Constructing Novel ceRNA Networks on MiR-616-3p Reveals LncRNAs as Potential Colorectal Cancer Prognostic Biomarkers

doi:10.21203/rs.3.rs-1455403/v1

Download PDF

Research Article

Constructing Novel ceRNA Networks on MiR-616-3p Reveals LncRNAs as Potential Colorectal Cancer Prognostic Biomarkers

https://doi.org/10.21203/rs.3.rs-1455403/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background. Long noncoding RNAs (lncRNAs) perform as competing endogenous RNAs (ceRNAs) to sponge microRNAs (miRNAs) lead to the advancement of cancer. In this work, we attempted to identify possible long noncoding RNA biomarkers as well as a new regulatory axis in colorectal cancer. This investigation was carried out in consideration of the critical oncogenic function played by miR-616-3p in many malignancies, as well as its regulatory involvement in several signaling pathways.

Material and Methods: Differentially expressed lncRNAs (DELs) were identified from Gene Expression Omnibus (GEO) database. Targeted mRNAs were recognized by the Targetscan database. ciBioPortal database and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis also were applied to identify mRNAs. CeRNA networks constructed via Cytoscape 3.7.1. QRT-PCR was utilized to verify these RNA molecules' expression levels in 30 CRC tissues compared to 30 adjacent tissue samples.

Results: Three lncRNAs and three mRNAs were discovered to have the greatest interaction with miR-616-3p. QRT-PCR revealed overexpression of (Linc01282, lnc-MYADM-1:1, ZNF347, XIAP) and downregulation of (Lnc-atp12a-1:1, SIK1 and miR-616-3p). In CRC, bioinformatic research revealed many unique ceRNA networks centered on miR-616-3p that were previously unknown.

Conclusion: This work reveals novel, previously unreported lncRNAs as prognostic biomarkers for CRC, as well as potential mRNAs as new therapeutic targets and predictive biomarkers for CRC. Furthermore, our analysis identified novel ceRNA networks that need be investigated further in CRC.

ceRNA

CRC

Long noncoding RNA

miR-616-3p

SIK1

XIAP

Considered the fourth most malignant, colorectal cancer (CRC) is the third most common cancer globally. (Xian and Zhao 2019) It is arduous to recognize tumors smaller than 1 cm with typical diagnostic methods. Therefore, early diagnosis is vital, and surgery is the only treatment for patients before metastasis. CRC genesis involves intricate steps that need the investigation of similar biomarkers. (Li et al. 2018, Zhang et al. 2019) The progression and development of colorectal cancer involve a set of intricate epigenetic and genetic alterations. Current findings have uncovered that almost 98% of human genome transcripts comprise noncoding RNAs (ncRNAs) without translated protein. (Bian et al. 2018) microRNAs (miRNAs) are a class of single-stranded, short, and conserved ncRNAs 18 ± 24 nucleotides in length that negatively modulate gene expression at the post-transcriptional level by binding to the 3ʹ-untranslated region (3ʹUTR) of target genes. (Zhou et al. 2020, Zhang et al. 2018) Several studies have suggested that miRNAs regulate many physiological processes in the cell, including migration, invasion, cell proliferation, and apoptosis. Aberrant miRNAs also participate in the initiation, promotion, angiogenesis, metastasis, and drug resistance of several types of cancers. (Zhang et al. 2018, Mo et al. 2020) Many studies apply putative miRNA target data for recognizing miRNA sponge interactions and hypergeometric examinations to identify candidate miRNA sponges. However, our research utilized experimentally validated miRNA-target interactions.

In terms of some former studies, miR-616 had a significant regulatory role in various cancer signaling pathways. For example, Bai Q et al. demonstrated that upregulation of SOX7 or downregulation of miR-616 could suppress the function of the Wnt/β-catenin signaling pathway in glioma. Thus, miR-616 performed as a tumor promoter in glioma and can regulate SOX7 gene and Wnt/β-catenin signaling by its oncogenic roles. (Bai et al. 2017) Wu Z-H et al. indicated that the impacts of miR-616-3p on angiogenesis and metastasis occurred by the downregulation of PTEN, one of the miR-616-3p direct targets. We suggest that the renewal of PTEN expression might obstruct miR-616-3p-induced angiogenesis. Overall, our results offer that the miR-616-3p-PTEN signaling axis may be a therapeutic target for gastric cancer. (Wu et al. 2018) Many studies have presented abnormal expression of miR-616-3p in various cancers such as ovarian, gastric, and breast cancer, resulting in cell invasion, migration, EMT, and maybe a potential therapeutic target. (Wu et al. 2018, Chen et al. 2018, Nguyen et al. 2020, Yuan 2019) However, no study has reported the involvement of miR-616-3p in CRC. Thus, it is the first time that miR-616-3p is evaluated as a candidate microRNA for further exploration in a novel ceRNA network.

