Comprehensive analysis of candidate signatures of long non-coding RNA LINC01116 and related protein-coding genes in patients with hepatocellular carcinoma

doi:10.21203/rs.3.rs-1747954/v1

Download PDF

Research Article

Comprehensive analysis of candidate signatures of long non-coding RNA LINC01116 and related protein-coding genes in patients with hepatocellular carcinoma

https://doi.org/10.21203/rs.3.rs-1747954/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 20 Jun, 2023

Read the published version in BMC Gastroenterology →

You are reading this latest preprint version

Background

Hepatocellular carcinoma (HCC) is a long-term malignancy that causes high morbidities and mortalities worldwide. Notably, long non-coding RNAs (LncRNAs) have been identified as candidate targets for malignancy treatments.

Methods

LncRNA LINC01116 and its Pearson-correlated genes (PCGs) were identified and analyzed in HCC patients. The diagnostic and prognostic value of the lncRNA was evaluated using data from The Cancer Genome Atlas (TCGA). Further, we explored the target drugs of LINC01116 for clinical application. Relationships between immune infiltration and PCGs, methylation and PCGs were explored. The diagnostic potentials were then validated by real time-PCR and Oncomine cohorts.

Results

LINC01116 and the PCG OLFML2B are differentially and highly expressed in tumor tissues (both P ≤ 0.050). We found that LINC01116, TMSB15A, PLAU, OLFML2B, and MRC2 have diagnostic potentials (all AUC ≥ 0.700, all P ≤ 0.050) while LINC01116 and TMSB15A have prognostic significance (both adjusted P ≤ 0.050). LINC01116 was enriched in the vascular endothelial growth factor (VEGF) receptor signaling pathway, mesenchyme morphogenesis, etc.. After that, candidate target drugs with potential clinical significance were identified: Thiamine, Cromolyn, Rilmenidine, Chlorhexidine, Sulindac_sulfone, Chloropyrazine, and Meprylcaine. Analysis of immune infiltration revealed that MRC2, OLFML2B, PLAU, and TMSB15A are negatively associated with the purity but positively associated with the specific cell types (all P < 0.050). Analysis of promoter methylation demonstrated that MRC2, OLFML2B, and PLAU have differential and high methylation levels in primary tumors (all P < 0.050). Validation results of the differential expressions and diagnostic potential of LINC01116 (RT-PCR) and OLFML2B (RT-PCR and Oncomine) were consistent with those obtained in the TCGA cohort (all P < 0.050, all AUC > 0.700).

Conclusions

Differentially expressed LINC01116 could be a candidate diagnostic and an independent prognostic signature in HCC. Besides, its target drugs may work for HCC therapy via the VEGF receptor signaling pathway. Differentially expressed OLFML2B could be a diagnostic signature involved in HCC via immune infiltrates.

Long non-coding RNA

LINC01116

protein-coding gene

signature

hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is the sixth most prevalent form of malignancy worldwide and the second leading cause of tumor-related deaths (1). China alone accounts for approximately 55% of the global HCC cases annually due to the chronic hepatitis B virus and liver cirrhosis (2). The morbidity and mortality of HCC are particularly high in China (3). Although many advanced treatments, including surgical resection, liver transplant, and comprehensive therapies, have been in clinical application, the 5-year overall survival (OS) rate of HCC patients is still unsatisfactory (4). On the other hand, many researches have used big data genomics and molecular biology to identify various carcinogenic factors and molecular modulatory mechanisms of HCC. However, many patients are diagnosed at an advanced stage, and the tumors are prone to recurrence, even after surgery (5, 6). The identification of novel candidate biomarkers for early diagnosis, prognostic surveillance, and studies on the molecular mechanisms of HCC is, therefore, of significance.

Non-coding RNAs (ncRNAs), including microRNAs and long non-coding RNAs (LncRNA), have been identified as oncogenes and tumor suppressors in various cancer types. Besides, ncRNAs have emerging roles as novel therapeutic targets (7, 8). LncRNAs are RNA molecules with a length of more than 200 nucleotides and do not code for proteins (9). They are often aberrantly expressed in various cancers, such as esophageal (10), bladder (11), and prostate (12), where they function as oncogenes or tumor suppressors. For instance, the LncRNA HOTAIR promotes cell migration and invasion by down-regulating the RNA binding motif protein 38 in HCC (13).

The lncRNA LINC01116, also known as TALNEC2, is located in the 2q31.1 region (14). Haibei Hu et al. demonstrated that LINC01116 is overexpressed in breast cancer, where it is associated with metastasis and is indicative of a poor prognosis (14). Jingliang Ye et al. found that the expression of LINC01116 is significantly up-regulated in glioma tissues and could serve as both a diagnostic biomarker and a therapeutic target for glioma (15). They further suggested that LINC01116 modulates tumorigenesis in glioma by targeting the vascular endothelial growth factor (VEGF) through microRNA-31-5p (15). The study by Jing Wu showed that LINC01116 is overexpressed in oral squamous cell carcinoma and nasopharyngeal carcinoma tissues, and is associated with OS and relapse-free survival rate of head and neck squamous cell carcinoma (HNSCC) (16). They showed that LINC01116 acts as a cancer-promoting oncogene via epithelial-mesenchymal transition (16). Nonetheless, the expression of LINC01116 in HCC, as well as its role in the diagnosis, prognosis, and as a potential molecular target for HCC therapy is obscure. Therefore, the present study explored the potential roles of LINC01116 in HCC to provide new insights into its application in HCC.

Data collection and genome-wide analysis to identify LINC01116-correlated mRNAs

The mRNA expression levels and clinical data of patients pathologically diagnosed with HCC were downloaded from The Cancer Genome Atlas (TCGA, https://www.cancer.gov/). We then performed a genome-wide analysis by Pearson correlation analysis to determine LINC01116-related protein-coding genes (PCGs) using the R 3.6.0 platform (https://www.r-project.org/).

