Regulation of CTGF Expression by miR-133b for the Treatment of Renal Interstitial Fibrosis in Old UUO Rats

Introduction: Renal interstitial brosis, an important pathological feature of kidney aging and chronic renal failure, is regulated by mesenchymal stem cells (MSCs). We have previously demonstrated the high expression of miR-133b in MSC-derived extracellular vesicles (MSC-EVs) from old rats, which mediated the inhibition of epithelial-mesenchymal transition (EMT) of renal tubules induced by transforming growth factor-β1 (TGF-β1). We investigated the effect of miR-133b for the treatment of geriatric renal interstitial brosis and evaluated its target genes. Methods: miR-133b expression induced during the EMT of HK2 cells by TGF-β1 at different concentrations (0, 6, 8, and 10 ng/mL) and time points (0, 24, 48, and 72 h) was detected using real-time polymerase chain reaction. The target genes of miR-133b were validated using a dual-luciferase reporter assay. In vitro experiments were performed to observe mRNA and protein expression of miR-133b targets, E-cadherin, α-smooth muscle actin (SMA), bronectin, and collagen 3A1 (Col3A1), in HK2 cells transfected with miR-133b under TGF-β1 stimulation. A 24-week-old unilateral ureteral obstruction (UUO) mouse model was established and injected with transfection reagent and miR-133b into the caudal vein. miR-133b (cid:0) target gene and other indexes mentioned above mRNA and protein levels and renal interstitial brosis were detected at 7 and 14 days. Results: miR-133b expression gradually decreased with an increase in TGF-β1 concentration and treatment time, and miR-133b mimic downregulated connective tissue growth factor (CTGF) expression. Dual-luciferase reporter assay conrmed CTGF as a direct target of miR-133b. miR-133b mimic transfection inhibited the TGF-β1-induced EMT of HK2 cells; this effect was reversed by CTGF overexpression. miRNA-133b expression signicantly increased (approximately 70-100


Background
Aging is characterized by signi cant changes in the structure and function of the kidney, even in the absence of age-related comorbidities. Renal interstitial brosis is an important pathological feature of kidney aging [1], and bone marrow mesenchymal stem cells (MSCs) play an important role in the regulation of renal interstitial brosis. Transplantation of the bone marrow from young mice into old mice was shown to signi cantly reduce renal brosis as well as the expression of aging markers in the recipient mice; bone marrow cells did not directly replace parenchymal cells but instead exerted paracrine effects on renal parenchymal cells [2]. Further, the injection of MSCs and MSC-derived extracellular vesicles (MSC-EVs) was shown to signi cantly alleviate renal interstitial brosis in a unilateral ureteral obstruction (UUO) model [3,4].
Aging can signi cantly alter the number of stem cells and their regenerative capacity and functions [5].
Our previous research revealed the signi cant differences in the expression pro les of microRNAs (miRNAs) of MSC-EVs derived from the bone marrow of young and old rats. We found that miR-133b-3p was highly expressed in the MSC-EVs of old rats and inhibited the epithelial-mesenchymal transition (EMT) of renal tubular cells induced by transforming growth factor (TGF)-β1 [6]. The inhibitory effects of MSC-EVs against renal brosis decreased with age. miR-133 of MSC-EVs derived from old rats was thought to exhibit important intervening effects on renal brosis [7]. miR-133 ameliorates cardiac brosis [8][9][10], reduces the TGF-β1-mediated EMT of bladder smooth muscle epithelial cells [11], and directly targets the connective tissue growth factor (CTGF) [9,11]. Whether CTGF is a target involved in the miR-133-mediated inhibition of the EMT of renal tubular epithelial cells is still unclear, and studies are warranted to investigate the effect of transfection with exogenous miR-133 in geriatric renal interstitial brosis.
Here, the EMT of the human renal proximal tubular epithelial cell line HK2 was stimulated by TGF-β1 in vitro to investigate the role of miR-133b and its target genes in this process. A UUO model was established using old C5BL/6J7 mice (aged 24 months) intravenously injected with an miR-133b transfection complex to verify the effect of miR-133b overexpression in geriatric renal brosis.

