CircRNA-associated ceRNA network reveals ErbB and Hippo signaling pathways in hypopharyngeal cancer
- Authors:
- Published online on: October 19, 2018 https://doi.org/10.3892/ijmm.2018.3942
- Pages: 127-142
-
Copyright: © Feng et al. This is an open access article distributed under the terms of Creative Commons Attribution License.
Abstract
Introduction
Hypopharyngeal carcinoma is a primary malignant tumor of the hypopharynx, accounting for 3-5% of the malignancies in the upper aerodigestive tract. Early diagnosis of hypopharyngeal cancer is hard because the early stages of hypopharyngeal carcinoma have no specific symptoms. Studies have reported that 60-80% of these patients had ipsilateral lymph node metastases and ≤40% of these patients have contralateral occult lymph node tumor deposits (1-3). Thus, the majority of patients with hypopharyngeal cancer have a poor prognosis and low survival rate (4). Therefore, identifying early stage indicators or biomarkers to improve patient survival is urgent.
Unlike normal linear RNA, the 3′ and 5′ ends of circular RNAs (circRNAs) are linked by covalent bonds and lack polarities or polyadenylated tails, thereby rendering them stable in tissues, serum and urine (5). Owing to this characteristic, the potential of circRNAs as biomarkers for human cancer has attracted significant focus. In addition, circRNAs are widely involved in cancer; ciRS-7 in HeLa cells (6), Hsa_ circ_001569 in colorectal cancer (7), circHIPK3 in several types of cancer (8), f-circM9, f-circPR in hematological malignancy (9), and circTCF25 in urinary bladder carcinoma (10). Previous studies have demonstrated that the main function of circRNAs is that they can function as a microRNA (miRNA) sponge, binding to miRNAs and regulating them and their downstream gene targets, through a competing endogenous (ce) RNA mechanism (11).
The present study comprehensively investigated the expression profile of circRNAs in HCa patients. The results identified a circRNA signature in HCa and suggested that a core miRNA-ceRNA network, regulating both the ErbB and Hippo signaling pathways, may have important roles in HCa progression.
Materials and methods
Patients and specimens
The study included three patients with HCa who underwent partial or radical cystectomies at the First Affiliated Hospital of Kunming Medical University (Kunming, China); samples were collected from March 2017 to October 2017. All three patients were male and their ages were 44, 54 and 56. Following surgery, the matched specimens were immediately preserved in liquid nitrogen until use. All patient samples were confirmed by pathological examination and none of the patients received neoadjuvant therapy. The study was approved by the Second Department of Otolaryngology Head and Neck Surgery of the First Affiliated Hospital of Kunming Medical University (Kunming, China). Written informed consent was obtained from all the participants in the study.
Total RNA isolation and quality control
Total RNA was isolated from samples using TRIzol reagent (Thermo Fisher Scientific, Inc., Waltham, MA, USA) following the manufacturer’s protocol. The quantity and quality of total RNA samples were measured using NanoDrop ND-1000 (Thermo Fisher Scientific, Inc.). RNA integrity was assessed and confirmed via electrophoresis using denaturing agarose gels. Isolated RNA samples were stored at −80°C prior to use.
Library preparation and sequencing
Total RNA from three matched HCa samples and adjacent normal tissues were treated with Epicenter Ribo-Zero rRNA Removal kit (Illumina, Inc., San Diego, CA, USA) and RNase R (Epicenter; Illumina, Inc.) to remove ribosomal and linear RNA. Then, the RNA-seq libraries were constructed using TruSeq Stranded Total RNA HT/LT Sample Prep kit (Illumina, Inc.). Sequencing was determined on Illumina Hiseq 2500 instrument with 2×150 bp paired reads.
Computational analysis of circRNAs
The clean reads were obtained after the raw reads were preprocessed with the FastQC quality control tool (12). CircRNAs were identified using CIRI (v.1.2) pipeline with default parameters (13). Genomic circRNAs were mapped to the human reference genome (GRCh37) by BWA (14). All circRNAs were annotated for circRNA-hosting genes with the application of GENCODE v24 (15). The identified circRNAs were converted to circRNA ID with web server circBase (16).
