MiR‐138 plays an important role in diabetic nephropathy through SIRT1–p38–TTP regulatory axis

Diabetic nephropathy (DN) is the main cause of chronic kidney disease (CKD) and is one of the most common and serious complications of diabetes mellitus (DM). Sirtuin 1 (SIRT1) and tristetraprolin (TTP) are two important protective factors in DN; however, the regulatory relationship between SIRT1 and TTP, and the underneath mechanism are interesting but still unclear. Identifying the key factors that regulate SIRT1 or TTP may be of great value to the understanding and treatment of the DN. In this study, through systematic experimental methods, we found that the expression of miR‐138 was significantly upregulated in DN clinical patient samples, and our experimental results suggested that miR‐138 could bind the 3ʹ‐UTR of SIRT1 and inhibit its expression in both cultured podocytes and db/db mice kidney tissues. Furthermore, our in vitro and in vivo experiments also indicated miR‐138 could target SIRT1 and affect TTP through p38 pathway. And downregulation of miR‐138 attenuated podocyte injury and showed some extent of therapeutic effects in DN mice models. Our findings revealed that the regulatory axis of miR‐138–SIRT1–p38–TTP might play a key role in DN. We believe that these findings may be of some value for deepening the understanding of DN and may serve as a reference for future treatment of this disease.

The occurrence and progression of DN are related to podocytes injury with the participation of many factors, such as stimulation of high glucose environment, changes in hemodynamics, and involvement of inflammatory factors (Forbes & Cooper, 2013;Rask-Madsen & King, 2010). Recently, more and more evidence shows that sirtuin 1 (SIRT1), which is a key molecule in glucose, lipid, and energy metabolism, is an important protective factor of DN (Cantó et al., 2009;H.-W. Liu et al., 2019;Wang et al., 2019aWang et al., , 2019b. The identification of the key factors regulating SIRT1 may be of great value to the understanding and treatment of the disease. It has been reported that  can regulate the expression of SIRT1 and participates in nerve regeneration in mammals (C.-M. Liu et al., 2013).
However, in DN, whether miR-138 plays a role in regulating SIRT1 and participates in the pathogenesis of this disease is still not clear.
As an important anti-inflammatory protein, tristetraprolin (TTP) recently has been identified as another protective factor of DN (Guo et al., 2018;F. Liu et al., 2015). TTP is an AU-enriched region binding protein that mediates the degradation of messenger RNA (mRNA) and regulates the expressions of posttranscriptional proteins (Patial & Blackshear, 2016). Decreased TTP expression leads to an increase in the expressions of inflammatory factors, which makes the podocytes remaining in an inflammatory state for a long time, resulting in losing normal morphology and function. Whether there is a regulatory relationship between SIRT1 and TTP and the underneath mechanism are also interesting but still unclear.
In this study, we attempt to clarify the issues mentioned above through in vivo and in vitro experiments, and provide an experimental reference for further understanding the pathological mechanism of DN and finding clues to improve its diagnosis and treatment in the future.

| Human studies
Renal tissue specimens of patients were collected. All patients provided informed consent, and the study was approved by an ethics committee. In total, 20 renal tissue specimens were collected: 18 from DN patients (15 with macroalbuminuria and 3 with microalbuminuria), 3 from DM patients, and 9 control samples from healthy renal tissue around the cancerous tissue of radical nephrectomy patients. Blood and urine samples were collected from 40, 18, 13, and 17

