MicroRNA-140-5p ameliorates the high glucose-induced apoptosis and inflammation through suppressing TLR4/NF-κB signaling pathway in human renal tubular epithelial cells

Abstract Hyperglycemia-induced renal tubular cell injury is thought to play a critical role in the pathogenesis of diabetic nephropathy (DN). However, the role of miRNAs in renal tubular cell injury remains to be fully elucidated. The aim of the present study was to investigate the role and mechanisms of miRNAs protecting against high glucose (HG)-induced apoptosis and inflammation in renal tubular cells. First, we analyzed microRNA (miRNA) expression profiles in kidney tissues from DN patients using miRNA microarray. It was observed that miRNA-140-5p (miR-140-5p) was significantly down-regulated in kidney tissues from patients with DN. An inverse correlation between miR-140-5p expression levels with serum proteinuria was observed in DN patients, suggesting miR-140-5p may be involved in the progression of DN. HG-induced injury in HK-2 cells was used to explore the potential role of miR-140-5p in DN. We found that miR-140-5p overexpression improved HG-induced cell injury, as evidenced by the enhancement of cell viability, and inhibition of the activity of caspase-3 and reactive oxygen species (ROS) generation. It was also observed that up-regulation of miR-140-5p suppressed HG induced the expressions of pro-inflammatory cytokines, such as tumor necrosis factor-α (TNF-α), interleukin (IL)-1β and IL-6 in HK-2 cells. In addition, TLR4, one of the upstream molecules of NF-κB signaling pathway, was found to be a direct target of miR-140-5p in the HK-2. Moreover, the HG-induced activation of NF-κB signaling pathway was inhibited by miR-140-5p overexpression. These results indicated that miR-140-5p protected HK-2 cells against HG-induced injury through blocking the TLR4/NF-κB pathway, and miR-140-5p may be considered as a potential prognostic biomarker and therapeutic target in the treatment of DN.


Introduction
Diabetic nephropathy (DN) is one of the most relevant microvascular complications of diabetes, which is the leading cause of end-stage renal disease throughout the world [1]. In China, DN has become a major public health problem, bringing huge social and economic burdens to individuals, families and society [2]. Although renal tubular cell injury has been considered as the most lethal attributes of DN, the mechanism underlying is still limited.
It is known that renal tubular cell injury is one of the main characteristics of DN and increasing evidences demonstrate the involvement of oxidative stress, inflammation and cell apoptosis during renal tubular cell injury [3]. Hyperglycemia triggers the generation of excessive reactive oxygen species (ROS), which leads to increased apoptosis of renal tubular epithelial cells, resulting in tubular injury [4,5]. Additionally, abnormal inflammation can directly destroy renal architecture, along with increased levels of inflammatory cytokines, such as interleukin (IL)-1β, tumor necrosis factor-α (TNF-α) and monocyte chemoattractant protein-1 (MCP-1) [6,7]. Therefore, therapeutic strategies that inhibit renal tubular epithelial cells apoptosis and inflammation are urgently required.
MicroRNAs (miRNAs) are ∼21 nucleotides, single-stranded, noncoding RNAs which induce mRNA cleavage, or translational repression by base-pairing with the 3 -UTR region of target mRNAs [8]. Cumulative studies have shown that miRNAs are involved in high glucose (HG)-mediated apoptosis in various types of cells. For example, inhibition of miR-183 prevented HG-elicited apoptosis by repressing peroxiredoxin 3 (PRDX3) expression in retinal pigment epithelial (RPE) cells [9]. Lin et al. [10] have shown that miR-199a-5p mediated HG-induced ROS production and apoptosis in pancreatic β-cells by targeting SIRT1. MiR-29c has been related to podocyte apoptosis by targeting Sprouty homolog 1 [11]. Notably, several studies have shown that various functions in renal tubular cell such as proliferation, cell adhesion and epithelial-mesenchymal transition (EMT) are finely regulated by miRNAs. For example, Tian et al. [12] have reported that up-regulated miR-485 suppressed apoptosis of renal tubular epithelial cells in mice via regulating the TGF-β-MAPK signaling pathway by inhibiting RhoA expression. However, few studies have been reported on whether miRNAs are involved in the regulation of hyperglycemia-induced renal tubular cell apoptosis and inflammation in DN disease.
In the current study, the miRNA expression profile was examined in kidney tissues from patients with DN. Furthermore, we established an HG-induced renal tubular cell injury model and investigated the role of miRNA-140-5p (miR-140-5p)/TLR4/NF-κB pathway in preventing HG-induced renal tubular cell injury.