Principal component analysis (PCA)
PCA was performed as previously described (17). A total of 4,634 distinct circRNAs with non-zero raw counts across the six samples were isolated and expressions of circRNAs were normalized with the reads per Million mapped reads (RPM) method and the expression matrix (each row represented a gene, each column represented a sample) were used for PCA. The prcomp package from R was used to perform PCA and the default parameters were used (18). The ggplot2 package from R was used to draw the scatter plot (19).
Normalization and differential expression analysis of circRNAs
Two steps were performed to normalize circRNA expression for depth. Firstly, the total back-spliced reads in a sample were counted and that number was divided by 1,000,000. This resulted in the ‘per million’ scaling factor. Secondly, the read counts were divided by the ‘per million’ scaling factor. This method normalized for sequencing depth, giving RPM. CircRNAs were isolated with RPM>0 across 6 samples and Mann-Whitney U test (20) (paired=T) followed by Benjamini-Hochberg multiple testing correction (21) were applied to identify the differentially expressed (DE) circRNAs. FDR<0.05 and a fold change of >2.0 or <0.5 were the selection criteria for significant DE circRNAs.
Functional enrichment analysis
Gene ontology (GO) term enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were conducted with web server DAVID 6.8 (22). P<0.05 was considered as statistically significant.
CeRNA network
The top 20 upregulated circRNAs and the top 20 downregulated circRNAs were used to survey miRNA targets with the web tool CircInteractome (23). Specifically, CircInteractome downloads the mature sequences of circRNAs from the UCSC browser mirror (http://genome.mdc-berlin.de) (24) and predicts miRNAs that target circRNA by surveying for 7-mer or 8-mer complementarity to the seed region, as well as the 3′end of each miRNA using the TargetScan algorithm (25). The complete miRNA list and sequences were taken from the miRBase (http://www.mirbase.org/) (16). miRNA downstream targets were isolated with mirPath 3.0 (26) which was also used for miRNA KEGG pathway analysis. The ceRNA network was displayed by Cytoscape (v3.5.1) (27).
Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)
Total RNA was extracted from pooled normal and tumor tissue samples using TRIzol (Thermo Fisher Scientific, Inc.), and 1 µg of total RNA was reverse transcribed into first-strand cDNA using a PrimeScript RT Reagent kit (Takara Bio, Inc., Otsu, Japan), according to the manufacturer’s protocols. qPCR was performed with a SYBR-Green real-time PCR kit (Thermo Fisher Scientific, Inc.) using the ABI StepOnePlus Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.). CircRNAs were analyzed with 18s rRNA as the internal standard and miRNA was analyzed with U6 as the internal standard. The reactions were prepared as follows: 7.5 µl SYBR Premixm Ex Taq II, 0.25 µl ROX Reference Dye II, 0.125 µl forward primer, 0.125 µl reverse primer, 5 µl RNase-free water, and 2 µl cDNA. The thermocycling conditions were: one step at 95°C for 30 sec, followed by 40 cycles of 95°C for 5 sec and 60°C for 30 sec, and a final step of 95°C for 15 sec, 60°C for 15 sec and 95°C for 15 sec. Primer sequences are listed in Table I; expression levels were quantified via the 2−ΔΔCq method (28).
 | Table IPrimer sequences used for reverse transcription-quantitative polymerase chain reaction analysis. |
Expression analysis of miR-548c-3p
Two methods were used to investigate the expression of miR-548c-3p among normal and tumor samples. The first was RT-qPCR, as detailed above. The second was in-silico analysis. The miRNA dataset of the esophageal carcinoma cohort from The Cancer Genome Atlas (TCGA) project (29) was exploited. There were 13 normal samples and 184 tumor samples in this dataset. Normalized miRNA expressions of miR-548c-3p were compared between normal and tumor samples. Mann-Whitney U test was applied to test the significance.