| Animal studies
All procedures involving animals were conducted in accordance with the guidelines prescribed by the Animal Care Committee of Zhengzhou University. Mice were purchased from the Model Animal Research Institute of Nanjing University. All mice were raised in a laboratory animal room at the Medical Science Research Institute, Henan Province. Five mice were raised per cage in IVC-II independent supply air isolation cages.
All bedding, feed, and drinking water were disinfected with ultraviolet light at appropriate dose intensity with a traffic limit device.
We used mice with a mutation of the leptin receptor gene in BKS.Cg-Dock7 m +/+Lepr db /J mice (db/db mice). We purchased 5-week-old mice and allowed a 1-week adaptation period. In diabetic mice, blood glucose increases spontaneously at Weeks 5 or 6, with proteinuria and renal impairment appearing at 8-10 weeks. A total of 60 db/db mice were randomly divided into three groups (n = 20 each): db/db group, no treatment/intervention; db/db + veh (vehicle) group, nonsense sequence of miR-138 encapsulated by lentivirus were injected into the blood through the mouse tail vein; and db/db + LV (lentivirus) group, miR-138 short hairpin RNA (shRNA) encapsulated by lentivirus were injected into the blood through the mouse tail vein. The first injections were performed when the mice were 10 weeks old, and the injection dose was 40 µl, and the titer was 2 × 10 7 pfu/ml. At 10 weeks, before miR-138 shRNA injection, three mice were killed in each group and marked as B-10w. After injection, they were killed after culture for 24 h, and marked as A-10w. The second injections were performed when the mice were 14 weeks old, at the same dose and titer to achieve the desired concentration in vivo. At 8 weeks post injection, that is, at 22 weeks old, model mice and control mice were euthanized and renal tissues were obtained. The control group included db/m mice of the same age and weight (n = 20).

| Cell culture
Cells in this experiment were derived from permanent mouse kidney podocyte cell lines, which were authorized by Peter Mundel of Mount Sinai Medical School in the United States.

| Culture, proliferation, and differentiation of podocytes
A podocyte cell culture medium (containing 10% fetal bovine serum, 5.6 mmol/L glucose, 4 ng/ml mouse γ-interferon, and RPMI-1640 medium) was prepared at 33°C. Microscopic observation showed that podocyte proliferation significantly slowed 1 week later, podocytes were irregular in shape, and numerous foot processes extended into the surroundings.

| Transfection
The differentiated mature podocytes were routinely digested before transfection, and evenly inoculated into sterile six-well plates. Transfection was performed when the cells grew to 40%-50% confluence, that is, after 24 h. To mediate transfection, 4-μl Lipofectamine 2000 was gently mixed with 200-μl Opti-MEM transfection medium and incubated for 5 min at room temperature. Then, 5 -μl each of miR-138 shRNA and Scramble were mixed with 200-μl Opti-MEM transfection medium, and then incubated at room temperature for 5 min. The shRNA transfection reagent mixture was produced by mixing Lipofectamine 2000 and shRNA solutions together at room temperature for 20 min. Then, the medium in the six-well plate was discarded, the plate was washed three times with sterile phosphate-buffered saline. After that, 1.6-ml Opti-MEM transfection medium and small interfering RNA transfection reagent mixture were added, the plate was gently shaken, and then placed in an incubator at 37°C and 5% CO 2 .

| Protein extraction and WB
The culture plate was placed on ice, and the cells were lysed with radioimmunoprecipitation assay (RIPA) lysate. Then, 1-ml RIPA was supplemented with 10-µl phenylmethanesulfonyl fluoride (PMSF; 100 mM; PMSF:RIPA, 1:100), 10-μl protease inhibitor cocktail, and 10-μl phosphatase inhibitor cocktail. After complete mixing, 100-μl solution was added to each well of a six-well plate and placed on ice for 30 min. To ensure complete lysis, a cell scraper was used to scrape the cell debris from the six-well plate.
Then, the lysate containing the cell debris was transferred into a 1.5-ml centrifuge tube using a pipette. All procedures were performed on ice as far as possible. Then, 5-μl supernatant after centrifugation (4°C, 12,000 rpm, 15 min) was used to detect protein concentration using a BCA Protein Concentration Assay Kit. Rabbit anti-mouse IL-18 polyclonal, rabbit anti-mouse Desmin polyclonal, rabbit anti-mouse Nephrin polyclonal, rabbit anti-mouse Podocin polyclonal, and rabbit anti-mouse GAPDH polyclonal antibodies were purchased from Santa Cruz Biotechnology or Abcam. All the grouping blots in Western blot results were cropped from different parts of the same gel or from different gels. 2.7 | RNA extraction and real-time reverse-transcription polymerase chain reaction (RT-PCR) One milliliter Trizol lysate was added to the six-well plate, the mixture was gently mixed with a pipette, and placed horizontally on ice for 10 min, which facilitated complete contact between Trizol and cells. Chloroform was pipetted into a nonenzymatic EP tube with an enzyme-free pipette tip (1-ml Trizol:200-μl chloroform); then, this mixture was rapidly shaken for 15 s, and set on ice for 5 min before being centrifuged in a cryogenic centrifuge at 4°C and 12,000 rpm for 10 min. This divided the sample into three layers; the upper layer was an RNA layer (colorless), which was transferred into a new nonenzymatic EP tube and centrifuged repeatedly under the same conditions. Finally, 30-μl diethylpyrocarbonate (DEPC) water was added to the precipitate and gently mixed by pipetting with a sterile pipette tip. Then, the complementary DNA was reverse-transcribed, and the corresponding fragments were amplified in an RT-PCR machine. To reduce human error, three replicate wells were set for each target gene.

