Deletion of the miR‐25/93/106b cluster induces glomerular deposition of immune complexes and renal fibrosis in mice

Abstract IgA nephropathy (IgAN), the most common form of primary glomerulonephritis, is caused by immune system dysfunction and affects only the kidneys. miRNA was involved in IgAN, in which their roles are still unknown. Herein, we found increased glomerular medulla size, proteinuria, kidney artery resistance, kidney fibrosis and immune complex deposition in 5‐month miR‐25/93/106b cluster knockout (miR‐TKO) mice. In vitro, the inhibition of miR‐25 cluster could promote cell proliferation and increase fibrosis‐related protein and transferrin receptor (TFRC) expression in human renal glomerular mesangial cell (HRMC). Luciferase assay revealed that inhibition of miR‐93/106b cluster could upregulate Ccnd1 expression through direct binding with the 3’UTR of Ccnd1. Conversely, inhibition of Ccnd1 expression prevented miR‐93/106b‐induced effect in HRMC. These findings suggested that miR‐25 cluster played an important role in the progression of IgAN, which provided new insights into the pathogenesis and treatment of IgAN.


| INTRODUC TI ON
IgAN is one of the most common primary glomerulonephritis, 1 about 15%-40% IgAN eventually progresses into end-stage renal disease (ESRD); therefore, early diagnosis and treatment are essential to prevent or delay the progression of IgAN. 2-5 So far, the diagnosis of IgAN still requires histologic diagnosis, mainly featured as abnormal IgA and complement C3 deposition or IgA co-deposition with or without IgM/G in the glomerular mesangium. 6,7 To date, the pathogenesis of IgAN is generally believed to be the four-strike theory: increased abnormal IgA(hit 1), the production of IgA autoantibodies(hit 2) and the antibody recognizes antigen to form an immune complex(hit 3), which is deposited in the mesangial region of the glomerulus and causes proliferation of mesangial cells(hit 4). 8 Although many researches have been made on the basis and clinical aspects of IgA nephropathy, the exact pathogenesis of IgA nephropathy has not been clarified.
MicroRNA is a single-stranded RNA molecule about 21 to 23 nucleic acids in length. It is mainly involved in the regulation of gene expression at the post-transcriptional level and played an important role in many physiological and pathological processes. 9,1011 MicroRNA is involved in every process of cells, and regulatory obstacles of microRNA are involved in a variety of human diseases including chronic kidney disease(CKD). 12,13 Many miRNAs were involved in IgAN, such as let-7b, 14 miR-148b, 15 miR-133a/b, 16 miR-223 17 and miR-106b. 16 Previously, we showed inhibition of miR-25 by commercial miRNA antagomir, which caused high blood pressure and kidney dysfunction. 18 It hinted us that miR-25 may be involved in CKD, but belongs to which type of CKD and the exact molecular mechanism under such CKD was unclear. Simultaneously, delivery of LNA-modified anti-miRNA oligonucleotides in vivo has also been reported to potentially cause off-target and short-term effects compared with miRNA-null mice. 19 Furthermore, many miRNAs located in same cluster, cotranscribed and targeted same seed sequences. These miRNAs always express functional redundancy, which is likely that deletion of single miRNA could have been made up for other increased miRNA existed in same cluster. 20 Herein, we employed miR-25/93/106b cluster knockingout (miR-TKO) mice and showed that 5-month-old miR-TKO mice induce deposition of glomerular mesangial immune complexes, increased glomerular fibrosis and renal arterial resistance, and activated renin-angiotensin-aldosterone system. Inhibition of miR-25 cluster promoted cell proliferation and fibrosis in human renal glomerular mesangial cell (HRMC) by targeting Ccnd1. These findings will provide new insights into the pathogenesis and therapeutic target of IgAN.

| Animal treatment
The miR-25 cluster knockout mice (#008460) were obtained from JAX laboratory. The miR-25 cluster genotypes were confirmed by PCR using primers including common: TCCACTGCTCTGGTGAGTGG,

| Cell culture
The human renal mesangial cell (HRMC) obtained from ScienCell

| Luciferase reporter assay
Ccnd1 gene 3'-UTR luciferase vector containing the miR-93/106b response elements was amplified by PCR from mouse cDNA.
Luciferase activity was measured using a SpectraMax M5 (Molecular Devices, Sunnyvale, CA, USA) and normalized by measuring βgalactosidase activity. The primers used to generate specific fragments for the mouse Ccnd1 gene 3'-UTR are listed in Table S1. KO, n = 10) from 5-mo-old male wild-type and miR-TKO mice; E and F, quantification of MRI T2 medulla and cortex distance from wildtype and miR-TKO mice; G and H, quantification of 24-h urine volume (n = 8) and proteinuria (n = 6) from wild-type and miR-TKO mice; I, quantification of ratio of kidney weight to bodyweight from wild-type (n = 6) and miR-TKO mice (n = 6); *** and ***indicate P < .05, P < .01 and P < .001, respectively. Data are shown as the mean ± SEM

| Western blotting and ELISA assay
The kidney tissues were homogenized in lysis buffer (Thermo Fisher was performed according to the manufacturer's instructions using a commercially available kit for mouse. An ELISA for renin, angiotensin 1/2, was performed using a commercially available kit for mouse (Santa Cruz) according to the manufacturer's instructions.

