Ferroptosis-related lncRNA NRAV affects the prognosis of hepatocellular carcinoma via the miR-375-3P/SLC7A11 axis

The RNA sequences of 422 patients (50 normal tissues and 374 tumor tissues) were extracted from the TCGA-HCC database, and the RNA sequences of 445 patients (202 normal tissues and 243 tumor tissues) were extracted from the ICGC-LIRI-JP database. Ferroptosis-related genes were acquired from FerrDb, an online source that supplies a comprehensive and updated checklist of ferroptosis markers, regulative molecules, and ferroptosis-disease associations [12]. In general, we identified 175 (Driver:76; suppressor:47; marker:75) ferroptosis-related genes (Table S1). Pearson correlation evaluation was used to identify the relationships between ferroptosis-related genes and lncRNAs. Connections with the value of correlation coefficients |R|> 0.5 and a P-value < 0.001 were considered statistically substantial. The clinicopathological attributes of HCC people, consisting of age, gender, stage, TMN, survival condition, and survival time, were accumulated. In this study, we identified significant differentially expressed lncRNAs (DElncRNAs) between HCC tumors and normal liver tissue (absolute log2FC >  = 1 and FDR < 0.05). After screening differentially expressed genetics (DEGs), we performed functional enrichment evaluation and KEGG analysis using upregulated and downregulated ferroptosis-related differentially expressed genetics (DEGs). Enrichment analysis using metascape for EDGs data.

Univariate Cox regression, LASSO, and multivariate Cox regression evaluations were used to establish the last variables for building the prognostic threat score model. stratified based on risk rating (Coefficient lncRNA1 × expression of lncRNA1) + (Coefficient lncRNA2 × expression of lncRNA2) + ⋯ + (Coefficient lncRNAn × expression lncRNAn). Based on the threat scores for OS, HCC clients with fibrosis were separated into high- or low-risk groups using the typical score as a cut-off, and also Kaplan–Meier contours were calculated for both teams.

To execute genetics established enrichment evaluations stabilized, all genetics matters were imported right into the GSEA software application as well as the Molecular Signature Database (MSigDB) Trademark, KEGG, and also Genetics Ontology gene libraries were utilized to identify enriched gene sets [13], which were then searched in the TCGA-HCC database. Enriched gene sets were selected based on statistical significance (incorrect exploration rate FDR q value < 0.25 and normalized p-value < 0.05) [14]. We built a nomogram incorporating typical medical variables such as age, gender, grade, tumor stage, and the risk rating originated from the prognostic trademark to assess the possible 3- and 5-year OS of clients with HCC.

At the same time, the CIBERSORT [15, 16], ESTIMATE [17], quanTIseq [18], xCell [19], MCP-counter (or mMCP-counter for mouse) [20], EPIC [21], and TIMER [22] formulas were analyzed, the CC or cell sorts of immune reactions in heterogeneous examples among risky as well as low-risk teams based on ferroptosis-related lncRNAs trademark. The differences in immune feedback under various formulas were revealed using a Heatmap. In enhancement, ssGSEA was utilized to quantify the tumor-infiltrating immune cell subgroups between both groups and assess their immune function. Many possible immune check factor molecules were determined that could be targeted and incorporated with potential medication therapy [23].

Liver cancer cell lines (HuH-7, SKhep1, hep3B, hepG2) were ordered from the Chinese Academy of Sciences (Shanghai, China) and cultured in DMEM medium containing 25 mM glucose, 4 mM L-glutamine and 1 mM pyruvate (Gibco, 12800). All cell lines were authenticated using short tandem repeat (STR) profiling (by GENEWIZ Co. Ltd. at Suzhou, China). After authentication, large frozen stocks were made for future use. All cell lines were used within 15 passages (less than 2 months)after reviving from the frozen stocks. The medium was supplemented with 10% FBS and 1% penicillin/streptomycin. Cells were kept at 37 °C in a humidified incubator with 5% CO₂. The trypan blue exclusion assay was used to detect cell growth. All cell lines were tested for mycoplasma contamination, and no cell lines were contaminated. The miR-375-3P mimic, inhibitor, and corresponding control oligonucleotides were purchased from Shangya (Fuzhou, China). The transfection of miR-375-3P mimics, inhibitors, or corresponding NCs into cells was performed using Lipofectamine 3000 (Invitrogen). The lentiviruses of Flag-SRSF5, NRAV, overexpression vector (vector), sh-NRAV, and empty vector (sh-NC) were constructed by Genechem (Shanghai, China). The cells were transfected with lentivirus according to the manufacturer's instructions. The sequences of shRNA against specific targets are available in Table S2.

