Mmu_circ_0000037 inhibits the progression of acute pancreatitis by miR‐92a‐3p/Pias1 axis

Abstract Background Acute pancreatitis (AP) is an inflammatory disease with high mortality. Previous study has suggested that circular RNAs are dysregulated and involved in the regulation of inflammatory responses in AP. This study aimed to investigate the function and regulatory mechanism underlying mmu_circ_0000037 in caerulein‐induced AP cellular model. Methods Caerulein‐treated MPC‐83 cells were used as an in vitro cellular model for AP. The expression levels of mmu_circ_0000037, microRNA (miR)‐92a‐3p, and protein inhibitor of activated STAT1 (Pias1) were detected by quantitative real‐time polymerase chain reaction. Cell viability, amylase activity, apoptosis, and inflammatory response were detected by 3‐(4,5‐Dimethylthiazol‐2‐yl)‐2,5‐diphenyltetrazolium bromide, Amylase Assay Kit, flow cytometry, and enzyme‐linked immunosorbent assays. The protein level was quantified by western blot analysis. The target interaction between miR‐92a‐3p and mmu_circ_0000037 or Pias1 were predicted by StarbaseV3.0 and validated by dual‐luciferase reporter assay and RNA immunoprecipitation assay. Results Mmu_circ_0000037 and Pias1 levels were decreased, whereas miR‐92a‐3p expression was elevated in caerulein‐induced MPC‐83 cells. Overexpression of mmu_circ_0000037 protected MPC‐83 cells from caerulein‐induced the decrease of cell viability, as well as the promotion of amylase activity, apoptosis and inflammation. MiR‐92a‐3p was targeted by mmu_circ_0000037, and miR‐92a‐3p overexpression rescued the effect of mmu_circ_0000037 on caerulein‐induced MPC‐83 cell injury. Pias1 was confirmed as a target of miR‐92a‐3p and mmu_circ_0000037 regulated the expression of Pias1 by sponging miR‐92a‐3p. Conclusion Mmu_circ_0000037 relieves caerulein‐induced inflammatory injury in MPC‐83 cells by targeting miR‐92a‐3p/Pias1 axis, providing a theoretical basis for the treatment of AP.


| INTRODUCTION
Acute pancreatitis (AP) is a serious inflammatory disease with rising incidence worldwide. 1 Cell death and inflammatory response are the classic pathologic features of AP. 2 With efforts, great progress have been gained on understanding the pathogenesis and management of AP. 3 However, the effective strategies remain limited for treatment of AP. Consequently, it is necessary to ascertain promising target for improving outcomes of patients with AP.
Circular RNAs (circRNAs) have been identified as vital elements in gene regulation under multiple physiological and pathological conditions. 4 CircRNAs are formed by reverse splicing of RNA and have a covalent closed-loop structure with high stability. 5 There is emerging evidence that abnormal expression of circRNAs participate in the regulation of inflammation infiltration and multiple diseases complication, including AP. [6][7][8] Thus, circRNAs have been deemed as perfect biomarkers for multiple diseases, including diagnosis, prognosis, and monitoring of treatment responses. For example, circRNA zinc finger protein 644 regulated the pathogenesis of AP by sponging miR-21-3p. 9 In addition, next-generation RNA sequencing identified several differentially expressed circRNAs and mmu_circ_0000037 was found to be significantly downregulated in the pancreatic tissues of three severe AP (SAP) mice. 9 However, the function and regulatory mechanism underlying mmu_circ_0000037 have not been clarified.
On the basis of competing endogenous RNA (ceRNA) theory, circRNAs can function as microRNA (miRNA) sponge to regulate the stability and translation of target messenger RNAs (mRNAs). 10 miRNAs are a form of noncoding RNAs that perform impressive roles in diseases by acting as potential targets of diagnosis and treatment. 11 Moreover, many miRNAs have been regarded as key diagnostic and prognostic targets for AP. 12 For example, increased miR-551b-5p had been reported to predict poor outcome of AP patients and was associated with inflammatory response. 13 Moreover, miR-148a could suppress inflammatory response and autophagy in caerulein-induced AP model. 14 High expression of miR-92a-3p was detected in AP rats and miR-92a-3p knockdown relieved AP pathological course. 15 In addition, miRNA microarray analysis revealed the overexpression of miR-92a-3p in taurolithocholic acid 3-sulfate-treated AR42J cells. 16 Zhang et al. 16 also indicated that miR-92a-3p could inversely affect the activation of trypsinogen to mediate the progression of AP. However, the regulatory mechanism of miR-92a-3p in AP still need further investigation.
Protein inhibitor of activated STAT1 (Pias1) is a wellstudied E3 small ubiquitin-like modifier ligases, which interacts with activated signal transducers and activators of transcription 1 (STAT1) to suppress its binding to DNA. 17 Pias1 was reported to manage multiple cellular processes, including inflammation. 18 In addition, silencing of Pias1 promoted inflammatory response and cell damage in caerulein-stimulated pancreatic acinar cells. 19 Chen et al. 20 suggested that Pias1 negatively regulated STAT1 expression, thereby remitting the severity of SAP. Thus, Pias1 may be a promising target for curing AP, whereas the function and mechanism of Pias1 remains need in depth exploration.
In the current research, an in vitro AP model was established by stimulating MPC-83 cells with caerulein. The expression levels of mmu_circ_0000037 in caerulein-induced MPC-83 cells were detected. Furthermore, functional experiments were performed to determine the function and regulatory mechanism of mmu_circ_0000037. This study aimed to seek the possible therapeutic biomarkers and provide theoretical support for AP treatment.