| Immunohistochemistry and periodic acid-Schiff (PAS) staining
Tissue samples were baked at 60°C for 30 min, dewaxed, hydrated, and placed in a plastic container containing Tris-EDTA Antigen Recover. The tissue was covered with 0.3% Triton X-100 and incubated for 15 min at room temperature. Then, it was blocked with 5% prostate-specific antigen, incubated with the primary antibody overnight at 4°C and the secondary antibody for 1 h at room temperature, and finally observed and photographed under a fluorescent microscope.

| Statistical analysis
Experiments were performed at least three times, and values reported are mean ± SD. Data were analyzed using IBM SPSS Statistics 19.0. Statistical significance was assessed using the Student's t test, one-way analysis of variance (ANOVA), the least significant difference t test, and two-way ANOVA. p < .05 were considered statistically significant.

| RESULTS
3.1 | miR-138 is highly expressed in DN patients and associated with podocyte injury In this study, we first focused on miR-138 and tried to determine whether miR-138 had changed in DN clinical case samples. This is LIU ET AL.
| 6609 an important question that needs to be answered at the most basic level, which was not clear before. To achieve this, subjects of the First Affiliated Hospital of Zhengzhou University from March 2016 to September 2017 were selected and separated into four groups: control, DM, DN1 (DN with microalbuminuria), and DN2 (DN with macroalbuminuria). We first analyzed the blood and urine indexes of each group (control, n = 40; DM, n = 18; DN1, n = 13; DN2, n = 17), and found that the changes of urinary albumin excretion rate, serum creatinine levels, and blood glucose levels were consistent with the characteristics of each group ( Figure S1A-C). Then, the results of PAS staining and electron microscopic (EM) examination also showed typical morphological changes of four groups ( Figure S1D). These data indicated that the grouping was appropriate for the present study. After further RT-PCR analysis of blood specimens, we found that the expression of miR-138 in the DM and DN groups (i.e., high-glucose environments) was significantly higher than that in the control group; moreover, we found that the expression of miR-138 increased with the severity of the disease, which was highest in DN2 group ( Figure 1a). Moreover, we found that the variation trend of podocyte-specific factor Desmin and inflammatory-related factors IL-18 in each group were consistent with that of miR-138 (Figure 1b,c). Additionally, Masson's stain of paraffin-embedded kidney specimens also showed that the degree of fibrosis in the DN group was higher than that in the DM and control groups (Figure 1d). These results suggested that miR-138 was highly expressed in DN patients and may be associated with inflammation, podocyte injury, and renal fibrosis in this disease.
3.2 | miR-138 can downregulate SIRT1 and promote podocytes injury in DN models Next, we tested whether miR-138 could target and downregulate SIRT1 in vitro and in vivo, which is an important protective factor of DN (Cantó et al., 2009;H.-W. Liu et al., 2019;Wang et al., 2019aWang et al., , 2019b. In cultured mice podocytes, we found that SIRT1 | 6611 a SIRT1-3ʹ-UTR-reporter, in which SIRT1-3ʹ-UTR was cloned downstream of the luciferase element, while SIRT1-3ʹ-UTR-mutareporter lost the predicted miR-138 binding site (Figure 2d). The dual-luciferase reporter assay showed that miR-138 could downregulate the expression of luciferase of SIRT1-3ʹ-UTRreporter but not SIRT1-3ʹ-UTR-muta-reporter (Figure 2e). In the in vivo experiments, we used 5-week-old male db/db mice as the DN model, which develop spontaneous renal impairment at 8-10 weeks of age, and male nondiabetic db/m mice were used as control ( Figure S2). Moreover, we found that the expression of SIRT1 was decreased at 22 weeks in db/db mice when compared with db/m mice, and knockdown of miR-138 by injecting shRNAexpression lentivirus caused obvious SIRT1 expression recovery ( Figure 2f). All these results suggested that miR-138 could target and downregulate SIRT1 in mice podocytes.
Additionally, in the in vitro experiments, we found that overexpression of miR-138 upregulated the expression of Desmin, which is the marker for podocyte injury (Figure 2b). In the in vivo experiments, knockdown of miR-138 could upregulate the expression of synaptopodin (SYNPO), which is a marker for differentiated podocytes and its downregulation indicated podocyte injury ( Figure 2f). Therefore, these results suggested that miR-138 might play a role in promoting podocyte injury in DN.

