Long intergenic noncoding RNA LINC00284 knockdown reduces angiogenesis in ovarian cancer cells via up-regulation of MEST through NF-κB1

Abstract

Ovarian cancer (OC) is one of the major causes of cancer-related mortality in women worldwide. Long noncoding RNAs might play a role as oncogenes or tumor suppressors. Therefore, we investigated the effect and underlying mechanisms of long intergenic noncoding RNA (LINC00) 284 on angiogenesis in OC cells. Expression of LINC00284 in OC tissues and cells was determined. Next, the interaction between LINC00284 and mesoderm-specific transcript (MEST) was evaluated. Subsequently, OC cells were transfected with overexpressed (oe)-LINC00284, silenced (si)-LINC00284, si–NF-κB1, oe-MEST, or si-MEST plasmids to investigate the underlying mechanism of LINC00284 in OC. Afterwards, the expression of matrix metalloproteinase (MMP)-2, MMP-9, B-cell lymphoma 2 (Bcl-2), Bcl-2–associated protein x (Bax), VEGF, and CD31 was determined to assess the effect of LINC00284 on OC cell proliferation, invasion, migration angiogenesis, and apoptosis. Finally, the effect of LINC00284 on tumorigenesis was investigated in nude mice models of OC. LINC00284 was highly expressed in OC. si-LINC00284 increased expression of MEST. si-LINC00284 or si–NF-κB1 led to the reduction in cell proliferation, migration, invasion, tube formation, angiogenesis, and tumorigenic ability and promoted apoptosis in OC by down-regulating MMP-2, MMP-9, Bcl-2, VEGF, and CD31 and up-regulating Bax. These effects were all reversed following the si-MEST. In vivo experiments found the same results, confirming the aforementioned findings. Taken together, LINC00284 is involved in angiogenesis during OC development by recruiting NF-κB1 and down-regulating MEST.—Ruan, Z., Zhao, D. Long intergenic noncoding RNA LINC00284 knockdown reduces angiogenesis in ovarian cancer cells via up-regulation of MEST through NF-κB1.

Ovarian cancer (OC) is one of the most fatal gynecologic carcinomas, accounting for 140,000 deaths and over 230,000 new cases reported annually (1). Most patients are diagnosed with OC in the middle or advanced stage of the disease after the development of intra-abdominal metastasis that is due to the insidious onset, asymptomatic nature of the disease and the lack of effective early detection (2). Despite the progress made in the treatment of OC through surgery and chemotherapy, the 5-yr survival rate remains to be 30% after the initial diagnosis (3). Angiogenesis is a critical process in tumor development and metastasis (4). Recent studies found that angiogenesis can serve as a novel therapeutic target for OC treatment (5). Moreover, the latest discoveries suggest that dysregulation of long noncoding RNAs (lncRNAs) is involved in diagnosis and prognosis of ovarian, lung, gastric, and liver carcinomas (1, 6, 7). Therefore, further understanding of the roles and underlying mechanisms of lncRNAs in OC is required in order to develop early detection methods and novel therapeutic targets for OC.

LncRNAs are recently discovered members of the noncoding RNA family that have been found with the ability to mediate gene expressions in terms of their transcription, posttranscription, and epigenesis (8). LncRNAs play a role in tumorigenesis as tumor inhibitors or oncogenes (9). For example, lncRNA metastasis-associated lung adenocarcinoma transcript 1 (MALAT1) plays a regulatory role in cell proliferation, invasion, and metastasis in OC (10). In addition, long intergenic noncoding RNA (LINC000) 92 exerts its effect on fibroblasts associated with cancer to promote glycolysis and progression of OC (11). In addition, LINC00284 was identified as a promising therapeutic target for gene therapy in OC in a previous study, and its effect has also been demonstrated in gastric cancer (12, 13). Moreover, the interaction between NF-κB and lncRNA (NKILA) has been reported in non-small cell lung cancer cells (14). NF-κB is an essential transcriptional factor that regulates various biologic processes, including immunologic reaction, apoptosis, and cell growth (15). NF-κB has been demonstrated to have significant proliferative function in cell lines of OC (16). Excessive activation of NF-κB pathway has also been observed in aggressive OC (16). According to bioinformatics analysis and dual luciferase reporter gene assay, mesoderm-specific transcript (MEST) was verified as a target gene of NF-κB1. MEST is an imprinted gene mapped to chromosome 7q32, and there is a loss of this imprinting of MEST in several kinds of human cancers, including colorectal cancer and lung cancer (17). In addition, the imprinted gene plays an important role in carcinogenesis through improper regulation of both tumor suppressors and oncogenes, resulting in cell immortalization (18). However, a number of recently conducted studies demonstrated the effect of LINC00284 and MEST on OC. Therefore, we conducted the present study with aims of exploring the effects of LINC00284 on OC and its underlying mechanism involving transcription factor NF-κB1 and the MEST gene.

MATERIALS AND METHODS

Ethics statement

This study was approved by the Ethics Committee of Shanghai Ninth People’s Hospital, School of Medicine, Shanghai Jiaotong University, and all patients signed informed consent prior to the study. All animal experiments were carried out in accordance with the principles and procedures of the National Institute of Animal Health Care Guidelines, and were approved by the Animal Committee of Shanghai Ninth People’s Hospital, School of Medicine, Shanghai Jiaotong University, with the maximum effort of minimizing animal number and pain on the experimental animals.