Long noncoding RNA (lncRNA) are described as transcripts with a length that extends beyond 200 nucleotides, without protein-coding potential. (Gao et al. 2020) LncRNAs have been considered a hotspot in biomedical research recently. They play crucial roles in the modulation of apoptosis, proliferation, cell cycle, differentiation, and else. (Liu et al. 2019) LncRNAs also act as competing endogenous RNAs (ceRNAs) that sponge miRNAs, affecting the expression of the target gene. (Xian and Zhao 2019) CeRNAs, as a vital class of post-transcriptional modulators, describes the pathogenesis of lncRNAs in cancer progression and tumorigenesis. They change the expression of crucial tumor-suppressive or tumorigenic genes by a microRNA-mediated mechanism. Since the introduction of this mechanism, it has been validated that ceRNA acts in the progression of different tumors and has been constructed in pancreatic adenocarcinoma, (Weng et al. 2020) breast cancer, (Tuersong et al. 2019) colorectal cancer, (Guo et al. 2020) lung cancer, (Zhu et al. 2019) and gastric cancer (Pan, Guo, et al. 2019) and so on. Thus, the ceRNA network might perform as a biomarker for predicting, diagnosing, and prognosis curative responses in CRC. (Hollstein et al. 2019)

Therefore, in the present study, we sought to elucidate lncRNA biomarkers and ceRNA networks in CRC. We used bioinformatics tools and analyzed data from the GEO, a public platform involving 15 CRC tumor tissues and 15 adjacent healthy tissues. We utilized some tissues from our patients, QRT-PCR was applied to confirm specific analytical results. This approach was beneficial in manifesting new potential lncRNA biomarkers and constructing novel ceRNA networks in CRC.

2.1. lncRNAmiRNA interaction analysis

To identify the lncRNAs targeted miR-616-3p, the lncRNAs to miR-616-3p binding (including the lncRNAs and miRNAs folded RNA predicted structure) was indicated via the RegRNA 2.0 database (regrna2.mbc.nctu.edu.tw/detection.html). (Chang et al. 2013) Furthermore, the minimum folding free energy was set under < 20 when predicting the miRNA target sites, and the system score was set to > 160. A high score shows a more potent binding ability. Besides, the lncRNA sequences related to Homo sapiens were already investigated applying the NCBI database.

2.2. Identification of miRNAtarget interactions

Interactions between miRNAs and mRNAs were obtained from the Targetscan database « http://www.targetscan.org/vert_72/ » (Agarwal et al. 2015) to predict mRNAs targeted screening miRNAs. By utilizing the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, (Kanehisa and Goto 2000) lncRNAs that play significant roles in the apoptosis pathway were determined. All mRNAs were obtained from the Targetscan database, then their HNG codes were received from the Ensemble database (https://asia.ensembl.org/). (Hubbard et al. 2002) They were analyzed in the cBioPortal database to identify mRNAs that play a crucial role in CRC. (Cerami et al. 2012)

2.3. Identification of differentially expressed lncRNAs

The function of dysregulated lncRNAs was investigated by a bioinformatics technique to predict the differentially expressed lncRNAs. Moreover, the gene expression profile dataset no. GSE117830; GPL15314 platform, deposited in the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/). (Clough and Barrett 2016) A total of 10 chips were involved in the dataset, including five colorectal cancer tissues (tumor group) and five paired adjacentnormal colorectal tissues (control group). DEmRNAs and DElncRNAs were analyzed in terms of a threshold of | log2 (fold change) | > 2 and | log2 (fold change) | < -2 and adjusted P-value < 0.001. Evaluations were executed with R version 3.5.3 (https://www.r-project.org/), and the program package “Limma, Gplot, GGplot, Plyr,” and the findings are demonstrated in a heatmap (red shows RNAs with high expression and green shows RNAs with low expression).

2.4. construction of the lncRNAmiRNAmRNA network

To better apprehend the correlations between DEmRNAs, lncRNAs, and miRNAs, the lncRNA-mediated ceRNA network in CRC was constructed. LncRNAs sponge miRNAs to bind miRNAs competitively to modulate the gene expression. A ceRNA network in terms of the recognized prognostic lncRNAs, miRNAs, and mRNAs was established. The Targetscan database was used to identify miRNA-mRNA interactions, and RegRNA 2.0 to identify LncRNA-miRNA interactions. We constructed and visualized the co-expression network by using Cytoscape 3.7.1.

2.5. Tissue collection

CRC tissues were collected from the Poursina Hakim Gastrointestinal Disease research center. Tumor samples were admitted by a pathologist to decline heterogeneity. Written informed consent was obtained from patients, and this study was approved by the Ethics Boards of the Isfahan University of Medical Sciences. All tissues were 60 in total (CRC tissues n = 30 and adjacent non-tumor samples n = 30). Patients have not received radiotherapy and chemotherapy before tissue sampling. Tissues were frozen in liquid nitrogen directly and maintained in a freezer at − 80°C.

2.6. qRT-PCR analyses

TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA) was utilized to extract total RNA from tissue samples. Thermo ND-1000 ultramicro nucleic acid protein meter (OD 260 nm; Thermo, Waltham, MA) and agarose gel electrophoresis were applied to assess RNA quantity. The amount of 9 µl of total RNA was used for quantitative real-time PCR (qRT-PCR). cDNA was constructed using a Transcriptor First Strand cDNA Synthesis Kit (Biofact, Korea). Real-time PCR analyses were performed with SYBR Green (Biofact, Korea), using the Rotor gene 6000 system, under the following conditions: (a) 95°C for 10 minutes; (b) 40 cycles of 95°C for 20 seconds and 56°C for 30 seconds; and (c) 72°C for 30 seconds. The relative gene expression was estimated by using the 2 − ΔΔCt method normalized to GAPDH. Each tissue sample was investigated in triplicate to obtain reliable outcomes.

2.7. Statistical Analysis

Statistical analyses were performed using GraphPad Prism 9 and also to plot the figures. A two-tailed P < 0.05 was regarded statistically critical. Aberrant gene expression between two groups was assessed by samples t-test and is two-sided. Kaplan-Meier method was applied to investigate survival rates.

3.1. MIR-616-3p expression is upregulated in CRC tissues and shows a positive association with gender and poor overall survival.