Analysis of expression levels and diagnostic potential

We explored the differential expressions and diagnostic potentials of the LncRNA LINC01116 and its top ten PCGs, as determined by correlation coefficient analysis. We first used the MERAV website (http://merav.wi.mit.edu/) to obtain the differential expression between tumorigenic and normal liver tissues for LINC01116 and 10 PCGs (17). The differential expressions were then analyzed using tumor and non-tumor data from the TCGA database. After that, we further explored the diagnostic potential of LINC01116 and the ten PCGs using receiver operative characteristic (ROC) curves in the TCGA database.

Prognostic analysis and conjoint analysis

The mRNA expressions of LINC01116 and the ten PCGs were divided into low and high expression groups by the median cutoff. Clinical data were then analyzed along these associations to assess the OS status. The Kaplan-Meier plot method and Cox hazard regression model were applied for univariate and multivariate analysis, respectively. Clinical factors related to OS were enrolled in the multivariate Cox regression model. The PCGs identified in the multivariate Cox regression model were determined as OS-related genes. Then, LINC01116 and these genes were used for conjoint analysis within their low or high expression groups.

Construction of predictive model using risk scores and nomogram

A risk score prediction model was constructed to predict patient survival based on its scores. A risk score prediction model was constructed using LINC01116 and OS-related PCGs as follows: risk scores= gene₁*β₁ + gene₂*β₂+ gene₃*β₃+… + gene_n*β_n(18, 19)_{. Where}β was the coefficient from the multivariate cox regression model, including LINC01116, PCGs, and clinical data. Then, low and high-risk groups were generated from the respective risk scores at the median cutoff. In addition, a nomogram was constructed using LINC01116, PCGs, and clinical data to predict survival probability at 1-5 years. Internal validation using c-index was further performed at 1-5 years.

Exploration of molecular mechanisms by gene set enrichment analysis (GSEA)

GSEA was performed to explore the potential molecular mechanisms, including gene ontology terms and metabolic pathways of OS-related PCGs and LINC01116. Analysis was conducted using a GSEA software (gsea2-2.2.4, https://www.gsea-msigdb.org/gsea/index.jsp), c2 curated gene sets (c2.cp.kegg.v7.0.symbols.gmt), and c5 gene ontology sets (c5.all.v7.0.symbols.gmt) (20, 21). Also, a false discovery rate (FDR) ≤ 0.25 was considered a significant enrichment.

Identification of potential drug targets of LINC01116

Analysis of the potential drug targets was further conducted to explore the clinical applications of LINC01116 for HCC. Then, a differential analysis was undertaken to obtain differentially expressed genes (DEGs) using the edgeR package in the R platform (22). DEGs were further used to obtain target drugs using the connectivity map website (cMAP, https://portals.broadinstitute.org/cmap/#). Negatively related drugs were considered drug targets.

Analysis of immune infiltration and methylation of PCGs

Analysis of immune infiltration was conducted via the Tumor Immune Estimation Resource database (TIMER, https://cistrome.shinyapps.io/timer/) (23, 24). Firstly, we analyzed the correlation between gene expression levels of PCGs and the extend of immune infiltration, including B cells, CD4⁺ T cells, CD8⁺ T cells, neutrophils, macrophages, and dendritic cells (This database did not recognize LncRNA). Then, analysis of somatic copy number alterations (SCNA) was performed to determine the correlations between the SNCA of PCGs and immune infiltration. Analysis of promoter methylation was applied to explore the relationships between gene expressions of PCGs and subgroups of clinical data via the UALCAN database (http://ualcan.path.uab.edu/) (This database did not recognize LncRNAs) (25). Clinical data, such as tumor types, gender, race, and tumor grade, were used for analysis.

Construction of competing endogenous RNA (ceRNA) network and validation of clinical significance

A ceRNA network was constructed based on negative regulation relationships between mRNA, miRNA, and lncRNA via LnCeVar database (http://ww.bio-bigdata.net/LnCeVar/index.jsp) (26). We further validated differential expressions and the diagnostic potentials of LINC01116 and PCGs by the oncomine database. The oncomine database was utilized to validate the differential expressions and diagnostic potentials of PCGs using scatter plots and ROC curves (this database did not recognize LncRNA).

Statistical analysis

Survival analysis, including the Kaplan-Meier, univariate, and multivariate Cox hazard regression model, was conducted using the SPSS statistical software package version 24.0. ROC curves, scatter plots, and survival plots were plotted using the GraphPad software version 8.0. A P value of ≤ 0.05 was considered statistically significant.

Clinicopathological characteristics of HCC patients and top ten PCGs

A flow diagram illustrating the process of the present study is shown in Figure 1. The study included a total of 370 HCC patients, and their clinicopathological characteristics, as obtained in the TCGA dataset, were previously reported (27). Several factors, including clinical factors, hepatitis B virus (HBV) status, tumor stage, and radical resection, were correlated to the OS (all P ≤ 0.05). The top ten PCGs of LINC01116-related by Pearson correlation are as follows: SOX2, BEND6, TMSB15A, PLAU, OLFML2B, NTNG1, SLC17A7, NTRK1, MRC2 and SLC7A3 (all correlation ≥ 0.800, all P< 1E-80, Table 1). Following Pearson correlation analysis, the genes associated with LINC01116 are shown in Supplementary Table 1.

Analysis of differential expressions and diagnostic potential

Differential expressions analysis using the MERAV database indicated that LINC01116, BEND6, PLAU, OLFML2B, SLC17A7, and SLC7A3 were significantly different from LINC01116 (Figure S1 A, C, E, F, H, K), while the others were not. Differential expression analysis showed that LINC01116 and OLFML2B were differentially expressed, with higher expression in tumor tissues (P=0.045, 0.019, Figure 2A, F). In terms of the diagnostic potentials of the various genes, only LINC01116, TMSB15A, PLAU, OLFML2B, and MRC2 were found to have the potential of aiding in HCC diagnosis (all AUC≥0.700, all P≤0.05, Figure 3A, D-F, J).