Tissue samples
Peripheral blood samples were obtained from 20 patients with DN and 20 healthy controls by standard protocol at Department of Neurosurgery, 150th Central Hospital of Chinese People's Liberation Army from January 2017 to June 2018. The kidney tissues were from patients diagnosed with DN or kidney carcinoma without diabetes as the control. All patients with DN were diagnosed based on the presence of diabetes, massive proteinuria, and other histological changes typical of DN. All experimental protocols were approved by the Ethics Committee of the 150th Central Hospital of Chinese People's Liberation Army. Informed consent was obtained from all patients.

miRNA microarray
Total RNA was isolated from three peripheral blood samples from patients with DN and three peripheral blood samples from healthy controls by miRNeasy isolation kit (Qiagen, Milan, Italy) according to the manufacturer's protocol. The RNA quantity was assessed by NanoDrop ND-1000 spectrophotometry (Thermo Fisher Scientific, Inc., Waltham, MA, U.S.A.). Total RNA (200 ng) was labeled using the miRCURY Hy3/Hy5 Power Labeling kit and hybridized on the miRCURY™ LNA array (v.16.0; Exiqon A/S, Copenhagen, Denmark) according to the manufacturer's protocol. Scanned images from Axon GenePix 4000B microarray scanner (Axon Instruments, Foster City, CA, U.S.A.) were imported into the GenePix Pro6.0 program (Axon Instruments) for grid alignment and data extraction. The miR-NAs with intensities ≥50 were used to calculate a normalization factor in all samples. The heatmap of the 50 miRNAs with the most marked differences was created using a method of hierarchical clustering by GeneSpring GX, version 7.3 (Agilent Technologies, Inc.).

MTT assay
Cell viability of HK-2 was evaluated 24 h after exposure to the various treatments. Briefly, 20 μl of MTT solution (5 mg/ml, Sigma, St. Louis, MO, U.S.A.) was added to each well and incubated for 4 h. Absorbance was measured at 560 nm by a microplate reader (Bio-Tek Instruments, Germany).

Flow cytometry assay
Apoptosis of HK-2 cells was evaluated 24 h after exposure to the various treatments. HK-2 cells (1 × 10 6 ) were digested with trypsin, harvested and washed in ice-cold PBS, and stained in 500 μl binding buffer with 5 μl PI and 5 μl AnnexinV-FITC (Nanjing KeyGen Biotech Co. Ltd., Nanjing, China) for 20 min in the dark. The fluorescence signals were collected by EPICS XL-MCL FACScan (BectoneDickinson, Mountain View, CA, United States) and then analyzed by FlowJo 8.7.1 software (Ashland, OR).

Caspase-3 activity assay
The caspase-3 activity in cell lysates was determined after exposure to the various treatments using a Caspase-3 Activity Assay kit (Beyotime Institute of Biotechnology), according to the manufacturer's protocol.

Measurement of ROS
ROS production in chondrocytes was measured using 2 ,7 -DCF diacetate (DCFH-DA; Sigma-Aldrich; EMD Millipore) as previously reported [14]. After exposure to the various treatments, HK-2 cells in six-well plate were stained with 20 μmol/l DCFH for 30 min at 37 • C. After that, cell nucleus was stained using DAPI staining. Cells were captured using a fluorescence microscope (IX-81; Olympus Corp., Shinjuku, Tokyo, Japan). The fluorescence intensity was analyzed by using the fluorescence plate reader (Titertek Plus MS 212, ICN, Eschwege, Germany) at Ex/Em = 488/525 nm. Fold-increases in ROS levels were determined by comparison with the control group.

Detection of malonaldehyde, superoxide dismutase and glutathione peroxidase
After designated treatment, the HK-2 were lysed using lysis buffer (Beyotime Biotechnology), the lysates were collected for the detection of superoxide dismutase (SOD) and malondialdehyde (MDA) and glutathione peroxidase (GPx). Then, the kits of SOD (cat no.S0103), MDA (cat no.S0131) and GPx (cat no.S0056) (Beyotime Biotechnology, Shanghai, China) were used to examine the level of SOD, the content of MDA and the activity of GPx according to the manufacturer's instructions.