In this study, we selected miR-616-3p as a candidate microRNA for further exploration in novel ceRNA networks due to some reasons. For example, former studies have reported that miR-616 had a significant regulatory role in various cancer signaling pathways, such as Wnt/β-catenin and PTEN signaling pathways. Besides, other investigations have suggested, dysregulated expression of miR-616-3p in different cancers such as ovarian, gastric, and breast cancer led to cell migration, invasion, and EMT. However, no study has reported the involvement of miR-616-3p in CRC. QRT-PCR was performed to investigate miR-616-3p expression. Findings illustrated that miR-616-3p was significantly down-regulated in CRC tissues as compared to that in non-tumor samples (95% CI, 0.3561 to 3.555; P < 0.0183) with (log2 Fold Change = 0.15) (Fig. 6A). After adjusting for sex, age, stage, location, and differentiation, multivariate analysis indicated no relation between these clinicopathological factors and miR-616-3p expression.

3.2. Differentially expression of SIK1, XIAP, and ZNF347 in CRC tissues.

miR-616-3p targeted mRNAs were identified by various bioinformatics approaches. Targetscan database (http://www.targetscan.org/vert_72/) was used to find targeted mRNAs related to miR-616-3p. All mRNAs had their specific HNG codes were obtained from the Ensemble database and were scrutinized in the cBioPortal database to identify mRNAs that play a crucial role in CRC (Fig. 2). Finally, SIK1, XIAP, and ZNF347 mRNAs indicated the most differential expression and interaction with miR-616-3p in CRC, and KEGG pathway analysis showed that candidate mRNAs were potentially involved in the apoptosis pathway.

The salt-inducible kinases (SIKs) are considered to control gene expression in two ways. (1) By restraining phosphorylation of the CREB-regulated transcriptional coactivators (CRTC), (2) by the transcriptional modulators class IIa histone deacetylases (HDAC). Therefore, by these factors, SIKs control the transcriptional programs. However, the SIKs action in cancer remains to be clarified. Between the SIKs, SIK1 has been investigated formerly as a modulator of anoikis and repression of metastatic potential in human breast cancer cells. Evaluation of The salt-inducible kinase 1 (SIK1) expression was performed via qRT-PCR, showed a significant down-regulation in the CRC tissues as compared to that adjacent non-tumor tissues (Fig. 6B, 95% CI, 0.2479 to 1.932; P < 0.013) and (log2 Fold Change = 0.62). Clinical parameters such as gender, stage, location, and differentiation were evaluated to find a positive correlation with the down-regulation of SIK1, but no correlations were found between them and SIK1 expression.

X-linked Inhibitor of Apoptosis Protein (XIAP) belongs to the Inhibitors of Apoptosis Proteins (IAPs) family members. Bioinformatic analysis showed that XIAP had a strong interaction with miR-616-3p. Expression analysis indicated that XIAP was markedly overexpressed in CRC tissues in comparison to their corresponding normal tissues (95% CI, -1.866 to -0.2748; P < 0.0101) with (log2 Fold Change = 2.85) (Fig. 6C). Upregulation of XIAP also demonstrated a positive association with sex (P < 0.0476). After adjusting for age, stage, location, and differentiation, multivariate analysis illustrated no relationships between these clinical parameters and XIAP expression.

Zinc Finger Protein 347 (ZNF347), located on chromosome 19, is one of the miR-616-3p targets. RT-PCR analysis indicated that ZNF347 expression levels were considerably upregulated in CRC tissues as compared to non-tumor pairs (P < 0.014) and with (log2 Fold Change = 3.46) (Fig. 6D). Furthermore, other clinical features showed no correlation with ZNF347 high expression in CRC.

3.3. Linc01282 and lnc-MYADM-1:1 were upregulated, and lnc-atp12a-1:1 was down-regulated in CRC tissues.

Differentially expressed lncRNAs in CRC tumors compared to non-tumor pairs (Figs. 3A, 3B) were identified from dataset no. GSE117830 (Table 1) and their interactions with miR-616-3p were assessed via RegRNA 2.0 database. RegRNA 2.0 is an integrated web server, can clarify the explicit mechanism and the function of lncRNAs and miRNAs in CRC. RegRNA 2.0 also was applied to evaluate the functional correlation between lncRNAs and miRNAs. Between the markedly DElncRNAs, using a threshold alteration frequency > 10%. Only three lncRNAs with minimum folding free energy were able to exert regulatory functions on miR-616-3p and are associated with CRC. It was predicted, hsamiR616-3p was targeted by Linc01282, Lnc-atp12a-1:1, and lnc-MYADM-1:

