MALAT1 shuttled by extracellular vesicles promotes M1 polarization of macrophages to induce acute pancreatitis via miR‐181a‐5p/HMGB1 axis

Abstract Acute pancreatitis (AP) is a serious condition carrying a mortality of 25–40%. Extracellular vesicles (EVs) have reported to exert potential functions in cell‐to‐cell communication in diseases such as pancreatitis. Thus, we aimed at investigating the mechanisms by which EV‐encapsulated metastasis‐associated lung adenocarcinoma transcript‐1 (MALAT1) might mediate the M1 polarization of macrophages in AP. Expression patterns of MALAT1, microRNA‐181a‐5p (miR‐181a‐5p) and high‐mobility group box 1 protein (HMGB1) in serum of AP patients were determined. EVs were isolated from serum and pancreatic cells. The binding affinity among miR‐181a‐5p, MALAT1 and HMGB1 was identified. AP cells were co‐cultured with EVs from caerulein‐treated MPC‐83 cells to determine the levels of M1/2 polarization markers and TLR4, NF‐κB and IKBa. Finally, AP mouse models were established to study the effects of EV‐encapsulated MALAT1 on the M1 polarization of macrophages in AP in vivo. MALAT1 was transferred into MPC‐83 cells via EVs, which promoted M1 polarization of macrophages in AP. MALAT1 competitively bound to miR‐181a‐5p, which targeted HMGB1. Moreover, MALAT1 activated the TLR4 signalling pathway by regulating HMGB1. EV‐encapsulated MALAT1 competitively bound to miR‐181a‐5p to upregulate the levels of IL‐6 and TNF‐α by regulating HMGB1 via activation of the TLR4 signalling pathway, thereby inducing M1 polarization of macrophages in AP. In vivo experimental results also confirmed that MALAT1 shuttled by EVs promoted M1 polarization of macrophages in AP via the miR‐181a‐5p/HMGB1/TLR4 axis. Overall, EV‐loaded MALAT1 facilitated M1 polarization of macrophages in AP via miR‐181a‐5p/HMGB1/TLR4, highlighting a potential target for treating AP.


| INTRODUC TI ON
Acute pancreatitis (AP) is a type of pancreatic inflammatory disease with sudden onset, and its incidence varies in different regions 1 with more females suffered from it than males. 2 The fatality rate of AP is closely related to the frequency and pathogenesis of the disease, and the fatality rate of first-episode AP is 14 times that of recurrent pancreatitis. 3 Gallstones and alcohol abuse are the key factors for AP, and the fatality rate of AP with biliary stones is twice higher than that of AP with alcoholism. 4 In addition, genetic factors, the use of certain drugs and damage to the pancreas are also causes of AP. 5 Interestingly, pancreatic acinar cells after injury may secrete chemical factors, cytokines and various cell adhesion factors (tumour necrosis factor [TNF], interleukin [IL]-6, IL-18, IL-1β, etc.) to induce the infiltration of immune cells, which contribute to the progression of AP. 6 Despite achievements in care, imaging and interventional approaches, AP is still related to significant morbidity and mortality. 7 Extracellular vesicles (EVs) are a heterogeneous family of membrane-limited vesicles and can be internalized via endocytosis or membrane fusion, releasing their contents into 'recipient' cells. 8 These EVs contain many proteins, sugars, lipids and multiple kinds of genetic materials, such as DNA, mRNA and non-coding (nc) RNAs. 9 Long noncoding RNA (lncRNA) metastasis-associated lung adenocarcinoma transcript-1 (MALAT1) can act as competitive endogenous RNA (ceRNA) to regulate its downstream target genes via interaction of microRNA response elements (MREs) and microRNAs (miRNAs). 10 Accumulating evidence proves that knockdown of MALAT1 promotes differentiation of macrophages into M1 subtypes in hepatocellular carcinoma. 11 Additionally, MALAT1 shuttled by exosomes from oxidized low-density lipoprotein-treated endothelial cells can promote the M2 polarization of macrophages in cardiovascular disease. 12 Depleted MALAT1 has been demonstrated to promote the M1 polarization of macrophages in inflammation of central nervous system. 13 What's more, downregulation of MALAT1 could inhibit LPS-induced activation of M1 macrophages and activates IL-4 induced M2 differentiation. 14 Recent study reveals that MALAT1 promotes AP through the miRNA-194/YAP1 axis. 15 Importantly, activated macrophages in AP can differentiate into proinflammatory M1 subtypes and secrete some cytokines and regulatory factors to induce local inflammation of the pancreas, systemic inflammatory responses or damage to functions of multiple tissues. 16,17 However, the specific mechanisms of MALAT1-mediated M1 polarization of macrophages remain unclear. This study attempts to explore how MALAT1 influences the occurrence of AP, and tries to reveal whether MALAT1 affects AP by regulating M1 polarization of macrophages.