| Ultrastructural analysis
Kidney tissues were fixed in 2.5% glutaraldehyde in 0.

| Magnetic resonance imaging
The animals were anaesthetized by inhalation of 2% isoflurane and a mixture of O2 and N2O. Bed temperature was maintained at 37.5℃ by applying warm water circulation. All MRI data were col-

| RNA extraction and analysis
For the tissue extraction, RNAs were extracted from the kidneys using a miRNA isolation kit (Ambion Inc) to separate into large and small RNAs according to the manufacturer's instructions.

| RNA-seq data analysis
The raw RNA-Seq data were filtered by trimmomatic (v0.33). 21 The clean reads were mapped using Tophat2 (v2.1.1) and then counted into different gene regions as the corresponding GTF files (mm9, GRCh38 or hg19, GRCh37 genome assembly) by featureCount high-sensitivity MRI, we found medullary was indeed thickened and cortico-medullary boundary was obscured in miR-TKO mice as compared to the wild-type mice ( Figure 1D, E). Although the medullary structure changed noticeably, the cortex thickness was not significantly different ( Figure 1F). Such kidney structure alteration reminded us measuring the urine feature, which showed decreased 24hour urine volume in miR-TKO mice ( Figure 1G), and it accompanied with increased proteinuria in miR-TKO mice compared with WT mice ( Figure 1H). We also found the ratio of kidney weight to bodyweight decreased in miR-TKO mice compared with WT mice ( Figure 1I).

| Deficiency of miR-25 cluster induces immune complex deposition in glomerular mesangial
In order to understand the basis for these structural and functional abnormalities, we performed standard histomorphometry, immuno-

Renin-Angiotensin-Aldosterone System
Hypertension, as target for treatment in clinic, always accompanied with IgAN. 25 Renal arterial resistance index (KRI, peak systolic and end-diastolic flow velocity/peak systolic flow velocity) is measured by colour Doppler and can reflect not only the onset and progression of renal disease, but also hypertension. 26 According to the ultrasound, it showed significantly increased KRI in miR-TKO mice compared with wild-type mice ( Figure 3A, B). This strongly reminded us that deficiency of miR-25 cluster maybe resulted in hypertension.
Abnormal RAS system was related to hypertension; here, we showed increased renin and AT1 expression in miR-TKO mice compared with WT mice ( Figure 3C). This variation was as same as upregulation of the RAS-related protein (renin, angiotensin 1, angiotensin 2, aldosterone) which was detected in the serum of miR-TKO mice as demonstrated by ELISA ( Figure 3D). We detected significantly elevated mRNA expression level of At1a, At1b and At1 in kidney of miR-TKO mice by qPCR compared with wild-type mice ( Figure 3E). MiR-TKO mice also existed increased renin, ATR (angiotensin 2 type 1 receptor), angiotensin and ACE2 (angiotensin-converting enzyme 2) protein expression level by WB measurement ( Figure 3F).  Figure 4A-D). By employing qPCR, we also analysed fibrosis-related genes and found increased mRNA expression levels of collagen 1a F I G U R E 3 Deficiency of miR-25 cluster associated with Increased KRI and Activated Renin-Angiotensin-Aldosterone System. A, Represented image of B-mode ultrasonography of the kidney from miR-TKO and wild-type mice; B, quantification of kidney artery resistance index (KRI) in 5-mo-old male WT and miR-TKO mice; C, representative images of renin and AT1 from 5-mo-old miR-TKO and wild-type mice, right panel is quantification of the renin and AT1-positive area within the glomeruli; D, quantification of serum renin-angiotensin 1/2-aldosterone-related protein in 5-mo-old miR-TKO and wild-type mice; E, Quantitative real-time PCR analysis for At1, At1a and At1b from 5-mo-old miR-TKO and wild-type mice; F, Western blot analyses for renin, ATR, angiotensin and ACE2 from 5-mo-old miR-TKO and wildtype mice. *, ** and *** indicate P < .05, P < .01 and P < .001, respectively. Data are shown as the mean ± SEM (I), collagen 4a (IV), Fn (Fibronectin), Acta2, Timp1 (tissue inhibitor of metalloproteinases 1) and Pai1 (plasminogen activator inhibitor type 1)