| Cell culture and treatment
The mouse pancreatic acinar cell line MPC-83 was procured from BeNa Culture Collection and maintained in RPMI-1640 medium (Thermo Fisher Scientific) containing 10% fetal bovine serum at 37°C with 5% CO 2 .
For establishment of AP cellular model, MPC-83 cells were exposed to 10 nmol/L caerulein (Sigma) for 8 h and then the cells were collected for RNA or western blot analysis. 21 For time-dependent expression analysis, MPC-83 cells were incubated with 10 nmol/L caerulein for 0, 4, 6, 8, and 10 h.

| Cell transfection
pCD-ciR overexpression vector containing mmu_circ_ 0000037 and empty vector (pCD-ciR), miR-92a-3p mimic and inhibitor, small interfering RNA (siRNA) against Pias1 and their negative controls (NC mimic, NC inhibitor, and si-NC) were procured from Genepharma. For cell transfection, MPC-83 cells were plated into the six-well plates until the cell density reached 60% and then the oligonucleotides were introduced into MPC-83 cells using Lipofectamine TM 2000 (Thermo Fisher Scientific).

| Enzymatic method measurement of amylase
MPC-83 cells (1 × 10 7 ) were harvested and resuspended in Amylase Assay Buffer, and then centrifuged at 1500g for 5 min at 4°C to collect supernatant. Cell samples were diluted to the appropriate concentration and then incubated with corresponding reagents in Amylase Assay Kit (Abcam). Then Amylase activity at 405 nm was assessed by a microplate reader (Bio-Rad).

| Flow cytometry
MPC-83 cells (5 × 10 5 /well) were cultured in 24-well plates in triplicates and stimulated with 10 nmol/L caerulein for 8 h. After washing with phosphate-buffered saline for three times, cells were resuspended and stained with Annexin V-fluorescein isothiocyanate and propidium iodide (Beyotime) for 15 min. The cells were then estimated by a flow cytometer (BD) and apoptosis rate was defined as the sum of the percentage of the early and late apoptosis.
2.7 | Quantitative real-time polymerase chain reaction (RT-qPCR) RNA was isolated using Trizol reagent (Thermo Fisher Scientific). Then, 1 μg RNA was used to synthesize complementary DNA (cDNA) using All-in-One TM miR-NA First-Strand cDNA Synthesis Kit (FulenGen) or Universal cDNA Synthesis Kit (Roche). The RT-qPCR assay was conducted with SYBR mix (TaKaRa) and specific primers using Bio-Rad CFX96 Real-time PCR Systems (Bio-Rad). The primer sequences were presented in Table 1. Relative expression levels of RNAs and mRNA were determined by 2 −ΔΔCt approach. 22 Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and U6 were used as endogenous controls.

