the mRNA level and protein level of VAMP2 in Huh7 cells transfected with si-NC, si-HULC, si-HULC + inhibitor NC, or si-HULC + miR-383-5p inhibitor (ANOVA). *p < .05. ANOVA, analysis of variance; HCC, hepatocellular carcinoma; HULC, highly upregulated in liver cancer; NC, negative control; qRT-PCR, quantitative real-time polymerase chain reaction; VAMP2, vesicle-associated membrane protein-2 target gene. 36 To probe whether HULC served as a molecular sponge of miR-383-5p, we predicted their targeting relationship by miRcode. As expected, miR-383-5p was identified as a direct target of HULC. Furthermore, we uncovered that miR-383-5p could abolish the functions of HULC on cell growth, apoptosis, and autophagy as well as chemosensitivity of Oxa. These results indicated that HULC exerted its function by sponging miR-383-5p in HCC cells.
To date, VAMP2 was reported to be aberrantly expressed in some tumors and played important regulative roles. 37,38 More importantly, Wang et al. proved that VAMP2 level was upregulated in HCC tissues and its upregulation reversed the anti-cancer role of miR-493-5p in liver cancer cells. 39 Here, we also uncovered that VAMP2 mRNA and protein expression were enhanced in HCC tissue samples, and its mRNA level was inversely associated with miR-383-5p abundance in HCC tissues. Using the online software starbase v3.0, it was predicted that VAMP2 might bind with miR-383-5p. Next, our results verified that VAMP2 could interact with miR-383-5p. Moreover, rescue experiments indicated that VAMP2 could abate the impact of miR-383-5p on the progression of HCC cells and chemosensitivity of Oxa in HCC cells. Additionally, HULC positively modulated VAMP2 expression via sponging miR-383-5p.
In vivo experiments implied that HULC overexpression promoted tumor growth and inhibited the chemosensitivity of Oxa through downregulating miR-383-5p and upregulating VAMP2. However, HULC has a small effect on tumor growth, and this effect does not necessarily translate into meaningful clinical benefits. Therefore, further research on the role of HULC in the clinic is needed in the future. Taken together, our findings suggested that HULC regulated the progression of HCC and chemosensitivity of Oxa via modulating miR-383-5p/VAMP2 axis. Although lncRNAs have great potential in the diagnosis and prognosis of HCC, when lncRNAs are used as a potential drug candidate or therapeutic target for HCC treatment, we also face a series of challenges, such as lncRNAs are difficult to pass through biofilms and lncRNAs can be quickly eliminated by the liver and kidneys. Presently, there are two main methods to solve these problems: One is to modify nucleic acid molecules to stabilize and avoid recognition by the immune system, the other is to use drug delivery systems, such as nanoparticle vector technology, protamine vector technology, adeno-associated viral vector technology.
Nevertheless, these delivery systems still have defects such as immune system rejection and reduced protein expression efficiency, and future research should focus on developing new methods to overcome these shortcomings.
In conclusion, we discovered that HULC and VAMP2 were overexpressed whilst miR-383-5p was lowly expressed in HCC tissues.
Functionally, HULC overexpression promoted the progression of HCC and attenuated the chemosensitivity of Oxa by regulating miR-383-5p/VAMP2 axis. Hence, our study might contribute to a better understanding of the regulatory mechanism of HCC progression and chemosensitivity of Oxa, offering a promising lncRNA-targeted therapy for HCC.

F I G U R E 8
Upregulation of HULC promoted tumor growth and inhibited chemosensitivity of Oxa by regulating miR-383-5p and VAMP2. Hep3B cells transfected with Vector or oe-HULC were inoculated subcutaneously into the nude mice. The mice were treated with Oxa (5 mg/kg) twice a week after injection for 3 days. (A,B) Tumor volume and weight were measured (ANOVA). (C) The expression levels of HULC, miR-383-5p, and VAMP2 were examined by qRT-PCR in resected tumor tissues (t-test). (D) Western blot assay was applied to detect the protein abundance of VAMP2 in resected tumor masses (t-test). (E) The protein levels of cyclinD1, cleaved-caspase-3, LC3I/II, and p62 were analyzed in resected tumor tissues using western blot analysis (t-test). *p < .05. ANOVA, analysis of variance; HULC, highly upregulated in liver cancer; LC3, light Chain 3; qRT-PCR, quantitative real-time polymerase chain reaction; VAMP2, vesicle-associated membrane protein-2

D I SCLOS U R E
The authors declare that they have no competing interests.

E TH I C S A PPROVA L A N D CO N S E NT TO PA RTI CI PATE
This study was approved by the ethical review committee of First Affiliated Hospital of Xi'an Jiaotong University. Written informed consent was obtained from all enrolled patients.

CO N S E NT FO R PU B LI C ATI O N
Patients agree to participate in this work.

DATA AVA I L A B I L I T Y S TAT E M E N T
The analyzed datasets generated during this study are available from the corresponding author on reasonable request.