A total of 78 paired specimens of PC and adjacent normal tissues were obtained from patients who had received surgical resection at Liyang People's Hospital (Liyang, China) between March 2017 and December 2019. Prior to surgery, none of the patients had received radiotherapy or chemotherapy. Patients that had been treated with chemotherapy or radiotherapy prior to surgery or lack of written informed consent were excluded from the study. The protocol of the present study was approved by the Ethics Committee of Liyang People's Hospital and written informed consent was provided by each patient. Immediately after collection, the tissues were frozen in liquid nitrogen and preserved at −80°C until use.

The PC cell lines (BxPC-3, AsPC-1, CaPAN-2, PANC-1 and SW1990) were purchased from the American Type Culture Collection and the human pancreatic duct epithelial (HPDE) cell line was purchased from The Cell Bank of Type Culture Collection of The Chinese Academy of Sciences. All the cell lines were cultured in DMEM, supplemented with 10% FBS (both from Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin-streptomycin, and incubated at 37°C in a humidified incubator with 5% CO₂.

Short hairpin (sh) RNAs targeting TTN-AS1 (sh-TTN-AS1#1 and -#2) were designed to downregulate TTN-AS1 and sh-negative control (NC) served as the control. miR-589-5p mimics (5′-UUAUGGUUUGCCUGGGACUGAG-3′) were purchased to increase miR-589-5p expression, with NC mimics (5′-UUCUCCGAACGUGUCACGUTT-3′) as the control. To overexpress FOXP1, the pcDNA3.1/FOXP1 plasmid was generated and empty pcDNA3.1 was the control. All the aforementioned plasmids were purchased from Shanghai GenePharma Co., Ltd., and Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) was used to transfect 50 nM sh-TTN-AS1#1, 50 nM sh-TTN-AS1#2, 50 nM sh-NC, 50 nM miR-589 mimics, 50 nM NC mimics, 4.0 µg pcDNA3.1/FOXP1 and 4.0 µg pcDNA3.1 into the BxPC-3 and AsPC-1 cell lines at room temperature for ~30 min. Reverse transcription-quantitative (RT-q) PCR analysis was used to confirm transfection efficiency. Subsequent experiments were performed 48 h post-transfection.

Total RNA was extracted from the PC tissues or cells using TRIzol^® (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Subsequently, total RNA was reverse transcribed into cDNA using the ReverTra Ace qPCR RT kit (Takara Biotechnology Co., Ltd.), according to the manufacturer's protocol. The SYBR Green real-time PCR Master Mix (Takara Biotechnology Co., Ltd.) was used for qPCR on an ABI 7500 real-time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.). The thermocycling conditions were as follows: Pre-denaturation at 95°C for 1 min, followed by 40 cycles of 95°C for 15 sec, 60°C for 30 sec and 72°C for 30 sec. GAPDH (for lncRNA and mRNA) or U6 (for miRNA) served as the internal references. Finally, the 2^−ΔΔCq method (25) was used to calculate and analyze relative target gene expression. The primer sequences were designed as follows: TTN-AS1 forward, 5′-CGATACCATTGAACACGCTGC-3′ and reverse, 5′-GGTTGAGGGTCCCAGTG-3′; miR-589-5p forward, 5′-CGCCTTGAATCGGTG-3′ and reverse, 5′-GTGCAGGGTCCGAGGT-3′; FOXP1 forward, 5′-AGGACTTGCACAAGCAGAAC-3′ and reverse, 5′-GTTGGCGTACACGGGCGGCT-3′; GAPDH forward, 5′-AGCCACATCGCTCAGACAC-3′ and reverse, 5′-GCCCAATACGACCAAATCC-3′; and U6 forward, 5′-GCTTCGGCAGCACATATACTAAAAT-3′ and reverse, 5′-CGCTTCACGAATTTGCGTGTCAT-3′.