Survival analysis and conjoint analysis

Survival analysis of LINC01116 and PCGs were performed using the univariate Cox hazard regression model. The model showed that LINC01116 and OLFML2B have prognostic significance (crude P= 0.044, 0.024, respectively. Table 2, Figure 4 A, F). We then conducted a multivariate cox regression model using prognosis-related clinical factors and these genes. Multivariate analysis revealed that LINC01116 and TMSB15A have prognostic significance (adjusted P=0.046, 0.003, respectively, Table 2). Further, conjoint analysis for LINC01116 and TMSB15A was performed and showed distinguished survival among groups a, b, and c (crude P= 0.032, adjusted P= 0.002, Table 3, Figure 4L).

Construction of predictive model using risk scores and nomogram

A risk score model was constructed using LINC01116, TMSB15A expressions, and HBV status, and tumor stage and radical resection via a multivariate cox hazard model (Figure 5A, Table 4). The identified elements of risk score include risk score rank, survival status, and heatmap of the expression of LINC01116 and TMSB15A. Risk scores were divided into low and high-risk groups at median cutoff. A Kaplan-Meier plot was drawn using low and high-risk groups (crude P=0.030, Figure 5B, Table 2). After that, time-dependent ROC curves were drawn at 1-5 year, which revealed similar prediction results (Figure 5C). A nomogram was constructed using the expressions of LINC01116, TMSB15A and HBV status, and tumor stage and radical resection based on the different points of each factor (Figure 6). Tumor stage I, radical resection, without HBV infection, low expression of LINC01116, and high expression of TMSB15A indicated lower points, which therefore suggested a better OS prediction at 1, 3-, and 5-years (Figure 6A). Internal validations were conducted using C-index for predicted and actual OS status (Figure 6B).

Exploration of molecular mechanisms via GSEA

We explored the potential molecular mechanisms of LINC01116 and TMSB15A that could be involved in HCC prognosis. We then analyzed gene ontology (GO) terms and the Kyoto encyclopedia of genes and genomes (KEGG) pathways to identify the specific mechanisms. Specifically, LINC01116 enriched in several GO terms, including cellular response to vascular endothelial growth factors stimulus, mesenchymal morphogenesis, dendritic cell differentiation, vascular endothelial growth factor receptor signaling pathway, vasculogenesis, and integrin-mediated signaling pathway. (Figure 7A-H). The enriched KEGG pathways participate in focal adhesion, cell adhesion molecular cams, chemokine signaling, TGF-β signaling, notch signaling, B cell receptor signaling, pathways in cancer, and MAPK signaling (Figure 7I-P). TMSB15A was enriched in GO terms involved in negative regulation of endothelial cell proliferation, blood vessel endothelial cell migration, stem cell division, mesenchyme development, and vasculogenesis. (Figure S2 A-H). TMSB15A was enriched in KEGG pathways involved in drug metabolism, other enzymes, peroxisome, propanoate metabolism, and steroid hormone biosynthesis (Figure S2 I-L).

Identification of candidate target drugs and interaction networks of LINC01116

Using |fold change|≥2 and P≤0.05, we identified a total of 171 up-regulated and 37 down-regulated genes. We then used these DEGs to construct interaction networks, including KEGG pathways and diseases (Figure 8). This interactive network was associated with metabolic diseases, peptide hormone metabolism, NODAL signaling, regulation of beta-cell development, WNT ligand biogenesis and trafficking, antimicrobial peptides, PI3K/AKT signaling in cancer, and signaling by the insulin receptor. After that, candidate target drugs were generated via the cMAP database using these DEGs and listed as follows: Thiamine, Cromolyn, Rilmenidine, Chlorhexidine, Sulindac_sulfone, Chloropyrazine, and Meprylcaine (Figure 9, Table 5). Two dimensional (2D) structures of these drugs are shown in Figure 9A-G. Our results show that the drugs have potential clinical significance, are negatively related to the expression of LINC01116, with its high expression indicating a poor outcome (Figure 9H).

Immune infiltration and promoter methylation analysis of PCGs

Due to the unavailability of LINC01116 in TIMER and UALCAN, only PCGs were conducted in the analysis of immune infiltration and methylation. The analysis of immune infiltration revealed that all the four PCGs (MRC2, OLFML2B, PLAU, TMSB15A) were negatively associated with the purity (all P<0.001, r<0, Figure 10). Meanwhile, all the four PCGs were positively associated with specific cell types, including B cell, CD8⁺ T cell, CD4⁺ T cell, macrophage, neutrophil, and dendritic cells (all P<0.050, r>0). Then, SCNA analysis indicated that all of the four genes were partially associated with SCNA among B cell, CD8⁺ T cell, CD4⁺ T cell, macrophage, neutrophil, and dendritic cells (Figure 11). Specifically, MRC2 and OLFML2B showed significance in arm-level gain and high amplification; PLAU showed significance in arm-level deletion, while TMSB15A exhibited significance in arm-level deletion and gain.

Analysis of promoter methylation demonstrated that MRC2, OLFML2B, and PLAU revealed differential and high methylation levels in primary tumors compared with normal (all P<0.001, Figure 12A, E, I). However, no significant differences were observed in TMSB15A (Figure 12M). Methylation analysis by gender demonstrated that MRC2, OLFML2B, and PLAU have differential and high methylation in HCC tissues of both male and female populations compared with healthy tissues (all P<0.050, Figure 12B, F, J). However, TMSB15A showed differential methylation between males and females (Figure 12N). Methylation analysis by race suggested that MRC2, OLFML2B, and PLAU have differential significance between normal and other races, including Caucasian, African-American, and Asian (Figure 12C, G, K). However, TMSB15A showed a difference between the Caucasian and Asian populations (Figure 12O). Methylation analysis by tumor grade suggested that MRC2, OLFML2B, and PLAU have differential significance between normal and tumor grades 1-3 (Figure 12D, H, L) while TMSB15A showed no difference between normal tissues and tumor grade (Figure 12P).

