METTL3/N6‐methyladenosine/ miR‐21‐5p promotes obstructive renal fibrosis by regulating inflammation through SPRY1/ERK/NF‐κB pathway activation

Abstract Renal fibrosis induced by urinary tract obstruction is a common clinical occurrence; however, effective treatment is lacking, and a deeper understanding of the mechanism of renal fibrosis is needed. Previous studies have revealed that miR‐21 impacts liver and lung fibrosis progression by activating the SPRY1/ERK/NF‐kB signalling pathway. However, whether miR‐21 mediates obstructive renal fibrosis through the same signalling pathway has not been determined. Additionally, studies have shown that N6‐methyladenosine (m6A) modification‐dependent primary microRNA (pri‐microRNA) processing is essential for maturation of microRNAs, but its role in the maturation of miR‐21 in obstructive renal fibrosis has not yet been investigated in detail. To address these issues, we employed a mouse model of unilateral ureteral obstruction (UUO) in which the left ureters were ligated for 3, 7 and 14 days to simulate the fibrotic process. In vitro, human renal proximal tubular epithelial (HK‐2) cells were transfected with plasmids containing the corresponding sequence of METTL3, miR‐21‐5p mimic or miR‐21‐5p inhibitor. We found that the levels of miR‐21‐5p and m6A modification in the UUO model groups increased significantly, and as predicted, the SPRY1/ERK/NF‐kB pathway was activated by miR‐21‐5p, confirming that miR‐21‐5p plays an important role in obstructive renal fibrosis by enhancing inflammation. METTL3 was found to play a major catalytic role in m6A modification in UUO mice and drove obstructive renal fibrosis development by promoting miR‐21‐5p maturation. Our research is the first to demonstrate the role of the METTL3‐m6A‐miR‐21‐5p‐SPRY1/ERK/NF‐kB axis in obstructive renal fibrosis and provides a deeper understanding of renal fibrosis.


| INTRODUC TI ON
Renal fibrosis is a common pathological change in hydronephrosis that is caused by urinary tract obstruction and leads to renal parenchyma destruction and renal function damage. To date, there is no effective intervention to completely restore renal function and reverse renal fibrosis, even if the obstruction is relieved by surgical treatment. A better understanding of the mechanism underlying progression of obstructive renal fibrosis is necessary to find an effective medicine or treatment procedure. In recent years, the focus of studies on renal fibrosis occurrence and development has gradually shifted from proteins and mRNAs to non-coding RNAs, including microRNAs, lncRNAs and circRNAs. microRNAs are well studied and widely found in the genomes of animals and plants. Most  its dysregulation has been found in many renal fibrosis models and clinical samples. [1][2][3] Most studies agree that functional miR-21 promotes renal fibrosis, 4-6 but its maturation process and downstream signalling pathway are still unclear and merit further exploration. SPRY1, a direct target of miR-21, inhibits the ERK/NF-κB pathway in angiotensin II-induced liver fibrosis 7 and bleomycin (BLM)-induced lung fibrosis in rats. 8 These findings indicate that miR-21 is involved in the fibrotic process in the lung and liver via the SPRY1/ERK/NF-kB signalling pathway. However, whether miR-21 promotes obstructive renal fibrosis via the ERK/NF-kB signalling pathway by targeting SPRY1 has not been determined.
Mature microRNAs originate from long primary transcripts called pri-microRNAs. N6-methyladenosine (m 6 A) modification labels pri-microRNAs in the nucleus 9 and is recognized and bound by nuclear reader protein, which recruits DGCR8 and the nuclear RNase III DROSHA to cleave the stem loop and produce pre-microRNAs. 10 Then, pre-microRNAs are exported to the cytoplasm for further splicing. The m 6 A modification is the most abundant RNA modification in eukaryotes and is highly conserved within mRNAs, microR-NAs and lncRNAs among many species. In RNA m 6 A modification, a methyl group is added onto the sixth N atom of the RNA base A. This process is catalysed by a core methyltransferase complex consisting of methyltransferase-like 3 (METTL3), methyltransferaselike 14 (METTL14) and Wilms tumour 1-associated protein (WTAP) and is reversed by demethylases, including fat mass and obesityassociated protein (FTO) and alkB homologue 5 (ALKBH5). 11,12 Thus, m 6 A RNA modification is a dynamic and reversible process.
The reader proteins include YT521-B homology (YTH) domain family, 13,14 heterogeneous nuclear ribonucleoproteins (HNRNPs) 14 and insulin-like growth factor 2 mRNA-binding proteins (IGF2BPs). 14,15 Among them, YTHDC1, 16 HNRNPA2B1 17 and HNRNPC 17 are predominantly found in the nucleus. HNRNPA2B1 has been found to be reader protein of m 6 A in pri-microRNAs and to promote maturation of microRNAs. 17 A previous study reported that m 6 A is involved in the metabolism of miRNA-126 and drives pulmonary fibrosis development. 18 However, to date, there have been no reports on the role of m 6 A modification in microRNA maturation during obstructive renal fibrosis development.
In the present study, we generated obstructive renal fibrosis models in mice and enhanced or inhibited the expression of miR-21-5p, METTL3, HNRNPA2B1 and ERK in HK-2 cells through transfection or chemical inhibition. We demonstrated that increased miR-21-5p levels induced by ureteral obstruction enhanced inflammation of the renal parenchyma by targeting SPRY1 and activating the ERK/NF-kB pathway, which resulted in extracellular matrix (ECM) deposition and progression of obstructive renal fibrosis. METTL3-mediated m 6 A modification promoted miR-21-5p maturation by promoting recognition and processing of pri-miR-21. These findings might provide novel information to further understanding of the mechanism underlying obstructive renal fibrosis.