ELISA
After exposure to the various treatments, the supernatant was carefully collected by centrifugation at 2000 rpm for 20 min. The concentrations of IL-6, IL-1β and TNF-α levels were analyzed by using IL-6 (cat no. p1330), IL-8 (cat no. p1640) and TNF-α (cat no. pt518) ELISA kits according to the kit instructions. All ELISA kits were purchased from Beyotime Biotechnology, Shanghai, China.

NF-κB activity assay
HK-2 cells (1 × 10 6 cells/well) were plated in six-well tissue culture plates for 24 h. Then, cells were transfected with 2.5 μg of an NF-κB reporter luciferase construct. Six hours later, cells were washed and then treated with miR-140-5p mimics for another 24 h, followed by 10 ng/ml HG for 24 h. Cells were then washed in PBS and harvested in 500 μl of 1× passive lysis buffer. Luciferase was quantified using Promega luciferase assay kit on a luminometer. Experimental values were recorded relative to untreated control samples.

Statistical analysis
Data were presented as mean + − S.D. GraphPad Prism 5 software (GraphPad Software, Inc., La Jolla, CA, U.S.A.) was used to perform all the statistical analyses. One-way analysis of variance followed by Tukey's post-hoc test was applied to compare differences between multiple groups. For correlation of miR-140-5p and TLR4 expression, the data were analyzed using Spearman's correlation. P≤0.05 was considered as statistically significant.

miR-140-5p was down-regulated in kidney tissues from patients with DN
First, microarray was performed to compare the miRNA patterns in kidney tissues from DN patients and healthy controls. The data identified 50 differentially expressed miRNAs (21 were up-regulated and 29 were down-regulated) ( Figure 1A). Among the aberrantly expressed miRNAs, miR-140-5p was selected for further investigation as its expression level was one of the lowest miRNAs in the DN patient group. Consistent with our results, miR-140-5p was also found to be down-regulated in kidney tissues from patients with DN [15], suggesting that miR-140-5p may be involved in the progression of DN. In addition, several studies have shown that miR-140-5p improved many types of cell injured models through regulation of inflammation and apoptosis [16][17][18][19][20][21]. However, whether miR-140-5p has a protective effect against hyperglycemia-induced tubular cell injury in DN remains to be elucidated. Thus, we selected miR-140-5p for further investigation. To further validate the expression pattern of miR-140-5p obtained from miRNA microarray assay, qRT-PCR was performed to detect miR-140-5p in ten kidney tissues from patients with DN and ten kidney tissues from healthy controls by qRT-PCR. Compared with the control group, miR-140-5p was significantly down-regulated in the kidney tissues from the patients with DN ( Figure 1B). It was also observed that the expression of miR-140-5p was markedly down-regulated in peripheral blood samples from the patients with DN, compared with that in the control group, suggesting that miR-140-5p may serve as a promising diagnostic marker for DN ( Figure 1C). More importantly, an inverse correlation between miR-140-5p levels and proteinuria was also observed in patients with DN ( Figure 1D). All data indicate that miR-140-5p may be involved in the pathogenesis of DN.

miR-140-5p is down-regulated in an HG-induced injury model of HK-2
To further examine the role of miR-140-5p in DN, we constructed an in vitro model through HG treatment in HK-2 cells, which is widely used for the research of DN [22]. Following treatment of the HK-2 cells with 30 mM HG at different times, the expression of miR-140-5p declined in a dose-dependent manner compared with control group, which was consistent with the results in the clinical samples ( Figure 2A). Subsequently, we transfected miR-140-5p mimics and miR-140-5p inhibitor into the cultured HK-2 cells and examined the effects on cell viability and apoptosis. The results showed that miR-140-5p was markedly increased (decreased) after miR-140-5p mimics (miR-140-5p inhibitor) transfection in HK-2 cells ( Figure 2B). Subsequently, the cell viability, caspase-3 activity, and the expression of apoptosis-associated proteins were evaluated. MMT assay showed that the cell viability was gradually decreased after HG stimulation at different times, while overexpression of miR-140-5p significantly attenuated the inhibitory effects of HG on the viability of HK-2 cells ( Figure 2C). Moreover, we confirmed that HG obviously elevated the activity of caspase-3 and the level of cleaved-caspase-3, compared with the control group in HK-2 cells, however, HG-induced elevation was significantly inhibited by overexpression of miR-140-5p ( Figure 2D,E). Taken together, these data indicated that the overexpression of miR-140-5p protected the HK-2 cells against HG-induced apoptosis, suggesting that miR-140-5p may be a key protective factor in DN.

