Deciphering The Molecular Mechanism of Long Non-Coding RNA HIFA-AS1 Regulating Pancreatic Cancer Cells

Background:HIFA-AS1, an antisense transcript of HIF1α gene, is a 652-bp long noncoding RNA (lncRNA) which globally expressed in multiple tissues of animals. Recent evidence indicated that the HIFA-AS1 was involved in tumorigenesis of several types of cancer, but there were no reports on pancreatic cancer (PC). Results: In order to investigate whether the HIFA-AS1 could mediate the PC or not, it was overexpressed in a PC cell line (PANC-1), and a series of experiments including cell viability detection, flow cytometry, transwell migration, clone formation and wound healing were performed. Functionally, the results indicated that overexpression (OE) of HIFA-AS1 could inhibit proliferation and shift, and promote apoptosis of PC cells. Moreover, to explore underlying molecular mechanism of anti-tumorigenic actions of HIFA-AS1 in PC cells, the iTRAQ (isobaric tags for relative and absolute quantification) quantitative proteomics analysis was implemented and the results indicated that OE of HIFA-AS1 globally affected the expression levels of multiple protein associated with metabolism of cancer. Moreover, the network analysis revealed that the most of these differentially expressed proteins (DEPs) were integrated, and severed essential roles in regulatory function. Conclusions: In summary, HIFA-AS1 may exhibit a potential therapeutic effect on PC, and our study provided useful information in this filed.


Introduction
Pancreatic cancer (PC) remains one of the most common causes of cancer-related mortality [1] at the seventh in humans worldwide [2], with 5-year overall survival rate of less than 5% [3]. In most cases, PC develops with usually clinically silent at the early stage, but the variable symptoms, local invasiveness, or metastases only develop at an advanced stage [4]. Nowadays, the therapeutic efficacy of PC treatment is still very limited, and far from satisfactory [5,6]. Hence, in order to enhance the cure rate of PC, it is necessary to investigate the molecular mechanisms, which would provide new opportunities to improve effective therapeutic strategies against PC.
Long non-coding RNAs (lncRNAs), a kind of non-coding RNAs transcripts, comprises longer than 200 bp without protein-coding potential [7][8][9]. Current studies have showed that lncRNAs could mediate gene expression via chromosome remodeling, transcription and post-transcriptional processes [10]. As so far, increasing evidences demonstrate that lncRNAs play an important role in regulating vital molecular mechanism [11] and biological functions of the cells [12,13], such as proliferation, migration, invasion, cell cycle and apoptosis [14][15][16]. Without a doubt, various expression of lncRNAs could contribute to tumor development and progression [17], but its regulatory mechanism had not been completely investigated.
HIFA-AS1 is an antisense transcript of HIF1α [18], and accumulating evidence has revealed that it plays a key role in proliferation and apoptosis of vascular smooth muscle cells [19][20][21], and human hepatic stellate cells [22]. Furthermore, it promotes tumor necrosis factor-α-induced apoptosis [23], thereby affecting the occurrence and development of thoracic aortic aneurysm [24]. HIF1A-AS1 can regulate starvationinduced hepatocellular carcinoma cell apoptosis, promoting hepatocellular carcinoma (HCC) cell progression [25]. Therefore, HIFA-AS1 has the capacities to affect the occurrence and development of multiple types of cancer, but there is no report on the molecular regulation mechanism of HIFA-AS1 in PC.
In the current study, to explore whether HIFA-AS1 could regulate PC or not, we constructed overexpression (OE) plasmids containing HIFA-AS1, and transferred them to a PC cell line. Subsequently, a series of experiments, including cell viability detection, flow cytometry, transwell migration, clone formation and would healing were conducted. And the experimental results showed that OE of HIFA-AS1 could inhibit proliferation and metastasis, and promote apoptosis of PC cells, comparing with the normal control (NC) cells. In order to further explore the molecular mechanism of HIFA-AS1 regulating PC cells, we collected cell samples from OE and NC groups for iTRAQ (isobaric tags for relative and absolute quantification) proteomics experiments.
Here we report the results.

Cell culture
The human pancreatic cancer cell line (PANC-1) was provided by Procell (Wuhan, China) and cultured in monolayers in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen, Carlsbad, USA). All media were supplemented with 10% fetal bovine serum (Hyclone UT, USA) in the presence of 100 U/ml penicillin and 50 μg/m streptomycin (Beyotime, Shanghai, China) with humidified atmosphere of 5% CO2 and 95% air at 37 °C.