| Animals
Sixty 8-week-old female C57BL/6 mice were provided by the experimental animal centre of the Medical College of Zhengzhou University and housed at a 22°C constant room temperature and 47% humidity with a 12-hours light-dark cycle and free access to standard laboratory chow and tap water. All experimental procedures on mice were performed in accordance with the National Institutes of Health guidelines and were approved by the Ethical Committee, Animal Care and Use Committee of the First Affiliated Hospital of Zhengzhou University (2020-KY-273).

| Mouse model of unilateral urethral obstruction (UUO) and experimental groups
A total of 60 mice were used in the current study and were randomly divided into two groups: 30 in the sham operation group and 30 in the UUO group. UUO was performed according to established procedures described in previous studies. 19 Briefly, each mouse was anaesthetized with inhaled isoflurane, and the left proximal ureter was exposed. Then, the ureter was ligated with 6-0 silk thread and severed. In the sham operation group, the left ureters of mice were exposed, but not ligated or severed. The 3rd, 7th and 14th days after surgery were the time points for killing. At each time point, a total of 10 mice in the UUO group were executed, and a total of 10 mice in the sham group were also executed to serve as controls. The left kidney specimens were collected, and inferior vena cava blood samples were collected for evaluation of renal function, including blood urea nitrogen (BUN) and serum creatinine (SCr). Part of the kidney tissue was frozen in liquid nitrogen for total RNA and protein extraction, and the remainder was fixed with 4% paraformaldehyde.

| Histopathological evaluation
Four per cent paraformaldehyde-fixed kidney specimens were embedded in paraffin, cut into 4μm sections on a rotary microtome (Leica, Heidelberg, Germany) and subjected to Masson's trichrome and haematoxylin and eosin (HE) staining. A Leica DM4B microscope equipped with Leica X software was used to examine the slides and take images. Tubulointerstitial impairment was evaluated according to the scoring criteria reported in a previous study 20 and included assessment of tubular atrophy, tubular necrosis, lymphocyte infiltration and interstitial fibrosis. The scores for each criterion were as follows: 0 = none; 1 = mild or <25%, 2 = moderate or 25% to 50% and 3 = severe or >50%. The blue Masson staining area was considered to indicate collagen deposition and was analysed using Image-Pro Plus 6.0 software. Six nonoverlapping fields in each section of the kidney cortex and medulla were selected for scoring or image analysis, and the results are expressed as the means ± SD.

| Immunohistochemistry (IHC)
A rotary microtome was used to cut the paraffin-embedded kidney specimens into 4μm sections, which were deparaffinized with xylene and rehydrated using graded ethanol (100%, 95%, 85% and 75%) and distilled water. The sections were incubated with 3% hydrogen peroxide for 10 minutes to block the activity of endogenous peroxidase and then heated with a microwave in 0.01 mol/L citrate buffer (pH 6.0) for 25 minutes for antigen retrieval. The specimens were washed three times with phosphate- (1:200) and anti-FN (1:1000) antibodies, overnight at 4°C. The next day, after being washed with PBS, the specimens were incubated with HRP-labelled goat anti-rabbit secondary antibody for 1 hour at room temperature. Finally, dehydration, clearing, 3,3′-diaminobenzidine (DAB) staining and neutral resin sealing were performed in sequence. In each section, six non-overlapping fields of the renal cortex were imaged at high magnification (100×). Image-Pro Plus 6.0 software was applied to assess the integral optical density (IOD) of the positive area, and the results are expressed as the means ± SD.

| Cell culture
The human renal proximal tubular epithelial cell line HK-2 was purchased from Beina Chuanglian Biotechnology Institute (Beijing, China). HK-2 cells were cultured in Dulbecco's modified Eagle's medium/nutrient mixture F-12 (DMEM/F12) supplemented with 10% (v/v) foetal bovine serum (FBS), 100 IU/mL penicillin and 10 mg/mL streptomycin in a humidified atmosphere of 5% CO 2 at 37°C. The medium was changed every 2 days, and the cells were subcultured before a confluent monolayer could be formed.
The HK-2 cells were seeded on 6-well plates at 1 × 10 5 cells and transfected when the cells reached 60%-70% confluence. After transfection for 48 hours, the cells were harvested and used in subsequent experiments.

