1 Introduction
Ovarian cancer is the 8th leading cause of cancer-related death in women worldwide, with 313 959 new cases and 207 252 deaths reported in 2020 [
1]. Given the lack of typical symptoms and early diagnostic strategies, over 80% of patients are initially diagnosed of stage III or IV advanced disease, and the 5-year overall survival rate of which is only 30.8%, threatening women’s health [
2,
3]. Despite the improved treatment outcome of ovarian cancer, the overall survival of patients remains unsatisfactory because of the high recurrence rate and metastasis. Therefore, additional studies on the precise molecular mechanism underlying the metastasis and recurrence of ovarian cancer must be conducted to identify potential therapeutic strategies.
Long noncoding RNAs (lncRNAs) are a class of transcripts that are greater than 200 nucleotides in length, and they cannot encode proteins [
4]. LncRNAs exert pivotal functions in various biological processes of cancer, including chromatin reprogramming, mRNA stability regulation, and post-translational modification, and serve as tumor suppressors or drivers by influencing various intracellular physiologic processes, such as cell proliferation, apoptosis, glycolytic reprogramming, invasion, migration, and chemoresistance [
5–
8]. LncRNAs modulate tumor progression through a variety of mechanisms, and a well-known mode is acting as microRNA (miRNA) sponges to diminish the inhibitory effect of miRNAs on their target genes [
9]. For example, lncRNA LINC00680 functions as a competitive endogenous RNA (ceRNA) by sponging miR-423-5p to increase PAK6 expression and promote esophageal squamous cell carcinoma progression [
10]. In addition, lncRNA FAM225A promotes nasopharyngeal carcinoma tumorigenesis by sponging miR-590-3p and miR-1275 to upregulate ITGB3 [
11]. Moreover, lncRNA CBR3-AS1 promotes adriamycin resistance in breast cancer cells through the activation of the JNK1/MEK4-mediated MAPK signaling pathway by sponging miR-25-3P [
12]. However, the clinical relevance and molecular mechanism of lncRNAs in ovarian cancer metastasis remain to be further investigated.
Sprouty-related EVH1 domain-containing protein 1 (SPRED1), which contains three structural domains (EVH-1 domain, c-Kit binding domain, and Sprouty domain), belongs to the Sprouty-related protein family [
13]. Loss-of-function mutation in SPRED1 is the direct cause of Legius syndrome [
14]. SPRED1 is known as a negative regulator of the Ras-MAPK signaling pathway [
15]. Low SPRED1 expression is associated with poor prognosis in patients with melanoma. Low levels of SPRED1 increase MAPK signaling activity and confer resistance to BRAF inhibition [
16]. In addition, SPRED1 deficiency can induce the transformation of chronic myelogenous leukemia (CML) to the accelerated phase (AP) and eventually to a blast crisis (BC) in a CML mouse model. Furthermore, the mRNA and protein levels of SPRED1 are significantly decreased in patients with AP/BC CML compared with those with CP CML [
17]. Moreover, SPRED1 is a validated target of miR-196a, which is induced by the estrogen receptor (ER). Thus, the upregulation of ER increases the expression level of miR-196a, which inhibits SPRED1 expression and leads to breast cancer progression [
18]. However, the function and mechanism of SPRED1 in ovarian cancer progression remain to be further elucidated.
In this study, five pairs of primary and metastatic tumor tissues from patients with ovarian cancer were analyzed to screen lncRNAs that may be involved in the metastasis of ovarian cancer. LOC646029 was identified as a key lncRNA, which suppresses the progression and metastasis of ovarian cancer. Patients with advanced ovarian cancer exhibited lower levels of LOC646029 than those with early-stage ovarian cancer. Moreover, LOC646029 suppressed the proliferative and metastatic phenotype of ovarian cancer in vivo and in vitro. Mechanistically, LOC646029 regulates SPRED1 expression by sponging miR-627-3p to activate the KRAS/MAPK pathway, thereby affecting cell proliferation and metastasis. Our results revealed that the LOC646029/miR-627-3p/SPRED1 axis suppresses ovarian cancer proliferation and metastasis, indicating the potential application of LOC646029 as a biomarker of ovarian cancer progression.
2 Materials and methods
2.1 Patient samples
Primary ovarian cancer tissues and corresponding metastatic tissues were obtained from patients at the Cancer Hospital, Chinese Academy of Medical Sciences. All samples were collected after surgical resection and immediately stored at −80 °C until total RNA or protein extraction. Ethical consent for this study was granted by the Ethics Committee of the Cancer Hospital, Chinese Academy of Medical Sciences, and informed consents were obtained from all patients.