Construction of ceRNA network and validations of clinical significance by oncomine database

A ceRNA network was constructed based on negative regulation relationship with LINC01116 (Figure 13). Specifically, LINC01116 was connected with miR-423-3P, miR-1908-5P, miR-744-5P, miR-1180-3P, miR-671-5P, GSK3B, FOXM1, TNIP2, PA2G4, BCL2L11, NKIRAS2, EEF1A2, TLE3. Then, validation by the Oncomine database suggested that OLFML2B, PLAU, and MRC2 have differential expressions and diagnostic potentials for HCC in the two datasets (all P<0.050, all AUC>0.700, Figure S3 C-H, K-P). However, TMSB15A showed diagnostic potentials in only one dataset (Figure S3 A-B, I-J).

Application of high-throughput sequencing technology and bioinformatics methodologies have led to the discovery that PCGs consist of approximately 2% of the entire human genome. The rest of the human genome comprises thousands of non-coding RNAs, including LncRNAs (28, 29). Recent evidence suggest that LincRNAs play crucial roles in the pathogenesis of multiple tumors. They do so by influencing several cellular functions, such as cell proliferation, differentiation, metastasis, and drug resistance (30, 31).

New discoveries involving lincRNAs have advanced our understanding on the initiation and progression of cancers. A study by Panzitt et al. revealed a novel mRNA-like to be the most up-regulated gene in HCC (32). Matok et al. found that HULC is up-regulated in colorectal carcinoma thereby accelerating metastasis of colorectal carcinoma cells to liver tissues indicating that HULC has an important role in HCC (33). Elsewhere, it was reported that HOTAIR associates with polycomb repressive complex 2, trimethylate H3K27 to repress the transcription levels of metastasis-related gene suppressors, therefore increase the invasiveness and metastasis of breast cancer (34, 35). Overexpression of HOTAIR in HCC tissues preducted high recurrence (36). The same study showed that knockdown of HOTAIR decreased the invasiveness and viability of HepG2 cells (36). Inversely, HepG2 cell line was led to a significant increase after the apoptotic stimuli TNF-α as well as chemotherapeutic drug cisplatin and doxorubicin (36).

Given the aforementioned evidence that lincRNAs participate in other cancers, we explored their roles in HCC. LINC01116 regulates diverse cancers, including glioma, HNSCC, breast cancer, osteosarcoma, epithelial ovarian cancer and oral squamous cell carcinoma. For instance, it was found that LINC01116 was not only significantly highly expressed in glioma tissues but also associated with an increased risk of recurrence and poor OS (15). The same study showed that LINC01116 regulates tumorigenesis of gliomas by targeting VEGF and modulating expression of VEGF by competitive adsorption of micorRNA-31-5p at the posttranscriptional level (15). Further analysis confirmed that LINC01116 may serve as a valuable auxiliary prognostic biomarker and prognostic indicator for glioma patients (15). In oral squamous cell carcinoma and nasopharyngeal carcinoma tissues, LINC01116 is not only up-regulated but also may serve as a valuable diagnostic biomarker and can predict the prognosis of HNSCC (16). Consistently, our study reveals that LINC01116 may serve as a diagnostic biomarker and prognostic indicator for HCC patients. Previously, Hao et al. reported that LINC01116, DUXAP8, LINC01138 and PCAT6 were dysregulated and significantly associated with poor prognosis of HCC (37).

In this study, GSEA analysis revealed that LINC01116 may regulate cellular responses to VEGF stimuli, VEGF receptor signailing pathway, mesenchyme morphogenesis, focal adhesion, cell adhesion molecular cams, chemokine signaling pathway, TGF β signaling pathway, notch signaling pathway, dendritic cell differentiation, and B cell receptor signaling pathway in cancer. These findings on VEGF stimuli, immune infiltration results, and VEGF receptor signailing pathway are parallel with those reported by Jingliang et al. (15, 16).

Studies have shown that lincRNAs regulate various cellular processes, such as cell cycle, immune surveillance and stem cell pluripotency (38, 39). Molecular mechanistic tests by GSEA analysis demonstrated that TMSB15A is enriched in stem cell division, mesenchyme development, vasculogenesis while LINC01116 modulates the differentiation of dendritic cells and B cell receptor signaling pathway. These results are consistent with of the aforementioned studies in stem cell and immune surveillance aspects. Further, immune infiltration analyses suggested that four PCGs (MRC2, OLFML2B, PLAU, and TMSB15A) were negatively associated with the purity while the four PCGs were positively associated with specific cell types, including B cell, CD8⁺ T cell, CD4⁺ T cell, macrophage, neutrophil and dendritic cell. These results are congruent with those mentioned previously in the above studies. MRC1 and MRC2 are crucial components of the innate immune system, and they contribute to defense against pathogenic bacterial infections (40). Moreover, they are highly expressed in liver tissues compared to spleen and kidney in chanllenged fish. MRC2 participates in lysosomal collagen degradation (41) and is required for T_reg differentiation in the ectopic lesion, especially for CD4^high T_reg (42).