Table 1

miR-616-3p targeted lncRNAs classified by expression status
microRNA	LncRNAs
Has-miR-616-3p	Upregulate AC094104.1, AL354861.2, CARMN, LINC00910, LINC01641, lnc-CCDC39-1:5, lnc-CDC27-2:3, lnc-CIAO1-6:1, lnc-EEF1AKMT1-3:6, lnc-LILRB4-1:2, lnc-MAN1B1-1:1, lnc-MTRNR2L11-2:2, lnc-NABP2-1:1, lnc-PALLD-4:1, lnc-PILRB-1:13, lnc-RPS24-3:24, lnc-SERPIND1-1:62, lnc-SLC24A2-2:1, lnc-SMG6-6:1, lnc-SPTBN2-6:1, lnc-SYN2-6:1, lnc-TAAR6-1:1, lnc-TRIM39-2:1, lnc-VMO1-1:1, lnc-WDR12-3:1, lnc-WFIKKN2-3:1, lnc-ZNF479-13:1, lnc-ZNF716-12:1, lnc-ZNF836-1:8, AC002463.1, AC006004.1, AC008278.1, AC016933.1, AC073316.2, AC103563.2, AL139246.4, AP000356.2, APTR, FAM138B, FAM138E, LINC00305, LINC00350, LINC00700, LINC00842, LINC01205, LINC01271, LINC01282, PRDM16-DT, SNHG14, HSALNT0023219, HSALNT0028808, HSALNT0050876, lnc-AIG1-8:14, lnc-ANKRD20A3-4:4, lnc-BDH1-5:7, lnc-CDH10-6:5, lnc-DDX39A-1:1, lnc-DUPD1-3:1, lnc-EXOSC1-1:1, lnc-FKBP5-1:2, lnc-FNBP1L-2:3, lnc-FRG2-9:1, lnc-FRMD1-3:1, lnc-FTCDNL1-1:1, lnc-GATA3-6:2, lnc-GRM3-2:3, lnc-HIVEP2-1:1, lnc-MARCKS-18:1, lnc-METTL16-3:2, lnc-MYADM-1:1, lnc-PALMD-1:1, lnc-PCDH1-2:1, lnc-PDLIM1-1:5, lnc-PRSS3-2:10, lnc-RAB19-3:1, lnc-RAPSN-2:1, lnc-SFTPA2-1:5, lnc-SLC47A2-1:4, lnc-ST8SIA4-7:1, lnc-STS-10:1, lnc-TGFA-1:1, lnc-TMEM121-6:1, lnc-TSSK3-2:2, lnc-TXNDC17-1:2, lnc-USP6NL-3:1, lnc-VWDE-2:1, lnc-ZNF726-5:1, NONGGOT009232.1, NONHSAT098681.2, NONHSAT123805.2, NONHSAT148170.2, NONHSAT159169.1, NONHSAT165589.1, NONHSAT196116.1, NONHSAT200082.1, NONHSAT214069.1, NONRATT014570.2, PLCE1-AS2:15 Downregulate lnc-C21orf59-1:1, lnc-PTPN2-4:1, lnc-TRPC4-2:1, lnc-HLA-DRB1-5:1, HSALNT0031418, lnc-LHFPL4-2:1, LY6E-DT, lnc-GJA9-1:2, lnc-MNX1-23:3, lnc-BPY2B-3:1, lnc-ATP12A-1:1

RegRNA2.0 software was applied to investigate the folded RNA predicted structure of the lncRNAs and miRNAs, and the partial reliability data of pair probabilities are represented in (Fig. 4). Hairpin loops are the most frequently anticipated secondary structure, which might facilitate the sticking of lncRNAs to miRNAs. Thus, RegRNA2.0 software showed that Linc01282, Lnc-atp12a-1:1, and lnc-MYADM-1:1 could bind to hsa‑miR-616-3p. The network constructed in this research indicated the unknown intricacy of noncoding RNA modulatory interactions and how lncRNAs might act as vital factors in the miRNA regulatory network to elucidate the role of these interactions in the CRC process. The binding sites of Linc01282/lnc-atp12a-1:1/lnc-MYADM-1:1/miR-616-3p/ and SIK1/XIAP/ZNF347 also were demonstrated in Fig. 5.

Linc01282, Lnc-atp12a-1:1, and lnc-MYADM-1:1 were the most differentially expressed lncRNAs which had − 21.8, -14.1, and − 13.5 minimum folding free energy, respectively (log2fold change (FC) of > 2, <-2 and P < 0.001). Therefore, dysregulate lncRNAs discovered in the GEO database were assessed by qPCR analysis to indicate the expression levels of the top three lncRNAs with the highest degree in 60 paired CRC tissues and adjacent non-tumor tissues from cases with colorectal cancer. RT-PCR evaluation showed that Linc01282 expression was significantly upregulated in the CRC tissues compared to that adjacent normal tissues diameter (95% CI, -2.218 to -0.2683; P < 0.0142) and (log2 Fold Change = 5.18) (Fig. 6E). Furthermore, overexpression of Linc01282 displayed a positive correlation with sex (P < 0.0188). The statistical analysis illustrated no significant association between Linc01282 expression and other clinical data.

Lnc-atp12a-1:1 was the second most differentially expressed lncRNA, had about − 14.1 minimum folding free energy, preceded by Linc01282. Lnc-atp12a-1:1 expression was significantly decreased in the tumor tissues as compared to the adjacent non-tumor samples (95% CI, 0.3710 to 1.882; P < 0.0049) with (log2 Fold Change = 0.44) (Fig. 6F). Lnc-atp12a-1:1 also showed a positive association with gender in CRC (P < 0.0266), although other clinical parameters showed no positive correlation with Lnc-atp12a-1:1 expression level (Table 2).

Table 2

Correlation between expression of Lnc-atp12a-1:1 in colorectal cancer with clinicopathologic features
Features All cases	Lnc-atp12a-1:1 P
Features All cases	Low High
Total 30 Age(years) ≤ 60 9 > 60 21 Gender Female 15 Male 15 Location Colon 20 Rectum 10 Differentiation Well/moderately 24 Poorly 6 TNM stage I–II 15 III–IV 15	21 9 0.0049 0.1220 5 4 16 5 0.0266 12 3 11 4 0.7373 15 5 6 4 0.79 15 6 4 4 0.8709 14 1 11 4

On the other hand, lnc-MYADM-1:1 was the third highly expressed lncRNA with − 13.5 minimum folding free energy. Its expression analysis illustrated that lnc-MYADM-1:1 considerably upregulated in the tumor tissues in comparison to the adjacent normal samples (95% CI, -2.149 to -0.2774; P < 0.0129) and (log2 Fold Change = 2.09) (Fig. 6G). Another significant feature was the positive correlation between elevated lnc-MYADM-1:1 expression and location of the Tumor (P < 0.0284).