Total RNAs were extracted from cell lines and tissues using Trizol (Invitrogen, USA). The RNA of nuclear and cytoplasmic was separated and extracted using PARIS Kit (Life technologies, USA) according to the manufacturer’s instructions. The expression levels of NRAV, U6, and GAPDH genes in various RNA samples obtained by qRT-PCR were measured.

Eighty-six matched HCC tissues and paracancerous tissues were obtained from patients with HCC diagnosed by The First Affiliated Hospital of Zhengzhou University. All experiments involving human samples and clinical data were approved by the Accreditation Committee of The First Affiliated Hospital of Zhengzhou University.

According to the manufacturer's protocol, total RNA from HCC cells, tissues, and matched non-cancerous tissues was isolated using the TRIzol Reagent (Thermo Fisher Scientific, Waltham, MA, USA). Reverse transcription was performed using the PrimeScript RT Reagent Kit (Takara, Dalian, China). Bulge-loop™ miRNA RT-qPCR Primers were applied to determine the level of miRNAs. The real-time PCR reactions were performed using StepOnePlus™ Real-Time PCR System (Thermo Fisher Scientific, MA, US). The program settings on temperature cycling were followed as instructed by the manufacturer. GAPDH was the cytoplasmic control, and U6 was the nuclear control. The sequences of primers are listed in Table S2.

Cells were seeded into 12-well plates at an initial density of 700 cells/well and cultured for 2 weeks. The colonies were fixed in 4% paraformaldehyde and stained with 0.1% crystal violet (Sigma, St. Louis, MO, USA). The visible colonies were counted using a light microscope. The number of colonies in triplicate wells were measured for each treatment group.

Transwell assays were conducted using 24-well plates with chamber inserts with 8.0 μm pores (Corning). The cells were digested 24 h after transfection and resuspended in a serum-free medium. Then, 3 × 10⁴ cells per well were placed into the upper chambers. μlLike a cell nutritional attractant, the lower chamber was filled with a 500 μl medium with 10% FBS. After incubation at 37 °C for 24 h, cells in the upper chamber were gently removed with a cotton swab. The migrated or invaded cells were fixed with 4% polyformaldehyde and visualized by staining with crystal violet for 20 min. Last, the migrated cells were imaged and quantified by capturing five randomly chosen microscopic fields using an inverted microscope (Olympus, Japan). All experiments were performed in triplicate.

For the Cell Counting Kit-8 (CCK-8) assay, the treated HCC cells were seeded into 96-well plates at a concentration of 3 × 10³ /well. Then, 10 μl of CCK-8 assay solution (Dojindo, Japan) was added and incubated in the dark for 2 h. The absorbance at 450 nm was measured every 24 h with a microplate reader (BioTek Instruments, USA). All experiments were repeated three times.

Following the manufacturer's protocol, RNA immunoprecipitation (RIP) assay was performed using Magna RIP™ RNA-binding protein immunoprecipitation kit (Millipore, Billerica, MA, USA). The cell extract was incubated with magnetic beads conjugated with anti-Argonaute 2 (AGO2) or anti-IgG antibody (Millipore, Billerica, MA, USA) for 6 h at 4 °C. The beads were washed and incubated with Proteinase K to remove proteins. Finally, isolated RNA was extracted using TRIzol Reagent (Thermo Fisher Scientific, Waltham, MA, USA), then the purified RNA was subjected to qRT-PCR analysis. The specific binding sites of NRAV and miR-375-3P were determined by annolnc2 (http://​annolnc.​gao-lab.​org).