2.2 Cell culture and reagents
The 293T cell line and human ovarian cancer cell lines (ES-2 and TOV-21G) were purchased from the American Type Culture Collection (Manassas, VA, USA). IOSE80 and A2780 cells were purchased from the National Experimental Cell Resource Sharing Platform (Beijing, China), and 3AO cells were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The 293T cells were cultured in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum (FBS), and IOSE80, ES-2, TOV-21G, 3AO, and A2780 cells were cultured in RPMI-1640 medium with 10% FBS. All cell media were supplemented with 1% penicillin/streptomycin mixture. All cells were maintained in a humidified cell incubator with 5% CO2 at 37 °C.
2.3 RNA sequencing (RNA-seq)
lncRNA-seq analysis was performed by Beijing Microread Genetics Co., Ltd. (Beijing, China). In brief, total RNA of five pairs of primary tumors and matched metastatic tumors was extracted with TRIzol and used as an input material for RNA sample preparations. rRNA was depleted, and other small RNAs were removed before RNA-seq library preparation. Then, the prepared libraries were sequenced by using Illumina NovaSeq 6000. Differential expression between primary and metastatic tumor samples was identified using the DESeq2 package with a cutoff value of log2 fold change > 1 and P value < 0.05. RNA-sequencing analysis was performed by Seqhealth Technology (Wuhan, China). In brief, total RNA of 3AO cells with LOC646029 overexpression or the corresponding empty vector was isolated using TRIzol. Three pairs of samples were generated from three biologically independent experiments. mRNA was enriched from total RNA with oligo-dT and then fragmented randomly. After reverse transcription, DNA fragments were ligated using an Illumina universal adapter for subsequent PCR amplification and sequenced via the Illumina NovaSeq 6000 platform. A fold change > 2 and P value < 0.05 were used as criteria for differentially expressed genes. Raw data can be accessed via GSE224350 and GSE224495.
2.4 Virus production and cell infection
HEK293T cells were used for lentiviral packaging and production. In particular, full-length cDNAs were cloned into the pLVX-IRES-neo vector (632184; Clontech, CA, USA) and cotransfected with lentiviral packaging vectors psPAX2 (12260; Addgene) and pMD2.G (12259; Addgene) at a 4:3:1 ratio using Hieff Trans Liposomal Transfection Reagent (40802; Yeasen, Shanghai, China) according to the manufacturer’s instructions. The supernatant containing the virus was collected 48 h after transfection and filtered through 0.45 µm pore-size filters (4614; Pall, Puerto Rico, USA) prior to use. Cells were infected with viruses in the presence of 8 mg/mL of polybrene (TR-1003; Sigma-Aldrich). Forty-eight hours after infection, cells were selected with 1 mg/mL of puromycin (A610593; Sangon Biotech, Shanghai, China) for 7 days before the subsequent experiments.
2.5 Cell transfection
Cells were seeded on six-well plates at 60%–70% confluence before transfection. The miRNA mimics, miRNA inhibitors, and siRNAs as well as the corresponding negative controls were purchased from RiboBio (Guangzhou, China). RiboFECT™ CP (RiboBio, Guangzhou, China) was used as the transfection reagent according to the manufacturer’s instructions. siRNA sequences targeting LOC646029 used in this study are shown in Table S1.
2.6 In situ hybridization (ISH)
RNA ISH assay was performed to detect LOC646029 in tissues of patients with ovarian cancer using the RNA in situ hybridization kit (Servicebio Technology, Wuhan, China) according to the manufacturer’s instructions. Written informed consents were obtained from all patients. The detailed clinical information of patients is shown in Table S2. The probes of LOC646029 were synthesized by the Servicebio Technology Co., Ltd., and labeled with digoxigenin at the 5′ and 3′ ends of the sequence. The probe sequence targeting LOC646029 is shown in Table S3. The H-scores were calculated in accordance with the percentage of positive cells and the staining intensity using the following equation: H-score = ∑ pi× i, where pi represents the percentage of positive cells (0–100%) and i represents the staining intensity (0, negative; 1, weak; 2, medium; and 3, strong). ISH staining was scored by two independent observers.