TGF-β1 stimulation and miR-133b transfection
The HK2 cells were cultured in Dulbecco's modi ed Eagle's medium (DMEM)/F12 (Corning, USA) supplemented with 5% fetal bovine serum. After reaching 50% con uency, the cells were synchronized in serum-free DMEM/F12 for 18 h and then stimulated with TGF-β1 at 6, 8, and 10 ng/mL concentrations for 24, 48, and 72 h. In the transfection experiment, miR-133b mimic and miRNA mimic control (GenePharma, China) were transfected into HK2 cells for 6 h, as per the instructions of jetPRIME® transfection reagent (Polyplus transfection, France). Following transfection, the cells were cultured in DMEM/F12 with 5% serum for 18 h and then incubated with DMEM/F12 with 5% serum and 8 ng/mL TGF-β1 (PeproTech, USA) for 48 h. RNA extraction and real-time polymerase chain reaction (PCR) Total RNA was extracted from HK2 cells and kidney tissues of each group by Trizol, and used to synthesize miR-133b and U6 cDNA using miScript II RT Kit (QIAGEN, China). Primers speci c for target genes were designed with reference to their mRNA-coding regions in GenBank using Primer 5.0 software. The primer sequences were veri ed on BLAST. The total RNA was used to synthesize cDNA of target genes using ReverTra Ace qPCR RT Master Mix kit (TOYOBO, Japan). The expression of the genes encoding miR-133b, CTGF, E-cadherin, α-smooth muscle actin (SMA), bronectin, collagen 3A 1 (Col3A1), U6, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was detected on a ABI-prism-7500 sequence detection system (Applied Biosystems, USA) using miScript SYBR Green PCR Kit (QIAGEN). The relative expression level was calculated using U6 or GAPDH as internal control.

Dual-luciferase reporter assay
The seed sequence for the binding between CTGF and miR-133b was searched using the bioinformatic software TargetScan. The sequence 5′-AUUUGUUGAGUGUGACCAAAA-3′ containing the 3′-untranslated region (UTR) of CTGF was synthesized and cloned into a luciferase reporter vector GP-miRGLO (GenePharma), which was termed as miRGLO-Wt-CTGF. A mutant sequence 5′-AUUUGUUGAGUGUUGGAUUAA-3′ of the target was also synthesized and cloned into the plasmid to obtain miRGLO-Mut-CTGF, which was used as the negative control.
293T cells from logarithmic growth phase were digested with pancreatin and seeded in 48-well plates for 24 h. After reaching 80% con uency, the cells were transfected the cell fusion reagent. According to the instructions of jetPRIME® transfection reagent, the synthesized miR-133b and miRNA mimic control (NC-miR) were respectively co-transfected with miRGLO-Wt-CTGF or miRGLO-Mut-CTGF into 293T cells. After 48 h, the cells were lysed using a passive lysis buffer (Promega, USA) and the cell lysate collected. The luciferase activity of the lysate was detected according to the steps indicated in the dual-luciferase reporter assay system (Promega).
Immuno uorescence staining HK2 cells were seeded at about 10 5 cells/well in six-well plates with glass cover slips and disinfected at high temperature and high pressure. The plates were placed in an incubator at 37 °C with 5% carbon dioxide for 6 h to allow the cells to adhere to the glass cover slips. After synchronization and transfection as mentioned above, the cells were incubated with DMEM/F12 containing 5% serum with or without 8 ng/mL TGF-β1 for 48 h. The cells were then xed with 4% paraformaldehyde at room temperature for 20 min and treated with 0.2% Triton X-100 for 2 min for permeabilization. The cells were blocked with 5% bovine serum albumin (BSA) at room temperature for 1 h and then treated with anti-E-cadherin rabbit monoclonal primary antibody (1:100 dilution) and α-SMA antibody (1:100 dilution) diluted in 5% BSA at 4 °C for overnight. After washing with phosphate-buffered saline (PBS), the cells were probed with an anti-rabbit uorescein isothiocyanate (FITC)-conjugated uorescent secondary antibody (Beyotime) (1:400 dilution) at room temperature in the dark for 2 h. The slides were then washed with PBS and treated with 4′,6-diamidino-2-phenylindole (DAPI) (ZSGB-BIO, China) uorescence nuclear staining mounting medium. The expression of α-SMA and E-cadherin in HK2 cells from each group was observed under a uorescence microscope (100×) with random elds of vision.
Experimental animals and establishment of a UUO model Animal care and experiments were performed according to the guidelines of the Institutional Animal Care and Use Committee of Chinese PLA General Hospital. A total of 28 female C5BL/6J7 mice (SPF grade), aged 24 months, weighing 20 ± 2 g, were provided by SPF (Beijing) Biotechnology Co., Ltd. Mice were randomly divided into sham (n = 8 ), UUO + NS-miR-133b (n = 10), and UUO + miR-133b groups (n = 10).
To establish a UUO model, each mouse was anesthetized by an intraperitoneal injection of pentobarbital (50 mg/kg) and the abdominal cavity was opened under sterile conditions. The left ureter was dissociated, double-ligated, and disconnected at 15 mm below the renal pelvis with 4 − 0 thread. The abdominal cavity was closed by layered suture. In the control group, the abdominal cavity was closed immediately after the ureter was dissociated.
The transfection reagent Entranster-in vivo (Engreen, China) was diluted in a 10% glucose solution at a nal glucose concentration of 5%. The transfection complex was prepared by mixing the two agents and incubating for 15 min. The UUO + miR-133b and UUO + NC-miR groups were administered with miR-133b and NC-miRNA transfection complex (100 µL/animal), respectively, by a caudal vein injection at 24 h before operation and once every 3 days thereafter. The sham group was given the same amount of normal saline by caudal vein injection. The mice were sacri ced at 7 and 14 days after UUO (4 mice from the sham group and 5 mice from the UUO group were sacri ced at each time point). The kidney tissue was collected from the obstructed side for western blotting, real-time PCR, and pathological analysis.

