lncRNA VASH1-AS1/miR-199a-5p/PDCD4 Axis Regulate Proliferation, Apoptosis, and Cell Cycle in PDAC Progress

Background: Pancreatic ductal adenocarcinoma (PDAC) is a highly malignant tumor with clinical characterized by short course, rapid progression and rapid deterioration. At present, the studies on the regulatory factors of PDAC are not enough. Methods: In this study, we applied transcriptome sequencing with PDAC tumor and normal tissues (5cases). Expression of miR-199a-5p was detected by qPCR in another 10 pairs of cancer and normal tissues. The miR-199a-5p mimic, miR-199a-5p mimic NC, miR-199a-5p inhibitor, and miR-199a-5p inhibitor NC were used to detect the function of miR-199a-5p in cell lines. Construction of double luciferase reporter vector of VASH1-AS1 and PDCD4 were used to conrm the binding of miR-199a-5p and 3’- UTR of human PDCD4 mRNA and VASH1-AS1, respectively. PDCD4 protein expression was detected by Western Blot. The cell apoptosis assay was performed using the Annexin V-FITC Kit. Cell proliferation was detected using CCK-8assay and EdU uorescence analysis. Cell cycle and apoptosis analyzed by ow cytometry at excitation and emission wavelengths of 488 and 530 nm, respectively. SPSS software (version 22.0; SPSS, Chicago, USA) and GraphPad Prism 5.0 (GraphPad software, San Diego, California, USA) were used for data analysis. Results: After data preprocessing, differentially expressed lncRNAs (606 up-regulated, 875 down-regulated), miRNAs (171 up-regulated, 188 down-regulated) and genes (4129 up-regulated, 3417 downregulated) were identied. With dada analysis, a new VASH1-AS1/miR-199a-5p/PDCD4 regulation mode was discovered. In PDAC, the expression level of miR-199a-5p is negatively correlated with VASH1-AS1, and the further binding was performed between miR-199a-5p and PDCD4. The overexpression of miR-199a-5p can promote the


Background
Pancreatic cancer (PC), a highly malignant tumor of the digestive system, was prone to recurrence after surgery. It had a poor prognosis, with a 5-year survival rate of less than 8% [1][2][3][4][5]. The incidence of pancreatic cancer had been on the rise. Research on the pathogenesis of PC is ongoing, but no breakthrough so far, which had caused di culties in diagnosis and treatment [6]. Approximately 70% of PDAC in ltrates to the matrix of normal pancreatic tissue. Although surgical resection was still the main way for PDAC treatment, it ass effective in only 10-15% of newly diagnosed patients. The number of surgical resections carried out is low due to di culty in early diagnosis, and lacked of other treatment options means that most patients with PDAC eventually die of metastasis [7].
In decades of miRNA research, more and more evidences showed that deregulation of miRNA expression levels were closely related to the development of tumors. In the study of PDAC, more than 80 types of miRNAs had been shown to regulate the proliferation and apoptosis through a variety of ways. Histone acetylation in the promoter zone can induce miR-21 up-regulation and associate with chemoresistance to gemcitabine and enhanced malignant potential in pancreatic cancer cells [8], miR-200a through regulating the DEK to suppress the metastasis of PDAC [9], miR-148a can target CCKBR and Bcl-2, then regulate growth and apoptosis of pancreatic cancer [10]. The expression pattern and mechanism of miRNA in various tumors are very different in previous reports. In liver cancer, ovarian cancer, prostate cancer, breast cancer, esophageal cancer, testicular cancer, and bladder cancer, miR-199a-5p is downregulated and inhibited tumor invasion and metastasis by regulating SMAD1, Wnt/β-catenin, HIF-1α and other pathways [11][12][13][14]. However, studies such as Li et al. have found that miR-199a-5p is highly expressed in gastric cancer tissues and target the mTOR signaling pathway to promote the cancer processing [15]. High expression of miR-199a-5p has also been found in pancreatic cancer, which can target and regulate the FOXA2 tumor suppressor gene, thereby promoting tumor cell proliferation and PDCD4 is a new tumor suppressor gene discovered in recent years. The higher expression of PDCD4 can increase the transcription of TIMP2 gene and inhibit the metastasis of breast cancer cells [17]. It can also interact with the transcription factor Twist1 to reduce the expression level of the target gene YB-1 of Twist1, thereby inhibiting cell proliferation [18].
In this study, we investigate whole transcriptome sequencing of the PDAC tumor and normal tissue samples (5cases). We discovered a new VASH1-AS1/miR-199a-5p/PDCD4 ceRNA network. Furthermore, we prove the relationships and functions in vitro between them using dual-luciferase report assay, CCK-8 assay, Flow Cytometry and EdU Fluorescence analysis.

