METTL3-mediated macrophage exosomal NEAT1 contributes to hepatic fibrosis progression through Sp1/TGF-β1/Smad signaling pathway

Hepatic fibrosis (HF) is caused by chronic hepatic injury and is characterized by hepatic stellate cells (HSCs) activation. Studies focusing on the function of exosomes derived from macrophages in HF progression are limited. This study aims to identify the roles of exosomal NEAT1 derived from macrophages on HF and the underlying mechanisms. Our studies showed that METTL3 targeted and enhanced NEAT1 expression in macrophages. Exosomal NEAT1 originating from LPS-treated macrophages promoted HSCs proliferation and migration, and induced the expression of fibrotic proteins including collagen I, α-SMA, and fibronectin. Macrophage exosomal NEAT1 contributed to HSCs activation by sponging miR-342. MiR-342 directly targeted Sp1 and suppressed its downstream TGF-β1/Smad signaling pathway, which eventually led to the inhibition of HSCs activation. Depletion of NEAT1 in the macrophage exosomes inhibited HF progression both in vitro and in vivo. Altogether, our study proved that silence of NEAT1 in the macrophage exosomes exerted protective roles against HF through the miR-342/Sp1/TGF-β1/Smad signaling pathway, suggesting a potential therapeutic target in HF treatment.


INTRODUCTION
Hepatic fibrosis (HF) results from chronic damage to the liver where excessive extracellular matrix accumulates [1]. It was reported that lipopolysaccharide (LPS)-treated THP-1 macrophages promoted hepatic stellate cells (HSCs) proliferation and activation and thus led to fibrosis progression [2]. Nevertheless, the mechanisms underlying macrophages regulate HSCs activation and fibrosis have not been fully understood.
MiRNA-342 (miR-342) was reported to inhibit the progress of renal interstitial fibrosis [11]. MiR-342 was also documented to be a potent suppressor in hepatocellular carcinoma [12]. NEAT1 may direct target miR-342 as predicted by the online prediction tool (Starbase).
However, whether miR-342 is involved in regulating HF remains to be further elucidated. The activation of the Sp1/TGF-β1/Smad pathway was associated with the progress of fibrogenesis of many organs, including atrial fibrosis, liver fibrosis, and endometrial fibrosis [13][14][15]. Our bioinformatics analysis reveals that miR-342 harbors predictive binding sites for Sp1 mRNA. But, the exact role of the miR-342 on the regulation of the Sp1/TGF-β1/Smad signaling pathway in HF is not clear.
In the present study, we aimed to investigate the functional role and underlying mechanism of METTL3-medicated macrophage exosomal NEAT1 in HF. We speculated that NEAT1 in exosomes secreted from LPS-activated macrophages suppressed miR-342 expression, which further activated the Sp1/TGF-β1/Smad signaling pathway and caused activation of HSCs, eventually led to accelerated progress of HF.

NEAT1 expression is increased in exosomes derived from LPStreated THP-1 macrophages
The exosomes from THP-1 macrophages were isolated and characterized using transmission electron microscope (TEM) and nanoparticle tracking analyzing (NTA) analyses (Fig. 1A, B). The exosomes showed typical rounded with an average diameter of 100-150 nm. The expression of specific exosomal markers CD9, CD63, and TSG101 was further quantified (Fig. 1C). As shown in Fig. 1D, the exosomes marked by PKH67 were situated in the cytoplasm of LX-2 cells. LX-2 cells incubated with exosomes from LPS-treated THP-1 macrophages showed a significantly higher NEAT1 expression (Fig. 1E). Taken together, NEAT1 expression was induced in exosomes derived from LPS-treated THP-1 macrophages.
Exosomes extracted from LPS-treated THP-1 macrophages enhance the activation of HSCs As indicated by Fig. 2A, the NEAT1 expression was significantly increased in LX-2 cells that were treated with conditional medium (CM) or exosomes derived from LPS-treated THP-1 macrophages. However, the treatment of GW4869 (an inhibitor of exosome generation) reversed the stimulation effect of CM. Similarly, the cell proliferation ( Fig. 2B), cell cycle transition from G1 phase to S phase (Fig. 2C), and migration ( Fig. 2D, E) of LX-2 cells were enhanced following the treatment of CM or exosomes, whereas the treatment of GW4856 reversed the influences of CM on LX-2 cells (Fig. 2B-E). Treatment of LX-2 cells with CM or exosomes largely increased the levels of collagen I, α-SMA and fibronectin (Fig. 2F). Similarly, Sp1, TGF-β1, phosphorylated-Smad2 and phosphorylated-Smad3 levels were also elevated in LX-2 cells after treatment with CM or exosomes. Nevertheless, GW4869 reversed the effects of CM treatment (Fig. 2F). Altogether, these data suggest that exosomes secreted from LPStreated THP-1 macrophages contributed to the activation of HSCs via Sp1/TGF-β1/Smad signaling pathway.