| Ethic statement
All participants signed informed consent, and this study was performed with the approval of the Ethics Committee of the Second Affiliated Hospital of Xi'an Jiaotong University. This study was carried out in strict accordance with the Declaration of Helsinki. All animal experiments were approved by the laboratory animal care committee of Xi'an Jiaotong University. All experimental procedures that involved animals were approved according to the guidelines of the Care and Use of Laboratory Animals by the National Institute of Health, China.
The 'limma' package of R language was utilized for differential analysis with normal samples as a control. The differentially expressed genes in AP were obtained with |logFC| >1, p value <0.05 as the threshold. The exoRBase database (http://www.exorb ase. org/exoRB ase/toIndex) was employed to retrieve the expression of MALAT1 in EVs. The starBase database (starbase.sysu.edu.cn/) was used to search the binding sites of MALAT1 and miR-181a-5p in humans and mice. Next, the starBase database and TargetScan database (http://www.targe tscan.org/vert_71/) were utilized to predict the target genes of miR-181a-5p in humans and mice, and the mirDIP database (http://ophid.utoro nto.ca/mirDI P/index.jsp#r) was used to further predict the target genes of miR-181a-5p in humans. The 'clusterprofiler' package of the R language was employed to perform KEGG pathway enrichment analysis on candidate target genes. The target binding sites of miR-181a-5p and high-mobility group box 1 protein (HMGB1) in humans and mice were attained through the TargetScan database.

| EV isolation
Serum samples (3 ml) were centrifuged at 2000 g (46962, Thermo Fisher) for 10 min and at 10,000 g for 30 min at 4℃. The obtained supernatant was resuspended in 8 ml phosphate-buffered saline (PBS) and ultracentrifuged at 120,000 g for 70 min in a 30% sucrose buffer. The sucrose fraction was recovered, washed with PBS, filtered through a 0.22µm filter and ultracentrifuged again at 120,000 g for 70 min. Then, EV precipitate was resuspended by an appropriate amount of PBS for later use or frozen at −80℃.

| Nanoparticle tracking analysis (NTA)
NTA was conducted using NanoSight analyser (NanoSight LM10-HS; Malvern, Worcestershire, UK). Briefly, EVs (10-20 μl) were diluted with PBS to the final volume of 1 ml and 1 ml EVs were injected into the sample pool using a 1-ml syringe. The focal length of the main engine knob was adjusted to see clear 'white bright spot', and the gain was adjusted for recording, 30 s per time. A small amount of sample was slowly injected into the sample pool using a syringe after 30 s of single recording, and the detection was repeated three times for each sample. The track of each EV in the screen was analysed and was automatically converted into the diameter and concentration of EVs according to the Brownian motion principle. The original concentration could be obtained via dilution ratio through conversion.

| EV uptake assay
A PKH67 green fluorescence kit (MINI67-1KT, Sigma-Aldrich) was used to label purified EVs from human serum. EVs were resuspended in 1 ml Diluent C solution, and then, 4 μl PKH67 ethanol dye solution was added into Diluent C solution to prepare 4 × 10 −6 M dye solution. Then, 1 ml EV suspension was mixed with PKH26 for 5 min and cultured with 2 ml of 1% bovine serum albumin (BSA) for 1 min to terminate staining. The labelled EVs were ultracentrifuged at 100,000 × g for 70 min at 4℃ and washed using PBS. EVs were ultracentrifugated again and resuspended in 50 μl PBS. PKH67-labelled EVs were incubated with THP-1 cells for 12 h. The cells were fixed with 4% paraformaldehyde and washed with PBS. The nuclei were stained with 4',6-diamidino-2-phenylindole (DAPI) (Sigma-Aldrich).
The uptake of labelled EVs was determined using a confocal microscope (Leica, Oskar Barnack, Germany).

| Extraction of MPC-83 cell-derived EVs
The MPC-83 cells were passaged after recovery. When reaching 70% confluence, MPC-83 cells were washed thrice with 0.01 mol/L PBS and cultured for 48 h in FBS medium without EVs. The supernatant was collected and centrifuged at 300 g and 2000 g for 10 and 30 min, respectively, to remove cells. Next, the supernatant was centrifuged at 10,000 g for 30 min to remove cell debris. The suspension was resuspended in sucrose buffer, centrifugated at 100,000 g (46962, Thermo Fisher) for 70 min, washed with PBS and centrifugated at 100,000 g for 70 min again. Then, EV precipitate was resuspended by an appropriate amount of PBS for later use or frozen at −80℃.  were transfected into RAW264.7 cells.

