Aggravation of acute kidney injury by mPGES-2 down regulation is associated with autophagy inhibition and enhanced apoptosis

The deletion of microsomal prostaglandin E synthase-2 (mPGES-2) does not affect in vivo PGE2 production, and the function of this enzyme remains unknown until now. This study investigated the expression and roles of mPGES-2 in LPS induced acute kidney injury (AKI) both in vitro and in vivo. We found that mPGES-2 was up-regulated in kidney of mice with LPS induced AKI. Inhibition of mouse mpges2 gene expression exacerbated LPS-induced renal dysfunction, renal tubular cell damage and apoptosis, while inhibited kidney autophagy. Further cellular experiments showed that over-expression of mPGES-2 resulted in increased autophagy and decreased apoptosis rate of renal tubular epithelial cells. In addition, treatment with autophagy inhibitor 3-methyladenine could reverse the above-mentioned results. On the contrary, interference of mPGES-2 expression by siRNA decreased autophagy level but significantly increased apoptosis of tubular epithelial cells and treatment with autophagy inducer rapamycin can reverse these results. Overall, our study shows that mPGES-2 can protect renal tubular epithelial cells by regulating autophagy levels and aggravation of acute kidney injury by mPGES-2 down regulation is associated with autophagy inhibition and enhanced apoptosis.


Results
Effects of LPS on kidney function and pathological changes of mice. The levels of serum urea nitrogen and creatinine in mice increased gradually with the prolongation of LPS treatment, showing significant difference from those of the normal control mice (Fig. 1A). The renal tissues of mice in the control group had clear structures without degeneration, atrophy, swelling and necrosis of the renal tubular epithelial cells or inflammatory infiltration. By comparison, mice treated with LPS for 12 h and 24 h exhibited marked edema, vacuolar degeneration and luminal narrowing of the renal tubular epithelial cells (Fig. 1B).

Expression of mPGES-2 in kidney of mice with LPS-induced AKI.
Western blotting results showed that mPGES-2 protein expression is very low in the kidney of the control mice, and increased significantly after LPS treatment, reaching its peak at 4 h and maintaining at levels higher than the control at 24 h of LPS stimulation (Fig. 1C). In addition, the expression of autophagy-related protein LC3B was also examined. Usually, the ratio of LC3-II/LC3-I is used to represent the autophagic level 20,21 . Our results showed that the ratio of LC3B-II/ LC3B-I increased with the increase of LPS stimulation time, reaching the peak at 8 h, maintained at the peak to 12 h and significantly decreased at 24 h of LPS stimulation (Fig. 1C). Since mice treated with LPS for 12 h exhibited markedly damage in the kidney and the highest autophagy level, this time point was used in the subsequent experiments. In order to further clarify the location of mPGES-2 protein in the kidney, immunohistochemical method was used to detect the distribution of mPGES-2 in kidney of AKI mice. The results showed that mPGES-2 was weakly expressed in the control mice. It is mainly located in renal cortex and cytoplasm of the proximal tubule epithelial cells, but little in glomeruli and renal medulla. In addition, LPS treatment significantly increased mPGES-2 expression in tubules of renal cortex and medulla, and mPGES-2 translocation was not observed in kidney of mice during AKI (Fig. 1D).