Ethics Statement
The tissues of fteen patients with PDAC (cases) and normal tissues (controls) were used. PDAC was diagnosed by histopathologic evaluation of biopsied pancreatic tissue or by endoscopy. The cancerous and para-cancerous tissues were collected before surgery or radiotherapy. The study was approved by the Biomedical Ethics Committee of the West China Hospital of Sichuan University and was conducted in accordance with the Declaration of Helsinki and its later amendments. All participants provided written informed consent prior to being enrolled in this study.

RNA Extraction and Library Construction
The PDAC tissue samples (

Data Analysis
Raw data in fastq format were processed with Perl scripts. Quality ltering of clean data through removing reads containing adapters, poly-N, too short reads and low-quality reads from raw data. At the same time, the number of clean reads, clean Q20, clean Q30, and GC contents of the clean data were calculated. All data analysis about miRNA, mRNA, lncRNA were based on clean data. Reference genome and gene annotation les were downloaded directly from the genome website (GRCh38.p13). The known miRNA sequences were obtained in miRBase. miRNA expression levels were obtained using miRDeep2.
Differential expression analysis of cancer compared with normal tissues was performed using DESeq2. miRNA target prediction was performed by TargetScan, picTar, microT, miRmap, RNA22, PITA and miRanda. Differential expressed mRNA and lncRNA analysis of cancer compared with normal tissues was performed using DESeq2. The screening criteria for signi cantly different genes are corrected Pvalues of 0.05, and log2 (fold change) ≥1. The target mRNA and lncRNA prediction of different expression miRNA uses Starbase database (version 3.0; starbase.sysu.edu.cn/index.php), which provides the prediction results of seven miRNA databases (TargetScan, picTar, microT, miRmap, RNA22, PITA and miRanda). The target genes were compared with different expression genes, and selected the overlap genes. The candidate DEL (different expression lncRNA)-DEM (different expression miRNA)-DEG (different expression mRNA) ceRNA network using Cytoscape software (version 3.8.2).

KEGG/GO pathway enrichment analysis
The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis uses clusterpro ler of R package. According to the KEGG annotations of genes, selecting all genes of human as background, and using P<0.05 as the signi cance threshold to obtain statistically signi cant highfrequency annotations relative to the background. In addition, Gene Ontology (GO) can be divided into three parts: Molecular Function, Biological Process and Cellular Component. We selected the all known cancer-associated pathways according the previously study to construct the scatter diagram of KEGG and GO pathways.

Cell Culture
Human pancreatic cancer cells CFPAC-1 was purchased from the American Type Culture Collection (ATCC, USA). Cells were cultured in IMDM (Gibco, USA) supplemented with 10% FBS and antibiotics (100 U/mL penicillin and 100 μg/mL streptomycin).

Real-time PCR
Total RNA was reverse transcribed to cDNA using a Reverse Transcription Kit (Takara Co., Ltd., Dalian, China). The designed primers were synthesized by Sangon Biotech Co., Ltd. (Shanghai, China). cDNA was ampli ed using SYBR® Premix Ex Taq™ (TaKaRa, Dalian). Gene expression levels were calculated by the ΔΔCt method with GAPDH and U6 as internal controls for mRNA, lncRNA and miRNA respectively. The primer sequences used in this manuscript were listed in Table3.

Flow Cytometry Analysis
Flow Cytometry Analysis in CFPAC-1 cells referred to the previous manuscript [19].

EdU Fluorescence Analysis
EdU Fluorescence Analysis in CFPAC-1 cells referred to the previous manuscript [21].
Statistical Analysis SPSS software (version 22.0; SPSS, Chicago, USA) and GraphPad Prism 5.0 (GraphPad software, San Diego, California, USA) were used for data analysis. The data screening standard was three times the mean ± standard deviation. T-test was used to evaluate differences between groups, and a P value of <0.05 was regarded as to be statistically signi cant.