Human subjects

Tumor tissue samples were collected from 72 patients diagnosed with OC from December 2012 to December 2015. The patients were between the ages of 24 and 78 yr, with the median age of patients being 51 yr, out of which 34 patients were over the age of 51 yr and 38 patients were under the age of 51 yr. There were 25 cases of stage I, 27 cases of stage II, and 20 cases of stage III. Meanwhile, 72 cases of paracancerous tissues were collected from corresponding OC patients as control. There was no significant difference in age among all groups (P < 0.05). All patients had complete data and did not receive radiotherapy before surgery.

Cell culture and transfection

Human OC cell SKOV3 (BNCC310551), A2780 (BNCC100884), ovcar3 (BNCC287606), H0-8910 (BNCC100717), Caov-3 (BNCC101649) (American Type Culture Collection, Manassas, VA, USA), and normal human ovarian epithelial cells (HOEpiC) were used for the purpose of this study. The expression of LINC00284 was detected in the above cell lines with the use of quantitative real-time PCR (qRT-PCR). The cell line with the highest expression of LINC00284 was selected for subsequent experiments.

LINC00284, NF-κB1, and MEST interference and overexpression sequences were constructed by Shanghai Sangon Biotech (Shanghai, China) based on the known sequences in the National Center for Biotechnology Information (Bethesda, MD, USA). OC cells in logarithmic phase were transfected with negative control (NC) plasmids [overexpressed (oe)-NC and silenced (si)-NC], or oe-LINC00284, si-LINC00284, si–NF-κB1, and oe-MEST plasmids. Target plasmids were purchased from Dharmacon Research (Lafayette, CO, USA). The procedure was briefly described as follows: 1 d before transfection, HO-9810 cells were inoculated into a 6-well plate at the density of 3 × 10⁵ cells/well. On the second day, when cell confluence reached 80%, cell transfection was conducted using Lipofectamine 2000 (Thermo Fisher Scientific, Waltham, MA, USA). Next, 250 μl Opti-MEM (Thermo Fisher Scientific) was added into two 1.5 ml eppendorf tubes. One tube was added with 4 nmol target plasmid, and another one was added with 10 μl Lipofectamine 2000, which were both allowed to stand at room temperature for 5 min. Afterwards, the above 2 tubes were uniformly mixed, allowed to stand at room temperature for 15 min, and added into the culture plate. Following incubation of the cells in a 5% CO₂ incubator at 37°C for 6 h, the culture medium was replaced. The cells were collected after 36–48 h of transfection.

Microarray expression analysis of OC-related lncRNAs

The data of OC-related chip were downloaded from Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo) database. The R language affy package (19) was used for standardized pretreatment of the chip expression data. The Limma package (20) was used to screen differentially expressed lncRNA. The corrected P value was expressed as adj.P.Val, and genes with |log2FC| >1.5 and adj.P.Val <0.05 were differentially expressed. A heatmap of lncRNA expression was drawn. LncMAP (http://bio-bigdata.hrbmu.edu.cn/LncMAP/) website was used to predict possible regulation mechanisms of lncRNA.

qRT-PCR

Total RNA was extracted by using Trizol (15596026; Thermo Fisher Scientific). UV spectrophotometry (UV1901; Shanghai AuCy Scientific Instrument, Shanghai, China) was used to detect purity and concentration of RNA. The purity of all samples (A₂₆₀/A₂₈₀ = 1.8–2.0) was adjusted into 50 ng/μl. RNA was then reversely transcribed to cDNA using a RT kit (RR047A; Takara Bio, Kusatsu, Japan). The samples were loaded using a SYBR Premix EX Taq Kit (RR420A; Takara Bio). qRT-PCR reaction was performed using ABI7500 quantitative fluorescence PCR system (Thermo Fisher Scientific). Primers of LINC00284, NF-κB1, MEST, VEGF, CD31, and β-actin were synthesized by Shanghai Sangon Biotech (Table 1). The samples were detected in triplicates. With β-actin used as the internal reference, changes in mRNA expression were calculated by means of relative quantification (2^−ΔΔCt method) (21).

TABLE 1.

Primer sequences for qRT-PCR

	Primer sequence, 5′–3′
Gene	Forward	Reverse
LINC00284	CCAGGGGATAAAACCCGCTT	TAAGCACCAAGTCACGCTGT
NF-κB1	GCACGACAACATCTCATTGG	TCCCAAGAGTCATCCAGGTC
MEST	TGTGGGTGTGGTTGGAAGTC	CCTCAAGGTCAGACCCTTCC
CD31	CAACGAGAAAATGTCAGA	GGAGCCTTCCGTTCTAGAGT
VEGF	TCCGGGTTTTATCCCTCTTC	TCTGCTGGTTTCCAAAATCC
β-actin	AATCTGGCACCACACCTTCTAC	TATCGTGTCGGACCTATCGTTG-

Western blot analysis

Total protein in cells was extracted and dissolved in 2× SDS loading buffer. Protein was then transferred onto a PVDF membrane with 10% SDS polyacrylamide gel. After blocking with 5% skim milk for 1 h, the membrane was incubated with diluted rabbit anti-human antibodies matrix metalloproteinase 2 (MMP-2) (ab37150, 1:1000), MMP-9 (ab73734, 1:1000), B-cell lymphoma 2 (Bcl-2) (ab182858, 1:2000), Bcl-2–associated protein x (Bax) (ab32503, 1:1000), VEGF (ab46154, 1:1000), CD31 (ab28364, 1:1000), NF-κB1 (ab220803, 1:1000), MEST (ab230114, 1:1000), and β-actin (ab179467, 1:5000) at 4°C overnight. The following day, the membrane was washed with Tris-buffered saline with Tween 20 and incubated with horseradish peroxidase–labeled secondary antibody goat anti-rabbit IgG (ab205718, 1:5000) for 1 h. The above antibodies were purchased from Abcam (Cambridge, MA, USA). Subsequently, the membrane was reacted with ECL (ECL808-25; Biomiga, San Diego, CA, USA) for 1 min and exposed in a gel imager. With β-actin used as an internal reference, the relative expression of individual protein was expressed by the ratio of the gray value of the target to the internal reference band.