Kidney tissue pathological examination
The kidney tissue was xed in 10% neutral formaldehyde, dehydrated with ethanol, embedded in para n, and cut into 2 µm sections. Morphological changes in the kidney tissue were observed by periodic acid Schiff (PAS) and Masson's trichrome staining. After staining with Masson's trichrome, 10 elds of vision were selected under a light microscope (400×). The area of each eld of vision and area of green collagen bers were measured using the Image-Pro Plus software. The relative area of collagen deposition was calculated was follows: area of green collagen ber/area of eld of vision × 100%.

Statistical analysis
The data were analyzed using the SPSS 17.0 software, and the results were expressed as mean + standard error of mean (SEM). The differences among the experimental groups were analyzed using the one-way analysis of variance (ANOVA) with completely random design. P < 0.05 was considered statistically signi cant.

Results
Inhibitory effect of TGF-β1 on miR-133b The real-time PCR results of miR-133b expression analysis showed that the stimulation of HK2 cells with TGF-β1 at 0, 6, 8, and 10 ng/mL for 48 h resulted in a gradual decrease in the expression of miR-133b in a TGF-β1 concentration-dependent manner. However, no signi cant difference was observed between the 8 and 10 ng/mL treatment groups (Figure.1A). The expression of miR-133b after stimulation of HK2 cells with TGF-β1 at 8 ng/mL for 0, 24, 48, and 72 h gradually decreased with an increase in the stimulation time; no signi cant difference was observed between the 48 and 72 h treatment groups ( Figure.1B).
Effect of miR-133b on the morphology of TGF-β1stimulated HK2 cells Morphological changes in cells were observed under an inverted microscope. HK2 cells had a round or an oval shape and were arranged in a cobblestone pavement-like pattern. After stimulation with 8 ng/mL TGF-β1 for 48 h, the cells became slender and fusiform, and their intercellular space signi cantly widened. Most transfected miR-133b cells could still maintain the morphology of epithelial cells, while the NC-miR-transfected cells had morphological changes similar to those observed for HK2 cells (Fig. 2).