Plasmid construction, Lentivirus package and transfection
HIFA-AS1 was cloned into the pcDNA3.1(+) vector using the restriction sites for KpnI (GGTACC) and XhoI (CTCGAG), and this 652 bp insert was verified by sequencing.
Two μg of plasmids containing HIF1A-AS1 were mixed with the lentivirus packaging plasmids pMDLg-pRRE, pMD2.G, and pRSV-Rev according to the previous standard protocol [26]. Subsequently, PANC-1 cells were infected with 20 multiplicity of infection (MOI) lentivirus for 24 h and incubated in fresh medium. The cells were washed with fresh complete media after 24 h and the efficiency OE of HIF1A-AS1 was verified by quantitative RT-PCR (qRT-PCR).

RNA extraction and qRT-PCR
Total RNA was extracted from the cells using TRIzol reagent (Ambion, Austin, USA) and further purified with two phenol-chloroform treatments, then treated with RQ1 DNase (Promega, Madison, USA) to digest DNA. The quality and quantity of the purified RNAs were determined using a Nano Photometer spectrometer with the absorbance at 260 nm/280 nm and next were verified by 1.2% agarose gel electrophoresis. The cDNA was synthesized with random primers with the Highcapacity cDNA Reverse-Transcription Kit (Takara, Dalian , China), and real time PCR was implemented for detecting gene expressions using the designed primers in Table 1 with SYBR Green I dye (Qiagen, Hilden, Germany). The PCR conditions were as follows: pre-denaturation at 95 °C for 1 min, 40 cycles of denaturing at 95 °C for 15 s, annealing at 60 °C for 30 s and elongation at 72 °C for 40 s. The relative expression of genes was analyzed by the 2 − △△ CT method with the Actin as an internal control [27].

Cell viability detection
The viability of PANC-1 cell was evaluated using the CCK-8 assay (Solarbio, Peking, China) according to the instructions of the manufacturer. The cells were slightly seeded into the 96-well plates with 100 μl suspension per well overnight. At 0, 24, 48 and 72 h, 10 μl of CCK-8 solution was added to each well, and then the plates were incubated for 0.5 h. At last, absorbance was measured at 450 nm by microplate reader (Bio-Rad, Hercules, USA).

Flow cytometric detection
The apoptosis of the PANC-1 cells were determined by flow cytometry with Annexin V-conjugated FITC Apoptosis detection kit (BD, Franklin Lakes, USA). After infection for 24 h, cells were harvested and washed twice with PBS, then re-suspended in 5μl FITC-conjugated anti-Annexin V antibody and in 500 μl binding buffer with 5μl Propidium iodide (PI). Apoptosis was measured with a FACS calibur flow cytometer MoFLO XDP (Beckman, Hercules, USA).

Transwell invasion
For the transwell assay, the properties of migration of cells were evaluated by using 24well transwell plates (Corning, NY, USA). About 1×10 4 cells per well were seeded into the upper chamber with serum free medium in triplicate. Medium containing 10% FBS (300 μl) was added to the DMEM of 5% CO2 and 95% air at 37°C. After incubation for 24 h, the medium were removed, and cells were fixed with 4% paraformaldehyde for 15 min, stained with 0.1% crystal violet for 20 min, and counted from five randomly chosen fields for each well by stereo microscope (Leica, Wetzlar, Germany).

Wound Healing Assay
After 12 h of the transfected cells were seeded in 6-well plates, confluent monolayers in each well were washed with PBS and created using a 200 μl sterile pipette tip to generate a wound. Wound healing was evaluated and photographed images were taken by 200 magnification a Zeiss microscope (Leica, Wetzlar, Germany) from each well at 0, 24, and 48 h post-injury time points after the wound was made.

Cell clone formation assay
PANC-1 cells were plated into 6-well plates (800 cells per well) and cultured for 10-14 days, then were digested at the logarithmic phase to make a single-cell suspension using culture medium. Cell were stained with 0.4% crystal violet (Bio Basic Inc., Markham, Canada). Finally, the number of colonies was calculated under an inverted microscope (Leica, Wetzlar, Germany).