| RNase R treatment and Actinomycin D (Act D) digestion assay
For RNase R treatment, total RNA (2 μg) was nurtured with or without RNase R (3 U/μg; Epicentre Technologies) at 37°C for 30 min, then the enrichment of mmu_circ_0000037 and Mfsd6 was assessed by RT-qPCR.
For Act D treatment, MPC-83 cells were incubated with 1 μM Act D for 0, 6, 12, and 24 h, respectively. Then, RNA was isolated at each time point and the expression of mmu_circ_0000037 and Mfsd6 was detected by RT-qPCR assay.

| RNA immunoprecipitation (RIP) assay
RIP assay was carried out to inspect the association between miR-92a-3p and mmu_circ_0000037 using the Magna RNA immunoprecipitation kit (Millipore). In brief, MPC-83 cells introduced with miR-92a-3p mimic or miR-NC were lysed in RIP lysis buffer. Then, cell lysates were cultured with anti-Argonaute 2 antibody (Ago2; Millipore) or immunoglobulin G (Millipore) and adsorbed on the magnetic beads. The mixture was incubated with protease K to isolate immunoprecipitated RNA. The enrichment of mmu_circ_0000037 in each group was measured by RT-qPCR.

| Statistical analysis
Each experiment was repeated three times. Data of three replications were displayed as mean ± SD. Statistical analysis was managed by GraphPad Prism 7 software (GraphPad) and analyzed by Student's t test or one-way analysis of variance followed by Tukey's test. p < .05 was deemed as significant difference.

| Mmu_circ_0000037 is downregulated in caerulein-induced MPC-83 cells
According to the results of circRNA sequencing analysis, mmu_circ_0000037 is downregulated in the SAP mice compared with healthy controls. 9 Nevertheless, the function and mechanism of mmu_circ_0000037 in AP has not been discussed. Mmu_circ_0000037 is located at chr1:52765007-52766599, which derived from the Mfsd6 gene ( Figure 1A). To verify the circular structure of mmu_circ_0000037, RNase R and Act D digestion assays were performed. As shown in  Figure 1B, RNase R treatment decreased Mfsd6 linear mRNA levels, whereas mmu_circ_0000037 was not digested. When cells were treated with Act D, the expression of mmu_circ_0000037 was not significantly decreased, while the expression of Mfsd6 was obviously decreased ( Figure 1C). These results confirmed the circular structure of mmu_circ_0000037. To investigate the influence of mmu_circ_0000037 in AP, MPC-83 cells were stimulated with 10 nmol/L caerulein for 0, 4, 6, 8, or 10 h to construct an in vitro cellular model for AP. The results indicated that mmu_circ_0000037 was significantly downregulated at 6, 8, or 10 h in MPC-83 cells during caerulein treatment ( Figure 1D). According to this results, MPC-83 cells stimulated with 10 nmol/L caerulein for 8 h were selected for subsequent study. Besides, we compared the difference of mmu_circ_0000037 expression between the caerulein treatment group and the nontreated (control) group, and the results showed that mmu_circ_0000037 was significantly reduced in caerulein-treated MPC-83 cells ( Figure 1E). These data indicated that mmu_circ_0000037 expression might be associated with AP progression.

| Overexpression of mmu_circ_0000037 mitigates caeruleininduced cell damage in MPC-83 cells
To scrutinize the influence of mmu_circ_0000037 in AP, we overexpressed mmu_circ_0000037 in MPC-83 cells and stimulated the cells with caerulein. RT-qPCR assay indicated that mmu_circ_0000037 was nearly 40-fold elevated by mmu_circ_0000037 overexpression vector in contrast to pCD-ciR group (Figure 2A). Caerulein treatment significantly decreased the expression of mmu_circ_0000037, whereas this effect was partly overturned by overexpression of mmu_circ_0000037 ( Figure 2B). Functional assay revealed that caerulein treatment suppressed cell viability ( Figure 2C), but elevated the activity of amylase ( Figure 2D) and apoptosis rate ( Figure 2E). However, these effects were partly reversed by overexpression of mmu_circ_0000037 in caerulein-induced MPC-83 cells ( Figure 2C-E).
Western blot analysis revealed that mmu_circ_0000037 overexpression partly overturned caerulein-induced