OLFML2B, highly expressed in ovary, was found to be the most destructive single nucleotide polymorphisn (up to 61) compared with OLFM2, OLFM4 and LPHN2, without mutations (43). It was, therefore, inferred that Olfactomedin protein modulates the immune function and development in the nerve system (43). Similarly, our study reveals that this protein plays an important role in the immune system. Immune-related signature PLAU was found differentially and highly expressed in esophageal squamous cell cancer and involved in the general immune response (44). PLAU was also found to predict poor prognosis of esophageal squamous cell carcinoma (44). In contrast, it is not know whether TMSB15A is related to immune functions. In this study, SCNA analysis indicated that four genes were partially associated with B cells, CD8⁺ T cells, CD4⁺ T cells, macrophages, neutrophils and dendritic cells. Specifically, MRC2 and OLFML2B showed apparent arm-level gain and high amplification; PLAU showed apparent arm-level deletion while TMSB15A showed significant arm-level deletion and gain. However, the association between these genes and SCNA has not been documented.

In addition, this study reveals that these four genes possess diagnostic potential for HCC. It was previously reported that OLFML2B was overexpressed in gastric cancer tissues compared to normal gastric tissues and exhibited moderate diagnostic potential (AUC = 0.867, P < 0.0001) and was associated with poor survival of gastric cancer (45). Similarly, our TCGA, Oncomine database findings reveal that OLFML2B is differentially and higly expressed, with diagnostic value in HCC. However, in our study, OLFML2B was not found to possess prognostic significance in HCC. Xiaohong et al. reported that MRC2 predicted poor prognosis of HCC by regulating TCGβ1 (46) but the study did not explore its diagnostic potential. No study has documented the diagnostic value of MRC2. For PLAU, it was found to be aberrantly expressed in HNSCC and that it can be used for diagnostic and prognostic purposes in HNSCC (47). Specific, it was associated with invasiveness of HNSCC cells (47). In this study, only its diagnostic potential and not prognostic significance was confirmed in HCC. Further investigations are advocated to confirm the prognostic value of MRC2 and PLAU in HCC.

The potential target drugs of LINC01116 were as follows: Thiamine, Cromolyn, Rilmenidine, Chlorhexidine, Sulindac_sulfone, Chloropyrazine, and Meprylcaine. Thiamine was to act on metabolic pathways such as glycosaminoglycan degradation in a pilot study on type 2 diabetes mellitus-related HCC (48). The study concluded that diabetes mellitus may influence the occurrence and progression of HCC by modulating various metabolic and immunity processes (48). Thiamine compromised the anticancer efficacy of methrotrexate by ameliorating diethyl nitrosamine-induced HCC in wistar strain rats (49). Sulindac_sulfone inhibited colon cancers in a k-ras (codon 12) mutation-independent manner (50). Chlorhexidine was exhibited superior anti-tumor properties than cranberry extract in oral cancer AW13516 and KB cell lines (51). Pyrazine diazohydroxide, was found to be a novel antineoplastic agent in a phase I and pharmacokinetic study (52). The clinical value of drugs targeting LINC01116 in liver cancer should be investigated further.

DNA methylation modulates cell differentiation and is involved in tumorigenesis (53). Previous evidence indicates that epigenetic markers can be used for prognostic and diagnostic purposes in oncology (54). Promoter methylation analysis demonstrated that MRC2, OLFML2B, and PLAU were differentially and highly methylated in primary tumor cells compared with normal cells. Moreover, MRC2, OLFML2B, and PLAU were differentially and highly methylated between tumor and normal tissues as well as between genders, races and tumor grads. However, the diagnostic and prognostic significance of these genes in HCC need to be further investigated.

This study has the following limitations. Our main findings need to be validated in other cohorts with more patients and clinical factors. In addition, in vivo and in vitro experiments should be performed to explore specific mechanisms of LINC01116 and PCGs in HCC. Thirdly, potential target drugs of LINC01116 for clinical application of HCC need future explores.

Ethics approval and consent to participate

Not available

Consent for publication

Not available

Data availability statement

The datasets generated and/or analyzed during the current study are available in the TCGA repository (https://cancergenome.nih.gov/).

Conflict of interest

The authors report there are no conflict of interest.

Funding

The present study was supported by The Key Scientific Research Project Plan of Henan University (Grant No.20A320037) and 2022 Henan Province Key R&D and Promotion Special Support Project (The role of OSBP2 in the regulation of malignant phenotype of pancreatic cancer).

Authors contribution

Conceptulization: Xiang-Kun Wang, Ren-Feng Li

Data curation: Xiang-Kun Wang, Kai-Luo, Hong-Lei Wang

Formal analysis: Xiang-Kun Wang, Ren-Feng Li

Funding acquisition: Ren-Feng Li

Investigation: Xiang-Kun Wang, Kai-Luo, Long Yu, Xu-Dong Zhang

Methodology: Xiang-Kun Wang, Kai-Luo, Long Yu, Hong-Lei Wang, Xu-Dong Zhang

Project administration: Ren-Feng Li

Resources: Xiang-Kun Wang, Kai-Luo, Long Yu, Xu-Dong Zhang

Software: Xiang-Kun Wang, Kai-Luo, Long Yu, Hong-Lei Wang, Xu-Dong Zhang

Supervision: Ren-Feng Li

Validation: Xiang-Kun Wang, Kai-Luo

Visualization: Xiang-Kun Wang, Kai-Luo, Long Yu, Xu-Dong Zhang

Original draft: Xiang-Kun Wang

Reviw & editing: Xiang-Kun Wang, Ren-Feng Li

Acknowledgement

None.