To measure cell viability, the BxPC-3 and AsPC-1 cell lines were plated into 96-well plates (2×10³ cells/well), and incubated for 0, 24, 48 and 72 h. Following which, 10 µl CCK-8 solution (Dojindo Molecular Technologies, Inc.) was added per well and the samples were incubation for an additional 2 h until the cells adhered. Finally, the absorbance at 450 nm was detected.

A total of 1,500 BxPC-3 and AsPC-1 cells per well were seeded into 6-well plates and the transfected cells were cultured for 2 weeks under normal culture conditions. Then, the cells were fixed in 4% paraformaldehyde for 10 min and stained using 0.5% crystal violet solution for 10 min at room temperature, respectively. After washing three times with PBS, images of the cell colonies were captured and counted under a light microscope (magnification, ×20).

A 24-well Transwell chamber (Corning, Inc.) containing a polycarbonate membrane filter (8-µm pore size) was used for the Transwell assays. To investigate invasion, the chamber was precoated with 100 µg Matrigel for 1 h at room temperature, whereas this step was omitted for the migration assay. Briefly, 5×10⁴ cells in serum-free medium were plated in the top chamber and the bottom chamber was filled with medium (500 µl), supplemented with 10% FBS. After incubating for 24 h at 37°C, the cells remaining in the upper side of the filter were removed using cotton-tipped swabs. The cells that had migrated or invaded into the lower side of the filter were fixed with 4% methanol for 20 min and stained with 0.1% crystal violet for 20 min, both at room temperature, and finally counted under an light microscope (magnification, ×200).

The StarBase 2.0 database (http://starbase.sysu.edu.cn/) was used to predict the binding sites between miR-589-5p and TTN-AS1 or FOXP1. Santa Cruz Genome Browser (http://genome.ucsc.edu) was used to predict the potential transcription factor of TTN-AS1.

The corresponding full-length sequence of TTN-AS1 or FOXP1 3′-untranslated region (UTR) with wild-type (WT) or mutated (Mut) miR-589-5p binding sites was ligated into the pmirGLO vector (Promega Corporation) to form pmirGLO-TTN-AS1-WT/Mut or pmirGLO-FOXP1-WT/Mut reporter vectors. Then, the constructed reporter vectors were co-transfected with miR-589-5p mimics or NC mimics into the BxPC-3 and AsPC-1 cells, using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). After 48 h, the relative luciferase activities were measured using a Dual-Luciferase Reporter Assay System (Promega Corporation). Firefly luciferase activity was normalized to Renilla luciferase activity.

The fragments of the TTN-AS1 promoter containing the FOXP1-binding site (WT or Mut) were ligated into the pGL3-basic vector (Promega Corporation). Subsequently, the recombinant construct was co-transfected with pcDNA3.1/FOXP1 plasmids into the BxPC-3 and AsPC-1 cell lines, using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). After 48 h, the relative luciferase activity was examined using the Dual Luciferase Reporter Gene Assay kit (BioTek Instruments, Inc.) to analyze firefly and Renilla luciferase activities.

The BxPC-3 and AsPC-1 cell lines were collected and fixed with 1% formaldehyde for 10 min at 37°C for cross-linking DNA and protein. Next, ultrasonication was used to generate DNA fragments (200–500 bp). Then, the cell lysates, with the DNA fragments, were immunoprecipitated with anti-FOXP1 (1:1,000; cat. no. ab93807) or anti-IgG (1:10,000; cat. no. ab172730) (both from Abcam). Subsequently, magnetic beads were used to capture the precipitated DNA fragments and the precipitated DNA was quantified using RT-qPCR.

