Long non-coding RNA UBE2CP3 enhances HCC cell secretion of VEGFA and promotes angiogenesis by activating ERK1/2/HIF-1α/VEGFA signalling in hepatocellular carcinoma

Antibodies against ERK (Proteintech, USA), p-ERK (Cell Signalling Technology, Beverly, MA, USA), P70S6K (Cell Signalling Technology), p-P70S6K (Cell Signalling Technology), HIF-1α (Abcam, Cambridge, UK), VEGFA (ABclonal, Wuhan, China) were used in Western blotting (WB). CD31 (Abcam) antibody was used for IHC. PD98059(Selleck Chemicals, Houston, USA) were used to inhibit p-ERK. Neutralizing antibody to VEGFA165 was used to block the effects of VEGFA in cell supernatant (R&D systems, Minneapolis, USA).

Two independent cohorts comprising a total of 94 HCC patients were enrolled in this study. In cohort 1, formalin-fixed, paraffin-embedded tissues from 48 HCC patients were used for IHC and ISH. All patients in cohort 1 were followed up for 5 years after surgery; the detailed clinical information for these patients is shown in Table 1. In cohort 2, fresh HCC samples and corresponding para-tumor tissues were obtained from 46 HCC patients who were undergoing routine surgery at Nangfang Hospital, Southern Medical University (Guangzhou, China); the clinical information for these patients is shown in Table 2. All patients provided written informed consent, and the research protocol was approved by the Nangfang Hospital (Guangzhou, China) Ethics Committee.

IHC assays were performed with anti-VEGFA antibody and CD31/periodic acid-Schiff (PAS) double-staining. The ISH probe used for detecting UBE2CP3-labelled digoxin was designed and synthesized by Exiqon (Shanghai, Chia). The probe sequence is listed in Additional file 1: Table S1. ISH was performed using an ISH Kit (Boster Bio-Engineering Company, Wuhan, China) in accordance with the manufacturer’s instructions. The scoring for staining intensity was as follows: 0 (negative staining), 1 (weak), 2 (medium), 3 (strong) (Fig. 1c). The score of staining extent was as follows: 0 (<10%), 1 (11%-25%), 2 (26%-50%), 3 (51%-75%), and 4 (76%-100%). The final UBE2CP3 expression score was calculated as the intensity score × the extent score, and it ranged from 0 to 12. Sections with a total score of 6 or higher were considered as the high expression group, and those with a score less than 6 were categorized as the low expression group. The IHC and ISH scores were evaluated by two pathologists in a blinded manner. When their opinions were inconsistence, a third pathologist who was also blinded to the patient information was asked to give the final score.

Human HCC cell lines (HepG2 and SMMC-7721) and HUVECs were purchased from the Cell Bank of Type Culture Collection (CBTCC, Chinese Academy of Sciences, Shanghai, China). Cells were grown in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, Gaithersburg, MD, USA) containing 10% foetal bovine serum (FBS) in a humidified incubator at 37 °C with 5% CO₂.

To obtain cell lines stably overexpressing UBE2CP3, HepG2 and SMMC7721 cells were infected with Lv-UBE2CP3 and Lv-control viruses (Land, Guangzhou, China). To study the knockdown effects of UBE2CP3 in vitro, HepG2 and SMMC7721 cells were transfected with shRNA-UBE2CP3 (Sh-UBE2CP3) or control (Sh-control) viruses (Obio Technology, Shanghai, China). The infection efficiency was confirmed by qRT-PCR.

Co-culture inserts (0.4-μm pores; Corning, USA) were placed into 6-well culture plates. HUVECs were added to the lower culture wells (2 × 10 ^[5] cells per well), and HepG2 and SMMC7721 cells (2 × 10 ^[5] cells per well) were placed in the inserts and cultured in DMEM supplemented with 5% FBS for 48 h; then, the HUVECs were harvested for biological function studies.

The supernatants from either UBE2CP3 overexpressing or UBE2CP3 knockdown HepG2 and SMMC-7721 cells were collected. Cell culture supernatants were centrifuged at 1,000 g for 10 min to remove the cells and cell debris; then, the supernatants were added to an Amicon Ultra filter device (Millipore, Billerica, MA, USA) and were centrifuged at 4,000 g for 30 min to obtain the concentrated, protein-containing liquid.