Laursen L. A preventable cancer. Nature. 2014;516(7529):S2-3.
Pascual S, Herrera I, Irurzun J. New advances in hepatocellular carcinoma. World journal of hepatology. 2016;8(9):421–38.
Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA: a cancer journal for clinicians. 2013;63(1):11–30.
Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. 2018;68(1):7–30.
Hefaiedh R, Sabbeh M, Ennaifer R, Romdhane H, Ben Nejma H, Belhadj N, et al. Percutaneous treatment versus hepatic resection for the treatment of small hepatocellular carcinoma. La Tunisie medicale. 2015;93(3):132–7.
Bosch FX, Ribes J, Diaz M, Cleries R. Primary liver cancer: worldwide incidence and trends. Gastroenterology. 2004;127(5 Suppl 1):S5-s16.
Khoury S, Tran N. Circulating microRNAs: potential biomarkers for common malignancies. Biomarkers in medicine. 2015;9(2):131–51.
Cui M, You L, Ren X, Zhao W, Liao Q, Zhao Y. Long non-coding RNA PVT1 and cancer. Biochemical and biophysical research communications. 2016;471(1):10–4.
Iyer MK, Niknafs YS, Malik R, Singhal U, Sahu A, Hosono Y, et al. The landscape of long noncoding RNAs in the human transcriptome. Nature genetics. 2015;47(3):199–208.
Wang K, Li J, Xiong G, He G, Guan X, Yang K, et al. Negative regulation of lncRNA GAS5 by miR-196a inhibits esophageal squamous cell carcinoma growth. Biochemical and biophysical research communications. 2018;495(1):1151–7.
Wieczorek E, Reszka E. mRNA, microRNA and lncRNA as novel bladder tumor markers. Clinica chimica acta; international journal of clinical chemistry. 2018;477:141–53.
Misawa A, Takayama KI, Inoue S. Long non-coding RNAs and prostate cancer. 2017;108(11):2107–14.
Ding C, Cheng S, Yang Z, Lv Z, Xiao H, Du C, et al. Long non-coding RNA HOTAIR promotes cell migration and invasion via down-regulation of RNA binding motif protein 38 in hepatocellular carcinoma cells. International journal of molecular sciences. 2014;15(3):4060–76.
Hu HB, Chen Q, Ding SQ. LncRNA LINC01116 competes with miR-145 for the regulation of ESR1 expression in breast cancer. European review for medical and pharmacological sciences. 2018;22(7):1987–93.
J Y, J Z, H C, J Q, L Z, Z W, et al. A novel lncRNA-LINC01116 regulates tumorigenesis of glioma by targeting VEGFA. International journal of cancer. 2020;146(1):248–61.
J W, Z C, L Z, J C, X L, Z G, et al. Knockdown of LINC01116 inhibits cell migration and invasion in head and neck squamous cell carcinoma through epithelial-mesenchymal transition pathway. Journal of cellular biochemistry. 2020;121(1):867–75.
Shaul YD, Yuan B, Thiru P, Nutter-Upham A, McCallum S, Lanzkron C, et al. MERAV: a tool for comparing gene expression across human tissues and cell types. Nucleic acids research. 2016;44(D1):D560-6.
Lossos IS, Czerwinski DK, Alizadeh AA, Wechser MA, Tibshirani R, Botstein D, et al. Prediction of survival in diffuse large-B-cell lymphoma based on the expression of six genes. The New England journal of medicine. 2004;350(18):1828–37.
Wang X, Liao X, Yu T, Gong Y, Zhang L, Huang J, et al. Analysis of clinical significance and prospective molecular mechanism of main elements of the JAK/STAT pathway in hepatocellular carcinoma. International journal of oncology. 2019;55(4):805–22.
Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences of the United States of America. 2005;102(43):15545–50.
Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S, Lehar J, et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nature genetics. 2003;34(3):267–73.
Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics (Oxford, England). 2010;26(1):139–40.
Li T, Fan J, Wang B, Traugh N, Chen Q, Liu JS, et al. TIMER: A Web Server for Comprehensive Analysis of Tumor-Infiltrating Immune Cells. Cancer research. 2017;77(21):e108-e10.
Li B, Severson E, Pignon JC, Zhao H, Li T, Novak J, et al. Comprehensive analyses of tumor immunity: implications for cancer immunotherapy. Genome biology. 2016;17(1):174.
Chandrashekar DS, Bashel B, Balasubramanya SAH, Creighton CJ, Ponce-Rodriguez I, Chakravarthi B, et al. UALCAN: A Portal for Facilitating Tumor Subgroup Gene Expression and Survival Analyses. Neoplasia (New York, NY). 2017;19(8):649–58.
Wang P, Li X, Gao Y, Guo Q, Ning S, Zhang Y, et al. LnCeVar: a comprehensive database of genomic variations that disturb ceRNA network regulation. Nucleic acids research. 2020;48(D1):D111-d7.
Wang X, Zhou X, Liu J, Liu Z, Zhang L, Gong Y, et al. Genomewide investigation of the clinical implications and molecular mechanism of long noncoding RNA LINC00668 and proteincoding genes in hepatocellular carcinoma. International journal of oncology. 2019;55(4):860–78.
Ponting CP, Oliver PL, Reik W. Evolution and functions of long noncoding RNAs. Cell. 2009;136(4):629–41.
Nagano T, Fraser P. No-nonsense functions for long noncoding RNAs. Cell. 2011;145(2):178–81.