miR-133b inhibited the EMT of HK2 cells induced by TGF-β1
Normal HK2 cells or HK2 cells transfected with miR-133b mimic/NC-miR for 24 h were stimulated with 8 ng/mL TGF-β1 for 48 h. The results of real-time PCR and western blotting showed that miR-133b signi cantly inhibited the downregulated E-cadherin mRNA and protein expression as well as the upregulated mRNA and protein expression of α-SMA, bronectin, Col3A1, and CTGF induced by TGF-β1 ( Fig. 4A and 4B).
Immuno uorescence results showed that the uorescence intensity of α-SMA increased and that of Ecadherin decreased in the cells treated with TGF-β1 as compared with that in control cells. In comparison with the TGF-β1 stimulation group, the cells treated with TGF-β1 and miR-133b showed a decrease in the uorescence intensity of α-SMA and an increase in the intensity of E-cadherin. Further, we failed to notice any signi cant change in the uorescence intensity of α-SMA and E-cadherin between the cells treated with TGF-β1 and NC-miR-133b and those treated with TGF-β1 alone (Table 1; Fig. 4C).

CTGF is a direct target gene of miR-133b
We used three different miRNA target gene prediction software (TargetScan, PiTar, and miRbase) and found CTGF as a potential target of miR-133b. The target site was located at 1027 to 1033 bp of CTGF 3′-UTR. The results of DAVID bioinformatic prediction suggested that CTGF is involved in TGF-β brosis and other signaling pathways.
After the transient transfection of miR-133b mimic into HK2 cells for 48 h, we performed real-time PCR and western blotting and found a signi cant increase in miR-133b expression (increased by about 21,566 times) (Fig. 3A), con rming the successful transfection of miR-133b. The expression of CTGF mRNA and protein signi cantly decreased following miR-133b transfection but showed no changes in the group transfected with an unrelated sequence ( Fig. 3B and 3C).
The cells were transfected with CTGF for 48 h, and then collected and analyzed by western blotting. In comparison with the HK2 cells transfected with GAPDH, those transfected with CTGF showed a signi cant upregulation in the expression of CTGF protein, con rming the successful transfection of CTGF (Fig. 5A). We stimulated HK2 cells with TGF-β1 at 8 ng/mL for 48 h and performed western blotting. We found that the miR-133b overexpression group showed a signi cant downregulation in the expression of CTGF, α-SMA, bronectin, and Col3A1 and a signi cant upregulation in the expression of Ecadherin as compared with the control group. The cells overexpressing miR-133b and CTGF showed a signi cant increase in CTGF expression. CTGF could revert the inhibition of α-SMA, bronectin, and Col3A1 expression and the increase in E-cadherin expression mediated by miR-133b overexpression.
Furthermore, the group overexpressing miR-133b and CTGF showed signi cantly higher CTGF and α-SMA levels and signi cantly lower E-cadherin levels than the control group. The group overexpressing miR-133b and GAPDH showed no signi cant difference in the expression of various proteins as compared with the miR-133b overexpression group (Fig. 5B).
The results of dual-luciferase reporter assay showed that miR-133b mimic signi cantly decreased the luciferase activity of miRGLO-Wt-CTGF but had no signi cant effect on the luciferase activity of miRGLO-Mut-CTGF (Fig. 3D).