| RNA m 6 A dot blot assays
The poly(A) + RNAs (400 ng) were heated at 55°C for 15 minutes for denaturation and then cooled on ice. Next, the samples were

| RNA immunoprecipitation (RIP)
A Magna RIP Kit (Millipore, MA, USA) was used to perform RIP as described in previous studies. Briefly, HK-2 cells were washed with pre-cooled PBS, gently scraped off the culture plate and then collected by centrifugation at 1500 rpm for 5 minutes at 4°C.

| Statistical analysis
GraphPad Prism 7 was used to determine statistically significant differences. All data are presented as the means ± SD and were analysed using unpaired t tests between two groups and one-way analysis of variance with Tukey's test for more than two groups. P < .05 was considered statistically significant.

| RE SULTS
According to the results of ultrasonography and pathological exami-

| Changes in kidney morphology and impairment of renal function induced by UUO in mice
After ligation of the left ureters in mice, ultrasonography revealed that the width of the renal pelvis gradually increased, while the renal cortex gradually decreased in thickness ( Figure 1A,B). Due to compensation by the right kidney, the BUN and SCr levels were roughly within the normal range (Table 2); however, with the accumulation of urine in the obstructed area, the upward trend within the normal range of BUN and SCr was still obvious ( Figure 1C,D).

| The expression of fibrosis indicators increased in a time-dependent manner in UUO mouse models
According to the qRT-PCR results, the mRNA expression levels of  in mice ( Figure 4A), which was accompanied by increased p-ERK1/2, p-NF-κB, IL-6 and TNFα protein levels and decreased SPRY1 protein levels ( Figure 4B,C). We increased the miR-21-5p level in HK-2 cells via transfection with a pGV514-miR-21-5p-mimic plasmid ( Figure 5A) staining showed that α-SMA expression was up-regulated by miR-21-5p mimic, and this effect was reversed by U0126 ( Figure 5H).

| m 6 A methylation is significantly increased in obstructed kidney tissue, and the increased expression of METTL3, a methylation enzyme involved in m 6 A formation, plays a major catalytic role in obstructed kidneys
The results of RNA m 6 A dot blot assays showed that m 6 A levels significantly increased after ligation of the left ureter for 3, 7 and 14 days ( Figure 6A). The expression levels of key m 6 A methyltransferase and demethylase genes were detected via qRT-PCR. The results showed that the METTL3 mRNA expression level was significantly increased in the UUO groups after ligation for 3, 7 and 14 days compared with the sham group, which was further confirmed by western blotting and IHC (Figure 6B,D-G). The levels of METTL14, WTAP and other genes were also altered, but there was no significant difference compared with levels in the sham group.
HK-2 cells were transfected with the pGV657-METTL3 plasmid to force METTL3 expression. The METTL3 protein level was increased after transfection compared to that in the control group ( Figure 6H).
As expected, overexpression of METTL3 also increased the global m 6 A modification level in HK-2 cells ( Figure 6I).

| METTL3 may drive obstructive renal fibrosis development by promoting miR-21-5p maturation in obstructive renal fibrosis
In the in vivo study, METTL3 was significantly up-regulated in the  Figure 7A,B,C). Moreover, qRT-PCR analysis revealed that mature miR-21-5p was increased and unprocessed pri-miR-21 was decreased in METTL3-overexpressing HK-2 cells ( Figure 7D,E). The sequence of hsa-pri-miR-21 was downloaded from Ensembl database.
Then, the m 6 A modification sites were predicted via SRAMP (http:// www.cuilab.cn/sramp/), which is a sequence-based m 6 A modification site predictor. We found 7 positions in hsa-pri-miR-21 that may be m 6 A modification sites, and the prediction score for 2 of the positions was distributed in the high confidence range ( Figure 7F). The results of a RIP assay with anti-DGCR8 antibody revealed that the level of pri-miR-21 binding to DGCR8 was significantly increased in the HK-2 cells with up-regulated METTL3 ( Figure 7G). Furthermore, as expected, a RIP assay performed with m 6 A antibody revealed that the level of m 6 A-modified pri-miR-21 was elevated in the HK-2 cells with METTL3 up-regulation ( Figure 7H).