Results
Identi cation of DELs, DEMs and DEGs The sequencing results showed that compared with the normal group, many genes showed signi cant differential expression. We identi ed the DELs, DEMs, and DEGs in cancer and normal tissues, with P < 0.01 and |logFC| ≥ 1 as the thresholds. A total of DELs (606 up-regulated, 875 down-regulated), DEMs (171 up-regulated, 188 down-regulated) and DEGs (4129 up-regulated, 3417 down-regulated) were identi ed (Table4) (Fig.1A). The heat-map result showed clear separation and consistency expression pro les between cancer and normal samples. With analysis of the signi cantly DEMs, they are divided into three expression patterns. Type I is highly expressed in tumors and normal tissues like miR-199 family. They may play important roles in promoting the tumor cells' development. Type II show common expression in tumors and normal tissues like miR-450a-5p, and the expression in tumor tissues is upregulated dozens to hundreds of times compared with normal tissues. These miRNAs may belong to tumor tissue-speci c miRNA, which are activated during the development of tumor cells, thereby regulating the related target genes to promote tumor development. Type III is lowly expressed in tumors, but highly expressed in normal tissues like miR-148a-5p. These miRNAs may belong to tumor suppressor genes, which can inhibit tumor development through target gene regulation (Fig.1B).

Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) pathway analysis
To better understand the role of miRNA in PDAC, 35 signi cantly different miRNAs were selected to perform GO terms enrichment. The results show that many differentially expressed genes are enriched in intracellular receptor signaling pathway, positive regulation of catabolic process and other GO pathways related to cell growth and signal transfer (Fig.1C). Particularly, the candidate target genes of miR-199 family can be enriched into Wnt, TNF, Pancreatic cancer, FoxO, HIF1 and other pathways, which are close related to tumors formation. All these results indicated that miR-199 family and target genes may play critical roles in the development of PDAC (Fig.1D).
Predicted miRNA-target analysis and ceRNA network construction 35 miRNAs were selected with more signi cant differences contain three expression patterns for mRNA and lncRNA prediction. Separately, 3 to 5 predicted mRNA/lncRNA with the highest credibility were screened out for ceRNA network construction. Then, 45 lncRNAs, 65 mRNAs and 20 miRNAs were used to construct a ceRNA network. According to the DELs and DEGs, the negative correlation between miRNA and mRNA, and the negative correlation between miRNA and lncRNA were used as the screening basis.
Furthermore, miR-199a-5p mimic and inhibitor were applied to modulate the expression of miR-199a-5p in CFPAC-1 cells (Fig. 2F&2G). With qPCR analysis, the expression of PDCD4 were markedly decreased by miR-199a-5p mimics treatment, while miR-199a-5p inhibitor signi cantly increased the transcription of PDCD4 (Fig.2H&2I). As shown in Fig.2J, compared with mimic NC, the expression of PDCD4 protein was signi cantly down-regulated by miR-199a-5p mimic treatment, while miR-199a-5p inhibitor signi cantly up-regulated the expression of PDCD4 protein compared with inhibitor NC treatment. All these results indicated that miR-199a-5p could affect the expression of PDCD4. The regulatory effect of miR-199a-5p on PDCD4 was further analyzed by luciferase reporter assay. As shown in Fig.2K, compared with mimic NC, miR-199a-5p mimic could signi cantly reduce the activity of PDCD4 in WT cells. Conversely, the miR-199a-5p mimic could lead to the down-regulation of PDCD4 in PDCD4-MUT cells.
Effect of miR-199a-5p on Proliferation, Cell Cycle and Apoptosis in CFPAC-1 Cell Fig. 3A&3B shown, compared with mimics NC, the cell proliferation rate was signi cantly increased with miR-199a-5p mimic. however, the cell proliferation rate was signi cantly decreased by miR-199a-5p inhibitor. As Fig.3C shown, the cell ratio of the G0/G1 phase and promoted cell cycle division were increased after miR-199a-5p mimic treatment compared with mimic NC, whereas miR-199a-5p inhibitor treatment can reduce the G0/G1 phase cell ratio and arrest the cell cycle effectively compared with inhibitor NC. Flow cytometric analysis suggested that the rate of apoptosis was markedly decreased after the expression of miR-199a-5p increased, while the rate of apoptosis was signi cantly increased associated with the inhibition of miR-199a-5p (Fig.3D). TO investigate the mechanism of cell proliferation, EdU uorescence analysis was performed and similar results indicated that compared with mimic NC, miR-199a-5p mimic activated the proliferation activity in CFPAC-1 cells signi cantly. In contrast, the miR-199a-5p inhibitor reduced the proliferation activity of CFPAC-1 cells compared to the inhibitor NC group effectively (Fig.3E).