2.7 Fluorescence in situ hybridization (FISH)
FISH was used to detect the subcellular localization of LOC646029. Cy3-labeled probes targeting LOC646029 were designed and synthesized by RiboBio (Guangzhou, China), and a FISH kit (C10910, RiboBio) was used to perform in situ hybridization in A2780 and TOV-21G cells according to the manufacturer’s instructions. Nuclei were stained with Hoechst 33342 (H21492, Thermo Scientific). All images were acquired using the confocal laser scanning system.
2.8 Real-time quantitative PCR (RT-qPCR)
Total RNA was extracted from tumor tissues or cells with TRIzol reagent (15596018; Invitrogen, CA, USA). For mRNAs, 1 μg of total RNA was collected for reverse transcription using a Quant script RT Kit (KR103, Tiangen Biotech, Beijing, China). For lncRNAs, 1 μg of total RNA was collected for reverse transcription using a lnRcute lncRNA First-Strand cDNA Kit (KR202; Tiangen Biotech, Beijing, China). RT-qPCR analysis was performed using PowerUp SYBR Green Master Mix (A25918; Applied Biosystems) and a StepOnePlus Real-Time PCR system (Applied Biosystems, CA, USA), and relative quantification (2−ΔΔCt) was used to calculate the foldchange. The primer sequences used for RT-qPCR are listed in Table S4.
2.9 Transwell assays
Cell invasion and migration were tested by Transwell chambers (3422; Corning, NY, USA). In brief, 3 × 104 cells were resuspended in 100 μL of RPMI-1640 without FBS and seeded into the upper well. Simultaneously, 650 μL of medium containing 10% FBS was added to the lower chamber well. After 24 h of incubation at 37 °C with 5% CO2, the remaining cells in the upper layer of the membrane were gently wiped clean with a cotton swab, and then the cells that passed through the membrane were stained with a methanol solution containing 0.5% crystal violet for 30 min. Representative images were photographed by microscopy, and the number of colored cells was counted for statistical analysis. For the invasion assay, a serum-free medium and Matrigel (Corning) were premixed at a ratio of 25:1, and 100 μL of the mixture was added to the upper chamber and incubated at 37 °C for 2 h before seeding the cells.
2.10 Cell counting kit 8 (CCK-8) assay
Cell proliferation was assessed by CCK-8 (Dojindo, Japan) analysis. A total of 2 × 103 cells were resuspended in 100 μL of medium and seeded in 96-well plates. After the cells had settled, the previous culture was replaced with 100 μL of medium containing 10% CCK-8 reagent, and then the cells were incubated for 1 h at 37 °C away from light. The absorbance of each well at a wavelength of 450 nm was measured by using a microplate reader (BioTek), and each sample contained five-well replicates. These experiments were independently repeated three times.
2.11 Nuclear and cytoplasmic separation of RNA
Nuclear and cytoplasmic fractions were separated using the PARIS™ Kit (AM1921, Thermo Fisher Scientific, Waltham, USA) according to the manufacturer’s instructions. In brief, A2780 and TOV-21G cells were lysed with cell fraction buffer on ice for 5 min and then centrifuged at 500× g for 5 min at 4 °C. The supernatant was collected as the cytoplasmic fraction, and the remaining sediment was resuspended using cell disruption buffer and lysed for 10 min at 4 °C as the nuclear fraction. GAPDH and U6 were used as distribution references for the cytoplasm and nucleus, respectively. The expression level of each gene was detected by RT-qPCR.
2.12 Western blot
Cells were lysed on ice for 10 min in RIPA buffer with a mixture of protease inhibitor cocktail (Roche, Basel, Switzerland) followed by ultrasonication for 20 s and then centrifuged at 14 000× g for 10 min. Protein concentrations were quantified using the PierceTM BCA Protein Assay Kit (Thermo Fisher Scientific). Equal amounts of proteins were separated by 10% SDS‒PAGE, and the proteins were subsequently transferred onto PVDF membranes (Millipore, Billerica, MA, USA). Then, the membranes were blocked using 5% skim milk for 1 h at room temperature followed by incubation at 4 °C overnight with a primary antibody. Afterward, the membranes were washed with TBST every 10 min for a total of 30 min. Finally, the membranes were incubated for 2 h at room temperature with the corresponding horseradish peroxidase (HRP)-labeled secondary antibodies. After three washes with TBST, the bands on the membranes were visualized using enhanced chemiluminescence HRP substrate (Thermo Fisher Scientific). The following antibodies were used for Western blot analysis: SPRED1 (ab271191, Abcam, MA, USA) and GAPDH (G9545, Sigma-Aldrich, Munich, Germany).