Sahu A, Singhal U, Chinnaiyan AM. Long noncoding RNAs in cancer: from function to translation. Trends Cancer. 2015;1(2):93–109.
Yarmishyn AA, Kurochkin IV. Long noncoding RNAs: a potential novel class of cancer biomarkers. Frontiers in genetics. 2015;6:145.
Panzitt K, Tschernatsch MM, Guelly C, Moustafa T, Stradner M, Strohmaier HM, et al. Characterization of HULC, a novel gene with striking up-regulation in hepatocellular carcinoma, as noncoding RNA. Gastroenterology. 2007;132(1):330–42.
Matouk IJ, Abbasi I, Hochberg A, Galun E, Dweik H, Akkawi M. Highly upregulated in liver cancer noncoding RNA is overexpressed in hepatic colorectal metastasis. European journal of gastroenterology & hepatology. 2009;21(6):688–92.
Rinn JL, Kertesz M, Wang JK, Squazzo SL, Xu X, Brugmann SA, et al. Functional demarcation of active and silent chromatin domains in human HOX loci by noncoding RNAs. Cell. 2007;129(7):1311–23.
Gupta RA, Shah N, Wang KC, Kim J, Horlings HM, Wong DJ, et al. Long non-coding RNA HOTAIR reprograms chromatin state to promote cancer metastasis. Nature. 2010;464(7291):1071–6.
Yang Z, Zhou L, Wu LM, Lai MC, Xie HY, Zhang F, et al. Overexpression of long non-coding RNA HOTAIR predicts tumor recurrence in hepatocellular carcinoma patients following liver transplantation. Annals of surgical oncology. 2011;18(5):1243–50.
Jiang H, Shi X, Ye G, Xu Y, Xu J, Lu J, et al. Up-regulated long non-coding RNA DUXAP8 promotes cell growth through repressing Kruppel-like factor 2 expression in human hepatocellular carcinoma. OncoTargets and therapy. 2019;12:7429–36.
Guttman M, Amit I, Garber M, French C, Lin MF, Feldser D, et al. Chromatin signature reveals over a thousand highly conserved large non-coding RNAs in mammals. Nature. 2009;458(7235):223–7.
Dinger ME, Amaral PP, Mercer TR, Pang KC, Bruce SJ, Gardiner BB, et al. Long noncoding RNAs in mouse embryonic stem cell pluripotency and differentiation. Genome research. 2008;18(9):1433–45.
Dong X, Li J, He J, Liu W, Jiang L, Ye Y, et al. Anti-infective mannose receptor immune mechanism in large yellow croaker (Larimichthys crocea). Fish & shellfish immunology. 2016;54:257–65.
East L, Isacke CM. The mannose receptor family. Biochimica et biophysica acta. 2002;1572(2–3):364–86.
Wei C, Mei J, Tang L, Liu Y, Li D, Li M, et al. 1-Methyl-tryptophan attenuates regulatory T cells differentiation due to the inhibition of estrogen-IDO1-MRC2 axis in endometriosis. Cell death & disease. 2016;7(12):e2489.
Li Q, Liu A, Gu X, Su Z. Olfactomedin domain-containing proteins: evolution, functional divergence, expression patterns and damaging SNPs. Molecular genetics and genomics: MGG. 2019;294(4):875–85.
Li Y, Lu Z, Che Y, Wang J, Sun S, Huang J, et al. Immune signature profiling identified predictive and prognostic factors for esophageal squamous cell carcinoma. Oncoimmunology. 2017;6(11):e1356147.
J L, Z L, X Z, T G, D Y. Bioinformatic exploration of OLFML2B overexpression in gastric cancer base on multiple analyzing tools. BMC cancer. 2019;19(1):227.
X G, K T, Z L, X Z. MRC2 expression correlates with TGFβ1 and survival in hepatocellular carcinoma. International journal of molecular sciences. 2014;15(9):15011–25.
Yang K, Zhang S, Zhang D, Tao Q, Zhang T, Liu G, et al. Identification of SERPINE1, PLAU and ACTA1 as biomarkers of head and neck squamous cell carcinoma based on integrated bioinformatics analysis. International journal of clinical oncology. 2019;24(9):1030–41.
Wei H, Wang J. The underlying pathophysiology association between the Type 2-diabetic and hepatocellular carcinoma. 2019;234(7):10835–41.
Saleem S, Kazmi I, Ahmad A, Abuzinadah MF, Samkari A, Alkrathy HM, et al. Thiamin Regresses the Anticancer Efficacy of Methotrexate in the Amelioration of Diethyl Nitrosamine-Induced Hepatocellular Carcinoma in Wistar Strain Rats. Nutrition and cancer. 2020;72(1):170–81.
de Jong TA, Skinner SA, Malcontenti-Wilson C, Vogiagis D, Bailey M, van Driel IR, et al. Inhibition of rat colon tumors by sulindac and sulindac sulfone is independent of K-ras (codon 12) mutation. American journal of physiology Gastrointestinal and liver physiology. 2000;278(2):G266-72.
Khairnar MR, Wadgave U, Jadhav H, Naik R. Anticancer activity of chlorhexidine and cranberry extract: an in-vitro study. Journal of experimental therapeutics & oncology. 2018;12(3):201–5.
Vogelzang NJ, Mick R, Janisch L, Berezin F, Schilsky RL, Ratain MJ. Phase I and pharmacokinetic study of a new antineoplastic agent: pyrazine diazohydroxide (NSC 361456). Cancer research. 1994;54(1):114–9.
Portela A, Esteller M. Epigenetic modifications and human disease. Nature biotechnology. 2010;28(10):1057–68.
Heyn H, Esteller M. DNA methylation profiling in the clinic: applications and challenges. Nature reviews Genetics. 2012;13(10):679–92.