VEGFA in the medium was measured by using the Quantikine human VEGF ELISA kit (R&D Systems, Minneapolis, USA) according to manufacturer’s instruction.

Total RNA from cultured cells or human tissue was extracted using TRlzol reagent (Takara, Dalian, China) according to the manufacturer’s instructions. The mRNA levels were measured by qRT-PCR using SYBR Green PCR Master Mix (Takara) by with an ABI 7500 Fast Real-Time PCR system. The expression of β-actin was used as the internal control for analysing the mRNA level of VEGFA. The expression of U6 was used as the internal control for analysing lncRNA UBE2CP3 expression in tissues and cells. Comparative quantification was determined using the 2^-ΔΔCT method. All samples were measured with at least three independent experiments, and the results are expressed as the means ± SD for comparative analysis. The primers used are listed in Additional file 1: Table S1.

Proteins from HepG2 and SMMC7721 cells and from the concentrated cell culture supernatant were extracted with protein extraction kits (Keygen Biotech, Jiangsu, China) containing protease and phosphatase inhibitors. The protein concentration was quantified using a bicinchoninic acid (BCA) protein assay kit (Keygen Biotech). Then, proteins were separated on a 10% sodium dodecyl sulfate-polyacrylamide gel and transferred onto polyvinylidene difluoride membranes. The membranes were blocked with 5% BSA for 1 h at room temperature and incubated with primary antibodies overnight at 4°C. Next, the membranes were washed with TBST three times and incubated with goat anti-rabbit secondary antibodies for 1 h at room temperature; the protein bands were visualized using a chemiluminescence detection system (ECL Plus Western Blot Detection System; Amersham Biosciences, Foster City, CA) according to the manufacturer's protocol. Quantification was completed by Image J software.

Following the manufacturer’s instructions, we used the Cell-Light EdU imaging detection kit (RiboBio, Guangzhou, China) for EdU incorporation assays. HUVECs were harvested from the co-culture system, and after being washed twice with PBS, the cells were seeded into 96-well plates at a density of 5000 cells/well. Six h later, we added EdU labelling medium and incubated the cells for approximately 60 min after they adhered to the plate surface; then, the cells were fixed with 4% formaldehyde for 15 min, treated with 0.5% Triton X-100 for 20 min and exposed to Apollo reaction cocktail for 30 min. Next, in each well, the cellular DNA was stained with Hoechst 33342 (5 up/mL) for 30 min. Finally, we visualized the results under a fluorescence microscope.

Cells were harvested from the co-culture system; after being washed twice with PBS, the cells were fixed in 1 ml of 70% ice-cold ethanol and were stored overnight at 4°C. Subsequently, the cells were stained with propidium iodide supplemented with RNaseA (Keygen Biotech) for 30 min at 37°C. The DNA content of the labelled cells was analysed using FACS flow cytometry (BD Biosciences Inc., Franklin Lakes, NJ, USA). Each experiment was performed three times.

For tube formation assay, 50 μl of Matrigel (BD Biosciences, San Jose, CA, USA) was added to each well of a 96-well plate. After polymerizing at 37°C for 30 min, HUVECs were harvested from the co-culture system and were seeded onto the Matrigel at the density of 30,000 cells per well. After incubation for 6 h in a humidified incubator at 37°C with 5% CO₂, tubules were visualized under an IX71 inverted microscope.

All cell migration assays in vitro were performed using Transwell chambers (8-μm pore size; Corning) in 24-well plates. After being suspended in DMEM with 10% FBS, HCC cells with UBE2CP3 overexpressed or knocked down along with the corresponding control groups were seeded at 50,000 cells per well into the bottom chamber. These HCC cells were cultured in the lower chamber for 24 h; then, in the upper chamber, we added a total of 30,000 HUVECs that had been harvested from the co-culture system. After an18-h incubation, the cells in the chamber were fixed with methanol, followed by staining with 0.1% crystal violet. Cells on the upper side of the Transwell membrane were wiped off with a cotton swab, and cells on the underside were photographed under a microscope and quantified.