Survival analysis
A Kaplan-Meier curve was used to examine the clinical relevance of miR-548c-3p levels in the patients’ outcomes (30). Patients were separated into two groups according to the median expression of hsa-miR-548c-3p using TCGA clinical and expression dataset. Differences between groups were analyzed using log-rank test (31) and two-tailed P-values <0.05 were considered statistically significant. Statistical analyses were performed using the survival package (version 2.39-5) in R (version 3.4.3).
Expression correlation of hsa-miR-548c-3p and its targeted genes
The miRNA and mRNA datasets of the esophageal carcinoma cohort from TCGA (29) were used for the correlation analysis. Common samples were isolated according to the sample barcodes. The Pearson correlation method was used to assess the expression association between hsa-miR-548c-3p and the targeted genes. Significance of association was determined by the R package cor.test (alternative=‘two.sided’, method=‘pearson’). Then, P-values were corrected with Benjamini-Hochberg procedure for multiple testing.
Statistical analysis
All statistical analyses were generated using R (32). The Pearson correlation method was used to assess the expression association. Significances of associations were determined by the R package cor.test. Mann-Whitney U test was used for comparisons between two groups. Benjamini-Hochberg procedure was applied for multiple testing. Log-rank test was used for Kaplan-Meier survival curves. P<0.05 was considered to indicate a statistically significant difference.
Results
Identification of DE circRNAs in HCa
To identify DE circRNAs in HCa, circRNA sequencing (Seq) was performed using three matched normal and HCa tissue samples, and an average of 90 million reads was achieved for each sample. A total of 4,634 distinct circRNAs with at least two unique back-spliced reads across six samples using CIRI pipeline (13) were identified and the expressions of circRNAs were normalized and represented by reads per million mapped reads (RPM) values. Genetic distances across 6 samples were evaluated using PCA (Fig. 1A), and the normalized expression level (RPM) of circRNAs across the six samples is illustrated in Fig. 1B. Following statistical analysis, 71 and 102 circRNAs were determined to be significantly upregulated and downregulated, respectively (Table II). The DE circRNAs between tumor and adjacent normal samples were presented in a heatmap (Fig. 1C). To confirm the circRNA-Seq results, RT-qPCR was performed to assess the expression of 19 of the above DE circRNAs in both normal and tumor samples. The results confirmed that 12 of them were consistently upregulated or downregulated with the circRNA-Seq results (Fig. 2).
Next, the distribution of circRNAs in different DNA elements and chromosomes was examined. The bar diagram of Fig. 3A demonstrates the % of back-spliced junction reads on intron, intergenic, and exon areas. The majority of circRNAs belonged to exonic, followed by intronic and intergenic elements (Fig. 3B). These dysregulated circRNAs are widely distributed in all chromosomes, including sex chromosomes X (Fig. 3C).
Functional enrichment analysis of genes producing DE circRNAs
To reveal the dysregulated pathways underlying HCa, first KEGG pathway enrichment analyses were performed for genes that matched DE circRNAs. The results demonstrated that genes containing downregulated circRNAs were enriched in endocytosis, ubiquitin-mediated proteolysis, and Janus kinase (JAK)/signal transducer and activator of transcription (STAT) signaling pathways (Fig. 4A), whereas there were no KEGG pathways enriched with genes producing upregulated circRNAs.
Next, GO term enrichment analyies was performed for genes that produced aberrantly expressed circRNAs. Biological processes, such as the establishment of spindle orientation, response to fungicide, positive regulation of transcription, cell division were significantly enriched (Fig. 4B), whereas genes producing downregulated circRNAs were related to autophagy, mitochondrion organization actin cytoskeleton organization, membrane fission, and cell-cell adhesion pathways (Fig. 4C). These results suggested that multiple pathways may contribute to HCa pathogenesis and progression.