2.13 RNA immunoprecipitation (RIP) assay
RIP assays were performed by using the Magna RIP RNA Binding Protein Immunoprecipitation Kit (17-700, Sigma-Aldrich). In brief, approximately 1 × 107 cells were harvested and resuspended in RIP lysis buffer with a protease inhibitor cocktail and RNase inhibitor. The magnetic beads were conjugated with anti-AGO2 or anti-IgG antibodies. Subsequently, the beads conjugated with antibodies were mixed with the lysate supernatant and incubated overnight at 4 °C. Proteinase K was used to digest the proteins in the RNA–protein complex, and the remaining RNA was purified and applied for relative quantification by RT-qPCR.
2.14 In vivo experiments
All animal protocols were approved by the Animal Care and Use Committee of the Chinese Academy of Medical Sciences Cancer Hospital. Female BALB/c nude mice (6 weeks old) were purchased from Nanjing Biomedical Research Institute of Nanjing University (Nanjing, China). For analysis of tumor metastatic capacity in vivo, 3AO cells stably expressing an empty vector or LOC646029 were established by lentiviral infection. Approximately 3 × 106 3AO cells in 100 μL of PBS were injected intraperitoneally into nude mice. Four weeks after injection, the mice were sacrificed, and the tumors were isolated for immunohistochemical (IHC) staining. For subcutaneous tumorigenicity, approximately 1 × 106 3AO cells in 100 μL of PBS were subcutaneously injected into nude mice. Tumor volume was calculated every 3 days since day 6. Twenty-one days after injection, animals were sacrificed, and tumors were weighed for statistical analysis.
2.15 Dual-luciferase reporter assay
The LOC646029 full-length sequence and the 3′-UTR of SPRED1, as well as their corresponding mutant (MUT), were constructed and inserted downstream of the luciferase reporter gene in the pmiR-GLO vector (Promega). Then, 293T and A2780 cells were seeded in 24-well plates, grown to 30%–50% confluence, and transfected with a reporter plasmid using Lipofectamine 2000. Wild-type (WT) or MUT luciferase reporter plasmids were cotransfected with miR-627-3p mimics or equivalent amounts of NC mimics, and the Dual Luciferase Reporter Assay System was used to measure the fluorescence intensity of firefly luciferase (LUC) and Renilla luciferase after 48 h.
2.16 Statistical analysis
Two-tailed Student’s t-test was used for comparisons between two groups. Overall survival curves were established using the Kaplan–Meier method, and the significance of differences was calculated using the log-rank test. All data were expressed as the mean ± SD. All statistical analyses were performed using GraphPad Prism 7.0, and P < 0.05 was considered statistically significant: * P < 0.05, ** P < 0.01, and *** P < 0.001; ns indicates not significant.
3 Results
3.1 Downregulation of LOC646029 correlates with metastasis and poor prognosis in ovarian cancer
In identifying lncRNAs that play critical roles in the metastasis of ovarian cancer, lncRNA-seq was performed in five pairs of ovarian cancer primary tumors and matched metastatic tumor tissues. The results indicated that 27 annotated lncRNAs were differentially expressed in primary and metastatic tumors, 18 of which were downregulated and nine of which were upregulated on the basis of the cutoff criteria of absolute log2 fold change > 1 and P value < 0.05. Notably, the lncRNA LOC646029 was downregulated more than 100-fold in metastatic tumors compared with primary tumors (M_normalize: 2.34 vs. P_normalize: 553.77, absolute log2 fold change = 7.88, adjusted P = 5.26E–12; Fig.1). Details of these differential lncRNAs are presented in Table S5. Another 38 cases of primary ovarian cancer tumors and corresponding metastatic tumors were collected to detect the expression of LOC646029 using RT-qPCR and to confirm these results. Consistent with the RNA-seq results, more than three-quarters of cases (30/38, 78.9%) exhibited decreased LOC646029 expression in metastatic tumors compared with their paired primary tumors (Fig.1). The expression of LOC646029 was also detected in 3AO, ES-2, TOV-21G, SKOV3, and A2780 cell lines as well as in the normal ovarian epithelial cell line IOSE80 using RT-qPCR. The results showed that LOC646029 expression was upregulated in most ovarian cancer cell lines except for 3AO cells compared with IOSE80 cells (Fig.1). Moreover, ISH analysis of 150 ovarian cancer cases revealed that low LOC646029 expression was significantly correlated with poor prognosis in patients with ovarian cancer (Fig.1 and 1E). Collectively, these results indicated that downregulation of LOC646029 in metastatic ovarian tumors might be associated with the progression of ovarian cancer.