The full-length wild-type (WT) sequence of NRAV and the indicated mutant NRAV containing the predicted miR-375-3P binding sites were separately synthesized and cloned into the dual-luciferase reporter vector (Genechem, Shanghai, China). The resulting dual-luciferase reporter plasmids (WT or Mut) were co-transfected with the miR-375-3P mimic or inhibitor into hepG2 or HuH-7 cells, respectively, using Lipofectamine 3000. After 48 h of incubation, the relative firefly luciferase activities concerning the corresponding Renilla luciferase activities were measured and analyzed using a Dual-Luciferase Assay System (Promega, Fitchburg, WI, USA) following the manufacturer's protocol.

Intracellular ferrous iron (Fe²⁺) level was determined using the iron assay kit (ab83366, Abcam) according to the manufacturer's instructions. Cells were collected and washed in ice-cold PBS, homogenized in 5 × volumes of iron assay buffer on ice, then centrifuged (13,000 × g, 10 min) at 4 °C to remove insoluble material. The supernatant was collected, and an iron reducer was added to each sample before mixing and incubating at room temperature for 30 min. Then, 100 μl of the iron probe was added to each sample, mixing and incubating the reaction for 1 h at room temperature in the dark. The absorbance at 593 nm was measured immediately using a colorimetric microplate reader.

Lipid ROS level was analyzed by flow cytometry using BODIPY-C11 (GLPBIO, GC40165) dye. Cells were seeded at 2.5 × 10⁵ per well in a six-well dish and grown overnight for 12 h. Cells were washed once with PBS. Cells were then stained with 2 ml medium containing 5 µM of BODIPY-C11 and incubated at 37 °C for 20 min in the dark. Cells were washed twice with PBS to remove excess labeling mixture, followed by resuspending in 500 μl fresh PBS (DPBS, Gibco). The cell suspension was filtered through a 0.4 μM cell filter and subjected to flow cytometric analysis to detect the amount of intracellular lipid ROS. The fluorescence intensities of cells per sample were determined by flow cytometry using the BD FACSAria cytometer (BD Biosciences). A minimum of 10,000 cells was analyzed for each sample. Data analysis was evaluated using the FlowJo Software.

For the subcutaneous xenograft mouse model, 10 4- to 6-week-old male athymic BALB/c nude mice (Shanghai SLAC Laboratory Animal Co,Ltd., China) were housed and fed in standard pathogen-free conditions. The male nude mice were randomly divided into two groups (n = 5 in each group) and inoculated with cells as follows: sh-NC stable transfected hepG2 Cell (2 × 10⁶ cells); sh-NRAV stable transfected hepG2 Cell (2 × 10⁶ cells); Cells were mixed with matrigel (1:2) and inoculated subcutaneously at the right rear back region. Tumors were measured weekly with calipers, and tumor volume was calculated as = (L × W²)/2. Five weeks later, the mice were euthanized and the tumors were isolated for further study. If a tumor grew to 2000 mm³, the experiment was terminated in advance. After measurement, the mice were euthanized with an overdose of 10% pentobarbital sodium (100 mg/kg; intraperitoneal injection), and death was confirmed by the disappearance of the heartbeat. All in vivo animal experiments were approved by the Committee on the Ethics of Animal Experiments at the Shanghai University of Traditional Chinese Medicine.

The experimental data were analyzed using statistical analysis software, including GraphPad Prism 8.0 software (GraphPad) and the R package (V.3.3.4). Differences between the indicated groups were compared using the Student's t-test and one-way analysis of variance (ANOVA) followed by Fisher's least significant difference (LSD) test. Correlations were evaluated by Pearson correlation analysis. The cumulative overall survival (OS) and progression-free survival (PFS) rates were calculated using the Kaplan–Meier method, and significance was evaluated with the log-rank test. A P-value < 0.05 was considered to indicate a statistically significant result.

The ferroptosis-related lncRNAs signature was constructed following the flowchart shown in Fig. 1A. We uncovered 174 ferroptosis-related DEGs (38 downregulated and 136 upregulated; Fig. 1B-D, Table S1). Enrichment analysis revealed the over-expressed genes were mainly involved in ferroptosis, cellular response to chemical, oxidative, and oxidative stress. (Fig. 1E).