| miR-138 could inhibit TTP through targeting SIRT1
We further investigated the in-depth mechanism by which miR-138 targets SIRT1 to affect podocytes. Although SIRT1 has been reported to have protective effects in podocyte injury during DN, it is worthwhile to explore new downstream pathways. As an important anti-inflammatory protein, TTP recently has been identified as another protective factor of DN (Guo et al., 2018;F. Liu et al., 2015). Whether there is a regulatory relationship between SIRT1 and TTP and the underneath mechanism is also interesting but still unclear. To investigate the direct relationship between SIRT1 and TTP, we utilized a SIRT1 agonist, SRT1720, which dose-dependently promoted the expression of SIRT1 protein ( Figure 3a). As shown in Figure 3c, the levels of SIRT1 and TTP were obviously upregulated in the HG + LV group when compared with HG group. More importantly, the protein level of TTP was indeed upregulated after the addition of SRT1720 in the HG + SRT1720 group (Figure 3b), and the mRNA level of TTP was also improved compared with HG group (Figure 3c). Additionally, EX527, a specific inhibitor of SIRT1, was found to downregulate the mRNA and protein levels of TTP (Figure 3b.c). As we have found that miR-138 could inhibit SIRT1, it should be able to find that knocking down miR-138 could upregulate TTP, which was confirmed in in vivo ( Figure 3d) and in vitro experiments (Figure 5f,g). These results suggested that SIRT1 could regulate TTP, and miR-138 could inhibit TTP through targeting SIRT1.