3.4. Construction of the global lncRNA-miRNA-mRNA triple network

Overall, a global ceRNA network was constructed using mentioned databases to confirm the association between lncRNA, miRNA, and mRNA markers. There were 56 common mRNAs interact with LncRNAs and miR-616-3p. Moreover, 3 LncRNAs (Linc01282, Lnc-atp12a-1:1, and lnc-MYADM-1:1) were obtained from RegRNA 2.0 that targeted miR-616-3p and showed the most interaction with miR-616-3p. As shown in Fig. 7, miRNA-mRNA and lncRNA-miRNA pairs were combined and constructed the global triple network.

3.5. Prognostic DElncRNA mediated ceRNA in sub-network

The prognostic DElncRNA mediated ceRNA network was investigated utilizing cytoscape3.6.1 to realize the exact interaction between prognostic lncRNAs, miRNAs, and mRNAs in CRC. Only 3 prognostics, 3 DElncRNAs, 1 DEmiRNA, and 3 DEmRNAs were displayed in the DElncRNA-mediated ceRNA sub-network.

Genes generally function in groups as networks, not alone. A new theory called the ceRNA network offers that various RNA transcripts could interact with one another by shared miRNA response elements (MREs). Evaluation of ceRNA analysis more In-depth can illuminate the roles of coding and non-coding RNAs. (Zhong et al. 2019) Invasion and metastasis in hepatocellular carcinoma are suppressed by knockdown of lncRNA ANRIL through modulating miR-122-5p expression. (Huang et al. 2019) Discovering the correlation between lncRNAs and their downstream targets would be an effective way to diagnose patients with CRC. In this study, the GSE117830 dataset was analyzed to identify DElncRNAs with the most interaction energy with has-miR-616-3p, in addition to target mRNAs involved in the apoptosis pathway CRC. We elucidated the role of these lncRNAs in CRC progression for the first time and constructed new ceRNA networks based on the GEO dataset.

In the present study, three unreported novel lncRNAs targeted miR-616-3p involved in CRC were manifested. The first one is Linc01282 (Long Intergenic Non-Protein Coding RNA 1282). It was elucidated that Linc01282 is located on chromosome X and 880 bp in length with six exons based on Lncipedia and NCBI databases and illustrated the most interaction energy with miR-616-3p. Linc01282 conspicuously upregulated in the tumor tissues compared to the non-tumor adjacent samples in real-time PCR analysis, as well as in bioinformatics analysis. Also, overexpression of Linc01282 was more prevalent among females, meaning that high expression of Linc01282 can be a prognostic biomarker for females. Therefore, Linc01282 can be considered as a novel diagnostic biomarker or potential new candidate for CRC therapy. Lnc-atp12a-1:1 is the other lncRNA located on chromosome 13, 1710 bp in length, and has 13 exons, showed the second most interaction energy with miR-616-3p. QRT-PCR analysis demonstrated that Lnc-atp12a-1:1 was down-regulated in CRC tissues than adjacent non-tumor tissues and also indicated a positive correlation with sex. It means the low expression of Lnc-atp12a-1:1 is more common among men. Thus, it can be regarded as a new prognostic biomarker and might have a CRC tumor suppressor function. Down-regulation of Lnc-atp12a-1:1 may be a potential diagnostic biomarker for males and is regarded as a typical characteristic in CRC progression. Lnc-MYADM-1:1 is the other lncRNA targeted miR-616-3p, located on chromosome 19, 555 bp in length, and has three exons. Real-time PCR analysis indicated that lnc-MYADM-1:1 was overexpressed in tumor tissues as compared to that normal samples. The findings also showed a positive relationship with the tumor location, meaning that upregulation of lnc-MYADM-1:1 was more prevalent in the colon than in the rectal. Thus, overexpression of lnc-MYADM-1:1 may be considered a diagnostic biomarker in CRC and the position of the tumor.

Similarly, Yu et al. (Yu et al. 2018) indicated that u50535 is a new lncRNA advanced metastasis, tumor progression, and often increased in CRC tissues and related to poor prognosis. In the process, u50535, through regulating the CCL20 signaling pathway, advances CRC tumorigenesis. Thus, this study introduced three novel prognostic biomarkers in CRC. However, more studies are needed to explore the exact practical functions of these lncRNAs in CRC.

As mentioned earlier, miR-616-3p has significant regulatory roles in different signaling pathways, and the aberrant expression of miR-616-3p has been reported in various cancers. However, the expression of miR-616-3p has not been investigated in CRC before. Our results showed that the relative expression of hsa-miR-616-3p in CRC tissues was significantly downregulated in real-time PCR analysis and can be considered a prognostic biomarker in CRC. Yuan et al. illustrated that miR-616 was upregulated in breast cancer (BC) and is pivotal in BC development. miR‑616 might be a possible therapeutic target for BC in miR‑616/TIMP2/MMPs axis. (Yuan 2019)