METTL3 facilitates NEAT1 expression in THP-1 macrophages
The m6A level was significantly upregulated in LPS-activated THP-1 macrophages (Fig. 4A). The mRNA and protein levels of METTL3 were significantly enhanced under this treatment (Fig. 4B, C). Moreover, the transfection efficiency of METTL3 was proved by Fig. 4D, E. METTL3 overexpression significantly enhanced the NEAT1 expression. In contrast, silence of METTL3 led to the downregulation of NEAT1 (Fig. 4F). NEAT1 was enriched by the METTL3 antibody (Fig. 4G). m6A RIP analysis illustrated that the knockdown of METTL3 suppressed the enrichment of NEAT1 with m6A ( Fig. 4H). Additionally, METTL3 overexpression led to a significant increase of luciferase activity in wild type of NEAT1 (NEAT1-WT) reporter, whereas it had no impact in mutant of NEAT1 (NEAT1-MUT) reporter group (Fig. 4I).

METTL3 promotes NEAT1 expression in Kupffer cells and NEAT1 expression is also increased in exosomes derived from LPS-treated Kupffer cells
We then isolated Kupffer cells from mouse liver and followed by overexpressing or silencing METTL3 expression ( Supplementary Fig.  1A, B). METTL3 overexpression significantly increased NEAT1 expression. In contrast, downregulation of METTL3 had the opposite effect on NEAT1 expression ( Supplementary Fig. 1C). NEAT1 was enriched by the METTL3 antibody in Kupffer cells ( Supplementary  Fig. 1D). These results indicated that METTL3 also promoted NEAT1 expression in Kupffer cells.
Exosomes from Kupffer cells were characterized by TEM ( Supplementary Fig. 1E), NTA ( Supplementary Fig. 1F) and western blot ( Supplementary Fig. 1G). The exosomes showed a typical rounded structure with an average diameter of around 100-150 nm and the expression of specific exosomal markers CD9, CD63, and TSG101 were also identified. As indicated by Supplementary Fig. 1H, a significantly increased NEAT1 expression was observed in exosomes isolated from LPS-treated Kupffer cells.   Exosomal NEAT1 from THP-1 macrophages promotes the activation of HSCs. A The NEAT1 expression level was determined by qRT-PCR in exosomes from THP-1 macrophages. B The NEAT1 expression level was determined by qRT-PCR in LX-2 cells. C The proliferation of the LX-2 was examined by the CCK-8 assay. D The cell cycle of the LX-2 was examined by the flow cytometry. E, F The migration of LX-2 was determined by the scratch assay (E) and the transwell assay (F). G The protein levels of collagen I, α-SMA, fibronectin, Sp1, TGF-β1, p-Smad2, Smad2, p-Smad3, and Smad3 were measured by western blot. *p < 0.05; **p < 0.01; ***p < 0.001.
Exosomes extracted from LPS-treated Kupffer cells enhance the activation of primary HSCs As indicated by Supplementary Fig. 2A, the NEAT1 expression was significantly enhanced in primary HSCs treated with exosomes derived from LPS-induced Kupffer cells. Similarly, the cell proliferation ( Supplementary Fig. 2B), and migration ( Supplementary Fig. 2C, D) of primary HSCs were increased following the treatment of exosomes derived from LPS-treated Kupffer cells. Treatment of primary HSCs with exosomes derived from LPS-activated Kupffer cells also increased the expression levels of collagen I, α-SMA, fibronectin, Sp1, TGF-β1, phosphorylated-Smad2 and phosphorylated-Smad3 ( Supplementary  Fig. 2E). These results indicated exosomal NEAT1 extracted from Kupffer cells also enhanced the activation of primary HSCs. We then overexpressed or knocked down NEAT1 expression in LX-2 cells and the success of the transfection was demonstrated in Fig. 5D. NEAT1 overexpression significantly reduced the expression of miR-342, whereas silence of NEAT1 enhanced the miR-342 level (Fig. 5E). The promotion effect of NEAT1 depletion on miR-342 expression in LX-2 cells was reversed after transfection with miR-342 inhibitor (Fig. 5F). The miR-342 expression was also significantly decreased in primary HSCs treated with exosomes derived from LPS-induced Kupffer cells ( Supplementary Fig. 2F).
Exosomal NEAT1 contributes to HF progression in rat The rat in the CCL 4 -induced group displayed elevated liver fibrosis level, as evidenced by accumulated collagen in livers by Masson and Sirus Red stainings (Fig. 10A). Moreover, the rat in the CCL 4 group showed significantly higher levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in serum as well as increased protein level of collagen I and α-SMA (Fig. 10B, C). Importantly, injection of NEAT1-depleted exosomes attenuated the CCL 4 -induced liver fibrosis and mitigated the AST and ALT levels as well as profibrotic makers, including collagen I and α-SMA. Liver fibrosis level in rat was further elevated following the injection of NEAT1 overexpressed exosomes (Fig. 10A-C). Furthermore, the CCL 4 group also showed the remarkably enhanced NEAT1 (Fig. 10D), METTL3, Sp1 (Fig. 10E) levels and activated TGF-β1/Smad signaling pathway (Fig. 10F), but a reduced miR-342 expression (Fig. 10D) in liver tissues. NEAT1-depleted exosomes significantly suppressed the CCL 4induced effects on METTL3, NEAT1, miR-342, and Sp1 expressions and deactivated the TGF-β1/Smad signaling pathway. Nevertheless, treatment with NEAT1 overexpressed exosomes displayed the opposite functions (Fig. 10D-F). In summary, silence of NEAT1 in macrophage exosomes attenuated HF progression induced by CCL 4 in vivo.