HUVECs were harvested from the co-culture system, and each group was washed, suspended in DMEM, and seeded at a density of 500,000 cells per well in 6-well culture plates. After culturing for 24 h and reaching 100% confluence, wounds were made with a 200-μl plastic pipette tip. After being photographed at h 0, co-culture inserts with HepG2 or SMMC-7721 cells (at a density of 200,000 cells per insert) were placed into 6-well culture plates. After incubating in a humidified incubator at 37°C with 5% CO₂ for 18 h, the wound width was measured and recorded using an IX71 inverted microscope. Lastly, the results from three wound healing assays were calculated and analysed.

Our animal investigations were performed in accordance with the institutional guidelines and were approved by the Animal Experimental Committee of Nanfang Hospital. Four-week-old male BALB/c nude mice were purchased from the Guangdong Experimental Animal Center of the Chinese Academy of Sciences and were bred and maintained in specific pathogen-free conditions. Cells from each group were resuspended in serum-free DMEM at a density of 50,000,000 cells per ml and then 0.1 ml of the suspension was injected into the back of the nude mice. Tumor tissues were obtained 5 weeks later for IHC.

After being cultured at 37 °C for 7 days, a window was opened on the shell of fertilized eggs to expose the CAM and then was covered with a filter paper disc (0.5 cm in diameter) containing HepG2 cells (18,000/disc) or PBS on the surface. Next, tape was used to cover the window for further incubation. Two days later, the CAM was fixed in 3.7% formaldehyde, and we visualized the results under a stereoscope.

Data were shown as the means±SD of at least three independent experiments. Statistical analysis was performed using SPSS 13.0 software (SPSS Inc., Chicago, IL, USA) and GraphPad Prism (GraphPad Software, Inc., La Jolla, CA, USA). The chi-Square test was used to examine the relationship between EV density, UBE2CP3 expression and clinicopathological characteristics. Differences between experimental groups were assessed by Student’s t-test or one-way ANOVA. Kaplan-Meier and log-rank tests were used to analyse survival time. Statistical significance was set at *P < 0.05, **P < 0.01, ***P < 0.001. P < 0.05 was considered statistically significant.

To identify the relationship between UBE2CP3 expression and HCC angiogenesis, we examined UBE2CP3 expression levels by ISH and qRT-PCR. We also concomitantly used CD31/PAS double-staining to measure EV density and used qRT-PCR to measure the CD31 mRNA level. The representative images are shown in Fig. 1a. UBE2CP3 expression was higher in HCC tissues than in para-tumor tissues (Fig. 1b, c and e) and was up-regulated in high EV density tissues compared with its level in low EV density tissues (Fig. 1b, d and f). Moreover, the levels of UBE2CP3 were positively correlated with CD31 mRNA levels (R ^[2]=0.3063, P<0.001 Fig. 1g). Kaplan-Meier and log–rank test analyses revealed that the high level of UBE2CP3 (Fig. 1h) and high EV density (Fig. 1i) were associated with reduced overall survival time (OS). When grouped by both UBE2CP3 expression and EV density, HCC patients in the high category had poor OS (Fig. 1j). The correlations among UBE2CP3 expression levels, EV density and the clinicopathological parameters of the HCC patients in cohort 1 are shown in Table 1. The correlations among UBE2CP3 expression levels, CD31 mRNA level and the clinicopathological parameters of the HCC patients in cohort 2 are shown in Table 2. Chi-square test revealed that both the levels of UBE2CP3 (P=0.003,Table 1; P=0.014, Table 2) and EV density (P=0.031,Table 1; P=0.015, Table 2) were significantly correlated with tumor numbers. Furthermore, the levels of UBE2CP3 were positively correlated with mortality (P=0.014,Table 1) and tumor invasion (P=0.037, Table 1; P=0.019, Table 2); and EV density was significantly correlated with tumor size (P=0.033, Table 1; P=0.026, Table 2).