Western Blotting Analysis
PANC-1 cells were collected and the homogenates were centrifuged for 30 min at 4 °C , 12,000 rpm with cell lysis buffer. Then the protein extracts were separated on 10% SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes.
Membranes were incubated at room temperature for 1 h in a 5% skim milk TBST blocking solution and incubated with agitation at 4 °C overnight with specific primary At last, protein bands were determined using the Western blotting detection system (GE Healthcare, Amersham, UK). The 100 μg protein were trypsin digested with trypsin each sample. After diluting the protein solution 5 times with 100 mM TEAB, trypsin were added by a mass ratio of 1:50 (trypsin: protein) overnight at 37 °C. The peptides were desalted with C18 column after enzymolysis, and the desalted peptides were vacuum freeze-dried.
The original MS/MS file data were analyzed using ProteinPilot Software v4.5 (AB Sciex, Shanghai, China). For protein identification, the Paragon algorithm, which was integrated into ProteinPilot, was used against the UniProt/SwissProt database for database searching. The parameters were set as follows: the instrument was TripleTOF 5600+, iTRAQ quantification, and cysteine modified with IAM, and biological modifications were selected as ID focus, trypsin digestion, quantitate, bias correction, and background correction was used for protein quantification and normalization. For calculation of the false discovery rate (FDR), an automatic decoy database search strategy was used to estimate FDR using the proteomics system performance evaluation pipeline software (PSPEP, integrated into the ProteinPilot Software). Unique peptides were used for iTRAQ labeling quantification, and peptides with global FDR values from fit less than 1% were considered for further analysis. Within each iTRAQ run, differentially expressed proteins (DEPs) were determined based on the ratios of differently labeled proteins and p values provided by ProteinPilot, the p values were generated by ProteinPilot using the peptides used to quantify the respective protein. For the determination of DEPs, fold changes (FC) were calculated as the average comparison pairs among biological replicates. Proteins with FC larger than 1.2 and a p < 0.05 were considered to be changes that are significantly different.

Bioinformatics and annotations
The biological and functional properties of all the identified proteins were analyzed by matching to NCBInr (http://www.ncbi.nlm.nih.gov/) and Swiss-Prot/UniProt The PPI network was constructed and visualized using Cytoscape software (version 3.5.1; www.cytoscape.org).

Statistical analyses
All data analysis was performed using the SPSS16.0 statistical software as the mean ± standard deviation. Statistically significant differences comparison between two groups means were analyzed by Student's t-test. P value <0.05 was considered to be statistically significant.

Data availability statement
The datasets generated and/or analysed during the current study are available in the Pr otemXchange repository(Accession No: IPX0003153000). For the interview, the data sets could also be obtained from a web link: https://www.iprox.org/page/PSV023.html;?url=1622727859237GpMj, with a code: E 5xT.

HIFA-AS1 regulates the apoptosis and proliferation of PC cells
To investigate the role of HIFA-AS1 in the PC cells, an OE vector containing the HIFA-AS1 was transfected into PANC-1 cells. The qRT-PCR experiment was applied to measure the efficiency of OE , and the results showed that expression levels of HIFA-AS1 in OE cells was about 10000-fold more as compared with the normal control (NC) cells (Fig. 1A), demonstrating a successful establishment of human PC cells with OE of HIFA-AS1.
To explore whether the HIF-AS1 regulate the proliferation of PC cells or not, the experiments including CCK-8, were conducted for PANC-1 cells from OE and NC groups. The results indicated that viability cells with HIF-AS1 OE was declined significantly, during varying time periods (0, 24, 48 and 72h) (P<0.001) (Fig. 1B and   C).
Furthermore, flow cytometry was used to determine whether HIF-AS1 could affect apoptosis of PANC-1 cells or not. It ( Fig. 1D and E) displayed that the OE of HIF1A-AS1 significantly promoted apoptosis of PC cells, and the number of apoptotic cells obviously increased about 50% compared with the NC group (Fig. 1D). The above results indicated that HIF1A-AS1 had the capacities to regulate proliferation and apoptosis of the PC cells. Further, western blot analysis reported that the expression levels of Cleaved caspase-3, Bax, P53, and PARP-1A protein were higher in OE group (Fig. 1F), thereby inducing apoptosis in pancreatic cancer.

HIFA-AS1 regulates the migration of PC cells
To further explore the role of HIF1A-AS1 in regulating metastasis of PANC-1 cells, the transwell migration assays were performed. It showed that the migration ability of cells from OE group was reduced about 50% (P<0.001) ( Fig. 2A and B). In terms of cell clone, the clones formed in NC had a greater cell number about 45% compared with OE ( Fig. 2C and D). Cell migration was detected using Wound-healing assay in PANC-1 cells. The results from invasion assay showed that OE of HIF-AS1 promoted cell invasion after transfection for 24 and 48 hours (Fig. 2E). And the difference of the wound width after 24 hours of transfection is the most significant compared with the comparison, which exceeds the control by about 20% (Fig. 2F), suggesting a functional role for HIF1A-AS1 in inhibiting metastasis of the PC cells.

