MAFG-AS1 aggravates the progression of pancreatic cancer by sponging miR-3196 to boost NFIX

Background A host of researches have demonstrated the regulation of long non-coding RNAs (lncRNAs) in the progression of pancreatic cancers (PC). In this study, our main task was to analyze the function of MAF bZIP transcription factor G antisense RNA 1 (MAFG-AS1) in PC. Methods RT-qPCR measured gene expression. Functional experiments, including EdU assay, flow cytometry analysis, TUNEL assay and transwell assay, assessed the biological changes of PC cells. RNA pull down assay, luciferase reporter assay and RIP assay verified the interaction between RNAs. Results MAFG-AS1 was lowly expressed in normal pancreatic samples but up-regulated in PC tissues and cell lines. Besides, MAFG-AS1 silence suppressed cell proliferation and migration whereas promoted cell apoptosis in PC. Mechanism assays verified that miR-3196 could bind with MAFG-AS1. Moreover, miR-3196 was discovered to be lowly expressed in PC cell lines, and its overexpression inhibited PC cell growth and migration. Importantly, nuclear factor I X (NFIX), overexpressed in PC cell lines, was validated to be positively modulated by MAFG-AS1 through absorbing miR-3196. Moreover, overexpression of NFIX could countervail the restraining effects of MAFG-AS1 knockdown on the growth and migration of PC cells. Conclusion MAFG-AS1 had an oncogenic function in the progression of PC via regulating miR-3196/NFIX pathway, and decreasing MAFG-AS1 expression could attenuate PC progression.


Background
Pancreatic cancer (PC) is a common tumor in the digestive system around the globe, characterized by intensive migration and high invasion [1]. Though great achievements have been made in comprehensive therapies of cancers, the morbidity and mortality of PC are still increasing, which makes it a great threat to human life and health [2]. The prognosis of PC is rather dismal in the past decade [3]. Surgery is a routine treatment for PC patients, but the recurrence rate is extremely high due to frequent metastasis [4]. The detailed cause of PC is still a riddle. Hence, it is necessary to study the pathological feature of PC and figure out the underlying mechanism to find out effective therapeutic ways.
Increasing essays certified that lncRNAs function as competing endogenous RNAs (ceRNAs) of messenger RNAs (mRNAs) to bind with microRNAs (miRNAs) so that mRNAs could be released from the binding of miRNAs. This regulation mode is known as the ceRNA mechanism [9]. For instance, LINC00511 boosted the progression of breast cancer through targeting miR-185-3p/E2F1/Nanog [10]. FER1L4 inhibited cell migration in osteosarcoma via sponging miR-18a-5p to modulate PTEN [11]. XIST facilitated the growth of thyroid cancer cells through sequestering miR-34a to modulate MET-PI3K-AKT pathway [12]. This study made investigations on whether MAFG-AS1 could be a ceRNA in PC.
The current study analyzed the role of MAFG-AS1 in the progression of PC. Moreover, we also assessed the modulatory mechanism of MAFG-AS1 in PC.

Sample collection
The PC tissues and paired non-cancerous samples were acquired from 65 PC patients who underwent surgery in Dongguan People's Hospital Hospital. Samples were processed with liquid nitrogen for quick-frozen, and then maintained at -80℃. Patients who received other treatment before operation were excluded. All the relevant patients signed the informed consents prior to surgery. This study was conducted with the approval of the Ethics Committee of Dongguan People's Hospital numbered (DGP2019-076).

RT-qPCR
TRIzol reagent (Invitrogen) was used to extract total RNA from cultured cells, and total RNA was reversely transcribed into cDNA by using PrimeScript ™ RT reagent kit (TaKaRa, Shiga, Japan). RT-qPCR analyses were carried out through using SYBR green Supermix (Thermo Fisher, Waltham, MA, USA) according to the manufacturers' instructions. Relative gene expression was calculated by 2 −ΔΔCt method. GAPDH and U6 acted as normalized genes. The experiment was repeated at least three times. The sequences of primers were listed in Table 1.

EdU
EdU assay was completed with BeyoClick ™ EdU Cell Proliferation Kit (Beyotime, Shanghai, China) based on the user guide. Cells were planted on 96-well plates (1 × 10 4 cells/well) and fixed by 4% paraformaldehyde for 30 min. 0.5% Troxin X-100 was utilized to permeate cells for 10 min. Results were acquired using fluorescent

Flow cytometry analysis
Cells (1 × 10 6 ) were first planted into 6-well plates for 48 h and then washed twice with PBS. After resuspension, the precooled 70% ethanol was employed to fix cells at 4 ℃ for 1 h. Annexin V-FITC/PI double staining kit (Invitrogen) was utilized for detecting apoptotic cells. At last, samples were analyzed with flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA). The experiment was repeated at least three times.