Infection efficiency of mPGES-2 gene 4-in-1 shRNA adenovirus and the confirmation of mPGES-2 gene knockdown in vivo.
The successful construction of mPGES-2 gene 4-in-1 shRNA adenovirus with green fluorescent protein (GFP) gene was determined by Sanger sequencing. To demonstrate the infection efficiency of mPGES-2 shRNA adenovirus in vivo, frozen slices of mouse kidneys were examined using fluorescent microscopy. After six days of mPGES-2 shRNA adenovirus administration, GFP were observed in the tubular cells of mouse kidneys, suggesting that mPGES-2 shRNA targeted the renal tubular cells ( Fig. 2A).
To confirm the mPGES-2 gene knockdown in vivo, the mRNA and protein of mPGES-2 in the kidney of mice were detected. The results showed that mPGES-2 mRNA and protein expression were significantly decreased after six days of mPGES-2 shRNA adenovirus administration compared with that of the control shRNA group (Fig. 2B,C) Inhibition of mPGES-2 increases mortality and renal impairment of mice with LPS-induced AKI. LPS treatment of mPGES-2 shRNA adenovirus infected mice resulted in a significant reduction in 72-h survival rate from 40% in the control mice to 10% (Fig. 3A). In addition, kidney function examination showed that serum urea nitrogen and creatinine levels were significantly higher in LPS-treated mice than those in untreated mice. Particularly, these increases were more significant in mPGES-2 shRNA adenovirus infected mice than in control mice (Fig. 3B,C). HE staining showed that control and mPGES-2 shRNA adenovirus infected mice without LPS treatment had clear and integrate renal structures. After LPS treatment, the renal tubular epithelial cells underwent edema and vacuolar degeneration and these phenomena were more sever in mPGES-2 shRNA adenovirus infected mice (Fig. 3D), showing statistically significant difference in histological scores (Fig. 3E). In addition, LPS treatment increased PGE 2 content in kidney of both control and mPGES-2 knockdown mice. However, there was no significant difference between them (Fig. 3F).
Effects of mPGES-2 knockdown on autophagy and apoptosis in kidney of mice with LPS-induced AKI. The expression of mPGES-2 was detected by western blot after mPGES-2 gene was knocked down by adenovirus infection. The results showed that the expression of mPGES-2 protein in kidney decreased significantly regardless of LPS treatment, while the expression of mPGES-2 was significantly increased in LPS-treated control mice. Then, expression levels of autophagy markers LC3/Atg8 and p62/SQSTM1 were detected by western blot. P62/SQSTM1 is a multifunctional protein that acts as a selective substrate in autophagy. It binds to ubiquitinated proteins, and forms complexes with LC3, resulting in p62 transportation into autophagolysosome and degradation. On the contrary, p62 is accumulated when autophagy is defective. Thus, p62 can be used as a marker to detect autophagic activity 22,23 . Our results showed that LPS treatment increased LC3B-II/ LC3B-I ratio and decreased p62 level in both mPGES-2 downregulated and control mice (Fig. 4A,B). But the increment of LC3B-II/LC3B-I ratio in mPGES-2 shRNA adenovirus infected mice was significantly lower than that in the control mice. These results suggest that the level of autophagy in kidney is decreased after mPGES-2 downregulation. In addition, autophagy level can also be observed using immunofluorescence 20 and transmission electron microscopy 20 . Because unbounded LC3-I is distributed in the cytoplasm and converted to LC3-II, which was localized on the autophagic membrane, after autophagy activation, under fluorescence microscopy, it is presented as bright fluorescence dots. Thus, the number of these small fluorescent dots can represent the level of intracellular autophagy. Immunofluorescence showed that autophagosome in the renal tubules was significantly increased after LPS stimulation, but this increase was significantly reduced in mPGES-2 shRNA adenovirus infected mice (Fig. 4C). Observation of typical autophagosome structure under transmission electron microscope has been the gold standard for autophagy detection 20 . Our electron microscopic results showed that LPS treatment significantly increased autophagic vacuoles in kidney. But this increase was significantly reduced in mPGES-2 shRNA adenovirus infected mice (Fig. 4D). TUNEL staining showed that LPS treatment significantly increased the number of brownish-yellow apoptotic cells in both control and mPGES-2 shRNA adenovirus infected mice. But this increase was more dramatic in mPGES-2 shRNA adenovirus infected mice than in control mice (Fig. 4E,F).

Effects of LPS on the injury of HK2 cells.
Because immunohistochemistry showed that mPGES-2 is mainly expressed in cortical renal tubular epithelial cells, renal tubular epithelial cell line HK-2 was used to examine the effects of mPGES-2 at cellular level. First, cell viability and LDH level in the medium were measured using CCK-8 assay and LDH cytotoxicity assay after 24 h of LPS treatment with different concentrations. The results showed that LPS decreased cell viability, but increased LDH level in medium in a dose dependent manner (Fig. 5A). Next, cell viability and LDH level in the medium were measured after 1000 ng/ml LPS treatment for different times. The results showed that LPS decreased cell viability, but increased LDH level in medium in a time dependent manner (Fig. 5B).
Expression of mPGES-2 in LPS-treated HK2 cells. First, HK-2 cells were treated with different concentrations of LPS for 12 h, and mPGES-2 mRNA and protein levels were detected. The results showed that the levels of mPGES-2 mRNA and protein increased gradually with LPS concentration increasing and reached significantly different level at the concentration of 1000 ng/ml (Fig. 5C). Therefore, this concentration was used in all    ScIeNtIfIc REPoRTs | 7: 10247 | DOI:10.1038/s41598-017-10271-8 still higher than the basal level at 24 h (Fig. 5D). Therefore, cells were treated with LPS for 12 h in all experiments except those indicated.