| miR-138 could target SIRT1 and affect TTP through p38 pathway
Then, what is the intermediate key regulator of miR-138 that affects SIRT1 and then TTP? Previous studies have shown that SIRT1 reduces phosphorylated p38 (p-p38) expression in astrocytes (Greenhill, 2017;Li et al., 2017) and neural progenitors (Rafalski et al., 2013). Additionally, it is known that TTP can be regulated by the p38 MAPK pathway (Thapar & Denmon, 2013;Tiedje et al., 2016). Based on these research studies, we hypothesized that in DN, high glucose stimulation could upregulate miR-138 level and inhibit SIRT1 expression, leading to p38 phosphorylation and the inhibition of TTP, and ultimately upre- Desmin levels (Figure 5c,f) were increased in the kidney tissues of db/ db mice compared with db/m mice at age 10 weeks before lentiviral injection. As expected, SIRT1 expression increased after 12 weeks of F I G U R E 2 MicroRNA-138 (miR-138) can downregulate sirtuin 1 (SIRT1) and might promote podocytes injury in diabetic nephropathy (DN). (a) The expression of SIRT1 in cultured mice podocytes after miR-138 mimics or short hairpin RNA (shRNA) introduction was tested by Western blot analysis. (b) The expressions of SIRT1 and Desmin in cultured mice podocytes in high glucose (HG) or normal glucose (NG) condition after miR-138 mimics transfection, tested by Western blot analysis. (c) The expression of SIRT1 in cultured mice podocytes after miR-138 mimics or inhibitor introduction was tested by Western blot analysis. (d) Schematic diagram of the plasmid vectors of SIRT1-3ʹ-UTRreporter and SIRT1-3ʹ-UTR-muta-reporter. The SIRT1-3ʹ-UTR was cloned downstream of the luciferase element in SIRT1-3ʹ-UTR-reporter. SIRT1-3ʹ-UTR-muta-reporter lost the predicted miR-138 binding site. (e) Results of the dual-luciferase reporter assay using the plasmid vectors of SIRT1-3ʹ-UTR-reporter SIRT1-3ʹ-UTR-muta-reporter and miR-138 mimics; *p < .05. (f) Immunofluorescence results of the expression and position of SIRT1 and synaptopodin (SYNPO) in glomeruli of mice models (scale bar = 20 μm). LV: lentivirus expressing miR-138 shRNA, which knockdown of miR-138; w, week; db/db: diabetic db/db mice (BKS.Cg-Dock7 m +/+Lepr db /J mice), which develop spontaneous hyperglycemia at 5-6 weeks of age followed by proteinuria and renal impairment at 8-10 weeks of age; db/m: nondiabetic mice, n = 20 in each group. The grouping blots in Western blot results were cropped from different parts of the same gel or from different gels lentiviral intervention (Figure 5g). Furthermore, we found that knockdown of miR-138 could decrease p38 phosphorylation (Figure 5g,i), and increase TTP expression (Figure 5g,h) in db/db mice. In addition, the expression of TNF-α was downregulated (Figure 5g,j), and the expression of Nephrin in podocytes was upregulated (Figure 5b,g), suggesting that the damage of podocytes was reduced. All these data indicated that miR-138-SIRT1-p38-TTP regulatory axis existed in podocytes and played a key role in podocyte injury in DN.

| Downregulation of miR-138 showed therapeutic effects in DN mice models
According to the above experimental results, miR-138 could target and downregulate SIRT1, and miR-138 could also downregulate TTP through SIRT1-p38 pathway. As SIRT1 and TTP are key protective factors in DN, negative intervention of miR-138 may have a certain therapeutic effect on DN. So, we further tested the related indexes of DN model mice treated with miR-138 shRNA virus, and found that downregulation of miR-138 had no significant effect on blood glucose and body weight in DN model mice (Figure 6a,b), but significantly reduced proteinuria and serum creatinine (Figure 6c,d). PAS staining and EM results also showed therapeutic effect after miR-138 shRNA virus treatment (Figure 6e). Furthermore, immunohistochemistry assay showed that downregulation of miR-138 suppressed podocyte injury marker Desmin, but increased the expression of Podocin, a marker for differentiated podocytes ( Figure S2).