Total protein was isolated from the PC cells using RIPA lysis buffer (Thermo Fisher Scientific, Inc.). Protein concentration was detected using a BCA assay kit (Sangon Biotech Co., Ltd.). The total protein (20 µg) was separated with 10% SDS-PAGE, then transferred to a PVDF membrane. After blocking with 5% skimmed milk for 2 h, the membrane was incubated with the primary antibodies against E-cadherin (cat. no. ab1416; 1:1,000), N-cadherin (cat. no. ab18203; 1: 1,000) and GAPDH (cat. no. ab9485; 1:1,000) overnight at 4°C. Subsequently, the membrane was incubated with goat anti-rabbit IgG H&L (HRP) secondary antibody (1:1,000; cat. no. ab205718; Abcam) and the bands were evaluated using an enhanced chemiluminescence kit (Pierce; Thermo Fisher Scientific, Inc.).

Data from independent triplicate experiments were analyzed via SPSS 20.0 (IBM Corp.) and are shown as the mean ± SD. The comparison between two groups was conducted using paired or unpaired Student's t-test and between multiple groups using one-way ANOVA followed by Tukey's post hoc test. The correlation between mRNA expression levels was analyzed using Pearson's correlation analysis. The χ² test was used to assess the associations between TTN-AS1 expression and the clinicopathological characteristics. P<0.05 was considered to indicate a statistically significant difference.

To examine the function of TTN-AS1 in PC, its mRNA expression levels were first detected using RT-qPCR. As shown in Fig. 1A, TTN-AS1 was found to be significantly increased in PC tissues. Furthermore, clinical data demonstrated that TTN-AS1 expression was associated with TNM stage and lymph node metastasis, while there was no significant association between TTN-AS1 mRNA expression level and age or sex (Table I). Then, TTN-AS1 mRNA expression level was analyzed in the PC and HPDE cell lines, the latter was used as the NC. The results of RT-qPCR indicated that the PC cell lines exhibited higher TTN-AS1 mRNA expression levels compared with that in the HPDE cell line (Fig. 1B). Furthermore, the BxPC-3 and AsPC-1 cell lines exhibited the highest TTN-AS1 mRNA expression level; thus, these two cell lines were selected for further experiments. Subsequently, TTN-AS1 expression was knocked down using shRNA in the BxPC-3 and AsPC-1 cell lines, and the transfection efficiency was confirmed using RT-qPCR (Fig. 1C). CCK-8 assays revealed that knock down of TTN-AS1 significantly suppressed the viability of both the BxPC-3 and AsPC-1 cell lines (Fig. 1D). In addition, colony formation assays further verified the inhibitory effect of TTN-AS1 knockdown on BxPC-3 and AsPC-1 cell proliferation (Fig. 1E). Lastly, Transwell and Matrigel assays demonstrated that TTN-AS1 knockdown reduced the migration and invasion abilities of the BxPC-3 and AsPC-1 cell lines (Fig. 1F and G, respectively). The aforementioned findings indicated that TTN-AS1 may act as an oncogene in PC.

A number of studies have found that lncRNAs play key roles in cancer by combining with miRNAs (26–28); thus, it was investigated whether potential miRNAs could interact with TTN-AS1. To predict potential miRNAs, StarBase database (http://starbase.sysu.edu.cn/) was used. As a result, TTN-AS1 was found to potentially combine with miR-589-5p under strict screening conditions (Pan-Cancer, 10 cancer types), and the binding sequence of TTN-AS1 on miR-589-5p is illustrated in Fig. 2A. Subsequently, miR-589-5p mRNA expression level was determined and found to be low in PC tissues and cell lines (Fig. 2B). In further experiments, miR-589-5p mRNA expression level was increased following transfection with miR-589-5p mimics (Fig. 2C). Subsequently, the plasmids containing the WT (TTN-AS1-WT) and Mut (TTN-AS1-Mut) miR-589-5p binding site were generated and ligated into dual-luciferase reporter vectors for luciferase activity assay. The results revealed that the luciferase activity of TTN-AS1-WT was inhibited by miR-589-5p overexpression, but that of TTN-AS1-Mut was unaffected (Fig. 2D), suggesting the direct binding of TTN-AS1 to miR-589-5p. Furthermore, Pearson's correlation analysis demonstrated the inverse association between TTN-AS1 and miR-589-5p expression levels in the PC tissues (Fig. 2E). Collectively, these findings indicated that TTN-AS1 directly interacted with miR-589-5p in PC.