Dual luciferase reporter gene assay

MEST-wild type (WT) and MEST-mutant (Mut) sequences were ligated to pGL3-Basic vector (Promega, Madison, WI, USA) to construct pGL3-MEST-WT and pGL3-MEST-Mut, which were then cotransfected into HEK293T cells with oe–NF-κB1 and oe-NC. Renilla luciferase was used as an internal reference. After 48-h transfection, the cells were lysed. Luciferase activity was detected by a Luciferase Assay Kit (K801-200; Biovision, Milpitas, CA, USA) with a dual luciferase reporter assay system (Promega).

FISH for subcellular location of LINC00284

The cells were inoculated in a 24-well culture plate at a density of 6 × 10⁴ cells/well. When cell confluence reached 85%, the cells were fixed with 1 ml 4% paraformaldehyde at room temperature for 15 min, washed with PBS twice, and reacted with 200 μl of 4 μg/ml protease K at room temperature for 5 min. Subsequently, the cells were reacted with 200 μl glycine/polybutylece terephthalate at room temperature for 5 min, washed with PBS twice, and then reacted with 200 μl acetylation reagent at room temperature for 10 min. After washing with PBS for additional 3 sessions, the cells were reacted with 200 μl prehybridization solution at 65°C for 1 h. Afterwards, the cells were added with 250 μl LINC00284 probe (Eurogentec, Liège, Belgium), and the culture plate was sealed with parafilm overnight at 65°C. The following day, the cell nucleus was stained with DAPI (diluted with phosphate-buffered saline with Tween-20 at the ratio of 1:800) in a 24-well culture plate for 5 min. Subsequently, the cells were sealed with anti-fluorescent quencher. Five random fields were selected for observation and imaging under a fluorescence microscope (Olympus, Tokyo, Japan) (22).

RNA-binding protein immunoprecipitation for binding of NF-κB1 and LINC00284

HO-9810 cells were lysed, after which the supernatant was collected. One part of the cell extract was taken as an input, and the rest was incubated with anti–NF-κB p105 antibody (ab32360, 1:5000; Abcam) and magnetic beads. The magnetic bead–antibody complex was washed and resuspension was carried out in 900 μl RNA-binding protein immunoprecipitation (RIP) wash buffer, followed by incubation with 100 μl cell extract at 4°C overnight. Next, the samples were placed on a magnetic base to collect magnetic bead-protein complex. Finally, the samples were separately detached by proteinase K in order to extract RNA for subsequent PCR detection. IgG (ab172730, 1:1000; Abcam) was used as a NC.

Chromatin immunoprecipitation assay for binding of NFκB1 and MEST

The cells were fixed with formaldehyde for 10 min to crosslink protein to DNA according to the instructions provided on the Chromatin Immunoprecipitation (ChIP) Kit (9003s; Cell Signaling Technology, Danvers, MA, USA). The cells were broken and chromatin was fragmentized by a sonicator at an interval of 10 s for 15 cycles (200–1000 bp). Next, the fragments were centrifuged at 4°C and 30,237 g and a part of DNA fragments were set aside as Input. The supernatant was then transferred into 3 tubes and respectively incubated with NC antibody against IgG (ab172730, 2.5 μg/ml; Abcam), positive control antibody against RNA polymerase II largest subunit (RPB1) (ab140509, 5 µg/mg of lysate; Abcam), and target protein-specific anti–NF-κB p65 antibody (acetyl K310, ab19870, 2.5 μg/ml; Abcam) (23) overnight at 4°C. DNA-protein complex was precipitated with protein agarose/sepharose and centrifuged at 12,000 g for 5 min, with the supernatant discarded. Afterwards, nonspecific complex was washed and de-cross-linked at 65°C overnight, and DNA fragments were extracted and purified using phenol/chloroform, with Input used as the internal reference. qRT-PCR was employed in order to detect binding of NF-κB1 and MEST using MEST-specific primers (24).

5-Ethynyl-2ʹ-deoxyuridine assay

The cells were incubated with 5-ethynyl-2ʹ-deoxyuridine (EdU) solution (cell culture medium and EdU solution were mixed at a ratio of 1000:1 for 2 h, fixed with 40 g/L paraformaldehyde for 30 min, incubated with glycine solution for 8 min, and rinsed with PBS containing 0.5% Triton X-100. Next, the cells were incubated first with Apollo staining reaction solution for 30 min and later Hoechst 33342 reaction solution for 20 min, which were then observed under a fluorescence microscope. Afterwards, the cells were photographed by 550-nm excitation channel of red light, with the red-stained cells regarded as proliferative cells. Subsequently, the cells were photographed by a 350-nm excitation channel of violet light, and all cells were stained as blue. Three fields were selected (original magnification, ×200) to count EdU-stained cells (proliferative cells) and Hoechst 33342–stained cells (total cells). The cell proliferation rate was calculated using the following formula: Cell proliferation rate = number of proliferative cells/number of total cells × 100%.