Targetscan database and KEGG pathway analysis were applied to identify CRC-specific mRNAs targeted miR-616-3p and may have functions in the apoptosis pathway. Three relevant mRNAs were selected named: ZNF347, SIK1, and XIAP. In this research, for the first time, we reported an unreported mRNA: Zinc Finger Protein 347 (ZNF347), located on chromosome 19. Our results showed that ZNF347 expression was significantly upregulated in tumor tissues compared to non-tumor tissues. ZNF347 overexpression in CRC may suggest a poor prognosis and may have an association with tumor progression. Inconsistent, Pan et al. reported the unreported up-regulation of EXOSC5 in CRC cell lines and tissues. EXOSC5 may develop tumorigenesis and cell growth through activating the ERK and Akt signaling pathways. (Pan, Pan, et al. 2019)

XIAP is another targeted mRNA of miR-616-3p that demonstrated overexpression in CRC tissues than normal samples. Consistently, Liang et al. showed that cell proliferation would be inhibited by losing XIAP that considerably sensitized CRC cells to PPAR; ligand-induced apoptosis. Therefore, activation of PPAR and limitation of XIAP simultaneously; may have a synergistic antitumor effect against colon cancer. (Qiao et al. 2008) Also, the contrast in XIAP expression between males and females was striking, showed that up-regulation of XIAP was more prevalent in females. Therefore, the expression of XIAP can be regarded as a prognostic biomarker in CRC, especially for women with CRC, and can be a therapeutic target for CRC.

Another miR-616-3p targeted mRNA is SIK1, showed significant downregulation in tumor tissues compared to normal samples, which may be correlated with poor prognosis. Similarly, Huang et al. showed that upregulated miR-17 might advance CRC development and function as a tumor suppressor miRNA by targeting SIK1. (Huang et al. 2019) Thus, downregulation of SIK1 may be a potential prognostic factor and a new candidate for CRC therapy.

In this study, we suggest this hypothesis that, based on the ceRNA network theory, bioinformatics tools showed that miR-616-3p is a direct target of Linc01282/ Lnc-atp12a-1:1/ lnc-MYADM-1:1 and located on the stem-loop structure of them. Also, every one of mentioned lncRNAs can potentially construct three district ceRNA networks with XIAP/ZNF347/SIK1 mRNAs. For example, Linc01282/miR-616-3p/XIAP axis. Thus, Linc01282/ Lnc-atp12a-1:1/ lnc-MYADM-1:1 can potentially regulate gene expression by competing miR-616-3p. However, further investigations are required to clarify the exact mechanisms of constructed ceRNA networks in this study.

Several interactions have already been verified. For instance, Liu et al. indicated that KCNQ1OT1 by sponging miR-329-3p could modulate CTNND1 expression positively, thereby expanding CRC development. (Liu et al. 2020) These researches also validated that our analytical outcomes are potentially reliable. Furthermore, we validated our bioinformatic analyzes and the possible associations between various genes involved in this research with clinical characteristics and their expression patterns by qRT-PCR. These findings were highly consistent and confirmed our analysis's reliability to find novel prognostic biomarkers and potential ceRNA networks in CRC.

Although these results would profit from a more significant cohort, extra molecular examinations, and clinicopathological method confirmation, we are confident that our findings supply novel prognostic and therapeutic targets to control CRC cases.

In conclusion, we demonstrated that miR-616-3p is a direct target of mentioned genes and has a strong interaction with them by bioinformatics analysis, constructing various ceRNA networks that may play crucial regulatory roles in different signaling pathways in CRC. The constructed CRC ceRNA networks bring to light unknown lncRNAs regulatory networks in CRC. QRT-PCR was used to validate aberrant expressions of these genes in CRC. Therefore, our findings provide a new approach recognizing potential lncRNAs, miRNA, and mRNAs biomarkers, possibly included in the regulation mechanisms, development, and CRC prognosis. These results also illustrated novel perceptions and a better comprehension of the miR-616-3p associated ceRNA networks in CRC. Further investigations and studies are required to entirely clarify the influences of these genes on the advancement of CRC to help understand the mechanism of CRC at the genetic level.

CRC Colorectal Cancer

lncRNAs Long non-coding RNAs

ceRNAs competing endogenous RNAs

miRNAs microRNAs

GEO Gene Expression Omnibus

KEGG Kyoto Encyclopedia of Genes and Genomes

DELs Differentially Expressed lncRNAs

3ʹUTR 3ʹ-untranslated region

ncRNAs non-coding RNAs

XIAP X-linked Inhibitor of Apoptosis Protein

IAPs Inhibitors of Apoptosis Proteins

SIKs salt-inducible kinases

CRTC CREB-regulated transcriptional coactivators

HDAC histone deacetylases

ZNF347 Zinc Finger Protein 347

PC Principle Component analysis

MREs miRNA response elements

BC Breast Cancer

Corresponding Author:

Simin Hemati

[email protected]

Ethics approval and consent to participate

the Ethics Boards of the Isfahan University of Medical Sciences

Ethical Approval Number: IR.MUI.REC.1395.1.175

Consent for publication

Yes

Availability of data and materials

Not applicable

Competing interests

None

Funding

None

Authors' contributions

Mohammad Abdolvand: Writing- Original draft preparation, Conceptualization, Methodology, Investigation. Zahra Mohammadi Chermahini: Software, RNA extraction. Ehsan Heidari: Writing- Reviewing and Editing, Software. Mohammad hassan Emami: Gastrointestinal subspecialty. Alireza Fahim: Gastrointestinal subspecialty. Hojjatolah Rahimi: Gastrointestinal subspecialty. Elham Amjadi: Pathologist. Shirin Abdolvand: PCR, qRT-PCR. Maliheh Roozbahani Darehsari: Writing- Reviewing and Editing. Mohammad Chehelgerdi: Writing- Reviewing and Editing. Simin Hemati: Manager of project, Methodology, Writing- Reviewing and Editing. Mansoor Salehi: Supervision, Validation.