To further verify the ceRNA hypothesis, the downstream target genes of miR-589-5p were investigated. Using StarBase, 10 potential candidate targets were identified (Fig. 3A) and the mRNA expression level of these genes in the miR-589-5p mimics-transfected cells was examined using RT-qPCR. The results revealed that FOXP1 expression was lower compared with that in the other 9 genes when miR-589-5p was overexpressed (Fig. 3B). Thus, FOXP1 was selected for subsequent analyses. As shown in Fig. 3C, PC tissues expressed higher expression levels of FOXP1 compared with that in adjacent normal tissues. RT-qPCR and western blot analyses also indicated that the mRNA and protein expression level of FOXP1 was upregulated in the PC cell lines compared with that in the HPDE cell lines (Fig. 3D and E, respectively). Furthermore, it was demonstrated that the luciferase activity of FOXP1-WT, but not that of FOXP1-Mut, was significantly reduced in the miR-589-5p mimics-transfected cells (Fig. 3F). In addition, FOXP1 expression in the PC tissues was found to be negatively correlated with miR-589-5p and positively correlated with TTN-AS1 mRNA expression levels using Pearson's correlation analysis (Fig. 3G). Taken together, these findings indicated that miR-589-5p directly targeted FOXP1 in PC.

According to previous reports, FOXP1 may act as a transcription factor and promote the transcription of lncRNAs (29,30). However, to the best of our knowledge, whether FOXP1 can transcriptionally activate the expression of TTN-AS1 has not been investigated to date. Using the University of California, Santa Cruz Genome Browser (http://genome.ucsc.edu/), FOXP1 was found to act as a potential transcription factor by binding to the TTN-AS1 promoter, and its DNA motif was obtained from the JASPAR database (http://jaspar.genereg.net) (Fig. 4A). Then, the pcDNA3.1/FOXP1 plasmid was transfected into the BxPC-3 and AsPC-1 cell lines to increase FOXP1 expression (Fig. 4B and C). Subsequently, TTN-AS1 mRNA expression level was found to be significantly increased by pcDNA3.1/FOXP1 transfection (Fig. 4D). ChIP assay revealed the direct interaction between FOXP1 and the TTN-AS1 promoter (Fig. 4E). Subsequently, the WT and Mut binding sites between FOXP1 and TTN-AS1 promoter were obtained, and a luciferase reporter assay revealed that FOXP1 overexpression increased the luciferase activity of the WT TTN-AS1 promoter reporter construct, while no notable changes were observed with the Mut TTN-AS1 promoter reporter (Fig. 4F). All these data indicated that FOXP1 directly binds to the TTN-AS1 promoter.

To verify whether TTN-AS1 promoted PC progression via FOXP1, rescue experiments were performed. Based on the results of the CCK-8 assay, FOXP1 upregulation counteracted the inhibitory effect of TTN-AS1 knockdown in PC cell viability (Fig. 5A). The results of the colony formation assay suggested that the suppressed proliferative ability in TTN-AS1 knockdown cells was restored by increasing the expression level of FOXP1 (Fig. 5B). Furthermore, TTN-AS1 knockdown inhibited the migration and invasion of the BxPC-3 and AsPC-1 cells, while the overexpression of FOXP1 recovered this effect (Fig. 5C and D). In addition, the results of western blot analysis demonstrated that TTN-AS1 knockdown notably increased and decreased E-cadherin and N-cadherin protein expression levels, while FOXP1 overexpression partially reversed this effect (Fig. 5E). In conclusion, TTN-AS1 upregulated FOXP1 to facilitate PC cell line migration and invasion.