TUNEL staining assay
Cells after transfections were fixed by 4% paraformaldehyde at 37℃ in darkness. Later, cells were permeated with 0.1% Triton x-100 for 5 min. Then PBS was applied to rinse cells thrice. In line with the manufacturer's instructions, cell apoptosis was assessed using One-Step TUNEL Apoptosis Assay Kit (Beyotime). DAPI was taken to stain nuclei. Apoptotic cells were observed through fluorescence microscope (Leica). The experiment was repeated at least three times.

Transwell assays
Cell migration assay was conducted by transwell chamber (Corning Incorporated, Corning, NY, USA), in line with the standard protocol. Cells were added into the upper chamber, and the lower chamber was supplemented with complete medium. After 24 h of incubation, the migrating cells were assessed in five random fields using optical microscope (Olympus). The experiment was repeated at least three times.

Luciferase reporter gene assay
For luciferase reporter gene assay, cells were co-transfected with indicated pmirGLO vectors (Promega, Madison, WI, USA) which were loaded with the full-length MAFG-AS1 or NFIX 3′-UTR sequence containing wildtype or mutant miR-3196 binding sites, in the presence of miR-3196 mimics or NC mimics. After 48 h of co-transfection, relative luciferase activity was monitored using Luciferase Reporter Assay System (Promega). The experiment was repeated at least three times.

FISH
RNA FISH KIT (Ribobio) was applied to confirm the subcellular location of MAFG-AS1. The RNA FISH probe mix was synthesized and created by RiboBio. The air-dried cells were treated with specific probes and hybridization buffer. DAPI (Beyotime) was used to stain cell nuclei. The fluorescent detection was completed by fluorescence microscope (Olympus). The experiment was repeated at least three times.

Subcellular fractionation
PARIS ™ Kit (Invitrogen) was utilized to isolate cytoplasmic and nuclear RNAs from SW1990 and PANC-1 cells, in light of the provided instruction. Briefly, cells were lysed with cell fractionation buffer, followed by the separation of cytoplasmic and nuclear fractions by centrifuge. U6 was used as the nuclear control and GAPDH as the cytoplasmic control. Expression levels of RNAs were analyzed by RT-qPCR. The experiment was repeated at least three times.

In situ hybridization (ISH)
ISH assay was implemented as per the user guide (Boster Bio-Technology Company). Briefly, the paraffin-embedded sample sections with 4 μm thickness were subjected to dewaxing via xylene, rehydration by diluted reagent grade ethanol, and digestion with protease K. Afterwards, the sections were processed with specific MAFG-AS1 probes labeled with digoxigenin at 37 ℃ overnight, followed by 30 min of the treatment with HRP-conjugated anti-digoxin antibody at 25 ℃. Diaminobenzidine was the HRP substrate for staining. The images were captured by an optical microscope (Olympus). The experiment was repeated at least three times.

RNA immunoprecipitation (RIP)
RIP assay was carried out using the Magna RIP ™ RNA-Binding Protein Immunoprecipitation Kit (Millipore, Billerica, MA, USA). The prepared cells were lysed in RIP lysis buffer, and then treated with RIP buffer containing magnetic beads conjugated with human Ago2 antibody or normal IgG antibody. Afterwards, the retrieved RNA was assayed by RT-qPCR. The experiment was repeated at least three times.

Pull down assay
The interaction between RNAs was examined by RNA pull down assay by use of Pierce Magnetic RNA-Protein Pull-Down Kit (Thermo Fisher), as guided by the provider. After collecting the extracts from cells, the magnetic beads were added to capture the RNA-RNA complex. At length, RNAs in the complex were eluted and finally analyzed by RT-qPCR. The experiment was repeated at least three times.

Statistical analysis
Bio-repeats were conducted thrice, and all results were exhibited as the mean ± SD. Data were analyzed by Student's t-test or one-way analysis of variance (ANOVA) using GraphPad PRISM 6 (GraphPad, San Diego, CA, USA). Differences were regarded as statistically significant with P < 0.05.

MiR-3196 hindered the progression of PC
The impacts of miR-3196 on the biological behaviors of PC cells were investigated as well. Data of EdU assays exhibited that overexpression of miR-3196 reduced the ability of PC cells to proliferate (Fig. 3a, **P < 0.01). Conversely, the apoptosis rate was increased when miR-3196 was up-regulated in PC cells (Fig. 3b, c, **P < 0.01). Meanwhile, cell migratory capacity was repressed under miR-3196 up-regulation (Fig. 3d, **P < 0.01). To sum up, miR-3196 inhibited the growth and migration of PC cells.