Acknowledgments

This research was performed with the financial support of Isfahan University of Medical Sciences and was supported by Poursina Hakim Gastrointestinal Disease research center. We are highly grateful to Shahrekord University, Iran, for sustaining this evaluation. Also, simulation study and Docking were performed at SaNa Zist Pardaz company.

Agarwal, Vikram, George W Bell, Jin-Wu Nam, and David P Bartel. 2015. "Predicting effective microRNA target sites in mammalian mRNAs." elife 4:e05005.
Bai, QL, CW Hu, XR Wang, JX Shang, and GF Yin. 2017. "MiR-616 promotes proliferation and inhibits apoptosis in glioma cells by suppressing expression of SOX7 via the Wnt signaling pathway." Eur Rev Med Pharmacol Sci 21 (24):5630-5637.
Bian, Zehua, Jiwei Zhang, Min Li, Yuyang Feng, Xue Wang, Jia Zhang, Surui Yao, Guoying Jin, Jun Du, and Weifeng Han. 2018. "LncRNA–FEZF1-AS1 promotes tumor proliferation and metastasis in colorectal Cancer by regulating PKM2 signaling." Clinical Cancer Research 24 (19):4808-4819.
Cerami, Ethan, Jianjiong Gao, Ugur Dogrusoz, Benjamin E Gross, Selcuk Onur Sumer, Bülent Arman Aksoy, Anders Jacobsen, Caitlin J Byrne, Michael L Heuer, and Erik Larsson. 2012. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. AACR.
Chang, Tzu-Hao, Hsi-Yuan Huang, Justin Bo-Kai Hsu, Shun-Long Weng, Jorng-Tzong Horng, and Hsien-Da Huang. 2013. "An enhanced computational platform for investigating the roles of regulatory RNA and for identifying functional RNA motifs." BMC bioinformatics.
Chen, Zhongbo, Jianqing Zhu, Yiming Zhu, and Junjian Wang. 2018. "MicroRNA-616 promotes the progression of ovarian cancer by targeting TIMP2." Oncology reports 39 (6):2960-2968.
Clough, Emily, and Tanya Barrett. 2016. "The gene expression omnibus database." In Statistical genomics, 93-110. Springer.
Gao, Weida, Hongbin Li, Yang Liu, Yao Zhang, Hong Zhao, and Fei %J Molecular Medicine Reports Liu. 2020. "Long non‑coding RNA FLVCR1‑AS1 promotes glioma cell proliferation and invasion by negatively regulating miR‑30b‑3p." 22 (2):723-732.
Guo, Li, Guowei Yang, Yihao Kang, Sunjing Li, Rui Duan, Lulu Shen, Wenwen Jiang, Bowen Qian, Zibo Yin, and Tingming Liang. 2020. "Construction and Analysis of a ceRNA Network Reveals Potential Prognostic Markers in Colorectal Cancer." Frontiers in Genetics 11:418.
Hollstein, Pablo E, Lillian J Eichner, Sonja N Brun, Anwesh Kamireddy, Robert U Svensson, Liliana I Vera, Debbie S Ross, TJ Rymoff, Amanda Hutchins, and Hector M Galvez. 2019. "The AMPK-related kinases SIK1 and SIK3 mediate key tumor-suppressive effects of LKB1 in NSCLC." Cancer discovery 9 (11):1606-1627.
Huang, Chengzhi, Jianhua Liu, Lishu Xu, Weixian Hu, Junjiang Wang, Muqing Wang, and Xueqing Yao. 2019. "MicroRNA-17 promotes cell proliferation and migration in human colorectal cancer by downregulating SIK1." Cancer management and research 11:3521.
Hubbard, Tim, Daniel Barker, Ewan Birney, Graham Cameron, Yuan Chen, L Clark, Tony Cox, J Cuff, Val Curwen, and Thomas Down. 2002. "The Ensembl genome database project." Nucleic acids research 30 (1):38-41.
Kanehisa, Minoru, and Susumu Goto. 2000. "KEGG: kyoto encyclopedia of genes and genomes." Nucleic acids research 28 (1):27-30.
Li, Meng, Lian‐mei Zhao, Suo‐lin Li, Jing Li, Bo Gao, Fei‐fei Wang, Sheng‐pu Wang, Xu‐hua Hu, Jian Cao, and Gui‐ying %J Cancer medicine Wang. 2018. "Differentially expressed lncRNAs and mRNAs identified by NGS analysis in colorectal cancer patients." 7 (9):4650-4664.
Liu, Haizhou, Shuyuan Wang, Shunheng Zhou, Qianqian Meng, Xueyan Ma, Xiaofeng Song, Lihong Wang, and Wei Jiang. 2019. "Drug resistance-related competing interactions of lncRNA and mRNA across 19 cancer types." Molecular Therapy-Nucleic Acids 16:442-451.
Liu, Xing, Yexiang Zhang, Yan Wang, Chao Bian, and Fengji Wang. 2020. "Long non-coding RNA KCNQ1OT1 up-regulates CTNND1 by sponging miR-329-3p to induce the proliferation, migration, invasion, and inhibit apoptosis of colorectal cancer cells." Cancer Cell International 20 (1):1-13.
Mo, Xiao-Mei, Peng-Fei Qin, Bing Wang, Feng-Hai Liu, and Hua-Hui Li. 2020. "miR‑421 promotes the viability of A549 lung cancer cells by targeting forkhead box O1." Oncology letters 20 (6):1-1.