In vitro lumen formation

Matrigel (75 μl) was added into a precooled 96-well plate and gently shaken evenly in order to allow the uniform distribution of the Matrigel on the plate while avoiding the formation of bubbles. This operation was conducted on ice. The plate was placed at 37°C for 60 min for solidification. Human microvascular endothelial cell line (HMEC)-1 suspension was added to the corresponding well of the 96-well plate at 2.5 × 10⁴ cells/well. Following cell adherence to the well, the culture medium was replaced with the supernatant of transfected HO-9810 cells for further culture. After 48 h of incubation in a 37°C incubator, lumen formation was obtained under a microscope. Three fields were randomly selected from each well to obtain images. Quantitative analysis was carried out with Angiogenesis Analyzer Plugin in Network Analysis Menu of ImageJ (National Institutes of Health, Bethesda, MD, USA). Tube length in each field was calculated to obtain the mean values. The angiogenesis level was expressed as tube length (mm) per square millimeter area (25).

Transwell assay

HO-9810 cells at the logarithmic growth phase were starved for 24 h. The following day, the cells were detached, centrifuged, and resuspended to a final concentration of 2 × 10⁵ cells/well. Next, 0.2 ml suspension was added to the apical chamber and 700 μL precooled DMEM cell culture medium containing 20% fetal bovine serum (FBS) was added to the basolateral chamber, which was incubated at 37°C with 5% CO₂. After 24 h, the cells on the apical chamber and basement membrane were wiped off using wet cotton swabs. The cells were fixed with methanol for 30 min, stained with 0.1% crystal violet for 20 min, and observed and imaged under an inverted microscope. Five fields (original magnification, ×100) were randomly selected, and the number of cells penetrating the membrane was counted.

Extracellular matrix gel (40 μl) was added to the polycarbonate membrane in each 24-well transwell chamber (serum-free medium was diluted at 1:9 to 1 mg/ml) and incubated with 5% CO₂ at 37°C for 5 h. Pure DMEM (70 μl/well) was then added for incubation at 37°C for 0.5 h. MEST cells were starved for 24 h without serum. The following day, the cells were detached, centrifuged, precipitated, and resuspended in fresh DMEM without FBS to a final concentration of 2.5 × 10⁵ cells/well. Cell suspension (0.2 ml) was added to the apical chamber where the basement membrane had been hydrated and 700 μl precooled DMEM containing 20% FBS was added to the basolateral chamber. The cells were then incubated for 24 h at 37°C with 5% CO₂. Chambers were removed on the following day, and the cells on the chamber and basement membrane were wiped with a wet cotton swab. The cells were then fixed with methanol for 30 min, stained with 0.1% crystal violet for 20 min, rinsed under running water, inverted, air dried, observed, and photographed under an inverted microscope. Five fields (original magnification, ×100) were randomly selected, and the mean number of cells penetrating the membrane was calculated.

Immunohistochemistry

The specimen was fixed in 10% formaldehyde, embedded in paraffin, and then cut into 4-μm serial sections. After being baked for 1 h in a 60°C incubator, the sections were dewaxed with xylene and dehydrated with gradient ethanol. Next, the sections were repaired with 0.1 M sodium citrate for 20 min and incubated with 5% bovine serum albumin blocking solution at 37°C for 30 min. Afterwards, the sections were incubated with rabbit MEST pAb (ab230114, 1:1000; Abcam) at 4°C overnight. The following day, the sections were incubated with biotinylated goat anti-mouse IgG (ab6789, 1:2000; Abcam) at 37°C for 30 min. Then, the sections were developed with diaminobenzidine (DA1010; Beijing Solarbio Science and Technology, Beijing, China), counterstained with hematoxylin, dehydrated, cleared, mounted, observed, and photographed under an optical microscope (XSP-36; Shenzhen Boshida Optical Instrument, Guangdong, China). Five high-power fields (original magnification, ×200) were randomly selected from each section, and 100 cells were counted from each field. The negative samples were those that had a percentage of positive cells <5% and positive samples were those with positive cells ≥5% (26). The cells were identified and counted by 2 individuals independently who were blinded to experimental groups.

Xenograft tumor in nude mice

Twenty-eight Balb/c nude mice (age, 4–6 wk; weight, 17–21 g; sex unlimited) (Hunan Silaike Jingda Laboratory Animal, Changsha, China) were raised in specific pathogen-free-grade condition. Nude mice were subcutaneously injected with 2 × 10⁷ cells stably transfected with small interfering RNA against LINC00284 or small interfering RNA against MEST or relative NC plasmids. The mice were then euthanized by anesthesia following the experiments. The tumor was then taken out in order to obtain some images. The size of the tumor was measured, and the volume of the transplanted tumor was calculated: V = (A × B²)/2 (A: long diameter, B: short diameter, unit: cubic millimeter). The mean volume of tumor at each time point (n = 7) was obtained.

Statistical analysis

Statistical analysis was conducted by SPSS21.0 (IBM, Armonk, NY, USA). All data had normal distribution and homogeneity of variance tests. Data were expressed as means ± sd if conformed to normal distribution and homogeneity of variance; otherwise, the data were expressed as interquartile ranges. Data with skewed distribution were tested by a nonparametric Wilcoxon signed rank test. One-way ANOVA was used for comparisons among multiple groups. Comparisons between 2 groups were analyzed using an unpaired Student’s t test with a post hoc test. Data comparison at different time points was performed using repeated measures ANOVA. The difference was statistically significant when P < 0.05.