Table 1

Top 10 protein-coding genes of Pearson correlation-related with *LINC01116*
LncRNA	PCG	Correlation	95% CI	P value
LINC01116	SOX2	0.820	0.780–0.850	1.87E-89
LINC01116	BEND6	0.810	0.780–0.850	4.16E-89
LINC01116	TMSB15A	0.810	0.770–0.840	4.18E-88
LINC01116	PLAU	0.810	0.770–0.840	1.41E-87
LINC01116	OLFML2B	0.810	0.770–0.840	5.51E-87
LINC01116	NTNG1	0.810	0.770–0.840	5.81E-87
LINC01116	SLC17A7	0.810	0.770–0.840	9.88E-87
LINC01116	NTRK1	0.800	0.760–0.840	1.44E-84
LINC01116	MRC2	0.800	0.760–0.830	1.35E-83
LINC01116	SLC7A3	0.800	0.760–0.830	1.69E-83
Note: LncRNA: long non-coding RNA, PCG: protein-coding gene, CI: confidence interval.

Table 2

Survival analysis of *LINC01116* and target genes in hepatocellular carcinoma
Variables	Patients	Overall survival
Variables	(n = 370)	No. of event	MST (days)	HR (95%CI)	Crude P value	HR (95%CI)	Adjusted P value^§
LINC01116
Low expression	185	55	2486	Ref.	0.044	Ref.	0.046
High expression	185	75	1423	1.430 (1.009–2.027)		1.493 (1.007–2.214)
SOX2
Low expression	185	63	1791	Ref.	0.136	Ref.	0.124
High expression	185	67	1560	1.305 (0.920–1.850)		1.365 (0.918–2.028)
BEND6
Low expression	185	63	1852	Ref.	0.956	Ref.	0.416
High expression	185	67	1685	1.010 (0.715–1.426)		0.852 (0.580–1.253)
TMSB15A
Low expression	185	69	1372	Ref.	0.112	Ref.	0.003
High expression	185	61	2116	0.754 (0.532–1.068)		0.540 (0.362–0.806)
PLAU
Low expression	185	57	2131	Ref.	0.564	Ref.	0.560
High expression	185	73	1624	1.108 (0.782–1.570)		0.892 (0.608–1.309)
OLFML2B
Low expression	185	51	NA	Ref.	0.024	Ref.	0.520
High expression	185	79	1423	1.503 (1.056–2.139)		1.139 (0.767–1.691)
NTNG1
Low expression	185	60	2116	Ref.	0.161	Ref.	0.644
High expression	185	70	1624	1.280 (0.906–1.809)		1.096 (0.742–1.621)
SLC17A7
Low expression	185	64	1791	Ref.	0.694	Ref.	0.828
High expression	185	66	1622	1.072 (0.759–1.512)		0.958 (0.650–1.411)
NTRK1
Low expression	185	66	1560	Ref.	0.389	Ref.	0.053
High expression	185	64	1694	0.859 (0.608–1.214)		0.681 (0.462–1.004)
MRC2
Low expression	185	65	1791	Ref.	0.953	Ref.	0.280
High expression	185	65	1685	0.990 (0.701–1.397)		0.805 (0.543–1.193)
SLC7A3
Low expression	185	64	2131	Ref.	0.544	Ref.	0.749
High expression	185	66	1490	1.113 (0.788–1.571)		0.938 (0.633–1.390)
Risk score model
Low risk	185	57	2116	Ref.	0.030	Ref.	0.018
High risk	185	73	1386	1.469 (1.038–2.078)		1.595 (1.082–2.351)
Note: §: P values were adjusted for radical resection, tumor stage and HBV infection; Bold indicates significant P values.
Abbreviations: NA: not available; MST: median survival time; HR: hazard ratio; 95%CI: 95% confidence interval.

Table 3

Joint-effect analysis of *LINC01116* and *TMSB15A* for overall survival
Group	LINC01116		TMSB15A	Overall survival
Group	expression	expression		Events /total	MST (Days)	Crude HR (95%CI)	Crude P value	Adjusted HR (95%CI)	Adjusted P value^ɸ
a	High		Low	40/87	1229	Ref.	0.032	Ref.	0.002
b	High		High	64/196	1852	0.687 (0.462–1.021)	0.063	0.591 (0.380–0.918)	0.019
	Low		Low
c	Low		High	26/87	2131	0.526 (0.320–0.865)	0.011	0.365 (0.207–0.643)	< 0.001
Note: ɸ: P values were adjusted for radical resection, tumor stage and HBV infection; Bold indicates significant P values.
Abbreviations: NA: not available; MST: median survival time; HR: hazard ratio; 95%CI: 95% confidence interval.

Table 4

Risk score model constructed by *LINC01116* and *TMSB15A*
Variables	β	SE	Wald	P value	HR (95% CI)
Tumor stage I			15.251	< 0.001
Stage II	0.378	0.267	1.995	0.158	1.459 (0.864–2.463)
Stage III + IV	0.896	0.231	15.099	0.000	2.450 (1.559–3.849)
Radical resection	0.215	0.356	0.364	0.547	1.240 (0.617–2.492)
HBV infection	-0.760	0.262	8.397	0.004	0.468 (0.280–0.782)
LINC01116	0.401	0.201	3.963	0.047	1.493 (1.006–2.215)
TMSB15A	-0.615	0.204	9.081	0.003	0.541 (0.362–0.806)

Table 5

Candidate pharmacological targets toward *LINC01116*
Name	PubChem CID	Mean	Enrichment	P value
Thiamine	1130	-0.783	-0.95	0.00022
Cromoglicic acid	2882	-0.831	-0.98	0.00087
Rilmenidine	68712	-0.396	-0.845	0.00103
Chlorhexidine	9552079	-0.402	-0.671	0.00891
Sulindac sulfone	5472495	-0.555	-0.932	0.00936
Chloropyrazine	73277	-0.394	-0.685	0.02158
Meprylcaine	4065	-0.391	-0.668	0.02741

No competing interests reported.

Supplementarymaterials.zip

Download PDF

Journal Publication

published 20 Jun, 2023

Read the published version in BMC Gastroenterology →

Editorial decision: Major revision
28 Feb, 2023
Reviews received at journal
27 Feb, 2023
Reviews received at journal
30 Jan, 2023
Reviewers agreed at journal
12 Jan, 2023
Reviewers agreed at journal
07 Jan, 2023
Reviewers invited by journal
04 Jan, 2023
Editor assigned by journal
04 Jan, 2023
Editor invited by journal
15 Jun, 2022
Submission checks completed at journal
15 Jun, 2022
First submitted to journal
11 Jun, 2022

You are reading this latest preprint version

Comprehensive analysis of candidate signatures of long non-coding RNA LINC01116 and related protein-coding genes in patients with hepatocellular carcinoma

Status:

Journal Publication

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Introduction

Materials And Methods

Results

Discussion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1