DISCUSSION
Previous studies demonstrated the continuous crosstalk between macrophages and HSCs [16]. Here, we for the first time demonstrated macrophage exosomal NEAT1 contributed to HSCs activation by sponging miR-342 and thus induced its downstream Sp1/TGF-β1/ Smad signaling pathway in HF. Thus, NEAT1-depleting macrophage exosomes and miR-342 may potentially be developed as new therapeutic targets for HF treatment.
NEAT1 played a crucial role in mediating inflammatory response in macrophages. For example, Li et al. showed that NEAT1 increased LPS-induced inflammatory response in macrophages [17]. METTL3 was suggested to drive pro-inflammatory M1 macrophage polarization [4]. Though both METTL3 and NEAT1 were involved in mediating inflammation, the relationship between METTL3 and NEAT1 was not documented. To the best of our knowledge, we proved for the first time that METTL3 directly bound to NEAT1. Upon LPS treatment, the METTL3 level was increased in THP-1 macrophages and resulted in enhanced NEAT1 expression. We also firstly proved that NEAT1 was highly expressed in exosomes derived from LPS-induced THP-1 macrophages and primary Kupffer cells, and exosomal NEAT1 promoted HSCs proliferation and migration as well as fibrogenesis of HSCs. Also, consistent with in vitro data, NEAT1 overexpressed macrophage exosomes promoted HF in vivo. To the best of our knowledge, this study first time indicated the impact of macrophage exosomal NEAT1 on liver fibrosis.
We further validated that NEAT1 could directly bind to miR-342 through dual-luciferase assay and RIP assay. MiR-342 was indicated as a tumor suppressor in hepatocellular carcinoma [12,18]. It was also documented by Jiang et al. that miR-342 suppressed renal interstitial fibrosis as evidenced by the decreased fibrosis markers [11]. However, the roles of miR-342 in HF were not reported before.
Here, we showed that knockdown of miR-342 reversed the stimulation effects of NEAT1 overexpressed exosomes on HSCs proliferation, migration, and fibrosis-associated markers. These data further confirmed that exosomal NEAT1 regulated HSCs activation by sponging miR-342. As far as we known, this is the first time that reported the regulatory effects of miR-342 on liver fibrosis.
Qu et al. showed that suppressing TGF-β1-induced Smad2/Smad3 phosphorylation attenuated fibrosis [15]. TGF-β1 can be activated by the transcriptional factor Sp1 [19]. Here, we proved that Sp1 was a direct target of miR-342. In addition, Sp1 overexpression rescued the inhibitory effects of miR-342 overexpression on HSCs proliferation, migration and fibrosis-related proteins. To our knowledge, this study for the first time investigated the interaction of miR-342 and Sp1.
In conclusion, our study showed that exosomal NEAT1 from macrophages promoted HSCs activation and induced the expression of fibrotic markers. Furthermore, we demonstrated the underlying network of NEAT1/miR-342/Sp1 axis via the TGF-β1/Smad signaling pathway, revealing a novel theoretical basis for the application of exosomal NEAT1 in HF treatment. Future studies will focus on determining whether other molecules such as circRNAs, miRNAs, or proteins presented in macrophage-derived exosomes could mediate HSCs activation and HF progression. G The protein levels of collagen I, α-SMA, fibronectin, Sp1, TGF-β1, p-Smad2, Smad2, p-Smad3 and Smad3 were measured by western blot. *p < 0.05; **p < 0.01; ***p < 0.001.