Nguyen, Vu Hong Loan, Chenyang Yue, Kevin Y Du, Mohamed Salem, Jacob O’Brien, and Chun Peng. 2020. "The Role of microRNAs in Epithelial Ovarian Cancer Metastasis." International Journal of Molecular Sciences 21 (19):7093.
Pan, Hongda, Chunmiao Guo, Jingxin Pan, Dongwei Guo, Shibo Song, Ye Zhou, and Dazhi Xu. 2019. "Construction of a competitive endogenous RNA network and identification of potential regulatory axis in gastric cancer." Frontiers in oncology 9:912.
Pan, Hongda, Jingxin Pan, Dongwei Guo, Shibo Song, Ye Zhou, and Dazhi %J Frontiers in oncology Xu. 2019. "Construction of a competitive endogenous RNA network and identification of potential regulatory axis in gastric cancer." 9:912.
Qiao, Liang, Yun Dai, Qing Gu, Kwok Wah Chan, Bing Zou, Juan Ma, Jide Wang, Hui Y Lan, and Benjamin CY Wong. 2008. "Down-regulation of X-linked inhibitor of apoptosis synergistically enhanced peroxisome proliferator-activated receptor γ ligand-induced growth inhibition in colon cancer." Molecular cancer therapeutics 7 (7):2203-2211.
Tuersong, Tayier, Linlin Li, Zumureti Abulaiti, and Shumei Feng. 2019. "Comprehensive analysis of the aberrantly expressed lncRNA‑associated ceRNA network in breast cancer." Molecular medicine reports 19 (6):4697-4710.
Weng, Wanqing, Zhongjing Zhang, Weiguo Huang, Xiangxiang Xu, Boda Wu, Tingbo Ye, Yunfeng Shan, Keqing Shi, and Zhuo Lin. 2020. "Identification of a competing endogenous RNA network associated with prognosis of pancreatic adenocarcinoma." Cancer cell international 20 (1):1-14.
Wu, Zhen-Hua, Chen Lin, Chen-Chen Liu, Wei-Wei Jiang, Ming-Zhu Huang, Xin Liu, and Wei-Jian Guo. 2018. "MiR-616-3p promotes angiogenesis and EMT in gastric cancer via the PTEN/AKT/mTOR pathway." Biochemical and biophysical research communications 501 (4):1068-1073.
Xian, Di, and Yu Zhao. 2019. "Lnc RNA KCNQ 1 OT 1 enhanced the methotrexate resistance of colorectal cancer cells by regulating miR‐760/PPP 1R1B via the cAMP signalling pathway." Journal of cellular and molecular medicine 23 (6):3808-3823.
Yu, Xihu, Zixu Yuan, Zuli Yang, Daici Chen, Taewan Kim, Yanmei Cui, Qianxin Luo, Zhihang Liu, Zihuan Yang, and Xinjuan Fan. 2018. "The novel long noncoding RNA u50535 promotes colorectal cancer growth and metastasis by regulating CCL20." Cell death & disease 9 (7):1-13.
Yuan, Chao. 2019. "miR‑616 promotes breast cancer migration and invasion by targeting TIMP2 and regulating MMP signaling." Oncology letters 18 (3):2348-2355.
Zhang, Guangle, Cong Pian, Zhi Chen, Jin Zhang, Mingmin Xu, Liangyun Zhang, and Yuanyuan Chen. 2018. "Identification of cancer-related miRNA-lncRNA biomarkers using a basic miRNA-lncRNA network." PloS one 13 (5):e0196681.
Zhang, Hui, Zhuo Wang, Jianzhong Wu, Rong Ma, and Jifeng Feng. 2019. "Long noncoding RNAs predict the survival of patients with colorectal cancer as revealed by constructing an endogenous RNA network using bioinformation analysis." Cancer medicine 8 (3):863-873.
Zhong, Min-Er, Yanyu Chen, Guannan Zhang, Lai Xu, Wei Ge, and Bin Wu. 2019. "LncRNA H19 regulates PI3K–Akt signal pathway by functioning as a ceRNA and predicts poor prognosis in colorectal cancer: integrative analysis of dysregulated ncRNA-associated ceRNA network." Cancer cell international 19 (1):1-13.
Zhou, Yifan, Xiaowen Cheng, Yufeng Wan, Tingting Chen, Qing Zhou, Zhengguang Wang, and Huaqing Zhu. 2020. "MicroRNA-421 Inhibits Apoptosis by Downregulating Caspase-3 in Human Colorectal Cancer." Cancer Management and Research 12:7579.
Zhu, Mingming, Wen Zhang, Jun Ma, Youguo Dai, Qi Zhang, Qin Liu, Burong Yang, Gang %J Experimental Li, and Therapeutic Medicine. 2019. "MicroRNA‑139‑5p regulates chronic inflammation by suppressing nuclear factor‑κB activity to inhibit cell proliferation and invasion in colorectal cancer." 18 (5):4049-4057.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Constructing Novel ceRNA Networks on MiR-616-3p Reveals LncRNAs as Potential Colorectal Cancer Prognostic Biomarkers

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials And Methods

2.1. lncRNAmiRNA interaction analysis

2.2. Identification of miRNAtarget interactions

2.3. Identification of differentially expressed lncRNAs

2.4. construction of the lncRNAmiRNAmRNA network

2.5. Tissue collection

2.6. qRT-PCR analyses

2.7. Statistical Analysis

3 Result

4 Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1