| RT-qPCR
RNA was extracted using TRIzol reagent (Invitrogen). LncRNA and mRNA were reversely transcribed into complementary DNA (cDNA) using a PrimeScript™ RT Master Mix Kit (Takara Bio Inc., Otsu, Shiga, Japan). miRNA was reversely transcribed into cDNA with PolyA tailing using PolyA tailing detection kit (B532451, Shanghai Sangon Biotechnology Co. Ltd., Shanghai, China, containing universal PCR reverse primers). The primers of lncRNA, mRNA and miRNA were purchased from Guangzhou RiboBio. U6 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were used as the internal reference for miRNA and lncRNA/mRNA, respectively. The 2 −ΔΔCt method was used to quantify the relative expression of target genes. The primers used are listed in Table S2.

| Western blot analysis
Cell was washed three times with PBS and lysed with radioimmunoprecipitation assay lysis +protease inhibitor (Roche) on ice for 30 min, followed by centrifugation at 15,000 g at 4℃ for 10 min.

| Northern blot analysis
A total of 30 μg of the indicated RNA was subjected to formaldehyde gel electrophoresis and then transferred to a Biodyne nylon membrane (Pall, NY). Biotin (Roche, Mannheim, Germany)-labelled MALAT1 cRNA probe was prepared using in vitro transcription from pSPT19-MALAT1 with the probe sequence of ACGAATTCAGGGTGAGG AAGTAAAAACAGGTCATCTATTCACAAAACTGA. Radio-labelled DNA oligonucleotides were regarded as detection probes for Northern blot detection of miRNA. For molecular markers, the Decade Marker System (Ambion) was used, and the probe sequence was ACTCACCGACAGCGTTGAATGTT. After pre-hybridization in ULTRAhyb buffer (Ambion, Grand Island, NY) for 60 min, the membrane was hybridized, washed and tested as described above.

| Animal experiments
A total of 32 healthy male C57BL/6J mice (6-8 weeks) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). C57BL/6 mice were kept in a specific pathogenfree animal facility with humidity of 60-65%, temperature of 22-25℃ and given free access to food and water under a 12-h light and dark cycle. Mice were acclimated 1 week before the experiment, and the health of the mice was observed before the experiment.
To establish AP mouse models, mice were intraperitoneally injected with 12 times of caerulein (50 μg/kg), once every one hour.
Mice were divided into 4 groups, control (blank control), AP (AP mice injected with saline), AP +sh-NC (AP mice injected with sh-NC) and AP +sh-MALAT1 (AP mice injected with sh-MALAT1). In vivo shRNA was purchased from Guangzhou RiboBio Co., Ltd. In order to silence the expression of MALAT1 in mouse pancreas, in vivo shRNA-MALAT1 (15 nmol/20 g) was injected through the common bile duct before intraperitoneal administration.

| Enzyme-linked immunosorbent assay (ELISA)
The cell supernatant or serum of AP mice was collected to detect the TNFα and IL-6 according to the instructions of TNF alpha Mouse ProQuantum Immunoassay Kit (A43658, Thermo Fisher Scientific) and IL-6 Mouse ProQuantum Immunoassay Kit (A43656, Thermo Fisher Scientific) though microplate reader (Invitrogen; Thermo Fisher Scientific, Inc.).

| Immunohistochemical staining
One hour after the last injections of caerulein, the mice were sac-
Next, the sections were stained with eosin (#AR-0731, Beijing Dingguo Changsheng Biotechnology), followed by gradient dehydration and sealing. The images were observed and photographed under a microscope (Olympus Optical Co., Ltd, Tokyo, Japan). The extent of cell injury and necrosis was quantified through computeraided morphological examination by experienced morphologists.

| Detection of serum amylase and lipase
Serum amylase and lipase were the most common serum markers in AP, which may reflect the severity of AP. After extraction of blood, the lipase detection kit (#A054-1-1, NanJing JianCheng Bioengineering Institute, Nanjing, China) and the amylase detection kit (#BC0615, Beijing Solarbio Science & Technology Co. Ltd., Beijing, China) were used to detect the activity of lipase and amylase in the serum of mice though microplate reader (Invitrogen; Thermo Fisher Scientific, Inc.).

| Flow cytometry
Peritoneal lavage was used to isolate peritoneal macrophages, and flow cytometry was used to analyse the phenotype of peritoneal macrophages. Briefly, peritoneal macrophages were washed with staining buffer (1% BSA in PBS containing 0.01% NaN 3 , Thermo The results were analysed with the BD FACS DIVA software (BD Biosciences).