CircRNAs regulate the ErbB and Hippo pathways through a miRNA-CeRNA network
The role of circRNAs as a miRNA sponge is the main mechanism of circRNA function in tumor cells (11,33). Therefore, we further investigated the roles of circRNAs in HCa progression through establishing a ceRNA network. Firstly, the top 20 upregulated and top 20 downregulated circRNAs were isolated and were converted to circRNA ID using circBase database (34). Secondly, miRNAs targeting DE-circRNAs were isolated with the web server CircInteractome (23). Specifically, CircInteractome downloaded the mature sequences of all of the reported circRNAs from the UCSC browser, then to characterize miRNA-circRNA interactions, CircInteractome incorporated the ability to search using the TargetScan algorithm, which predicts miRNAs that target circRNA by surveying for 7-mer or 8-mer complementarity to the seed region as well as the 3′end of each miRNA (23). A total of 191 and 182 miRNAs were putatively identified as the targets of upregulated and downregulated circRNAs, respectively. Networks consisted of circRNAs and miRNAs were displayed using Cytoscape software (27). The results demonstrated extensive interactions between miRNAs and upregulated (Fig. 5A), and downregulated circRNAs (Fig. 5B). Then, KEGG pathway enrichment analysis was performed for the miRNAs targeted by the top 40 DE circRNAs, in order to explore the altered biological processes using mirPath 3.0 (26). Genes targeted by miRNAs were significantly enriched in multiple signaling pathways, including the ErbB, the Hippo, the Ras, the transforming growth factor (TGF)-β, the phosphoinositide 3-kinase/AKT serine/threonine kinase and the Wnt signaling pathways (Fig. 5C).
To get further insight into the function of circRNAs in the ErbB and Hippo signaling pathways, miRNA-ceRNA networks were constructed corresponding to the two pathways using Cytoscape. For the miRNA-ceRNA network regulating the ErbB pathway, there were 33 circRNAs, 43 miRNAs and 74 ErbB pathway genes (Fig. 6A). In the ErbB miRNA-ceRNA network, we isolated a subnetwork consisting of circRNAs (hsa_circ_0008287 and hsa_circ_0005027), miRNAs (hsa-miR-548c-3p) and 38 ErbB pathway genes which had the most interaction between miRNAs and targeted genes (Fig. 6B). Hsa_circ_0008287 and hsa_circ_0005027 were significantly downregulated in tumor samples compared with normal (Figs. 2 and 6C). In a similar manner, the miRNA-ceRNA network regulating the Hippo pathway was constructed, consisting of 33 circRNAs, 43 miRNAs and 110 Hippo pathway genes (Fig. 7A). In the Hippo miRNA-ceRNA network, we also isolated a subnetwork consisting of circRNAs (hsa_circ_0008287 and hsa_circ_0005027), miRNAs (hsa-miR-548c-3p) and 61 Hippo pathway genes, which had the most interaction between miRNAs and targeted genes (Fig. 7B).
To further investigate the important role of this subnet-work in tumor progression, the miRNA and mRNA datasets of the esophageal carcinoma cohort from TCGA (29) were exploited. The esophageal carcinoma cohort contains 13 normal samples and 184 tumor samples. In this cohort, the miRNA hsa-miR-548c-3p expression between normal and tumor samples was detected, and its clinical relevance to patient survival was analyzed. The results suggested that hsa-miR-548c-3p was highly expressed in tumor samples compared with normal samples (Fig. 8A and B), and its high expression was significantly associated with lower survival in patients with esophageal carcinoma (Fig. 8C). These findings suggested that hsa-miR-548c-3p is an oncogenic miRNA, which is consistent with the hypothesis that in tumor samples circRNAs were downregulated resulting in more oncogenic hsa-miR-548c-3p being released, and highly expressed hsa-miR-548c-3p may promote HCa progression through downstream target genes. To confirm the negative regulation of hsa-miR-548c-3p on the ErbB and Hippo pathway genes, the expression correlation of hsa-miR-548c-3p and its targeted genes were also analyzed. Many of the targeted genes were negatively correlated with hsa-miR-548c-3p levels, which supported a negative regulatory role of hsa-miR-548c-3p on the ErbB and Hippo pathways (Table III). The present results demonstrated that circRNAs regulate HCa progression through multiple pathways and identifying a miRNA-ceRNA network that regulated the ErbB and Hippo signaling pathways.