3.2 LOC646029 inhibits cell proliferation and metastasis in ovarian cancer
Next, we explored the biological function of LOC646029 in OC. Based on the expression level of LOC646029 in different ovarian cancer cell lines, 3AO and ES-2 cells were transfected with LOC646029-overexpressing plasmids or the corresponding empty vector, whereas two independent siRNA sequences were used to silence LOC646029 in TOV-21G and A2780 cells (Fig. S1A and S1B). Notably, the overexpression of LOC646029 significantly inhibited the proliferation of 3AO and ES-2 cells, whereas the knockdown of LOC646029 dramatically promoted the proliferation of A2780 and TOV-21G cells (Fig.2 and 2B). In further determining the effect of LOC646029 on cell invasion and migration, Transwell assays were performed in ovarian cancer cells. The results showed that the overexpression of LOC646029 significantly reduced the invasion and migration of 3AO and ES-2 cells. Meanwhile, the knockdown of LOC646029 significantly promoted invasion and migration in A2780 and TOV-21G cells (Fig.2 and 2D, Fig. S1C and S1D). Moreover, wound healing assays indicated that the overexpression of LOC646029 impeded the migratory ability of 3AO and ES-2 cells, whereas the knockdown of LOC646029 had the opposite effect in TOV-21G and A2780 cells (Fig.2 and 2F, Fig. S1E and S1F). In further elucidating the effect of LOC646029 on ovarian cancer metastasis in vivo, an abdominal metastasis model was established using BALB/c nude mice with stable overexpression of LOC646029 or corresponding control 3AO cells. Compared with the control group, the LOC646029 overexpression group had a markedly decreased size and number of tumor nodes (Fig.2 and 2H). Furthermore, the result of subcutaneous tumorigenesis showed that the overexpression of LOC646029 suppressed tumor growth (Fig.2). Compared with the control group, the LOC646029-overexpressed group showed a significant decrease in tumor size and tumor weight (Fig.2, Fig. S1G). Collectively, these results indicate that LOC646029 inhibits ovarian cancer proliferation and metastasis in vivo and in vitro.
3.3 miR-627-3p is a direct target of LOC646029, which promotes ovarian cancer metastasis and proliferation
Accumulating evidence suggests that lncRNAs are widely expressed in tumor cells, which play pivotal roles in gene regulation. One of the known functions of lncRNAs located in the cytoplasm is acting as a molecular sponge for microRNAs [
9,
19]. In confirming the subcellular localization of LOC646029, we performed FISH and nuclear-plasma RNA fractionation followed by RT-qPCR in A2780 and TOV-21G cells. The results showed that LOC646029 was primarily located in the cytoplasm (Fig. S2A and S2B). Considering the cytoplasmic localization of LOC646029, we hypothesize that LOC646029 inhibits ovarian cancer progression by sponging miRNAs. A bioinformatic algorithm (miRDB [
20]) was used to predict potential target miRNAs bound by LOC646029. We selected the top 10 target miRNAs for validation based on the binding score, and the results showed that miR-627-3p was the most significantly downregulated compared with the other nine candidates after the overexpression of LOC646029 (Fig.3). In addition, miR-627-3p expression was significantly upregulated after silencing LOC646029 using two independent siRNAs (Fig.3). Based on the predicted binding sites, we further investigated whether LOC646029 exerts its function by binding to miR-627-3p through a luciferase reporter assay with WT or mutated LOC646029 sequences. miR-627-3p overexpression significantly reduced the luciferase activity with the WT LOC646029 binding site, but it had no effect on MUT LOC646029 in A2780 and 293T cells (Fig.3). In general, miRNAs function as RISC components, which bind to Argonaute-2 (AGO2). Accordingly, we performed an anti-AGO2 RIP assay, and the results showed that LOC646029 and miR-627-3p were significantly immunoprecipitated by anti-Ago2 antibodies compared with IgG (Fig.3).