| Immunofluorescence staining
The distribution of the M1 phenotype in the pancreas was assayed by immunofluorescence staining. Pancreatic tissues were cut into sections which were dewaxed. CD68 (ab955, Abcam) expression was detected by anti-mouse Alexa-Flour647 (P0191, Biotechnology Co., Shanghai, China), and CD86 (ab119857, Abcam) expression was detected by anti-rabbit FITC (Beyotime). Then, the samples were stained with DAPI to visualize the nuclei. The distribution of M1 macrophages in the pancreas was examined by a fluorescence microscope (Olympus Optical Co., Ltd, Tokyo, Japan).

| Statistical analysis
Data analysis was performed using the SPSS 21.0 software (IBM, Armonk, NY, USA). All measurement data are presented as mean ±standard deviation. Unpaired t test was used for comparisons of independent samples between two groups. For multiple group comparisons, one-way analysis of variance (ANOVA) and Tukey's post hoc tests were used. A value of p < 0.05 was considered significant.

| MALAT1/miR-181a-5p/HMGB axis is involved in the occurrence and development of AP
Differential analysis of AP-related lncRNA microarray data set GSE12 1038 yielded 1549 differentially expressed genes ( Figure 1A).
Through further gene type annotation, we found that there were 3 lncRNAs (Table S3); among them, we found highly expressed MALAT1 in serum-derived EVs through exoRBase ( Figure 1B). At the same time, MALAT1 was found highly expressed in AP samples in the GSE12 1038. Further prediction revealed that MALAT1 was capable of binding to miR-181a-5p in humans and mouse ( Figure 1C).
The starBase database was used to predict target genes of miR-181a-5p in humans and mouse, and 410 candidate target genes were obtained after intersection of prediction results ( Figure 1D).
Through KEGG pathway enrichment analysis ( Figure 1E), we found that 410 candidate target genes were mainly enriched in PI3K-AKT signalling pathway and animal autophagy-related pathways.
Existing studies indicate that regulation of autophagy can affect macrophage polarization. 18,19 In the autophagy pathway, there was a candidate target gene HMGB1. Accumulating evidence demonstrates that HMGB1 regulates TLR4 expression to facilitate AP occurrence and development, and TLR4 promotes M1 polarization of macrophages. [20][21][22] In humans and mouse, there were binding sites among MALAT1, miR-181a-5p and HMGB1 ( Figure 1F). Thus, it can be concluded that the involvement of the MALAT1/miR-181a-5p/ HMGB axis affects the occurrence and development of AP.

| MALAT1 is upregulated in pancreatic cellderived EVs and serum EVs in AP patients
MALAT1 is reported to induce AP progression. 15 In addition, EVs can aggravate the inflammatory responses of AP mice and cells. 23 To

| EVs containing MALAT1 promote M1 polarization of macrophages
It has been found that M1 polarization of macrophages aggravates AP. 24 Our previous experimental results also revealed that MALAT1 exerted an important effect on AP. Therefore, we tried to analyse

| MALAT1 upregulates HMGB1 expression by competitively binding to miR-181a-5p
The results of bioinformatic analysis showed that MALAT1 could bind to miR-181a-5p ( Figure 4A). As a negative regulator of AP,  miR-181a-5p could repair AP damage. 25 AGO2 pull-down assay presented ( Figure 4B,C) that the enrichment degree of MALAT1 in complex pulled down by AGO2 was higher in RAW264.7 cells treated with miR-181a-5p mimic than in cells without treatment. Dualluciferase reporter gene assay exhibited that luciferase activity of MALAT1-WT was inhibited by miR-181a-5p mimic (p < 0.05), while no evident difference was found in MALAT1-MUT (p > 0.05) ( Figure 4D).

| Silencing of MALAT1 attenuates pancreatic tissue injury in AP mice
AP mouse models were established to explore the effects of MALAT1 on the occurrence and development of AP in vivo. It was found that levels of serum amylase and lipase expression significantly increased in serum of AP mice, and they decreased in AP mice injected with sh-MALAT1 ( Figure 6A,B). Observation from H&E staining exhibited that pancreatic tissues in AP mice showed mesenchymal congestion, oedema, inflammatory cell infiltration, focal or confluent necrosis and haemorrhage ( Figure 6C).

| DISCUSS ION
In the present study, we found that MALAT1 was highly expressed in the serum, pancreatic tissues and pancreatic cells of patients with AP as well as pancreatic cell-derived EVs, which indicated that   In addition, MALAT1 could regulate the macrophages polarized to M1 phenotype in AP, but the specific molecular mechanism warranted further investigation and discussion.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interests.

F I G U R E 7
The molecular mechanism of MALAT1 shuttled by EVs in affecting M1 polarization of macrophages involved in the occurrence and development of AP via miR-181a-5p/HMGB1 axis

DATA AVA I L A B I L I T Y S TAT E M E N T
The data sets generated and/or analysed during the current study are available from the corresponding author on reasonable request.