A prognostic risk evaluation model based on only 6 ferroptosis-related lncRNAs was then constructed using the optimal penalty parameter (λ = -1.8) for the LASSO model from the ferroptosis-related lncRNAs group. The cvfit and lambda curve are shown in Fig. 2A, B. Kaplan–Meier curve analysis also demonstrated that patients with the high-risk lncRNA signature had much poorer survival than those with the low-risk lncRNA signature (P < 0.001, Fig. 2C). Figure 2D demonstrated the variation of risk scores between the low-risk and high-risk groups and more fatalities and fewer years of survival in the high-risk group. As shown in Fig. 2E, the RNA expression of the six ferroptosis-associated lncRNAs was lower in the low-risk group than in the high-risk group. Meanwhile, the AUC of the signature lncRNAs was 0.795. Then, we demonstrated that the 6-gene signature was an independent prognostic factor for HCC patients superior to traditional clinicopathological factors and verified their survival prediction ability in an external HCC cohort in TCGA (Fig. 2F, G). The AUC predictive value of the novel lncRNAs signature for 1, 3, and 5-year survival rates was 0.795, 0.749, and 0.748, respectively (Fig. 2H).

Subsequently, we used univariate Cox analysis, a Robust likelihood-based survival model, and multivariate Cox analysis to construct a signature composed of lncRNAs and tumor stage, independent prognosis factors of OS of HCC patients (P < 0.001; Fig. 3A, B). The relationship between the novel lncRNAs and Ferroptosis-related genes is shown in Fig. 3C. The heatmap for the association between ferroptosis-related lncRNAs prognostic signature and clinicopathological manifestations was also analyzed (Fig. 3D). The hybrid nomogram incorporating clinicopathological characteristics and the novel ferroptosis-related lncRNAs prognostic signature (Fig. 3E) was stable and accurate, thus may be applied in surgical pathology or clinical management of HCC patients. Gene set enrichment analyses (GSEA) revealed that most of the novel ferroptosis-related lncRNAs of high-risk groups' prognostic signature regulated cell cycle and tumor-related pathways (Fig. 3F). On the other hand, most of the novel ferroptosis-related lncRNAs of low-risk groups' prognostic signatures regulated multiple metabolic processes of cells (Fig. 3G). The heatmap of immune responses based on CIBERSORT, ESTIMATE, quanTIseq, xCell, MCP-counter (or mMCP-counter for mouse), EPIC, and TIMER algorithms is shown in Figure S1A. To further explore the relationships between the risk scores and immune cells and functions, we quantified the enrichment scores of 16 immune cell subpopulations and their related functions with the ssGSEA R package, which revealed that APC co-stimulation, cytolytic activity, MHC class I, type I INF response and type II INF response were significantly different between the low-risk and high-risk groups (Fig. S1B). Given the importance of checkpoint inhibitor-based immunotherapies, we further explored the difference in the expression of immune checkpoints between the two groups. We found a significant difference in the expression of HHLA2, TNFRSF14, and CTLA4 among other groups of patients (Fig. S1C).

Next, to further confirm the role of lncRNAs in the HCC prognostic model. By analyzing the data of 424 cases of HCC in the TCGA database, the expression levels of 6 ferroptosis-related lncRNAs are shown in Fig. 4A. Through the prognostic analysis of three molecules in the Gepia database, we found that NRAV was significant for OS (Fig. 4B), while DANCR, MKLN1-AS1, and ZFPM2-AS1 showed little difference (Fig. S2A-C), and no difference between AL137186.2 and LNCSRLR. So we selected NRAVs with the most significant difference in HCC prognosis for further validation. NRAV is located on chromosome 12. In different HCC cell lines, the expression of NRAV is shown in Fig. 4C. Compared with Paracancer, NRAV is generally upregulated in cancer and can be stably expressed in various commercial HCC cells (Fig. 4D). The subcellular localization of NRAV in hepatoma cells was detected by nuclear and cytoplasmic separation experiments, and it was found that NRAV was located not only in the cytoplasm but also in the nucleus (Fig. 4E). With the deepening of the research, researchers have found that many noncoding RNAs can also work by translating into proteins. To study the action mode of NRAV, we use software to predict and analyze the protein-coding ability of NRAV. ORF reading frame prediction shows a reading frame, and ORF12 is greater than 300 bp (Fig. S2D), arousing our strong curiosity. However, it is a pity that coding-potential computer (CPC) analysis shows that the coding possibility of NRAV is very little (Fig. S2E), and we can not detect any special protein bands through in vitro translation experiments, while the control gene can detect band (Fig. 4F), which also supports that NRAV is a noncoding RNA.