| Mmu_circ_0000037 acts as a sponge for miR-92a-3p
As circRNA regulated cell function by acting as miRNA sponges, we further forecasted the probable target miRNA of mmu_circ_0000037 by StarbaseV3.0. In addition, miR-92a-3p was predicted to be a target of mmu_circ_0000037. The complementary binding sites between mmu_circ_0000037 and miR-92a-3p are exhibited in Figure 3A. The expression of miR-92a-3p was  Figure 3B). In contrast with the NC mimic group, transfection of miR-92a-3p considerably decreased the luciferase activity of mmu_circ_0000037-WT group, whereas it had little effect on the luciferase activity of mmu_circ_0000037-MUT group ( Figure 3C). Ago2-RIP assay indicated that miR-92a-3p mimic increased the enrichment of mmu_circ_0000037 in Ago2-immunoprecipitate group ( Figure 3D). The expression of miR-92a-3p was elevated in caerulein-stimulated MPC-83 cells ( Figure 3E), but it was downregulated by mmu_circ_0000037 overexpression ( Figure 3F). Moreover, miR-92a-3p mimic partly overturned the suppression effect of mmu_circ_0000037 overexpression vector on miR-92a-3p expression ( Figure 3G). Altogether, mmu_circ_0000037 acts as miR-92a-3p sponge to negatively regulate miR-92a-3p expression.

| Pias1 is a target of miR-92a-3p in MPC-83 cells
We then predicted the potential targets of miR-92a-3p by using StarbaseV3.0 database and the complementary miR-92a-3p-binding fragments were found in the 3′UTR of Pias1 sequence. Then, Pias1 sequence harbored the WT-or MUT-type miR-92a-3p-binding motifs were inserted into the dual-luciferase reporter vectors to confirm the interaction between them ( Figure 5A). As presented in Figure 5B, high expression of miR-92a-3p diminished the luciferase activity of Pias1 WT-3′UTR group in contrast with transfection of NC mimic, whereas had no significant influence on the luciferase activity of Pias1 MUT-3′UTR group. Besides, we further investigated Pias1 expression in caerulein-induced MPC-83 cells. As shown in Figure 5C,D, Pias1 mRNA and protein levels were decreased with time of caerulein treatment. Additionally, the mRNA and protein levels of Pias1 in MPC-83 cells was remarkably lowered by miR-92a-3p overexpression ( Figures 5E,F). After confirmation that miR-92a-3p inhibitor indeed reduced miR-92a-3p expression ( Figure 5G), we detected Pias1 expression and confirmed that the mRNA and protein levels of Pias1 were elevated by miR-92a-3p inhibitor ( Figures 5H,I). Meanwhile, mmu_circ_0000037 overexpression significantly elevated Pias1 at mRNA and protein levels, whereas miR-92a-3p partly overturned this effect in caerulein-induced MPC-83 cells ( Figure 5J,K). The performance maintained that mmu_circ_0000037 upregulated Pias1 expression by absorbing miR-92a-3p. To determine if miR-92a-3p acted as a sponge of Pias1, siRNA against Pias1 was constructed and functional rescue experiment was conducted. As displayed in Figure 5L,M, Pias1 levels were decreased more than half in cells with si-Pias1 transfection, suggesting a successful knockdown of Pias1. In addition, Pias1 mRNA and protein levels in caerulein-induced MPC-83 cells were elevated by miR-92a-3p downregulation, while these effects were partly overturned by Pias1 knockdown (Figure 5N,O). Taken together, mmu_circ_0000037 modulated Pias1 expression via performing as a ceRNA of miR-92a-3p.