| Deletion of miR-25 cluster induces kidney fibrosis
in 5-month-old miR-TKO mice kidneys ( Figure 4E). Consistent with previous results, we detected upregulated protein level of COLL1A1, FN, a-SMA in 5-month-old miR-TKO kidney by employing WB, compared with WT mice ( Figure 4F). Our results showed that inhibition of miR-25/93/106b increased the HRMC proliferation ( Figure 5A). Inhibition of miR-25/93/106b also increased mRNA and protein expression levels of collagen 1a (I), collagen 4a (IV), Fibronectin and Acta2 which was analysed by qPCR and WB in HRMC ( Figure 5B-E). Interestingly, upregulation of TFRC was also detected in the HRMC treated by miR-25 cluster inhibitor ( Figure 5F, G).

F I G U R E 5
Inhibition of miR-25 cluster promotes HRMC proliferation and fibrosis. A, Cell proliferation level analysis in HRMC treated by miR-25 mimic or inhibitor, miR-93 mimic or inhibitor, and miR-106b mimic or inhibitor; B-F, quantitative real-time PCR analysis of Coll1a1, Coll4a1, Fn, Acta2 and Tfrc level in HRMC treated by miR-25 mimic or inhibitor, miR-93 mimic or inhibitor, and miR-106b mimic or inhibitor; G, Western blot analysis for TFRC, COLLA1, FN and α-SMA proteins in HRMC treated by miR-25 mimic or inhibitor, miR-93 mimic or inhibitor, and miR-106b mimic or inhibitor; *, ** and *** indicate P <.05, P <.01 and P < .001, respectively. Data are shown as the mean ±SEM

| MiR-93/106b regulates HRMC proliferation and fibrosis by targeting CCND1
Through the RNA-seq results, 536 genes were significantly altered, and 19 of these genes were miR-25/93/106b cluster potential targets analysed by crossing this cluster target database ( Figure 6A).
Through the analysis of overlapped 19 genes by bioinformatic protein network interaction, it showed the weight of Ccnd1, a potential target, was the most significant ( Figure 6B). Luciferase assay showed that overexpression of miR-93 and miR-106b significantly inhibited F I G U R E 6 miR-93/106b targets Ccnd1 in HRMC. A, The number of genes in the kidney with significant changes in 5-mo-old miR-TKO mice and those identified as potential miR-25/93/106b targets using the Targetscan database; B, PPI analysis from differential gene between miR-TKO mice and miR-93/106b targeting genes; C, Luciferase activity in HEK 293A cells that were transfected with the indicated 3'-UTR reporter constructs showing binding of miR-93/106b with the 3'-UTR of Ccnd1; D, Western blot analysis for CCND1 from 5-mo-old miR-TKO and wild-type mice; E and F, quantitative real-time PCR and western blot analysis for Ccnd1 from overexpression or inhibition of miR-25, miR-93 and miR-106b treated HRMC. *, ** and *** indicate P < .05, P < .01 and P < .001, respectively. Data are shown as the mean ± SEM    Figure 6C). The upregulated protein expression level of Ccnd1 was also detected in 5-month-old miR-TKO mice as demonstrated by Western blot ( Figure 6D). We also found that the protein and mRNA expression levels of Ccnd1 were increased in HRMC with miR-93 and miR-106b inhibitor compared with scrambled miRNA and miR-25 inhibitor transfection ( Figure 6E and F). These data identified both miR-93 and miR-106b could bind to the 3'-UTR of its putative target Ccnd1.
To explore the mechanism of Ccnd1 underlying the involvement of miR-25/93/106b in HRMC proliferation and fibrosis, we successfully decreased protein and mRNA level of Ccnd1 in HRMC cells with siR-Ccnd1 ( Figure 7A, B). We explore whether decreased protein level of Ccnd1 suppresses cell proliferation and fibrosis of HRMC with transfecting miR-25/93/106b inhibitor. We found that inhibition of Ccnd1 could suppress miR-25/93/106b inhibitor-induced cell proliferation by CCK8 measurement ( Figure 7C). Meanwhile, we also found that inhibition of Ccnd1 could decrease mRNA and protein expression level of fibrosis-related gene (Coll4a1, Coll1a1, Fn and Acta2) and Tfrc in HRMC treated by miR-25/93/106b inhibitor ( Figure 7D-I).

| D ISCUSS I ON
In the present study, we showed increased IgA/G and C3 deposition in glomeruli mesangium, which is consistent with serum IgA. We

CO N FLI C T O F I NTE R E S T
All the authors declared no competing interests.