| Table IIIExpression correlation of hsa-miR-548c-3p and its targeted genes in The Cancer Genome Atlas esophageal carcinoma cohort. |
Discussion
HCa is clinically difficult to diagnose and has a poor prognosis, therefore, identifying early stage molecular biomarkers has become urgent. CircRNAs, which are stable and easier to extract and detect, are considered ideal candidates for early-stage biomarkers. This is the first report on the expression profile of circRNAs in HCa. In the present study, a number of aberrantly expressed circRNAs in HCa samples were identified. Pathway enrichment results revealed that circRNAs may regulate HCa progression through multiple signaling pathways, especially the ErbB and Hippo signaling pathways. These results provided several potential biomarkers and therapeutic targets for HCa.
The ceRNA hypothesis was described as a way that RNAs communicate with each other, via competing for binding to miRNAs and regulating the expression of each other to construct a complex post-transcriptional regulatory network (35,36). mRNAs and long non-coding (lnc) RNAs may all serve as ceRNAs (37). It has been demonstrated that circRNAs can also function as miRNA sponges (6,11). The present study demonstrated that aberrantly expressed circRNAs have extensive interactions with miRNAs, and those miRNAs exerted their effect on multiple cancer-related pathways. These data indicated that the circRNA-associated ceRNA network may have crucial roles in HCa progression.
The activation of ErbB oncogenes has been described in various types of human tumors, including hypopharynx carcinomas, and it has been correlated with a poor prognosis. For example, one study describing the molecular alterations in hypopharynx carcinomas demonstrated that ErbB1 was amplified in 29% of patients with hypopharyngeal squamous cell carcinomas (38). In addition, ErbB1 amplification is correlated with a hypopharyngeal primary site (39). Another study reported that v-erbB stained positively in 62.5% of hypopharyngeal squamous cell carcinomas samples but negatively in normal mucosa (40). The present ceRNA network analysis demonstrated that a circRNA (hsa_circ_0008287 and hsa_circ_0005027)/miRNA (hsa-miR-548c-3p) axis may have important roles in ErbB-mediated tumor progression (Fig. 6).
Another pathway that is likely to be associated with hypopharynx carcinomas is the Hippo signaling pathway. The Hippo pathway has generated considerable interest in recent years because of its involvement in several key hallmarks of cancer progression and metastasis (41). Regulation of Hippo signaling can be an attractive alternative strategy for cancer treatment (42-44). Previously, ACTL6A and p63 were demonstrated to cooperatively promote head and neck squamous cell carcinoma, through activation of the Hippo/Yes-associated protein 1 (YAP) pathway and YAP activation can predict poor patient survival (45). The present ceRNA network analysis demonstrated that a circRNA (hsa_circ_0008287 and hsa_circ_0005027)/miRNA (hsa-miR-548c-3p) axis may have important roles in Hippo-mediated tumor progression (Fig. 7).
Extensive evidence has suggested that miRNAs have important roles in breast cancer. The miR-548 family has been demonstrated to be involved in the pathogenesis of several cancers. For example, miR-548-3p was significantly downregulated in breast cancer and overexpression of miR-548-3p inhibited the proliferation and promoted the apoptosis of breast cancer cells (46). Overexpression of miR-548c-3p was also confirmed in prostate epithelial stem cells and in castration-resistant prostate cancer cells (45). Overexpression of miR-548c-3p in differentiated cells induced stem-like properties and radio-resistance (45). Re-analyses of published studies further revealed that miR-548c-3p is significantly overexpressed in castration-resistant prostate cancer cells and is associated with poor recurrence-free survival, suggesting that miR-548c-3p is a functional biomarker for prostate cancer aggressiveness (47). The present results demonstrated that miR-548c-3p may have important roles in HCa progression through modulating the ErbB and Hippo pathways. Due to the crucial roles of miR-548c-3p in multiple types of cancer, development of novel gene therapies based on miR-548c-3p might be encouraged.
Taken together, the present study indicated that hsa_ circ_0008287 and hsa_circ_0005027 were downregulated in HCa and competitively bound miR-548c-3p with ErbB and Hippo signaling pathway genes. Further studies are warranted on the roles of hsa_circ_0008287, hsa_circ_0005027, and miR-548c-3p as potential diagnostic biomarkers and therapeutic targets for HCa.