RESULTS

LINC00284 is up-regulated in OC tissues and cells

Microarray analysis was conducted on OC-related microarray dataset GSE38666 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE38666) (27) to screen out OC-related differentially expressed genes, and the results identified a high expression of LINC00284 in OC (Fig. 1A). qRT-PCR showed that the expression of LINC00284 in OC tissues was significantly higher than that in normal ovarian tissues (P < 0.05, Fig. 1B). Compared with normal HOEpiC, 5 human OC cell lines (SKOV3, A2780, ovcar3, HO-8910, Caov-3) had higher LINC00284 expression (P < 0.05), with the highest expression detected in HO-9810 (Fig. 1C). FISH was conducted to evaluate the location of LINC00284 in OC cells, which suggested that LINC00284 was located in the nucleus (Fig. 1D). Because human OC cell line HO-9810 had the highest LINC00284 expression, it was selected for subsequent experiments.

Figure 1.

LINC00284 is highly expressed in OC cells and tissues. A) Heat map of OC-related microarray dataset GSE38666. B) mRNA expression of LINC00284 in normal ovarian tissues and OC tissues detected by qRT-PCR (n = 72). *P < 0.05 vs. normal ovarian tissues. C) mRNA expression of LINC00284 in normal HOEpiC and human OC cell lines (SKOV3, A2780, ovcar3, HO-8910, Caov-3) assessed by qRT-PCR. *P < 0.05 vs. normal ovarian epithelial cell line. D) The location of LINC00284 in OC cells detected by FISH. The above data are measurement data and expressed as means ± sd. Comparison between 2 groups was analyzed by a nonpaired Student’s t test, whereas comparison among multiple groups was analyzed by ANOVA, n = 72. The experiment was repeated 3 times.

Down-regulated LINC00284 suppresses OC cell proliferation, migration, invasion, lumen formation, and tumorigenesis and promotes apoptosis

To investigate the role of LINC00284 in OC cells, HO-9810 was transfected with oe-NC, oe-LINC00284, si-NC, or si-LINC00284 to detect cell proliferation (EdU assay), apoptosis (TUNEL staining), migration (transwell assay), invasion (transwell assay), and in vitro lumen formation ability (in vitro lumen formation assay). The results from qRT-PCR showed that the introduction of oe-LINC00284 resulted in a significant increase in LINC00284 expression, whereas the introduction of si-LINC00284 led to the opposite results (Fig. 2A). In addition, when compared with cells introduced with relative oe-NC and si-NC plasmids, the cells introduced with oe-LINC00284 plasmid had increased proliferation, migration, and invasion and number of lumen formation as well as decreased apoptosis, whereas cells introduced with si-LINC00284 plasmid exhibited the reverse tendency (Fig. 2B–F).

Figure 2.

si-LINC00284 inhibits HO-9810 cell proliferation, migration, invasion, tube formation, and tumorigenic ability and promotes apoptosis in OC. A) mRNA expression of LINC00284 after transfection of oe-LINC00284 and si-LINC00284 plasmids detected by qRT-PCR. B) HO-9810 cell proliferation in OC determined by EdU assay (original magnification, ×200). C) Apoptosis of HO-9810 cells measured by TUNEL assay. Green indicates positive apoptosis cells, and blue indicates cell nucleus (original magnification, ×400). D) HO-9810 cell migration in OC determined by transwell assay (original magnification, ×200). E) HO-9810 cell invasion in OC evaluated by transwell assay (original magnification, ×200). F) Lumen formation of HMEC-1 cell detected by in vitro lumen formation assay (original magnification, ×100). G) Protein bands and expression of MMP-2, MMP-9, Bcl-2, and Bax determined by Western blot analysis. H) Protein bands and expression of VEGF and CD31 measured by Western blot analysis. *P < 0.05 compared with cells transfected with oe-NC plasmid, ^#P < 0.05 compared with cells transfected with si-NC plasmid. Data expressed as means ± sd. Comparison among multiple groups was analyzed by ANOVA. Cell experiments were repeated 3 times.

In addition, Western blot analysis was conducted to determine expression of cell migration-related proteins (MMP-2 and -9), apoptosis-related proteins (Bax and Bcl-2), and angiogenesis-related proteins (VEGF and CD31). The results revealed that relative to cells introduced with si-NC plasmid, the cells introduced with si-LINC00284 plasmid presented with decreased expression of MMP-2, MMP-9, Bcl-2, VEGF, and CD31 and increased Bax expression, whereas the opposite results were found in cells introduced with oe-LINC00284 plasmid in contrast to cells introduced with oe-NC plasmid (P < 0.05) (Fig. 2G, H). The above experiments were conducted in OC cell line ovcar3, the results of which were in line with OC cell line HO-8910 (Supplemental Fig. S1). Moreover, EdU assay, in vitro lumen formation assay, and Western blot analysis were carried out on normal HOEpiC. The results revealed that oe-LINC00284 did not result in any significant effects on the growth and lumen formation ability of normal ovarian epithelial cells (Fig. 3). The above findings demonstrated that si-LINC00284 resulted in the inhibition of OC cell proliferation, migration, invasion, lumen formation, and tumorigenesis while promoting apoptosis.

Figure 3.