METTL3-mediated m 6 A in pri-miR-21 and promote maturation of miR-21-5p
The recognition of m 6 A in pri-miRNAs by reader proteins occurs in the nucleus; thus, the possible reader proteins of m 6 A in pri-miR-21, including YTHDC1, HNRNPA2B1 and HNRNPC, were evaluated via qRT-PCR, but the results did not reveal any obvious changes ( Figure 6C). Then, HK-2 cells were transfected with the PGV102-shHNRNPA2B1#1 plasmid and PGV102-shHNRNPA2B1#2 plasmid to inhibit HNRNPA2B1 expression. The protein levels of HNRNPA2B1 were decreased after transfection ( Figure 6J). Then, qRT-PCR analysis revealed that mature miR-21-5p was decreased It has been well documented that miR-21 is involved in fibrosis of the kidneys and other organs, including the heart, 21 liver 22 and lungs. 23 According to previously published studies, 24,25 the miR-21 level is increased in renal fibrosis tissue and serves as an important biomarker of fibrosis in plasma 26 and urinary tissue. 27 Ample evidence supports miR-21 as a promoting factor during renal fibrosis, and targeting miR-21 likely has a therapeutic effect and attenuates renal fibrosis development. 28,29 In our study, after ligation of the left ureter for 7 and 14 days, the morphology of obstructed kidneys changed dramatically. The SCr and BUN levels showed an upward trend within the normal range, which was related to the extent of impairment of the left kidney and was consistent with the results of a previous study. 30 The expression of fibrosis indicators, including α-SMA, collagen I and FN, was significantly elevated and the miR-21-5p level was significantly increased in the 7-day and 14-day groups, which was consistent with previously reported studies. miR-21 exerts pro-fibrotic effects via complex downstream signalling pathways. 5 The mitogen-activated protein kinase (MAPK) signalling pathway is well reported and is strongly associated with fibrosis progression.
ERK is a member of the MAPK family and has important roles in many signalling cascades. The ERK signalling pathway participates in the fibrotic process in many organs, including the artery, 31 In the UUO groups in our study, p-ERK1/2 and p-NF-κB protein expression was up-regulated, while the expression of SPRY1 was inhibited. Infiltration of inflammatory cells was observed, and inflammation foci were formed in the 7-day and 14-day groups.
The TNFα and IL-6 protein levels were significantly increased.
Masson's trichrome staining showed higher collagen deposition in areas of inflammation foci. Therefore, we propose that the pro- Representative bands and fold changes in α-SMA, collagen I and FN protein expression in HK-2 cells transfected with pGV657-METTL3 plasmid or cotransfected with miR-21-5p inhibitor, demonstrated by Western blotting (mean ± SD, n = 3). C, Immunofluorescence staining of α-SMA and METTL3 (400×) in HK-2 cells showing that METTL3 enhancement increased α-SMA expression, whereas cotransfection with the miR-21-5p inhibitor weakened this effect; scale bar represents 20 μm. (D and E) miR-21-5p and pri-miR-21 levels in HK-2 cells transfected with pGV657-METTL3 plasmid determined by qRT-PCR (mean ± SD, n = 3). F, Seven possible m 6 A modification sites exist in hsa-pri-miR-21, including 2 high confidence positions based on prediction with SRAMP. G, The levels of pri-miRNA-21-5p binding to DGCR8 in HK-2 cells with METTL3 up-regulation determined by qRT-PCR. H, Immunoprecipitation of m 6 A-modified RNA in HK-2 cells with METTL3 up-regulation, followed by qRT-PCR to assess the pri-miR-21 m 6 A modification level. (I and J) miR-21-5p and pri-miR-21 levels in HK-2 cells transfected with PGV102-shHNRNPA2B1#1 plasmid and PGV102-shHNRNPA2B1#2 plasmid measured via qRT-PCR (mean ± SD, n = 3). K, Mode pattern of the METTL3-m 6 A-miR-21-5p-SPRY1/ERK/NF-kB regulatory network in obstructive renal fibrosis *P < .05, compared to the oeVector group. # P < .05, compared to the oeMETTL3 group study 17 published in CELL in 2015, HNRNPA2B1 is a reader protein of m 6 A modification sites in pri-miRNAs and has similar effects on alternative splicing as METTL3 modulation. In the study, HNRNPA2B1 depletion in MDA-MB-231 and HEK293 cells caused a reduction in the expression level of mature miRNAs and resulted in accumulation of specific pri-miRNAs in the nucleus. In the present study, we knocked out HNRNPA2B1 in HK-2 cells and found similar results: The expression level of miR-21-5p decreased, and pri-miR-21 accumulated in the cells. Therefore, HNRNPA2B1 might be a reader protein of m 6 A in pri-miR-21 and recruit DGCR8 for processing. These results confirm that METTL3 mediates m 6 A upregulation in UUO mice and indeed enhances miR-21-5p maturation by m6A-dependent pri-miR-21 processing, thus driving obstructive renal fibrosis development.

| CON CLUS ION
In summary, our research revealed a significant increase in m

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.