Next, we studied the carcinogenic function of NRAV in HCC. To evaluate the biological function of NRAV in HCC cells, we designed and constructed three siRNA targeting NRAV to change their expression specifically. QRT-PCR results showed that all three siRNAs effectively silenced NRAV (Fig. 5A). The introduction of the recombinant human NRAV gene successfully realized the overexpression of the gene (Fig. 5B). The results of CCK-8 and clone formation experiments showed that knockout of the NRAV gene effectively inhibited the proliferation of hepG2, while up-regulation of NRAV significantly enhanced the proliferation of HuH-7 (Fig. 5C-F). Transwell's experiment showed that the down-regulation of NRAV significantly decreased the migration ability of hepG2, while the up-regulation of NRAV enhanced the migration ability of HuH-7 cells (Fig. 5G, H). Next, we constructed a CDX model to examine the carcinogenic function of NRAV in vivo. Compared with the control group, the tumor volume and weight in the knockout group decreased significantly, and the tumor growth rate slowed (Fig. 5I-K). To sum up, these experimental results show that NRAV has a carcinogenic function in HCC.

Iron accumulation was increased in hepG2 cells infected with sh-NRAV lentivirus compared with control and decreased in HuH-7 cells infected with oe-NRAV compared with Vector (Fig. 6A). Analysis of ROS by flow cytometry showed a similar trend as the iron accumulation results (Fig. 6B, C). These results suggested that NRAV plays a negative regulatory role in ferroptosis in HCC. Furthermore, western blot analysis in Fig. 6D revealed that SLC7A11 and GPX4 expressions were decreased in hepG2 cells infected with sh-NRAV lentivirus compared with control and increased in HuH-7 cells infected with lv-NRAV compared with Vector. The opposite trends were observed in ACSL4 expressions, and the most obvious change is SLC7A11. The potential downstream miRNA of NRAV was found through the annolnc2 database, in which miR-375-3P was reported to regulate the key molecule SLC7A11 of ferroptosis [24]. However, the specific mechanism by which NRAV regulates the miR-375-3P/SLC7A11 axis has not been explored. Through bioinformatics methods, we discovered the binding site of NRAV and miR-375-3P (Fig. 6E). We conducted immunoprecipitation assays using the AGO2 antibody and found that the AGO2 antibody was able to enrich both endogenous miR-375-3P and NRAV (Fig. 6F). To verify the binding of NRAV and miR-375-3P, we constructed two NRAV luciferase reporter plasmids: wild type and miR-375-3P binding site mutant (Fig. 6G). The luciferase activity of the wild type can be significantly inhibited by miR-375-3P mimics, but the luciferase activity of the mutant has no significant change (Fig. 6H).

Moreover, the knockout of NRAV significantly increased miR-375-3P expression, while the overexpression of NRAV significantly reduced the expression of miR-375-3P (Fig. 6I, J). western blot analysis confirmed that the miR-375-3P inhibitor and mimic reversed the alterations in SLC7A11 caused by alterations in NRAV(Fig. 6K). These results prove that NRAV promotes iron export and ferroptosis resistance by regulating the miR-375-3P/SLC7A11 axis.

Through the analysis of q-PCR results of clinical samples, the results showed that the expression of NRAV was negatively correlated with miR-375-3P and positively correlated with SLC7A11 (Fig. 7A, B). To further study the clinical significance of NRAV in HCC, the clinical characteristics of patients with high NRAV expression in our hospital showed that the clinical outcome of HCC patients with high NRAV expression was worse than that of patients with low NRAV expression (Fig. 7C, D). The pattern of ferroptosis regulated by NRAV is shown in Fig. 7E. These results suggest that the high expression of NRAV may predict a poor prognosis in patients with HCC.