LINC00284 has no impact on normal HOEpiC. A) mRNA expression of LINC00284 after HOEpiC cells are transfected with oe-LINC00284 and si-LINC00284 plasmids detected by qRT-PCR. B) HOEpiC cell proliferation in OC assessed by EdU assay (original magnification, ×200). C) Lumen formation of HMEC-1 cell examined by in vitro lumen formation assay (original magnification, ×100). D) Protein bands and expression of VEGF and CD31 measured by Western blot analysis. *P < 0.05 compared with cells transfected with oe-NC plasmid, ^#P < 0.05 compared with cells transfected with si-NC plasmid. Data expressed as means ± sd. Comparison among multiple groups was analyzed by ANOVA. The experiment was repeated 3 times.

si-LINC00284 increases the expression of MEST via NF-κB1

The lncMAP website (http://bio-bigdata.hrbmu.edu.cn/LncMAP/) found that LINC00284 could regulate expression of MEST through transcription factor NF-κB1 (Fig. 4A). RIP assay was conducted to assess the binding between LINC00284 and NF-κB1. The results showed that LINC00284 bound more anti–NF-κB1 compared with that combined with IgG (P < 0.05) (Fig. 4B). According to bioinformatics analysis, MEST was a target gene of NF-κB1. qRT-PCR was therefore applied to examine the expression of MEST in an OC cell line HO-8910 and a normal HOEpiC. The results showed that MEST expression was decreased in HO-8910 cells compared with that in HOEpiC (Fig. 4C).

Figure 4.

LINC00284 decreases the expression of MEST through NF-κB1. A) Bioinformatic analysis predicts that MEST is a target gene of LINC00284. B) Binding between LINC00284 and NF-κB1 detected by RIP. *P < 0.05 compared with IgG. C) mRNA expression of MEST in HO-8910 cells and normal ovarian cells measured by qRT-PCR. *P < 0.05 compared with normal HOEpiC. D) mRNA expression of MEST after HO-8910 cells are transfected with oe-LINC00284 and si-LINC00284 plasmids assessed by qRT-PCR. *P < 0.05 compared with cells are transfected with oe-NC, ^#P < 0.05 compared with cells transfected with si-NC. E) Binding between NF-κB1 and MEST determined by dual luciferase report gene assay. *P < 0.05 compared with cells transfected with oe-NC. F) protein band and quantitative analysis regarding binding between NF-κB1 and MEST verified by ChIP. NF-κB1 (lane 2), RPB1 (lane 3), and IgG (lane 4) are used for ChIP. RPB1 is regarded as positive control. *P < 0.05 compared with IgG. G) Expression of MEST after the rescue of NF-κB1 in si-LINC00284–transfected cells. *P < 0.05 compared with cells transfected with si-LINC00284 and si-NC plasmid. Data expressed as means ± sd. Comparison between 2 groups was analyzed by Student’s t test, whereas comparison among multiple groups was analyzed by ANOVA. The experiment was repeated 3 times.

In order to further verify the effect of LINC00284 on MEST, mRNA expression of MEST in transfected HO-9810 was detected by qRT-PCR. mRNA expression of MEST was decreased in cells transfected with oe-LINC00284 plasmid compared with those transfected with oe-NC plasmid (P < 0.05), whereas the expression of MEST in cells transfected with si-LINC00284 plasmid was increased in comparison to those transfected with si-NC plasmid (P < 0.05) (Fig. 4D).

Dual luciferase reporter gene assay was conducted to detect whether NF-κB1 could regulate MEST promoter activity. Luciferase activity of MEST-WT was reduced following the treatment of oe–NF-κB1 plasmid compared with the treatment of oe-NC plasmid (P < 0.05), but luciferase activity in MEST-Mut presented with no significant changes (P > 0.05), which demonstrated that NF-κB1 could lead to the inhibition in the activity of MEST promoter (Fig. 4E). ChIP was applied to detect the binding between NF-κB1 and MEST. The results revealed that NF-κB1 bound to more MEST than IgG (P < 0.05), suggesting that NF-κB1 could specifically bind to MEST protein (Fig. 4F). Moreover, the expression of MEST was decreased in cells transfected with both si-LINC00284 and oe–NF-κB1 plasmids compared with those transfected with si-LINC00284 and si-NC plasmid (P < 0.05) (Fig. 4G). The above results revealed that LINC00284 resulted in the inhibition in the transcriptional expression of MEST through NF-κB1.

LINC00284/NF-κB1 plays a crucial role in OC cells by regulating MEST

To determine the interaction among MEST, LINC00284, and NF-κB1 in OC cells, HO-9810 cells were transfected with si-LINC00284, si–NF-κB1, si-MEST, oe-LINC00284, oe-MEST, or related NC plasmids alone or in combination. A series of in vitro experiments including Western blot analysis, EdU assay, TUNEL staining, transwell assay, and in vitro lumen formation assay was performed to assess transfection efficiency, cell proliferation, apoptosis, migration, invasion, and in vitro lumen formation following transfection of HO-9810 cells.

The results from the Western blot analysis showed that the introduction of si–NF-κB1 and oe-MEST plasmids led to a significant decrease in the expression of NF-κB1 and increased expression of MEST (P < 0.05) (Fig. 5A, B). In addition, when compared with relative matched controls, the cells transfected with si–NF-κB1 or oe-MEST or combined oe-LINC00284 and si–NF-κB1 plasmids presented with decreased cell proliferation, migration, invasion, and number of lumens and increased apoptosis, whereas cells transfected with si–NF-κB1 or si-MEST plasmids displayed the opposite results (P < 0.05) (Fig. 5C–L).

Figure 5.