MAFG-AS1 promoted the progression of PC via elevating NFIX
Then, rescue assays were performed to validate whether MAFG-AS1 modulated NFIX to influence the progression of PC. In this case, NFIX expression was enhanced in PC cells by the transfection of pcDNA3.1-NFIX (Fig. 5a, **P < 0.01). Also, it was proved that the falling trend of NFIX expression induced by MAFG-AS1 depletion was then recovered under the co-transfection of pcDNA3.1-NFIX (Fig. 5b, **P < 0.01). Seen from the results of EdU assays, PC cell proliferation was suppressed by MAFG-AS1 silence, while such descending tendency was neutralized after NFIX recovery (Fig. 5c, **P < 0.01). Besides, the apoptosis of PC cells accelerated by MAFG-AS1 knockdown was reversed by the overexpression of NFIX (Fig. 5d, e, **P < 0.01). Likewise, the falling trend of cell migratory ability imposed by down-regulated MAFGAS1 was offset by up-regulated NFIX (Fig. 5f, **P < 0.01). In conclusion, MAFG-AS1 facilitated the oncogenic phenotypes of PC cells by enhancing NFIX expression.

Discussion
Mounting essays have illustrated that lncRNAs have crucial functions in the biological progression of various cancers. For example, lncRNAs FTX facilitated the progression of gastric cancer via sponging miR-144 to elevate ZFX [13]. EPIC1 was reported to interact with YAP1 to facilitate the growth of PC cells [14]. DILC had oncogenic contributions to the progression of gallbladder carcinoma [15]. In previous studies, MAFG-AS1 has been introduced to promote the progression of lung adenocarcinoma [16] and breast carcinoma [17]. Presently, MAFG-AS1 was first detected to be highly expressed in PC tissues and cell lines. The following functional assays demonstrated that MAFG-AS1 knockdown suppressed cell proliferation and migration while promoted cell apoptosis in PC. This was similar to the finding of a former study that MAFG-AS1 boosted cell proliferation and migration in hepatocellular carcinoma [18]. Based on the above findings, we inferred MAFG-AS1 as an oncogene in PC. Substantial literatures analyzed the role of lncRNAs in ceRNA mechanism in the development of multiple cancers [19]. As an example, lncRNA ATB was depicted as a ceRNA of CTNNB1 to bind with miR-200a [20]. MiR-NAs have been recognized as the main regulators of cancer development as well. For instance, miR-511 was described to suppress breast cancer cell proliferation and invasion through targeting FGF4 [21]. In this study, we found miR-3196 could bind with MAFG-AS1. MiR-3196 was measured to be lowly-expressed in PC cell lines, and the up-regulation of miR-3196 repressed PC cell proliferative and migratory capacities, which was consistent with its function in breast cancer [22]. Moreover, many researches proved that miRNAs can bind to the 3′ UTR of mRNAs to silence these mRNAs. In this research, we assumed NFIX was the downstream target of miR-3196 and demonstrated that NFIX expression was positively regulated by MAFG-AS1 and negatively regulated by miR-3196. Existing research verified that NFIX acted as an oncogene in gastric cancer [23]. The current study discovered NFIX was highly expressed in PC cells. Besides, up-regulation of NFIX could restore the suppressive impacts of MAFG-AS1 silence on PC cell growth and migration.
Our research unveiled a ceRNA network of MAFG-AS1/miR-3196/NFIX axis in PC, highlighting MiR-3196 bound with NFIX 3′ UTR. a The expression of three mRNAs was assessed by RT-qPCR in cells with or without MAFG-AS1 inhibition. b NFIX expression was detected by RT-qPCR in cells with miR-3196 mimics or NC mimics. c NFIX expression was measured by RT-qPCR in cells transfected with sh-NC, sh-MAFG-AS1#1 or sh-MAFG-AS1#1 plus miR-3196 inhibitor. d NFIX expression was detected in PC cell lines through RT-qPCR. e RIP assay results showed that MAFG-AS1, miR-3196 and NFIX were enriched by Ago2 antibody. f RNA pull down examined miR-3196 bound with NFIX. g Luciferase reporter assays validated the binding between miR-3196 and NFIX. *P < 0.05, **P < 0.01 MAFG-AS1 as a promising target for PC treatment. However, the deficiency of in vivo evidences is the main limitation of the current work, and this will be solved in the future. The clinical significance of MAFG-AS1 in PC needs to be further testified. Moreover, we will further focus on the reason why MAFG-AS1 is upregulated in PC and the mechanism whereby NFIX affects PC development.

Conclusion
In conclusion, our study validated that MAFG-AS1 facilitated the progression of PC by targeting miR-3196/NFIX pathway. Importantly, targeting MAFG-AS1 might be a potential way to treat patients with PC.