Acknowledgments
Not applicable.
Funding
This work was funded by Yunnan Applied Basic Research Projects (grant no. 2016FB038).
Availability of data and materials
The sequencing data have been deposited in the Gene Expression Omnibus (GEO) database under the accession number GSE111423.
Authors’ contributions
CF designed experiments and helped analyze the data. YaL, YuL and XC collected the samples. DL analyzed data. HZ interpreted the results and wrote the manuscript. XH designed experiments and interpreted the results. All authors read and approved the final manuscript.
Ethics approval and consent to participate
The study was approved by the Ethics Committee of the First Affiliated Hospital of Kunming Medical University (Kunming, China). Written informed consent was obtained from all the participants in the study.
Patient consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Abbreviations:
|
HCa |
hypopharyngeal cancer |
|
DE |
differentially expressed |
|
ceRNAs |
competing endogenous RNA |
|
RPM |
reads per Million mapped reads |
|
GO |
gene ontology |
|
KEGG |
Kyoto Encyclopedia of Genes and Genomes |
|
PCA |
principal component analysis |
References
|
Pingree TF, Davis RK, Reichman O and Derrick L: Treatment of hypopharyngeal carcinoma: A 10-year review of 1,362 cases. Laryngoscope. 97:901–904. 1987. View Article : Google Scholar : PubMed/NCBI | |
|
Chan JY and Wei WI: Current management strategy of hypopharyngeal carcinoma. Auris Nasus Larynx. 40:2–6. 2013. View Article : Google Scholar | |
|
Lagha A, Chraiet N, Labidi S, Rifi H, Ayadi M, Krimi S, Allani B, Raies H, Touati S and Boussen H: Larynx preservation: What is the best non-surgical strategy. Crit Rev Oncol Hematol. 88:447–458. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Carrasco Llatas M, Lopez Molla C, Balaguer Garcia R, Ferrer Ramírez MJ, Guallart Doménech F, Estellés Ferriol JE, Fernández Martínez S and Dalmau Galofre J: Hypopharyngeal cancer: Analysis of the evolution and treatment results. Acta Otorrinolaringol Esp. 60:3–8. 2009.In Spanish. View Article : Google Scholar : PubMed/NCBI | |
|
Chen LL and Yang L: Regulation of circRNA biogenesis. RNA Biol. 12:381–388. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Memczak S, Jens M, Elefsinioti A, Torti F, Krueger J, Rybak A, Maier L, Mackowiak SD, Gregersen LH, Munschauer M, et al: Circular RNAs are a large class of animal RNAs with regulatory potency. Nature. 495:333–338. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Xie H, Ren X, Xin S, Lan X, Lu G, Lin Y, Yang S, Zeng Z, Liao W, Ding YQ and Liang L: Emerging roles of circRNA_001569 targeting miR-145 in the proliferation and invasion of colorectal cancer. Oncotarget. 7:26680–26691. 2016.PubMed/NCBI | |
|
Zheng Q, Bao C, Guo W, Li S, Chen J, Chen B, Luo Y, Lyu D, Li Y, Shi G, et al: Circular RNA profiling reveals an abundant circHIPK3 that regulates cell growth by sponging multiple miRNAs. Nat Commun. 7:112152016. View Article : Google Scholar : PubMed/NCBI | |
|
Guarnerio J, Bezzi M, Jeong JC, Paffenholz SV, Berry K, Naldini MM, Lo-Coco F, Tay Y, Beck AH and Pandolfi PP: Oncogenic role of fusion-circRNAs derived from cancer-associated chromosomal translocations. Cell. 165:289–302. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Zhong ZY, Lv MX and Chen JX: Screening differential circular RNA expression profiles reveals the regulatory role of circTCF25-miR-103a-3p/miR-107-CDK6 pathway in bladder carcinoma. Sci Rep. 6:309192016. View Article : Google Scholar : PubMed/NCBI | |
|