LINC00284 inhibits the expression of MEST by recruiting NF-κB1. A, B) Protein bands and expression of NF-κB1 and MEST after HO-9810 cells are transfected with si–NF-κB1 and oe-MEST plasmids detected by Western blot analysis. C, D) HO-9810 cell proliferation examined by EdU assay (A: original magnification, ×200, B: original magnification, ×100). E, F) HO-9810 cell apoptosis detected by TUNEL assay (original magnification, ×400). G, H) HO-9810 cell migration assessed by transwell assay (original magnification, ×200). I, J) HO-9810 cell invasion examined by transwell assay (original magnification, ×200). K, L) Lumen formation in HMEC-1 cells examined by in vitro lumen formation assay (original magnification, ×100). M, N) Protein bands and expression of MMP-2, MMP-9, Bcl-2, and Bax in HO-9810 cells determined by Western blot analysis. O, P) Protein bands and expression of VEGF and CD31 in HO-9810 cells evaluated by Western blot analysis. *P < 0.05 vs. cells transfected with si-NC or both si-LINC00284 and si-NC plasmids, ^#P < 0.05 vs. cells transfected with oe-NC or both oe-LINC00284 and si-NC plasmids. Data expressed as means ± sd. Comparison among multiple groups was analyzed by ANOVA. Cell experiments was repeated 3 times.

Subsequently, Western blot analysis was conducted to determine protein expression of MMP-2, MMP-9, Bax, Bcl-2, VEGF, and CD31. The results from Western blot analysis showed that in comparison with the matched controls, cells transfected with si–NF-κB1 or oe-MEST or combined oe-LINC00284 and si–NF-κB1 plasmids had lower expression of MMP-2, MMP-9, Bcl-2, VEGF, and CD31 and higher expression of Bax, whereas cells transfected with si–NF-κB1 or si-MEST plasmids presented with opposing results (P < 0.05) (Fig. 5M–P). These findings suggested that LINC00284 and NF-κB1 played a vital role in OC cells by regulating MEST.

si-LINC00284 inhibits OC development in vivo

Transfected OC cells were injected into nude mice in order to determine the effects of LINC00284 on tumor growth in vivo. There was a significant difference in tumor volume observed 12 d after the injection (Fig. 6A). The injection of si-LINC00284 plasmid resulted in slower tumor growth relative to the injection of si-NC plasmid (P < 0.05). Tumor growth increased following the injection of both si-LINC00284 and si-MEST plasmids when compared with the injection of si-LINC00284 and si-NC plasmid (P < 0.05). These results showed that si-LINC00284 inhibited tumor growth, which could be restored by si-MEST.

Figure 6.

si-LINC00284 inhibits OC tumor growth in nude mice. A) Tumor volume in different groups (n = 7). B) Protein bands and quantitative analysis of MMP-2 and MMP-9 protein expression in HO-9810 cell examined by Western blot analysis. C) Representative images and quantitative analysis of CD31 protein expression in tumor tissue measured by immunohistochemistry (original magnification, ×400; scale bars, 25 μm). *P < 0.05 vs. mice injected with cells transfected with si-NC plasmid, ^#P < 0.05 vs. mice injected with cells transfected with both si-LINC00284 and si-NC plasmid. Data expressed as means ± sd. Comparison among multiple groups was analyzed by ANOVA, and tumor volume at different time points was analyzed by repeated measurements of ANOVA. The experiment was repeated 3 times.

Next, the results of Western blot analysis showed that protein expression of MMP-2 and MMP-9 was decreased in mice injected with OC cells transfected with si-LINC00284 plasmid and increased in those injected with OC cells transfected with combined si-LINC00284 and si-MEST when compared with matched controls (P < 0.05) (Fig. 6B). Furthermore, immunohistochemistry was conducted to determine protein expression of angiogenesis-related factor CD31. The results revealed that CD31 expression was decreased in mice injected with OC cells transfected with si-LINC00284 plasmid and increased in mice injected with OC cells transfected with si-LINC00284 and si-MEST plasmids when compared with matched controls (P < 0.05) (Fig. 6C). These results suggested that si-LINC00284 suppresses tumor growth of OC in vivo by promoting MEST expression.

DISCUSSION

OC is the fifth primary cause of cancer-related death among women and accounts for nearly 6% of cancer-related mortality (28). Mounting evidence has demonstrated that lncRNAs have a close relationship with tumor development and progression (29, 30). Several lncRNAs have been shown to play vital roles in pathologic processes in OC, including cell growth, metastasis, and apoptosis (31). Here, we investigated the effect of LINC00284 on angiogenesis in OC and found that si-LINC00284 inhibited angiogenesis by up-regulating MEST via NF-κB1.

Initially, our study found a high expression of LINC00284 in OC cells, highlighting the first indication of LINC00284 involvement in OC. A number of recently conducted studies illustrated that abnormal lncRNA expressions are involved in the oncogenesis and development of various cancers, serving as tumor promoters or suppressors (32). A prior study reported that there exists a close link between LINC00284 and the overall survival of patients with papillary thyroid cancer (33). LINC00284 has been found to be highly expressed in human gastric cancer (13). Moreover, there is a high expression in lncRNA HOST2 in OC (34). Our findings also suggest that MEST was the target gene of NF-κB1 and that the silencing of LINC00284 led to the elevation of the expression of MEST via NF-κB1. LncRNA has been previously identified as a multipotent modulator of the NF-κB signaling pathway (35). However, few studies have mentioned the target relationship between MEST and NF-κB1 as well as the correlation between lncRNA and MEST.