MATERIALS AND METHODS Cell culture
The human monocytic leukemia cell line THP-1 was purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). The human HSCs LX-2 and 293T cells were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). THP-1 cells were maintained in RPMI 1640 medium (Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco) and 1% penicillin-streptomycin (Gibco). THP-1 cells were incubated for 24 h at 37°C with PMA (25 ng/mL, Sigma-Aldrich) to differentiate into macrophages. THP-1-derived macrophages were further treated with 10 ng/mL LPS for 12 h. The LX-2 and 293T cells were incubated in DMEM (Gibco) supplemented with 10% FBS and 1% penicillin-streptomycin. All the cell lines included in this study have been authenticated by STR profiling and tested for mycoplasma contamination.

Isolation of primary Kupffer cells and HSCs
Primary Kupffer cells and HSCs from mouse liver were isolated based on a previously reported protocol [20]. All procedures were approved by the Animal Care and Use Committee of the Second Xiangya Hospital of Central South University (Changsha, Hunan, China). The used 8-week-old male BALB/c mice were obtained from Shanghai SLAC Laboratory Animal Center (Shanghai, China).

Cell treatments
LX-2 cells were seeded into six-well plates. Cells were treated with either 50 µg of exosomes from THP-1 cells diluted in 2 mL DMEM medium or 2 mL conditioned medium (CM) from LPS-activated THP-1 cells or CM from LPSactivated THP-1 cells treated with GW4869 (Sigma-Aldrich), which was well accepted to inhibit exosome generation [21], before performing further experiments. Primary HSCs isolated from mouse liver tissues were also seeded into six-well plates, and treated with 50 µg of exosomes from LPS-activated Kupffer cells. Cells were also transfected with lentivirus pLCDH vectors (GeneCopoeia) harboring NEAT1, shNEAT1, METTL3, or shMETTL3 to overexpress (OE) or silence NEAT1/METTL3 expression. Lentiviral and packaging vectors were co-transfected into 293T cells to generate lentivirus. 48 h after the transfection, the supernatant was collected and concentrated using Lenti-Pac Lentivirus Concentration Solution (GeneCopoeia) according to the manufacturer's instruction. After determining the titer, lentivirus was transduced into target cells at a multiplicity of infection (MOI) of 50 in the presence of polybrene (8 µg/mL, Sigma-Aldrich) overnight. shSp1, pcDNA-3.1-Sp1, miR-342 mimics, miR-342 inhibitor or their related negative controls (GenePharma) were also transfected into cells for 48 h using Lipofectamine 2000 (Invitrogen) without serum following the manufacturer's instruction.

Exosome isolation and characterization
To isolate the exosomes from the cells, total exosome isolation reagent (Cat# 4478359, Thermo Fisher Scientific) was used based on a previously published study [2]. In brief, THP-1 macrophages or Kupffer cells were cultured in culture plates until they reached 70% confluence. Then, cells were washed with phosphate-buffered saline (PBS, Gibco) and further incubated with 10% exosome-depleted FBS (Gibco) for 24 h. Afterward, the cell culture medium was collected and then was centrifuged at 2000 × g for 30 min, and the supernatant was filtered through a 0.22 mm filter (Sigma-Aldrich). The exosome isolation reagent (Invitrogen) was added to the medium and incubated overnight at 4°C, following with centrifugation at 10,000 × g at 4°C for 1 h. The exosome pellets were resuspended in 150 μL PBS and measured using NTA and TEM.