Overexpression of miR-140-5p attenuated the ROS generation in HG-induced HK-2 cells
It was suggested that ROS contributes to tubular injury via the induction of renal tubular epithelial cells apoptosis in DN [23]. Thus, inhibition of ROS may be an effective therapeutic strategy for protection against HG-induced tubular injury [24]. Thus, to determine whether miR-140-5p attenuated the HG-induced renal tubular injury through suppression of the ROS generation, DCFH-DA assay was used to analyze ROS production of HK-2 cells. The results indicated that HG gradually increased the ROS production time-dependently, while overexpression of miR-140-5p significantly decreased ROS production caused by HG in HK-2 cells ( Figure 3A). In addition, the commonly used biomarkers of oxidative stress including MDA, SOD and GPx [25] were employed for the assessments of miR-140-5p on oxidative stress. As show in Figure 3B-D, the content of MDA was significantly increased, but the activities of SOD and GPx were notably decreased after HG stimulation compared with that in control group. However, overexpression of miR-140-5p abolished HG-induced oxidative stress in HK-2 cells. These data indicated that miR-140-5p up-regulation protected the HK-2 against HG-induced ROS production.

Overexpression of miR-140-5p attenuated the HG-induced inflammatory response in HK-2 cells
Increasing evidence supports that inflammation is engaged in the DN pathogenesis [26]. Therefore, we also investigated the effect of miR-140-5p up-regulation on the HG-induced inflammatory response in HK-2 cells. As shown in Figure 4A-C, the levels of pro-inflammatory cytokines, such as IL-6, IL-8 and TNF-α were significantly increased after HG stimulation, whereas the increased pro-inflammatory cytokines induced by HG was markedly reduced by overexpression of miR-140-5p. These data suggest that miR-140-5p up-regulation inhibited HG-induced inflammatory response in HK-2 cells. constructs containing the TLR4 wild-type or mutated 3 -UTRs and miR-140-5p mimics, mimics NC, miR-140-5p inhibitor or inhibitor NC, and then the luciferase activity was analyzed using the dual-luciferase reporter assay system. Data represent the mean + − SD of three independent experiments. **P<0.01 vs mimics NC; ## P<0.01 vs inhibtor NC. (F) HK-2 cells were transfected with miR-140-5p mimics, mimics NC, miR-140-5p inhibitor or inhibitor NC for 24 h and then exposed to HG for 24 h, the expression of TLR4 was measured by qRT-PCR. Data represent the mean + − SD of three independent experiments. *P<0.05, **P<0.01 vs control group; ## P<0.01 vs HG+ mimics NC; && P<0.01 vs HG + inhibitor NC.