miR-133b inhibited renal interstitial brosis and reduced renal function loss in old UUO mice
The miR-133b or NC-miRNA transfection complex was administered to UUO + miR-133b or UUO + NC-miR mice by intravenous injection, respectively, and the expression of miR-133b was detected by real-time PCR. At 7 and 14 days after UUO establishment, the expression of miR-133b was signi cantly higher in the UUO + miR-133b group than in the sham and UUO + NC-miR groups, con rming the successful transfection of miRNA-133b. The expression of miRNA-133b in the UUO + NC-miR group was signi cantly lower than that in the sham group (Fig. 6A).
PAS and Masson staining techniques were used to observe the morphology of the kidney tissue. In comparison with the sham group, the UUO + NC-miR group showed interstitial edema, tubular dilatation, interstitial in ammatory cell in ltration, and a slight increase in collagen level at 7 days after UUO operation on the ureter ligation side of the kidney. At 14 days after UUO operation, the interstitial cell population increased and collagen expression was signi cantly upregulated. In comparison with the UUO + NC-miR group, the UUO + miR-133b group showed a signi cant decrease in the level of collagen in the renal interstitium at 7 and 14 days after UUO operation (Fig. 6B).
We observed the relative area of collagen deposition in the kidney tissue and found it to be signi cantly higher in all UUO model animals than in the sham group. The value observed for UUO + miR-133b group was signi cantly lower than that reported for the UUO + NC-miR group at 7 and 14 days after UUO operation (Table 2). Legend: After Masson trichrome staining, 10 elds for each animal were observed under a light microscope (400×), and the area of each eld as well as the area of green collagen ber were measured by Image-Pro Plus software. The relative area of collagen deposition was calculated as follows: Area of green collagen ber/Area of eld of vision × 100%. *: P < 0.01 versus sham group; #: P < 0.01 versus UUO + NC-miR group.
The results of real-time PCR and western blotting showed that the mRNA and protein levels of CTGF, bronectin, Col3A1, and α-SMA in the UUO + NC-miR group were signi cantly higher and those of Ecadherin were signi cantly lower than the values reported in the sham group. Further, the changes were more signi cant at 14 days as compared with those at 7 days after UUO operation. At both 7 and 14 days after UUO operation, the mRNA and protein levels of CTGF, bronectin, Col3A1, and α-SMA signi cantly decreased and those of E-cadherin signi cantly increased in the UUO + miR-133b group as compared with the values observed for the UUO + NC-miR group. However, the mRNA and protein levels of CTGF, bronectin, Col3A1, and α-SMA in the UUO + miR-133b group were still signi cantly higher and those of Ecadherin were signi cantly lower than the values for the sham group.
The results of renal function test showed no signi cant difference in the blood urea nitrogen (BUN) and serum creatinine (sCr) levels between the three groups at day 0. At day 7 following UUO operation, BUN and sCr levels signi cantly increased in UUO mice as compared with those in sham mice. The BUN level in the UUO + miR-133b group was signi cantly lower than that in the UUO + NC-miR group. At 14 days after UUO operation, BUN and sCr levels in UUO mice were signi cantly higher than those in the sham mice and those in the UUO + miR-133b group were signi cantly lower than the values reported for the UUO + NC-miR group (Table 3).