In investigating the physiological function of miR-627-3p, we used RNA mimics or inhibitors to overexpress or silence miR-627-3p (Fig.3). CCK8 assays showed that the upregulated expression of miR-627-3p significantly promoted cell proliferation in A2780 and TOV-21G cells, whereas the downregulation of miR-627-3p resulted in a decrease in the proliferative capacity of 3AO cells (Fig.3–3H). Transwell assays indicated that the migration and invasion of A2780 and TOV-21G cells were markedly enhanced by increasing the expression level of miR-627-3p but significantly reduced by decreasing the expression level of miR-627-3p in 3AO cells (Fig.3–3K). Collectively, these results suggested that miR-627-3p can directly bind to LOC646029 and promote ovarian cancer cell metastasis and proliferation.
3.4 SPRED1 is a crucial downstream target of miR-627-3p to suppress proliferation, invasion, and migration in ovarian cancer
In investigating the target gene of miR-627-3p, four bioinformatic databases (miRDB, TargetScan, miRWalk, and mirDIP) were queried to identify potential candidate target genes. The results predicted by bioinformatic methods are displayed in Table S6. A total of 19 mRNAs exhibited the propensity to be regulated by miR-627-3p after overlapping the predicted results (Fig.4). Next, we validated these 19 candidates through RT-qPCR and found that SPRED1 might be a potential target gene of miR-627-3p (Fig.4, Fig. S2C). The upregulation of miR-627-3p expression using microRNA mimics significantly reduced SPRED1 expression at the mRNA and protein levels (Fig.4). In confirming whether miR-627-3p targets the 3′-UTR of SPRED1 mRNA, WT or MUT SPRED1 3′-UTR was cloned into the luciferase reporter vector pmiR-GLO and then cotransfected with miR-627-3p mimics or NC into 293T and A2780 cells, respectively. The results showed that compared with the NC, miR-627-3p cotransfection significantly reduced the relative luciferase activity in the SPRED1-WT group, but it had no effect on the SPRED1-MUT group (Fig.4). In addition, we observed that the overexpression of miR-627-3p reduced SPRED1, which was reversed by the exogenous expression of SPRED1 (Fig.4 and 4F). Considering that miR-627-3p negatively regulates the expression of SPRED1, we determined whether the cancer-promoting phenotype of miR-627-3p is achieved by downregulating SPRED1. CCK8 assays illustrated that the upregulation of miR-627-3p enhanced proliferative viability in A2780 and TOV-21G cells; however, this effect was effectively alleviated by the overexpression of SPRED1 (Fig.4 and 4H). Next, a Transwell assay was applied to measure the invasion and migration of ovarian cancer cells. The results showed that the overexpression of miR-627-3p significantly promoted A2780 and TOV-21G cell invasion and migration, whereas SPRED1 overexpression partially abolished the miR-627-3p effect (Fig.4 and 4J, Fig. S2D and S2E). Our data support that SPRED1 is a direct target of miR-627-3p, which inhibits the malignant phenotype of ovarian cancer.
3.5 LOC646029 inhibits ovarian cancer progression by regulating miR-627-3p/SPRED1 signaling
In investigating whether LOC646029 functions as a tumor suppressor by regulating SPRED1 expression, we determined the mRNA and protein expression levels of SPRED1 in A2780 and 3AO cells after transfection with LOC646029 siRNA or pcDNA-LOC646029 by RT-qPCR and Western blotting. The results showed that the mRNA and protein levels of SPRED1 were decreased in the A2780 cells transfected with LOC646029 siRNA, whereas the ectopic expression of LOC646029 showed the exact opposite effect in 3AO cells (Fig.5 and 5B). Next, we aimed to examine whether the upregulation of SPRED1 by LOC646029 in ovarian cancer cells was mediated by sponging miR-627-3p; thus, we knocked down LOC646029 followed by transfection of the miR-627-3p inhibitor. Knocking down LOC646029 significantly reduced SPRED1 expression, which was rescued by the miR-627-3p inhibitor (Fig.5). In addition, we overexpressed LOC646029 and cotransfected miR-627-3p mimics into 3AO cells. Consequently, the upregulation of LOC646029 significantly increased the expression level of SPRED1, whereas the transfection of the miR-627-3p mimics partially reversed this upregulation (Fig.5). CCK-8 assays demonstrated that knocking down LOC646029 enhanced the proliferative ability of A2780 and SKOV3 cells, whereas the miR-627-3p inhibitor attenuated this effect (Fig.5 and Fig. S3B). Moreover, Transwell assays showed that knocking down LOC646029 increased the invasiveness of A2780 and SKOV3 cells, and this elevation could also be counteracted by miR-627-3p inhibitors (Fig.5 and 5G, Fig. S3C and S3D). In addition, the overexpression of LOC646029 repressed cell proliferation in 3AO cells, and subsequent transfection with miR-627-3p mimics reversed this trend (Fig.5). Furthermore, miR-627-3p mimics similarly counteracted the repression of ovarian cancer cell invasion and migration by LOC646029 (Fig.5 and 5J). We also detected epithelial–mesenchymal transition (EMT)-related proteins, and the result showed that A2780 and SKOV3 with LOC646029 knockdown exhibited high levels of N-cadherin and low expression level of E-cadherin, which were reversed by the miR-627-3p inhibitor. Similar results were observed for the overexpression and rescue groups (Fig.5 and 5D, Fig. S3A). The expression level of SPRED1 in xenograft tumors derived from control cells was lower than that in tumors derived from LOC646029-overexpressing 3AO cells (Fig.5). We also generated survival curves by Kaplan–Meier Plotter with data from patients with ovarian cancer in TCGA and found that patients with low SPRED1 expression had a poorer prognosis than those with high SPRED1 expression (Fig.5). Collectively, these results suggest that LOC646029 functions as a miR-627-3p sponge to regulate the expression level of SPRED1, thereby regulating the development of ovarian cancer.