Hansen TB, Jensen TI, Clausen BH, Bramsen JB, Finsen B, Damgaard CK and Kjems J: Natural RNA circles function as efficient microRNA sponges. Nature. 495:384–388. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Brown J, Pirrung M and McCue LA: FQC Dashboard: Integrates FastQC results into a web-based, interactive, and extensible FASTQ quality control tool. Bioinformatics. 33:3137–3139. 2017. View Article : Google Scholar | |
|
Gao Y, Wang J and Zhao F: CIRI: An efficient and unbiased algorithm for de novo circular RNA identification. Genome Biol. 16:42015. View Article : Google Scholar : PubMed/NCBI | |
|
Li H and Durbin R: Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 25:1754–1760. 2009. View Article : Google Scholar : PubMed/NCBI | |
|
Harrow J, Frankish A, Gonzalez JM, Tapanari E, Diekhans M, Kokocinski F, Aken BL, Barrell D, Zadissa A, Searle S, et al: GENCODE: The reference human genome annotation for the ENCODE project. Genome Res. 22:1760–1774. 2012. View Article : Google Scholar : PubMed/NCBI | |
|
Kozomara A and Griffiths-Jones S: miRBase: Integrating microRNA annotation and deep-sequencing data. Nucleic Acids Res. 39:D152–D157. 2011. View Article : Google Scholar : | |
|
Lever J, Krzywinski M and Atman N: Points of Significance: principal component analysis. Nat Methods. 14:641–642. 2017. View Article : Google Scholar | |
|
Hotelling H: Analysis of a complex of statistical variables into principal components. J Educ Psychol. 24:417–441. 1933. View Article : Google Scholar | |
|
Ginestet C: ggplot2: Elegant graphics for data analysis. J R Stat Soc Ser A. 174:245. 2011. View Article : Google Scholar | |
|
Harris T and Hardin JW: Exact Wilcoxon signed-rank and Wilcoxon Mann-Whitney ranksum tests. Stata J. 13:337–343. 2013. | |
|
Benjamini Y and Hochberg Y: Controlling the false discovery rate: A practical and powerful approach to multiple testing. J Roy Stat Soc B Met. 57:289–300. 1995. | |
|
Huang da W, Sherman BT and Lempicki RA: Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 4:44–57. 2009. View Article : Google Scholar : PubMed/NCBI | |
|
Dudekulay DB, Panda AC, Grammatikakis I, De S, Abdelmohsen K and Gorospe M: CircInteractome: A web tool for exploring circular RNAs and their interacting proteins and microRNAs. RNA Biol. 13:34–42. 2016. View Article : Google Scholar | |
|
Casper J, Zweig AS, Villarreal C, Tyner C, Speir ML, Rosenbloom KR, Raney BJ, Lee CM, Lee BT, Karolchik D, et al: The UCSC genome browser database: 2018 update. Nucleic Acids Res. 46:D762–D769. 2018. | |
|
Agarwal V, Bell GW, Nam JW and Bartel DP: Predicting effective microRNA target sites in mammalian mRNAs. Elife. 4:75542015. View Article : Google Scholar | |
|
Vlachos IS, Zagganas K, Paraskevopoulou MD, Georgakilas G, Karagkouni D, Vergoulis T, Dalamagas T and Hatzigeorgiou AG: DIANA-miRPath v3.0: Deciphering microRNA function with experimental support. Nucleic Acids Res. 43:W460–W466. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B and Ideker T: Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 13:2498–2504. 2003. View Article : Google Scholar : PubMed/NCBI | |
|
Livak KJ and Schmittgen TD: Analysis of relative gene expression data using real-time quantitative PCR and the 2(T)(−Delta Delta C) method. Methods. 25:402–408. 2001. View Article : Google Scholar | |
|
Cancer Genome Atlas Research Network Analysis Working Group, Asan University, BC Cancer Agency; et al: Integrated genomic characterization of oesophageal carcinoma. Nature. 541:169–175. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Lancar R and Funck-Brentano C: Survival analysis example based on an event history model from a clinical trial in cardiology. Rev Epidemiol Sante Publique. 47:613–618. 1999.In French. | |