Our ISH and IHC results indicated that UBE2CP3 may participate in the HCC angiogenic process. The indirect effects of UBE2CP3 on ECs were studied by establishing a co-culture system using Transwell chambers with a 0.4-μm pore size. HCC cells that had been transfected with Lv-UBE2CP3 or Sh-UBE2CP3 virus (the infection efficiencies are shown in Additional file 2: Figure S1A) were seeded into the upper chamber, and ECs were seeded into the lower chamber. HCC cells and ECs from the co-culture system were separated, but the cytokines and growth factors were able to communicate with each other (Additional file 2: Figure S1B). In the co-culture system, ECs were exposed to HCC cells for 48 h. Following this exposure, ECs were removed from the co-culture system for further studies. Since angiogenesis involves EC proliferation, migration and tube formation, we analysed the cell cycle and performed EdU, Transwell, wound healing and tube formation assays to gain insight into the role of UBE2CP3 in cell proliferation, migration and tube formation. We observed that compared with the control group, EC cells co-cultured with HCC cells stably overexpressing UBE2CP3 had significantly enhanced proliferation (Fig. 2a, b), migration (Fig. 2c, d) and tube formation potential (Fig. 2e). Conversely, ECs co-cultured with the UBE2CP3-knockdown HCC cells had decreased proliferation (Fig. 2a, b), migration (Fig. 2c, d) and tube formation abilities (Fig. 2e). Our results suggested that the dysregulation of UBE2CP3 in HCC cells may play an indirect effect on ECs proliferation, migration and tube formation in vitro.

ECs co-cultured with UBE2CP3-dysregulated HCC cells had alterations in cell proliferation, migration and tube formation abilities. We hypothesized that cytokines and/or growth factors may participate in this process. As shown in Additional file 2: Figure S1C, 12 kinds of common angiogenic factors were detected using qRT-PCR, the mRNA levels of VEGFA (up-regulated fold=4.04, P=0.008), Ang2(up-regulated fold=1.82, P=0.031) were increased in UBE2CP3 overexpressing HepG2 cells and the mRNA levels of VEGFA(down-regulated folds=0.31, P=0.008), Ang2(down-regulated folds=0.63, P=0.037), and PDGFC(down-regulated folds=0.49, P=0.038) were decreased in UBE2CP3 knocking down HepG2 cells. Since VEGFA is one of the most important growth factors in angiogenesis and is the most obvious changing cytokine in UBE2CP3 overexpressing or knocking down HepG2 cells, we decided to investigate the role and the underlying mechanism of VEGFA in UBE2CP3 inducing angiogenesis in HCC. The VEGFA levels in the Lv-control, Lv-UBE2CP3, sh-control, and sh-UBE2CP3 cell supernatants from the co-culture system after concentrated by an Amicon Ultra filter device were determined by WB and ELISA assay. As expected, a higher level of VEGFA was observed in the UBE2CP3 overexpressing HCC cell supernatant than in the supernatant of the control cells. In contrast, in the co-culture system, knocking down UBE2CP3 induced a decrease in the supernatant VEGFA level (Fig. 3a, b). To investigate whether VEGFA take part in UBE2CP3 inducing EC proliferation, migration and tube formation, VEGFA neutralizing antibodies were used to block the VEGFA effects in the co-culture system. When adding VEGFA neutralizing antibodies, the indirect effects of UBE2CP3 on EC proliferation, migration, and tube formation in the co-culture system were significantly reduced. (Fig. 3c, d, e). Meanwhile, the WB assays with HCC cell protein showed higher levels of phosphor -ERK (p-ERK), phosphor-p70S6K (p-p70S6K), HIF-1α, and VEGFA in the UBE2CP3 overexpressing cells than in the control cells. In contrast, knocking down the expression of UBE2CP3 reduced the levels of p-ERK, p-p70S6K, HIF-1α, and VEGFA (Fig. 3b, c) compared with their expression in the control cells (Fig. 3f). After treated with 100μM PD98059 for 48h, the level of p-ERK was significantly reduced, and the levels of p-p70S6K, HIF-1α, and VEGFA can also be reduced by the inhibitor (Fig. 3g).

To determine the effects of UBE2CP3 on HCC angiogenesis, we injected HepG2 cells that were stably transfected with Lv-UBE2CP3 or Lv-control into chick chorioallantoic membranes (CAM). Chick embryos injected with cells overexpressing UBE2CP3 had an increase in new vessel density (Fig. 4a). In contrast, when chick embryos were injected with cells with inhibited UBE2CP3 expression, a decrease in new vessel density was observed (Fig. 4b). Consistent with these results, mice injected with cells overexpressing UBE2CP3 had a higher EV density in the tumor tissue than the mice injected with Lv-control cells (Fig. 4c). Mice injected with cells knocking down UBE2CP3 had a lower EV density in the tumor tissue than the mice injected with sh-control cells (Fig. 4d).