TLR4 was a direct target of miR-140-5p in HK-2 cells
Considering the protective roles of miR-140-5p in HG-induced HK-2 cell injury, it may be beneficial to identify the downstream targets of miR-140-5p. Using TargetScan 7.0 (targetscan.org/) and miRanda (microrna.org/). TLR4, one of the upstream molecules of NF-κB signaling pathway, was identified as a potential target of miR-140-5p ( Figure 5A). We also measured the expression of TLR4 in kidney tissues from ten patients with DN and ten healthy controls. As shown in Figure 5B, TLR4 was significantly up-regulated in the kidney tissues from the patients with DN, compared with that in the control group. Moreover, an inverse relationship between miR-140-5p expression levels and TLR4 in kidney tissues was observed ( Figure 5C). To further identify whether the TLR4 levels was regulated by miR-140-5p, the effect of miR-140-5p mimics and inhibitor on the protein levels of TLR4 in HK-2 cells were examined by Western blot. As shown in Figure 5D, the protein expression of TLR4 was markedly down-regulated after miR-140-5p mimics transfection, but up-regulated by miR-140-5p inhibitor transfection. To experimentally validate whether TLR4 was a direct target of miR-140-5p, a luciferase reporter assay was performed. As shown in Figure 5E, the luciferase activity of the TLR4-3 UTR wt reporter plasmid was significantly reduced in HK-2 cells transfected miR-140-5p mimics, but increased in HK-2 cells transfected miR-140-5p inhibitor, compared with NC group. However, the luciferase activity of TLR4-3´UTR mut reporter plasmid showed no significant change ( Figure 5E), suggesting that miR-140-5p regulated TLR4 expression by binding TLR4-3 UTR. These results demonstrated that TLR4 is a direct target of miR-140-5p. In addition, in the HG-treated HK-2 cells, the results also showed that the mRNA level of TLR4 was markedly up-regulated after HG-stimulation and overexpression of miR-140-5p attenuated the promoting effect of HG on the expression of TLR4, while miR-140-5p inhibitor enhanced this effect of HG ( Figure 5F). These results indicated that miR-140-5p may inhibit renal tubular epithelial cell apoptosis and inflammatory response by targeting TLR4 in HG-induced cell injury model.

miR-140-5p blocked the HG-induced activation of NF-κB pathway
Since TLR4 is one of the upstream molecules of NF-κB signaling pathway, which is associated with the inflammatory response [27,28], further experiments were designed to examine whether miR-140-5p affects the activation of NF-κB pathway in HG-treated HK-2 cells. Levels of IκBα, p-IκB-α and nuclear p-p65 were assessed by performing Western blot. The results showed that the level of p-IκB-α and nuclear p-p65 was obviously increased after HG stimulation ( Figure 6A). However, the increased expressions of p-IκB-α and nuclear p-p65 were markedly reduced by miR-140-5p overexpression in HK-2 cells. In addition, it was also found that the NF-κB activity was markedly increased after HG stimulation compared with control group, whereas this promoting effect was attenuated by miR-140-5p overexpression ( Figure 6B). All data indicate that miR-140-5p may prevent HG-induced cell injury through blocking the activation of NF-κB pathway.