| DISCUSSION
Due to the limited understanding of AP pathophysiology, the treatment of this disease remains not specific. Hence, it is vitally important to investigate the mechanism underlying AP to find novel therapeutic strategy. CircRNAs have reported to perform important roles in many cell biology processes, including cell proliferation, apoptosis, and inflammation. 23,24 In current research, MPC-83 cells were exposed to caerulein to establish the in vitro cell model for AP. The expression levels of mmu_circ_0000037 and Pias1 were diminished, whereas miR-92a-3p was elevated in caerulein-stimulated MPC-83 cells. Besides, gain-or loss-of-function assays were inhibitor, miR-92a-3p inhibitor + si-NC, or miR-92a-3p inhibitor + si-Pias1. *p < .05, **p < .01, ***p < .001.
carried out to verify the function of these RNAs and gene in caerulein-induced MPC-83 cells. In addition, the modulation network among mmu_circ_0000037, miR-92a-3p, and Pias1 was constructed by functional recovery experiments. Our study provided clues to reveal the pathogenesis of AP. Recent research disclosed that severe pathological changes in pancreatic acinar cells occupy a prominent place in AP occurrence and development. 25 Apoptosis is a common programmed cell death, which has a significant influence during the course of AP. 26,27 Bcl-2 family controls the intrinsic apoptotic pathway, 28 in which Bcl-2 protein is associated with antiapoptotic response and Bax mediates proapoptotic process by leading to irreparable damage to mitochondria. 29,30 Moreover, caspase 3 is also a main mediator of apoptosis and its activation (c-caspase 3) is essential for apoptotic progression. 31 Inflammatory response is another feature of AP, 32 and some inflammatory cytokines such as TNF-α, IL-6, and IL-1β are vital to inflammatory diseases. 33 In addition, the increased amylase and lipase activity are another leading course of AP. 34 Functional experiments revealed that stimulation of caerulein induced MPC-83 cell apoptosis, increased amylase activity, and promoted the release of inflammatory factors, suggesting the successful establishment of AP model in vitro.
Mmu_circ_0000037 is a newly discovered circRNA, which was reported to be lower expressed in the pancreatic tissues from the mice with SAP. 9 In accordance with previous research, our data showed that mmu_circ_0000037 was lower expressed in caeruleininduced MPC-83 cells. Besides, we confirmed that mmu_circ_0000037 overexpression significantly relieved caerulein-stimulated cell injury by regulating cell viability, amylase activity, apoptosis, and inflammation. The above data showed that mmu_circ_0000037 could decrease amylase activity, apoptosis, and inflammation of caerulein-stimulated MPC-83 cells, which might slow down AP progression.
For investigation of miR-92a-3p promising target in AP progression, we performed bioinformatics analysis and confirmed that Pias1 contained the complementary binding sites of miR-92a-3p in 3′UTR sequence. Pias1, a transcriptional coregulator that selectively mediates STAT1-dependent gene expression, exhibits a prominent preference for inflammatory response. 17 Previous study showed that downregulation of Pias1 could enhance the expression of inflammatory mediators and induce cell injury in AP. 19,20 These data provided evidence that Pias1 might be a potential therapeutic target for AP. In our research, we found that miR-92a-3p could target Pias1 and Pias1 was downregulated in caerulein-stimulated MPC-83 cells. In addition, rescue experiments revealed that miR-92a-3p inhibitor inhibited caerulein-induced AP cell injury by targeting Pias1 in vitro. Importantly, we pointed out that mmu_circ_0000037 could upregulate Pias1 expression by sponging miR-92a-3p, which improved the conclusion that mmu_circ_0000037 sponged miR-92a-3p to regulate Pias1, thereby mediating the progression of AP.
In conclusion, our research reveals for the first time the expression and function of mmu_circ_0000037 in caeruleininduced MPC-83 cells. Our data showed that mmu_-circ_0000037 participated in the pathogenesis of AP, which could inhibit the apoptosis and inflammatory response of caerulein-stimulated MPC-83 cells by sponging miR-92a-3p to regulate Pias1 expression (Figure 7). Our results enhance the understanding of AP pathogenesis and provide the potential therapeutic targets for AP.

AUTHOR CONTRIBUTIONS
Hua Chen and Jun Tu performed the research. Lei He, Ning Gao, and Weiqiang Yang designed the research study. Lei He, Ning Gao, and Weiqiang Yang contributed essential reagents or tools. Hua Chen and Jun Tu analyzed the data. Hua Chen and Jun Tu wrote the paper.