Exosome uptake experiment
The exosomes from macrophages were marked by the PKH67 Fluorescent Cell Linker Mini Kit (Sigma-Aldrich) according to its manufactory's instruction. The exosomes marked by PKH67 and HSCs were co-cultured for 12 h and subsequently fixed by 4% paraformaldehyde and stained by 2,4-diamino-5-phenylthiazole (DAPI, Sigma-Aldrich). The fluorescent images were taken with a Zeiss LSM 800 laser scanning confocal microscope (Zeiss, Germany). The measuring process was conducted by an assessor blind to treatment allocation. Fig. 8 MiR-342 suppresses HSCs activation by negatively mediating Sp1. A The cell cycle of the LX-2 was examined by the flow cytometry. B, C The migration of LX-2 was determined by the scratch assay (B) and the transwell assay (C). D The LX-2 cell proliferation was examined by the CCK-8 assay. E The protein levels of collagen I, α-SMA, fibronectin, Sp1, TGF-β1, p-Smad2, Smad2, p-Smad3, and Smad3 were measured by western blot. *p < 0.05; **p < 0.01; ***p < 0.001.
RNA immunoprecipitation (RIP) assay and m6A RNA RIP assay THP-1 cells, Kupffer cells, or LX-2 cells were collected, and RIP assay was performed using EZ-Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore) to identify the interaction of METTL3, NEAT1 and miR-342. In brief, RIP lysis buffer was added to the cells that were incubated with magnetic beads conjugated to either human anti-METTL3 antibody (1:50, ab195352, Abcam), human anti-Ago2 antibody (1:50, ab186733, Abcam) or the nonspecific anti-IgG antibody (1:50, ab172730, Abcam) at 4°C overnight. Afterward, protein K buffer was added for another 2 h at 4°C. After centrifuging five times, the co-immunoprecipitated RNA was isolated and used for further qRT-PCR analysis.
m6A RIP assay was performed utilizing the Magna m6A RNA PIP Kit (Millipore) based on the manufacturer's instruction. In brief, THP-1 cells that were transfected with shMETTL3 or shNC were incubated in lysis buffer, then incubated with anti-m6A antibody (1:100, ab208577, Abcam), which was conjugated with protein A/G magnetic beads at 4°C for 12 h. After centrifuging five times, the m6A-modified RNA was enriched and subjected to qRT-PCR analysis.

m6A RNA methylation assay
The total RNA was isolated from THP-1 cells and the level of m6A RNA was measured with m6A RNA methylation Assay Kit (Abcam) according to the manual's instruction.

Cell proliferation detection by Cell counting kit-8 (CCK-8) assay
The proliferation of LX-2 cells and primary HSCs was assessed utilizing the CCK-8 assay (Dojindo, Japan). In brief, cells were seeded into 96-well plates with a density of 2000 cells/well. Following 12 h incubation to let the cells attach, 10 µL of CCK-8 solution was added into each well. Following 2 h incubation, the absorbance at 490 nm was obtained by a microplate reader (Thermo Fisher Scientific).
Cell cycle assay LX-2 cells were trypsinized and washed twice with PBS (Gibco). The cells were fixed with 70% ethanol. 24 h later, the cells were stained with propidium iodide (PI) by the CycleTEST PLUS DNA Reagent Kit (BD Biosciences) based on the manufacturer's instructions. Subsequently, the cells were analyzed using the flow cytometer (Becton Dickinson, USA) and the cell cycle was analyzed by the Cell Quest Modfit software.
Scratch assay LX-2 cells and primary HSCs were seeded onto the 24-well plates until they reached a confluent monolayer. A 10 µL pipette tip (Eppendorf, Germany) was used to create a scratch. Afterward, the new cell culture medium without serum was replaced. Following 24 h of culture, the migrated distance was calculated under a light microscope (Zeiss, Germany). The measuring process was conducted by an assessor blind to treatment allocation.
Transwell migration assay LX-2 cells and primary HSCs were trypsinized, washed with PBS and were resuspended in a serum-free medium in a density of 5 × 10 5 cells/mL. The cells were seeded into the upper chamber of transwell inserts (Millipore) with 8 µm pores. DMEM containing 10% FBS was added to the bottom chamber. 48 h later, the migrated cells were fixed in 4% paraformaldehyde, stained with 1% crystal violet. The stained cells were counted under the light microscope (Zeiss, Germany). The measuring process was conducted by an assessor blind to treatment allocation.