3.6 LOC646029/miR-627-3p/SPRED1 axis mediates the inhibition of Ras signaling
In comprehensively understanding the inhibitory effects of LOC646029 on the proliferation and metastasis of ovarian cancer cells, we used RNA-seq to analyze the gene expression profile after the overexpression of LOC646029. We set the threshold as an absolute value of log
2 fold change > 1 and
P < 0.05, and a total of 511 differentially expressed genes were identified, 47 of which were upregulated and 464 of which were downregulated (Fig.6). Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis revealed that LOC646029 affects multiple signaling pathways associated with cancer metastasis and progression, including ECM–receptor interactions, the Hippo signaling pathway, and the TGF-beta signaling pathway (Fig.6). Gene set enrichment analysis showed that LOC646029 expression was negatively correlated with the EMT and KRAS signaling, and we confirmed that their mRNA levels were regulated by LOC646029 (Fig.6 and 6D). Point mutations in genes encoding RAS proteins are common in various cancer types, which result in the hyperactivation of the RAS and downstream target pathways [
21,
22]. SPRED1 is known as a negative regulator of RAS signaling [
15,
23]. Given the upregulation of SPRED1 expression by LOC646029 through sponging of miR-627-3p and the negative correlation between LOC646029 and KRAS signaling, LOC646029 might suppress KRAS activation through the upregulation of SPRED1 in ovarian cancer. The knockdown of LOC646029 increased the phosphorylation level of ERK in TOV-21G and A2780 cells, which was in turn reversed by the miR-627-3p inhibitor (Fig.6 and 6F). In addition, the ectopic expression of miR-627-3p in 3AO cells counteracted the reduction of ERK phosphorylation by LOC646029 (Fig.6). Collectively, these results demonstrated that the LOC646029/miR-627-3p/SPRED1 axis might contribute to ovarian cancer progression by impeding KRAS signaling.
4 Discussion
Ovarian cancer is an insidious disease with no evident clinical symptoms in the early stage. The early diagnosis rate of ovarian cancer is lower than that of other tumors because of the lack of effective early diagnostic means [
24]. Approximately 80% of patients with ovarian cancer are initially diagnosed at an advanced stage with extensive abdominal and pelvic metastases [
25]. Despite surgical tumor reduction in combination with platinum/paclitaxel treatment, most patients eventually develop drug resistance, leading to recurrence and poor prognosis. Therefore, the identification of new biomarkers and molecular mechanisms during ovarian cancer development is necessary.