Discussion
Tubulointerstitial brosis is a chronic and progressive process affecting the kidney tissue during aging as well as in chronic kidney disease regardless of the underlying cause [12]. Kidney brosis is characterized by the EMT of tubular epithelial cells [13] that contributes to both the destruction of the tubular epithelial compartment and accumulation of interstitial broblasts [14]. TGF-β signaling is thought to play a predominant role in this process [15]. CTGF is a direct downstream early response factor of TGF-β, also known to potentiate TGF-β signaling by directly binding TGF-β1 through its CR domain [16].
The miRNA miR-133 was rst experimentally characterized in mice, and its homologs were identi ed in several other species, including invertebrates such as the fruit y Drosophila melanogaster [17]. In the human genome, miR-133 genes include miR-133a-1, miR-133a-2, and miR-133b located on chromosomes 18, 20, and 6, respectively [18]. miR-133 is necessary for the proper development and function of skeletal and cardiac muscles, and its aberrant expression has been linked to many diseases of the skeletal and cardiac muscles. It is identi ed as a key factor in cancer development [17][18][19][20] and is also known to alleviate cardiac brosis in many animal models [8][9][10]. Lentiviral transfection of miR-133 was found to reduce renal interstitial brosis in old UUO mice [7] and renal brosis in diabetic rats [21]. In the present study, we induced the high expression of miR-133b in the kidney tissue through an intravenous injection of an miR-133b transfection complex. It was found that miR-133 downregulated the mRNA and protein expression of CTGF, bronectin, Col3A1, and α-SMA and upregulated the mRNA and protein levels of Ecadherin, thereby signi cantly alleviating renal interstitial brosis and reducing the loss of renal function in old UUO mice. We, thus, recon rmed the effect of miR-133 on geriatric renal interstitial brosis.
The effect of miR-133 on the EMT induced by TGF-β1 is controversial. miR-133 was previously shown to inhibit the EMT of HK2 cells induced by TGF-β1 [6,7]. TGF-β1 downregulated the expression of miR-133a/b in the bladder smooth muscle epithelial cells in a concentration-dependent manner, and Page 12/20 transfection with miR-133 mimics resulted in the attenuation of the TGF-β1-induced α-SMA, extracellular matrix subtype, and brotic growth factor expression [11]. Treatment of primary murine and human hepatic stellate cells with TGF-β1 resulted in a signi cant downregulation in the expression of miR-133a.
On the other hand, the overexpression of miR-133a in primary murine hepatic stellate cells was shown to decrease the expression of collagen [22]. However, miR-133b was found to be overexpressed in TGF-β1treated HK-2 cells, and miR-133b inhibition attenuated the TGF-β1-induced EMT of HK-2 cells [21]. Our experiments con rmed the TGF-β1-mediated downregulation of miR-133 expression in a concentrationand time-dependent manner, and showed that the overexpression of miR-133b signi cantly inhibited the downregulation in the mRNA and protein expression of E-cadherin as well as the upregulation in the mRNA and protein levels of α-SMA, bronectin, Col3A1, and CTGF induced by TGF-β1. The inhibitory effect of miR-133 on the EMT of HK2 cells induced by TGF-β1 was further clari ed.
Only a few studies have evaluated the effect of target genes of miR-133 on the alleviation of tissue brosis. Only a single renal brosis study con rmed Sirtuin-1 as a target of miR-133b in HK-2 cells and showed that the inhibition of miR-133b expression resulted in the attenuation of the TGF-β1-induced EMT and renal brosis through the upregulation of Sirtuin-1 expression [21]. Further, CTGF was found as a direct target of miR-133 during the EMT of cardiomyocytes [9] and bladder smooth muscle epithelial cells [11]. The overexpression of miR-133 signi cantly downregulated the mRNA and protein levels of CTGF, and CTGF overexpression could reverse the inhibitory effect of miR-133b on the TGF-β1-induced EMT of HK2 cells. The results of dual-luciferase reporter assay also con rmed that CTGF is a direct target of miR-133. Therefore, our study con rmed for the rst time that CTGF is a target of miR-133 and is involved in ameliorating renal brosis.
CTGF not only potentiates TGF-β signaling by directly binding TGF-β1 but also modi es various growth factors and cytokines. Each domain of CTGF can bind to multiple ligands, including insulin-like growth factor-1, bronectin, bone morphogenetic factors, α5β3 integrin, low-density lipoprotein receptor-related protein 1, vascular endothelial growth factor (VEGF), Wnt, integrins, heparan sulfate proteoglycan, receptor-related proteins, and epidermal growth factor receptor [22]. These cytokines can participate in kidney aging and renal brosis through various signaling pathways. Therefore, miR-133 may play different biological roles by downregulating the expression of CTGF and in uencing the processes of kidney aging and renal brosis. These effects need to be con rmed through future studies.

Conclusions
We showed that CTGF is a target gene of miR-133 involved in ameliorating renal brosis and clari ed the miR-133b-mediated inhibition of the EMT of HK2 cells induced by TGF-β1 that resulted in the alleviation of renal interstitial brosis in old UUO mice. These observations serve as a basic research evidence for the development of new drugs based on miR-133 toameliorate kidney aging and renal interstitial brosis.
Declarations by real-time PCR. C: CTGF expression was analyzed by western blotting. D: Dual-luciferase reporter experiment con rmed CTGF as a direct target gene of miR-133b.