The summary of iTRAQ proteomics analysis
To explore the molecular mechanism of HIFA-AS1 mediating the proliferation, apoptosis and shift of PANC-1 cells, an iTRAQ was applied to uncover altered protein expressions and signaling pathways.
In total, the quality of the data obtained from the iTRAQ was analyzed using parameters such as coefficient of variation about repeatability, distribution of unique peptide, peptide length, and distribution of coverage ( Table 2). First of all, for the repeatability, there is a little difference concentration of CV data between NC and OE groups, and the cumulative percentages of CV were 7.81 % and 7.29 % respectively, indicating that the PANC-1 samples in each group are more reproducible (Fig. 3A).
In accordance with unique peptide determined as the peptide identified only for one protein, the presence of the corresponding protein can be uniquely determined. Then for the distribution of unique peptide number, the two-coordinate distribution map showed the number of unique peptides contained in all the proteins identified in this assay. For example, when the x-axis, left y-axis and right y-axis are 2, 646 and 26.25 respectively, it means that there are 646 proteins with 2 as the unique number of peptides, which account for 26.25% of the total number of proteins obtained (Fig. 3B).
Subsequently, the length of the identified peptides was analyzed. The average length of the polypeptide was 11.56 and within a reasonable range. Moreover, the length of the identified peptides was mainly concentrated between 7 and 15, and 9 was the maximum number (Fig. 3C). In addition, the protein identification coverage could reflect the overall accuracy of the identification results indirectly. The different colored pie represented the percentage of proteins with different identification coverage ranges. It showed that 37.21% proteins were with the peptide coverage less than 10%, and 39.51% proteins had more than or equal to 20% of the peptide coverage, with the average protein identification coverage being 19.53% (Fig. 3D).
A total of 4872 proteins were identified in all samples, and 4738, 2475 and 2539 ones were annotated successfully by GO, COG and KEGG, respectively (Fig. 3E).
Particularly, the GO enrichment for the 4738 annotated proteins was carried out, including cellular localization (CC) (Data not shown), molecular functions (MF) (Data not shown) and biological processes (BP). The BP classification showed that most of these proteins were enriched in cellular process (13.04%), metabolic process (11.26%), biological regulation (8.55%), regulation of biological process (8.13%), cellular component organization or biogenesis (7.00%) and so on (Fig. 3F).

Exploration of DEPs and functional analysis
On basis of the relative quantitative results, 338 DEPs were found in OE VS NC according to FC and p value (FC ≥ 1.2 or ≤ 0.83, p ≤ 0.05), and the up-regulated and down-regulated ones were 183 ( Table 3) and 155 (Table 4), respectively. The protein abundance distribution graph, and the volcano plot showed the proportion of DEPs in the total identified proteins ( Fig. 4A and 4B). A hierarchical clustering analysis of DEPs was also performed (Fig. 4C).
According to these results indicate that the inhibitory effects of HIF1A-AS1 on the proliferation, apoptosis and migration of PANC-1 cells may be related to its capability to regulate protein interactions, catalytic activity and enzyme regulator activity. The top 10 pathway metabolic function types were different in all up-regulated and downregulated differential proteins by KEGG. Four types were same containing metabolic pathway, regulation of actin cytoskeleton, microbial metabolism in diverse environment, protein processing in endoplasmic reticulum. Pathway analysis revealed that "Metabolic pathways" at the second of up-regulated genes and at the first in downregulated genes in the enrichment results. Moreover, OE of HIF1A-AS1 may exhibit anticancer effects by regulating the pathways associated with metabolism of cancer.
Therefore these results suggested that HIF1A-AS1 might affect RNA polymerase to control the transcription of downstream tumor-associated genes as to antagonize the proliferation, apoptosis and migration of PC cells.

Construction of DEPs Protein-protein interaction (PPI) network
PPI network of common DEPs was constructed by the STRING online database and Cytoscape software to analyze the interactions of DEPs because the String database could identify interactions between known proteins and predictive proteins (Fig. 5). A total of 338 DEPs (155 down-regulated and 183 up-regulated) were filtered into the DEPs PPI network complex. The wonderful network suggested that these DEPs might work together to regulate apoptosis, proliferation and invasion of PC cells. These proteins are expected to become targets for the treatment of PC.