Accumulating evidence has shown that lncRNAs mediate multiple biological processes, including chemoresistance, cell cycle regulation, and metabolic homeostasis, in a variety of human cancers [
7,
26–
28]. Despite many studies conducted on lncRNAs, further investigation of the lncRNAs that control the progression of ovarian cancer is necessary. In this study, we screened differentially expressed lncRNAs by comparing primary ovarian cancer tissues and paired metastatic tissues through transcriptome sequencing, among which LOC646029 was significantly downregulated in metastatic tissues. We assessed other 38 pairs of specimens, and more than three-quarters of which showed low expression of LOC646029 in metastatic tumors, indicating that LOC646029 was potentially correlated with ovarian cancer progression and metastasis. To our knowledge, no studies have focused on LOC646029; thus, we explored the biological function of LOC646029 in ovarian cancer
in vivo and
in vitro. Our results showed that the overexpression of LOC646029 significantly inhibited cell proliferation, invasion, and migration, whereas the downregulation of LOC646029 exhibited the opposite effect. LOC646029 reduced the metastatic ability of ovarian cancer cells in the peritoneal cavity of nude mice, indicating the tumor-suppressive role of LOC646029 in ovarian cancer and its potential value as a biomarker for prognostic prediction in patients with ovarian cancer. However, given the intricate mechanism of patient progression and varying responses to treatments, relying on a single lncRNA as the most clinically useful biomarker may not be optimal. While some lncRNAs have demonstrated promising prognostic and predictive potential, their practical clinical application still faces significant challenges and requires further research and standardization. Furthermore, although the downregulation of LOC646029 has been observed in metastatic ovarian cancer tissues, this conclusion still requires validation in a larger cohort of specimens to establish its robustness and generalizability. Therefore, reliable non-invasive methods must be developed for detecting LOC646029, particularly for patients who cannot undergo repeated tumor biopsies but require continuous monitoring.
The functional mechanism of lncRNAs is closely related to their intracellular localization, and cytoplasmic lncRNAs often serve as sponges for miRNAs in the form of ceRNAs [
29–
31]. FISH and nuclear-cytoplasmic RNA separation revealed that LOC646029 was primarily distributed in the cytoplasm, suggesting that it may function as a miRNA sponge. Bioinformatic analysis showed two potential binding sites for miR-627-3p in the sequence of LOC646029. Dual-luciferase reporter and RIP assays showed that LOC646029 directly bound to miR-627-3p, and the overexpression of LOC646029 downregulated miR-627-3p expression in ovarian cancer cells. Collectively, these results demonstrated that LOC646029 functions as an endogenous sponge of miR-627-3p in ovarian cancer cells.
miRNAs are involved in the initiation and progression of tumors by modulating the stability of target mRNAs. For example, in esophageal squamous cell carcinoma, exosome-derived miR-339-5p promotes radiosensitivity by regulating CDC25A expression [
32]. Vescarelli
et al. reported that miR-200c sensitizes Olaparib-resistant ovarian cancer cells by targeting Neuropilin 1 [
33]. A recent study showed that miR-210 promotes gastric cancer cell invasion and migration by suppressing the expression of DRD5, STK24, and MNT [
34]. Several studies have demonstrated that miR-627-3p is closely associated with multiple malignant phenotypes by promoting or repressing its target genes [
35–
38]. In our study, we found that miR-627-3p promoted the proliferation and metastasis of ovarian cancer cells, which directly opposed the function of LOC646029. Hence, LOC646029 exerts tumor-suppressive functions by sponging miR-627-3p to regulate target genes.
Bioinformatic analysis revealed that SPRED1 is a potential target for miR-627-3p, which was validated by subsequent RT-qPCR and dual-luciferase reporter assays. In addition, the overexpression of SPRED1 antagonized the oncogenic phenotype of miR-627-3p in ovarian cancer cells, indicating a critical negative regulatory relationship between miR-627-3p and SPRED1. SPRED1 belongs to the Sprouty-related protein family, and it is a negative regulator of RAS signaling, which is consistent with our experimental results. SPRED1 loss-of-function mutations or deletions are a direct driver of diseases, such as Legius syndrome, pediatric acute myeloblastic leukemia, and mucosal melanoma [
39–
41]. Data from TCGA demonstrated that low SPRED1 expression was associated with poor prognosis in ovarian cancer. Furthermore, we found that the overexpression of LOC646029 significantly increased the expression level of SPRED1, which was counteracted by transfection with miR-627-3p. In addition, LOC646029 and SPRED1 were found to be positively correlated in xenograft tumors.
Our study identified lncRNA LOC646029, which functions as a tumor suppressor and inhibits ovarian cancer progression. Low LOC646029 expression was associated with poor prognosis in patients with ovarian cancer. LOC646029 functions as a miR-627-3p sponge to upregulate SPRED1 mRNA expression, thereby repressing cell proliferation and metastasis in ovarian cancer (Fig.6). The LOC646029/miR-627-3p/SPRED1 axis may serve as a potential biomarker for ovarian cancer progression and prognosis.
PDF
(3864KB)