Discussion
In the present study, we found that miR-140-5p was significantly down-regulated in kidney tissues and the peripheral blood of patients with DN, and miR-140-5p level was inversely correlated with proteinuria. Using an HG-induced HK-2 cell injury model, miR-140-5p overexpression inhibited HG-induced apoptosis, ROS generation, as well as inflammatory response through inhibition of the TLR4/NF-κB signaling pathway. These data indicated that miR-140-5p/TLR4/NF-κB axis may be an effective therapeutic target for DN patients.
Increasing evidences have demonstrated that miRNAs play a key role in the pathogenesis of DN [29,30]. For example, Wang et al. [31] found that miR-424 was down-regulated in renal tissues of DN rats and up-regulation of miR-424 obviously improved the symptoms in DN rat models by targeting Rictor. Yu et al. [32] showed that miR-370 was overexpressed in a rat model of DN and up-regulation of miR-370 promoted mesangial cell proliferation and extracellular matrix (ECM) by suppressing CNPY1 in DN rat models. Yao et al. [33] demonstrated that miR-874 was markedly down-regulated and miR-874 overexpression dramatically attenuated the inflammatory response in DN through targeting TLR4 in rats. These findings suggest that miRNAs maybe a potential therapeutic target for DN treatment. In our study, using an miRNA microarray assay, we found that miR-140-5p were significantly down-regulated in kidney tissues of patients with DN. Moreover, miR-140-5p levels were inversely correlated with proteinuria levels in DN patients. These results imply the potential role of miR-140-5p in human DN.
Multiple factors have been implicated in the progression of DN, but inflammatory cytokines and apoptosis in renal tubular cells may be critical in this disease [6, 34,35]. Hyperglycemia leads to increased inflammatory cytokines of renal tubular epithelial cells and these inflammatory cytokines, such as IL-1, IL-6, IL-18, and TNF-α can directly damage renal tubular cells [36]. In addition, the influence of glucose-induced ROS production in the apoptosis of tubular cells has been described, which finally lead to DN progression [37]. Notably, several studies have shown that miRNAs are implicated in the inflammatory response and apoptosis of renal tubular cells. For example, Huang et al. [22] found that inhibition of miR-125b improved HG-induced HK-2 cell injury through blocking ROS formation and apoptosis. Yang et al. [38] showed that miR-374a suppressed the inflammatory response in HK2 cells through negative regulation of MCP-1 expression. However, whether miR-140-5p has anti-inflammatory and anti-apoptotic activities in DN remains unknown.
It has previously been reported that miR-140-5p is involved in cellular anti-inflammatory and anti-apoptotic processes in various types of cell models [39]. For example, Yang et al. [18] have found that miR-140-5p overexpression reduced the levels of inflammation in the in vitro model of acute lung injury (ALI) via blocking the TLR4/MyD88/NF-κB signaling pathway. Another study reported the inhibitory effects of miR-140-5p on cellular oxidative stress in acute kidney injury (AKI) by activating Nrf2/ARE pathway [40,41]. Based on the above researches, we hypothesized that miR-140-5p may affect the progression of DN through the regulation of renal tubular cell apoptosis, oxidative stress and inflammatory response. In the present study, using HG-induced injury model, we found that treatment of HK-2 cells with HG significantly decreased the expression of miR-140-5p in a time-dependent manner, which provided support for the possible role of miR-140-5p in HG-induced cell injury. Moreover, it was observed that up-regulation of miR-140-5p suppressed HG-induced apoptosis, ROS generation and inflammatory response, suggesting the protection role of miR-140-5p against HG-induced cell injury.
TLR4, an important regulator of the NF-κB signaling pathway, has been shown to regulate cell apoptosis, inflammatory response and oxidative damage of renal tubular epithelial cell in hyperglycemia [42,43]. Notably, a previous study showed that miR-140-5p inhibits pulmonary inflammatory response in ALI by targeting TLR4 [18]. However, whether TLR4 mediated the inhibitory effects of miR-140-5p on renal tubular cell apoptosis, oxidative stress and inflammatory response remained to be elucidated. In the present study, TLR4 was confirmed as a direct target gene of miR-140-5p in HK-2 cells. TLR4 is an important regulator of the NF-κB signaling pathway and the latter has been implicated in DN in recent years [44,45]. For example, Wang et al. [46] showed that umbelliferae (Umb) improved renal function through suppressing inflammatory response by blocking the activation of NF-κB pathway in a streptozotocin (STZ)-induced DN rat model. Zhu et al. [47] reported that Berberine ameliorated DN through relieving STZ-induced renal injury, inflammatory response via inactivating TLR4/NF-κB pathway in rats. To determine whether miR-140-5p suppressed the NF-κB signaling pathway by targeting TLR4 in HG-induced HK-2 injury model, the effects of miR-140-5p on the expressions of key kinases in NF-κB pathway were examined. The data showed that the NF-κB signaling pathway was induced in HG-treated HK-2 cells, as assessment by the expressions of key kinases in NF-κB pathway, and the overexpression of miR-140-5p blocked the activity of this pathway induced by HG, suggesting that the protective effect of miR-140-5p on HG-induced HK-2 cell injury may be mediated by the TLR4/NF-κB signaling pathway.
However, there are still some limitations in the present study. For example, multiple sources of ROS, including mitochondria, NADPH oxidases and NOS uncoupling, contribute to DN [48][49][50]. Interestingly, some reports indicate that ROS produced by mitochondria play a critical role in hyperglycemia-induced diabetic vascular complications [51,52]. In this study, we only detected global cellular ROS but not mitochondrial ROS. In future studies, we will distinguish the specific mechanism of ROS production. Moreover, although we elucidated the role and mechanisms of miR-140 protecting against HG-induced injury in renal tubular cells and obtained partial data from kidney tissues of DN patients, we did not further validate its therapeutic effects in mouse models. In future, we will further confirm if elevating miR140-5p in DN mice in vivo would have a therapeutic effect against DN.
In conclusion, we found overexpression of miR-140-5p attenuated HG-induced renal tubular epithelial cell injury by blocking the TLR4/NF-κB signaling pathway. These findings indicate that enhancing the expression of miR-140-5p may be a potential therapeutic approach for the treatment of DN.