| DISCUSSION
In this study, we first found that the expression of miR-138 was significantly upregulated in DN clinical patients' samples, and then verified miR-138 could bind the 3ʹ-UTR of SIRT1 and inhibit F I G U R E 3 MicroRNA-138 (miR-138) could inhibit tristetraprolin (TTP) by targeting sirtuin 1 (SIRT1). (a) Western blot results of the protein level of SIRT1 in cultured mice podocytes after treatment of SRT1720 in different concentrations. SRT1720 is a specific SIRT1 agonist. (b) Western blot results of the protein levels of TTP and Podocin in cultured mice podocytes after treatment of SRT1720 and EX527. EX527 is a specific inhibitor of SIRT1. (c) Reverse-transcription polymerase chain reaction results of RNA expression levels in cultured mice podocytes after different treatment. U6 or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were used as endogenic control. *p < .05 vs. NG; The development of DN is closely related to podocyte injury (Koshikawa et al., 2005;Zhou et al., 2016). However, podocytes are terminally differentiated cells that cannot regenerate after injury, which is an irreversible step in the development of DN.
Our findings revealed that the regulatory axis of miR-138-SIRT1-p38-TTP might play a key role in podocytes injury of DN. These results might be of potential value in deepening the understanding of the molecular mechanism of podocyte injury in DN and proposing targeted interventions. Of course, our research is still relatively simple, and more in-depth and systematic large sample research is needed in the future. SIRT1 is an important protective factor in DN, which can regulate metabolism and modulate metabolic diseases such as diabetes (Cantó et al., 2009;Lagouge et al., 2006;H.-W. Liu et al., 2019;Wang et al., 2019aWang et al., , 2019b. Cellular studies have shown that SIRT1 modulates fat accumulation, regulates mitochondrial biogenesis, and activates fatty acid oxidation. Mouse studies have revealed the importance of physiological effects of SIRT1, including its effect on metabolism during caloric restriction. F I G U R E 4 MicroRNA-138 (miR-138) could inhibit tristetraprolin (TTP) by targeting sirtuin 1 (SIRT1) in vitro. (a) Western blot results of the protein levels in cultured mice podocytes after treatment of miR-138 short hairpin RNA (shRNA). (b-e) Grayscale analysis results of the Western blot results in (a); *p < .05. (f) Schematic display of miR-138-SIRT1-p38-TTP regulatory axis. In diabetic nephropathy (DN), high glucose stimulation could upregulate miR-138 level, which inhibited SIRT1 expression, SIRT1 inhibited the phosphorylation of p38 to inhibit TTP. Both SIRT1 and TTP are protective factors that reduce podocytes injury in DN. The grouping blots in Western blot results were cropped from different parts of the same gel or from different gels.
Transgenic mice overexpressing SIRT1 are protected from some pathological conditions including insulin resistance and glucose tolerance (Cantó et al., 2009). The identification of key factors regulating SIRT1 may be of great value in understanding and treating the disease. MicroRNAs play an important role in the control of renal function (Putta et al., 2012), with more than 16 miRNAs modulating SIRT1 expression (Mitomo et al., 2008).
However, important miRNAs that play an important role in DN through regulating SIRT1 are still not fully understood. Here, we showed that miR-138 was upregulated under high glucose stimulation, and it could negatively regulate SIRT1 in podocytes and DN model mice. Further experiments indicated miR-138 could be a potential target for DN therapy; however, new SIRT1 regulating factors still need to be identified in the future. Given the key role of SIRT1 in many fields, such as stem cells (Igarashi & Guarente, 2016), metabolism (Arumugam & Kennedy, 2018), and cancer (Brooks & Gu, 2009), miR-138 may play a role in a broader area, which, of course, requires more research support.
Increased DN inflammation is associated with downregulation of the RNA-binding protein tristetraprolin (TTP), also known as zinc finger protein 36 homolog (ZFP36). TTP has been reported to regulate the stability of multiple target mRNAs, including pro-inflammatory F I G U R E 5 MicroRNA-138 (miR-138) could inhibit tristetraprolin (TTP) by targeting sirtuin 1 (SIRT1) in vivo. (a) Schematic diagram of miR-138 treatment in diabetic db/db mice. Db/db mice were divided into three groups: db/db mice, vehicle short hairpin RNA (shRNA) db/db mice, and db/db + LV mice (n = 20 in each group). miR-138 downregulated lentivirus (LV) was injected into the tail vein of db/db mice at 10 and 14 weeks. (b) Western blot results of the protein levels in db/m mice (n = 20, nondiabetic) or db/db mice treated with miR-138 downregulated lentivirus (LV) as shown in (a). (c) Western blot results of the protein levels in db/m mice or db/db mice before being treated with miR-138 downregulated lentivirus (at 10 weeks). (d-f) Grayscale analysis results of the Western blot results in (c). *p < .05. (g) Western blot results of the protein levels in db/m mice or db/db mice after treatment of miR-138 downregulated lentivirus (at 22 weeks). (h-j) Grayscale analysis results of the Western blot results in (f). *p < .05. The grouping blots in Western blot results were cropped from different parts of the same gel or from different gels.
Therefore, as an important anti-inflammation protein, TTP not only plays an important role in DN, but also plays a key role in many other diseases. Research on the factors and mechanisms of TTP regulation may have extensive application value in various fields. Here, our in vitro and in vivo experiments also indicated that miR-138 could target SIRT1 and affect TTP through p38 pathway. Our findings revealed that the regulatory axis of miR-138-SIRT1-p38-TTP might plays a key role in DN. These findings may be of some value for DN understanding and future treatment. Whether this regulatory axis also plays a role in other diseases and bioprogress remains to be systematically investigated in the future.