In addition, our study showed that LINC00284 silencing inhibited proliferation, migration, and invasion while promoting apoptosis in OC cells by elevating MEST and decreasing NF-κB1. However, oe-LINC00284 in normal OC HOEpiC did not affect the growth of normal cells. This may be due to the short transfection time; the effect of LINC00284 on normal cells was less obvious than that of cancer cells, so it did not play a corresponding role in a certain period of time. Therefore, the specific reasons need to be further studied to explore the relevant mechanisms for further explanation. Recently, depleted lncRNA TUG1 has been demonstrated to result in the suppression of OC cell proliferation, invasion, and increased apoptosis (36). Moreover, it has been found that reduced lncRNA MNX1-AS1 led to the suppression of OC cell proliferation, migration, and promoted apoptosis (37). Furthermore, Fu et al. (38) also found that si-lncRNA EWSAT1 contributes to the down-regulation of proliferation, clonogenicity, and invasion of OC cells. Furthermore, the interaction between NF-κB and lncRNA suppresses cell proliferation, migration, and invasion while inducing apoptosis in melanoma by down-regulating NF-κB (39). NF-κB is involved in various cellular processes, including cell growth, apoptosis, angiogenesis, and differentiation (40). In addition, the NF-κB was found to be up-regulated in recently diagnosed advanced OC (41). NF-κB also regulates several cell apoptosis and invasion-associated proteins, including Bcl-2 and MMP-9 (42). A previous study has shown that the suppression of NF-κB resulted in reduced cell invasion, proliferation. and elevated apoptosis by down-regulating MMP-2, MMP-9, Bcl-2 while also up-regulating Bax in gallbladder cancer (39). MEST is an imprinted gene that’s been found to be involved in the development and invasion of certain tumors (43, 44). The loss of imprinting of PEG1/MEST is linked to malignant transformation and tumor formation, particularly in non-small cell lung cancer (45). Therefore, the aforementioned findings were consistent with our results that deficiency of LINC00284 results in the inhibition of proliferation, migration, and invasion while promoting apoptosis in OC cells by increasing MEST and decreasing NF-κB1.

The final finding from our study revealed that silencing of LINC00284 inhibited angiogenesis in OC cells through the up-regulation of MEST and down-regulation of NF-κB1. Tumor angiogenesis is a complex process that involves multiple factors including VEGF, noncoding RNAs, and their related pathways (46). VEGF, a potent angiogenic factor, has been found to be particularly down-regulated following the inhibition of NF-κB (47). Thus, the crucial role of NF-κB in angiogenesis can be established (48). Meanwhile, NF-κB signaling has been considered as the most notable putative paracrine effector of angiogenesis in mesenchymal stem cells exosomes (49). Moreover, the inactivation of NF-κB represses angiogenesis in human pancreatic cancer (50). In addition, lncRNAs have also been found to play a crucial functional role in gene expression during angiogenesis (51). For example, Zhou et al. (52) found that up-regulated lncRNA CASC2 leads to the suppression in the invasion and angiogenesis in gastric cancer. Another study showed that the knockdown of lncRNA TUG1 results in the inhibition of cell proliferation, migration, tube formation, and angiogenesis (53). In addition, MEST might be associated with oncofetal angiogenesis through its expression in extraembryonic tissues during formation and development of choriocarcinoma/trophoblast tumor (54). All of the aforementioned findings conformed to our results that LINC00284 knockdown suppresses angiogenesis in OC cells through up-regulation of MEST and down-regulation of NF-κB1.

In summary, our results demonstrated that the down-regulation of LINC00284 inhibited OC proliferation, migration, invasion, and angiogenesis through the up-regulation of MEST via transcription factor NF-κB1 (Fig. 7). Thus, LINC00284 inhibition may be a potential therapeutic target for OC. However, owing to the limitation of expenditure and time, the intrinsic reason of the high expression of LINC00284 in OC, and the effect of LINC00284 in hypoxia has not been clearly clarified but will be included in our future study. In addition, our experiments only conducted on OC cell lines HO-8910 and ovcar3, so further studies are required in order to elucidate the underlying mechanism of LINC00284 in OC by using more OC cell lines.

Figure 7.

Schematic diagram shows proposed effects of LINC00284 in OC. LINC00284 inhibits MEST by recruiting NF-κB1 to MEST promoter, leading to OC cell invasion, migration, and angiogenesis by increasing expression of MMP-2, MMP-9, VEGF, and CD31.

Supplementary Material

This article includes supplemental data. Please visit http://www.fasebj.org to obtain this information.

Glossary

Bax

Bcl-2–associated protein x

Bcl-2

B-cell lymphoma 2

ChIP

chromatin immunoprecipitation

EdU

5-ethynyl-2ʹ-deoxyuridine

FBS

fetal bovine serum

HMEC

human microvascular endothelial cell line

HOEpiC

human ovarian epithelial cell

LINC00284

long intergenic noncoding RNA

lncRNAs

long noncoding RNAs

MEST

mesoderm-specific transcript

MMP

matrix metalloproteinase

Mut

mutant

NC

negative control

OC

ovarian cancer

oe

overexpressed

RIP

RNA-binding protein immunoprecipitation

RPB1

RNA polymerase II largest subunit

si

silenced

WT

wild type

AUTHOR CONTRIBUTIONS

Z. Ruan designed the study; D. Zhao collated the data, carried out data analyses, and produced the initial draft of the manuscript; and both authors contributed to drafting the manuscript and read and approved the final submitted manuscript.