Animal study
All experiments were approved by the Animal Care and Use Committee of the Second Xiangya Hospital of Central South University (Changsha, Hunan, China). Male Sprague-Dawley rats (6 week old, 180 ± 20 g) were obtained from Shanghai SLAC Laboratory Animal Center (Shanghai, China) and randomized into 6 groups (n = 10 per group): the control group, the CCL 4 treatment model group, the CCL 4 + OE-NC-Exo group, the CCL 4 + OE-NEAT1-Exo group, the CCL 4 + shNC-Exo group, and the CCL 4 + shNEAT1-Exo group. For the model groups, 0.4 mL of CCL 4 /olive oil (3:2, v/v) per 100 g body weight was subcutaneously injected into rats in a frequency of 2 times per week for 5 weeks. For exosome treatment groups, the indicated exosomes were intravenously injected through the tail vein into rats twice a week for 5 weeks. Rats were sacrificed after 5 weeks, and the liver tissues and serum were collected for further experiments. The measuring process was conducted by an assessor blind to treatment allocation.

Histological analysis
Liver tissues were fixed in 4% paraformaldehyde and were subsequently embedded in paraffin. Then the samples were sliced into 5 µm sections and were performed with histological analysis as previously reported [23]. Briefly, Masson staining of the sections was performed using a Masson's Trichrome Stain Kit (Sigma-Aldrich) and imaged using a light microscope. Sirius Red staining was performed using a Picro Sirus Red Stain Kit (ab150681) from Abcam according to the manufacture's instruction. In brief, sections were incubated in Picro Sirus Red solution for 60 min and followed by two times quickly rinsing in acetic acid solution and absolute alcohol. The images were also taken using a light microscope (Zeiss, Germany). The measuring process was conducted by an assessor blind to treatment allocation.

Enzyme-linked immunosorbent assay (ELISA) assay
The ALT and AST levels in the rat serum were measured using ELISA kits (Abcam) based on the manufacturer's protocol.

Immunohistochemistry (IHC) staining
Liver tissues were fixed in 4% paraformaldehyde and were subsequently embedded in paraffin. IHC staining was performed on 5 μm slices and stained using the protocol as described previously [24]. In brief, the sections were deparaffinized in xylene and rehydrated in graded alcohols, followed by blocking the endogenous peroxidase activity in 0.3% hydrogen peroxide. The sections were then incubated with primary antibodies against collagen I (1:100, ab270993, Abcam), α-SMA (1:200, ab124964, Abcam), METTL3 (1:100, ab195352, Abcam), and Sp1 (1:500, SAB4502837, Sigma-Aldrich) at 4°C overnight, followed by the incubation with the biotinylated anti-rabbit secondary antibody (1:1000, ab6720, Abcam) for 1 h at room temperature. Afterward, the avidin peroxidase conjugate (1:2000, ab59653, Abcam) was added for 1 h at room temperature. Finally, the target signal was developed utilizing DAB substrate (Vector Labs), and the slices were stained with hematoxylin. The images were taken under a light microscope (Zeiss, Germany). The measuring process was conducted by an assessor blind to treatment allocation.

Real-time polymerase chain reaction (qRT-PCR)
The total RNA was extracted from cells or tissues using the Trizol reagent (Beyotime). 2 µg of total RNA was reversely transcribed using TaqMan ® microRNA reverse transcription Kit (Thermo Fisher Scientific) or PrimeScript RT Kit with gDNA Eraser (Takara) based on the manufacturer's instructions. qPCR was carried out utilizing Taqman ® microRNA assay Kit (Thermo Fisher Scientific) or Taqman ® Universal PCR mixture Kit (Thermo Fisher Scientific) based on the manufacturer's manual. The relative mRNA levels were quantified using the 2 −△△Ct method and normalized to GPDAH or U6. Primer sequences were purchased from Sangon Biotech (Shanghai).

Statistical analysis
All data were analyzed by GraphPad Prism 6.0 and presented as the mean ± standard deviation (SD). All the data meet the assumption of normal distribution. All experiments were performed in at least three biological replicates, and each biological replicate contained three technical replicates. Statistical differences were performed using unpaired two-tailed t test for comparison between two groups and one-way analysis of variance followed by Tukey post hoc test for multiple comparisons. P < 0.05 was considered statistically significant.

DATA AVAILABILITY
All data generated or analyzed during this study are included in this article. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

MATERIAL AVAILABILITY
Supplementary figure legends and original western blots were shown in Supplemental Materials.