Discussion
PC remains one of the deadliest cancer types and worlds' most aggressive malignancies [28]. Accumulating reports have reported that the potential of lncRNAs as diagnostic or prognostic biomarkers ubiquitously dysregulated and have crucial regulatory roles in tumor cells, including PC [29]. However, the regulatory mechanisms of multiple lncRNAs are elusive in many kinds of cancers such as thoracic aortic aneurysm and HCC. Herein, we first aimed to explore the molecular mechanisim of HIF1A-AS1 regulating PC.
In the present study, we investigated the biological function of HIF1A-AS1 on proliferation, apoptosis, and metastasis of PC cells. Consistently, it was found that HIF1A-AS1 was a suppressor of cell growth and progression in PC. Firstly, upregulation of HIF1A-AS1 inhibited cell growth and promoted apoptosis in PANC-1 cancer cells. Moreover, HIF1A-AS1 inhibited cell migration. Actually, the function of HIF1A-AS1 in other tumors has been reported. For instance, higher expression of HIF1A-AS1, a novel diagnostic predictor, could be clinically functioned as a potential biomarker in colorectal carcinoma [30]. Besides, the levels of HIF1A-AS1 were significantly increased in tumor tissues or serum from non-small cell lung cancer patients [31]. Above reports researched with clinical samples, but may be the opposite with this study using PC cells. Therefore, this study indicated HIF1A-AS1 as an important role for a novel mechanism in the PC modulated progress, could develop as a potential therapeutic target or biomarker for PC prevention and control.
The quantitative proteomics analysis revealed that the expression levels of SOD2 was up-regulated. Previous studies have revealed that SOD2 has both tumor suppressive and promoting functions, which are primarily related to its role as a mitochondrial superoxide scavenger and H2O2 regulator [32]. SOD2 is role as both a tumor suppressor in early tumorigenesis and as a tumor promoter during metastatic progression [33].
Therefore, HIF1A-AS1 could regulate SOD2 to inhibit the metabolic developmental process in PC. In addition, Mx1 expression was inversely correlated with prostate cancer [34]. However, in this study the expression levels of MX1 was positive correlation with PC. MX1 [35], the members of IFN-stimulated genes [36], HIF1A-AS1 could regulate MX1 of type I interferon-mediated signaling pathway to restrain PC.
According to previous reports, IFIT1 [37], OE of IFIT1 involved in a variety of biological processes, such as cell proliferation, migration and tumor growth [38,39].
In this review, we hypothesize that HIF1A-AS1 could inhibit IFIT1 to regulate the cell proliferation, migration to restrain PC to some extents. IFIH1, may play an important role in enhancing natural killer cell function and may be involved in growth inhibition and apoptosis in several tumor cell lines [40,41]. In a word, the involvement of IFIH1 in the growth inhibition and apoptosis of PANC-1 cancer cells was regulated by HIF1A-AS1. ISG15, induce natural killer cell proliferation, act as a chemotactic factor for neutrophils and act as an IFN-gamma-inducing cytokine lays a substantial role in the antiviral state induced by IFN [42,43]. Further, we showed that inhibition of PC by HIF1A-AS1 and ISG15 is sufficiently. In line with these studies, the present findings demonstrated that HIF1A-AS1 might regulate some tumorigenesis of PANC-1 cancer cells via targeting interferon-mediated signaling pathway, ubiquitin system and H2O2 regulator all closely related to metabolic pathways in cancer. In addition, tumorsuppressive role HIF1A-AS1 positively regulated the expression of SOD2, and negatively regulated the five gene expression related to metabolic regulation to suppress PC growth and progression.
In summary, these findings demonstrated that HIF1A-AS1 could inhibit cell growth and progression of PANC-1 cells. In the future studies about diagnostic specificity and sensitivity of HIF1A-AS1, may be used as a potential biomarker and guidance for early diagnosis of PC. Although clinical applications need to be further explored, these results further provided insight into the molecular mechanisms associated with the tumorigenesis and scientific experimental basis for the treatment of PC.

Ethics approval and consent to participate
Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
The datasets generated and/or analysed during the current study are available in the Pr otemXchange repository(Accession No: IPX0003153000). For the interview, the data sets could also be obtained from a web link: https://www.iprox.org/page/PSV023.html;?url=1622727859237GpMj, with a code: E 5xT.

Competing interests
The authors declare that they have no competing interests.        Table Legends Table 2. The all original data from the iTRAQ was analyzed using parameters.

Tables and
These data were uploaded in Table 2. These data were uploaded in Table 3. These data were uploaded in Table 4. Table 2，Table 3 and Table 4.

Supplementary. The all original data included
These data were uploaded in Supplementary. Figure 1 Overexpression (   Protein-protein interaction (PPI) network based on the differentially expressed proteins (DEPs). The total of 338 DEPs were ltered into the DEPs PPI network complex using the STRING online database. The round nodes indicate individual proteins.

Figure 6
The quantitative RT-PCR (qRT-PCR) validation of some differentially expressed proteins (DEPs) obtained from iTRAQ analysis in Overexpression (OE) of HIF-AS1 and normal control (NC) groups.