|
Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep. 50:163–170. 1966.PubMed/NCBI | |
|
R Development Core Team: R: A language and environment for statistical computing Vienna, Austria: The. R Foundation for Statistical Computing; 2011 | |
|
Guo JU, Agarwal V, Guo H and Bartel DP: Expanded identification and characterization of mammalian circular RNAs. Genome Biol. 15:4092014. View Article : Google Scholar : PubMed/NCBI | |
|
Glazar P, Papavasileiou P and Rajewsky N: circBase: A database for circular RNAs. RNA. 20:1666–1670. 2014. View Article : Google Scholar : PubMed/NCBI | |
|
Tay Y, Rinn J and Pandolfi PP: The multilayered complexity of ceRNA crosstalk and competition. Nature. 505:344–352. 2014. View Article : Google Scholar : PubMed/NCBI | |
|
Salmena L, Poliseno L, Tay Y, Kats L and Pandolfi PP: A ceRNA hypothesis: The rosetta stone of a hidden RNA language. Cell. 146:353–358. 2011. View Article : Google Scholar : PubMed/NCBI | |
|
Karreth FA and Pandolfi PP: ceRNA cross-talk in cancer: When ce-bling rivalries go awry. Cancer Discov. 3:1113–1121. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Rodrigo JP, González MV, Lazo PS, Ramos S, Coto E, Alvarez I, García LA and Suárez C: Genetic alterations in squamous cell carcinomas of the hypopharynx with correlations to clinico-pathological features. Oral Oncol. 38:357–363. 2002. View Article : Google Scholar : PubMed/NCBI | |
|
Rodrigo JP, Ramos S, Lazo PS, Alvarez I and Suarez C: Amplification of ERBB oncogenes in squamous cell carcinomas of the head and neck. Eur J Cancer. 32A:2004–2010. 1996. View Article : Google Scholar : PubMed/NCBI | |
|
Otsu M, Hayashi Y, Amatsu M and Itoh H: Immunohistochemical study of p53, EGF, EGF-receptor, v-erb B and ras p21 in squamous cell carcinoma of hypopharynx. Kobe J Med Sci. 40:139–153. 1994.PubMed/NCBI | |
|
Ma Y, Yang Y, Wang F, Wei Q and Qin H: Hippo-YAP signaling pathway: A new paradigm for cancer therapy. Int J Cancer. 137:2275–2286. 2015. View Article : Google Scholar | |
|
Santucci M, Vignudelli T, Ferrari S, Mor M, Scalvini L, Bolognesi ML, Uliassi E and Costi MP: The Hippo pathway and YAP/TAZ-TEAD protein-protein interaction as targets for regenerative medicine and cancer treatment. J Med Chem. 58:4857–4873. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Guo L and Teng L: YAP/TAZ for cancer therapy: Opportunities and challenges (Review). Int J Oncol. 46:1444–1452. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Liu AM, Xu MZ, Chen J, Poon RT and Luk JM: Targeting YAP and Hippo signaling pathway in liver cancer. Expert Opin Ther Targets. 14:855–868. 2010. View Article : Google Scholar : PubMed/NCBI | |
|
Saladi SV, Ross K, Karaayvaz M, Tata PR, Mou H, Rajagopal J, Ramaswamy S and Ellisen LW: ACTL6A Is Co-amplified with p63 in squamous cell carcinoma to drive YAP activation, regenerative proliferation, and poor prognosis. Cancer Cell. 31:35–49. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Shi Y, Qiu M, Wu Y and Hai L: MiR-548-3p functions as an anti-oncogenic regulator in breast cancer. Biomed Pharmacother. 75:111–116. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Rane JK, Scaravilli M, Ylipää A, Pellacani D, Mann VM, Simms MS, Nykter M, Collins AT, Visakorpi T and Maitland NJ: MicroRNA expression profile of primary prostate cancer stem cells as a source of biomarkers and therapeutic targets. Eur Urol. 67:7–10. 2015. View Article : Google Scholar |