Effect of PDCD4 on Cell Proliferation, Cell Cycle and Apoptosis in CFPAC-1 cells
To further prove whether PDCD4, the potential down-stream target of miR-199a-5p, can affect the biological function in CFPAC-1 cells, the si-PDCD4 and PDCD4 overexpression lines were generated in CFPAC-1 cells. Western blot analysis (Fig.4A) showed that the expression of PDCD4 was signi cantly increased by PDCD4 overexpression plasmid, whereas the expression of PDCD4 protein was inhibited by PDCD4 silencing. As Fig.4B-4C shown, the cell proliferation rate was signi cantly increased after si-PDCD4 transfection compared with si-NC. However, the cell proliferation rate was obviously decreased by PDCD4 overexpression compared with PDCD4 negative control. As Fig.4D shown, overexpression of PDCD4 can increase the G0/G1 phase cell ratio and affect the cell cycle effectively, and inhibiting the expression of PDCD4 can reduce the G0/G1 phase cell ratio and arrest the cell cycle effectively. Flow cytometric analysis suggested that the rate of apoptosis markedly decreased by PDCD4 silencing, while the rate of apoptosis was signi cantly increased after PDCD4 overexpressed (Fig.4E). As shown in Fig.4F, compared with PDCD4 NC, PDCD4 overexpression signi cantly inhibited the proliferation activity of CFPAC-1 cells. si-PDCD4 effectively enhanced the proliferation activity of CFPAC-1 cells, suggesting a negative regulatory relationship between PDCD4 and tumor proliferation in CFPAC-1 cells. Therefore, all above results indicated that PDCD4 could negatively promote the proliferation and cell cycle division of CFPAC-1 cells, and positively regulate the apoptosis in CFPAC-1 cells, which may lead to the negative regulation between miR-199a-5p on PDCD4.

Discussion
The role of ceRNA in the formation of tumors was becoming more and more important. LncRNA can competitively bind to miRNA, thereby affect the expression of mRNA which may affect the tumor formation. Cui et al. had found that LINC01133 inhibits Gastric cancer formation and metastasis by acting as a ceRNA for miR-106a-3p to regulate APC expression and the Wnt/β-catenin pathway, suggesting that LINC01133 may serve as a potential prognostic biomarker and anti-metastatic therapeutic target for Gastric cancer [22].
Through whole-transcriptome sequencing and bioinformatics analysis, we conducted a ceRNA network with VASH1-AS1/miR-199a-5p/PDCD4 axis. The relationships and functions of this axis were further veri ed by experiments. Interestingly, lncRNA VASH1-AS1 is a new lncRNA, and the role of VASH1-AS1 in tumor formation is rarely reported. In another hand, miR-199a-5p is a special kind of miRNA, and its expression pattern is completely opposite in different tumors. It plays a role as tumor suppressor in many tumors, but it plays a role as tumor-promoting gene in PDAC. PDCD4 has been reported to be a tumor suppressor that can inhibit tumor transformation, formation, and translation [23][24]. Therefore, if miR-expression level of these markers can be detected in the blood in the early stage, which may help the early diagnosis of PDAC.
In summary, we found a novel mechanism can in uence the progression of PDAC associate with VASH1-AS1/miR-199a-5p/PDCD4 axis. The functional and clinical application of VASH1-AS1/miR-199a-5p/PDCD4 axis needed further research, and it may contribute to early PDAC diagnosis and help with PDAC treatments. All participants provided written informed consent prior to being enrolled in this study.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analyzed during this study are included in this published article.  Effect of miR-199a-5p on CFPAC-1 cell proliferation, cycle and apopotosis. To investigate the effect of miR-199a-5p on the biological function of CFPAC-1 cells. miR-199a-5p mimic and miR-199a-5p inhibitor were transfected into CFPAC-1 cells for 48 h. A-B: The proliferation rate of CFPAC-1 cells was determined by CCK-8 assay. C: The cell cycle was measured by the assay kit after transfection with miR-199a-5p mimic and miR-199a-5p inhibitor. D: The apoptosis rate of CFPAC-1 cells was detected using the ow cytometry. E: The proliferation activity of CFPAC-1 cells was detected by EdU uorescence analysis.
Values are expressed as means±SD. *P<0.05 vs mimic NC or inhibitor NC group.