Effects of mPGES-2 overexpression on LPS-induced viability, autophagy and apoptosis of HK-2 cells.
To investigate the effect of mPGES-2 on LPS-induced viability, autophagy and apoptosis of renal tubular epithelial cells, HK-2 cells were transiently transfected with control and mPGES-2 plasmid, and expression of mPGES-2 was verified by both western blot and qRT-PCR. The results showed the levels of mPGES-2 mRNA and protein were significantly increased in HK-2 cells transfected with mPGES-2 plasmid (Fig. 6A). In addition, these cells were treated with LPS to explore the effects of mPGES-2 on LPS-induced viability, autophagy and apoptosis of HK-2 cells. The results showed that mPGES-2 overexpression significantly increased the ratio of LC3B-II/LC3B-I, but decreased the expression of p62 after LPS treatment (Fig. 6B). Similarly, immunofluorescence showed that mPGES-2 overexpression significantly increased the number of autophagosomes after LPS stimulation (Fig. 6C). In addition, mPGES-2 overexpression significantly increased cell viability and decreased LDH release (Fig. 6D,E). Flow cytometry showed that mPGES-2 overexpression significantly reduced apoptosis of HK-2 cells (Fig. 6F,G).

Effects of interfering mPGES-2 expression on LPS-induced viability, autophagy and apoptosis of HK-2 cells.
The expression of mPGES-2 was significantly inhibited in HK-2 cells transfected with mPGES-2 siRNA at both mRNA and protein levels (Fig. 7A). In contrast to mPGES-2 overexpressed HK-2 cells, mPGES-2 downregulation significantly reduced the ratio of LC3B-II/LC3B-I (Fig. 7B). Immunofluorescence showed that mPGES-2 downregulation significantly reduced the number of autophagosomes after LPS stimulation (Fig. 7C). Moreover, mPGES-2 downregulation significantly decreased cell viability, but increased LDH release (Fig. 7D,E). Flow cytometry showed that mPGES-2 downregulation significantly increased apoptosis of HK-2 cells induced by LPS (Fig. 7F,G). These results suggest that mPGES-2 can promote autophagy, inhibit apoptosis and promote proliferation of renal tubular epithelial cells, and inhibit the release of toxic substance LDH.

Effects of 3-MA and rapamycin on LPS-induced autophagy and apoptosis of HK-2 cells.
To further clarify that mPGES-2 exerts its protective effect on renal tubular epithelial cells through regulating autophagy, mPGES-2 overexpressed HK-2 cells were treated with autophagy inhibitor 3-MA, and mPGES-2 downregulated HK-2 cells were treated with autophagy-inducer rapamycin. The results showed that 3-MA treatment significantly decreased LPS-induced autophagy (Fig. 8A), but increased LPS-induced apoptosis in both control and mPGES-2 overexpressed HK-2 cells (Fig. 8B). In contrast, rapamycin treatment significantly increased LPS-induced autophagy (Fig. 8C), but decreased the ratio of apoptosis (Fig. 8D) in both control and mPGES-2 downregulated HK-2 cells. In addition, we found that treatment of 3-MA significantly decreased mPGES-2-induced expression of PI3KC3 protein, while treatment of rapamycin, an autophagy inducer, further increased LPS-induced the expression of PI3KC3 protein (Fig. 9A,B).  Fig. 1A-C). mPGES-2 overexpression significantly increased the ratio of LC3B-II/LC3B-I and reduced apoptosis of HK-2 cells. On the contrary, mPGES-2 interference significantly reduced the ratio of LC3B-II/LC3B-I and promoted apoptosis of HK-2 cells (Supplementary Fig. 1D-G). Our results suggest that mPGES-2 also ameliorates H 2 O 2 induced damage by regulating autophagy of HK-2 cells.

Discussion
At present, the biological functions of mPGES-2 in vivo are not clear. This study for the first time explored the expression and localization of mPGES-2 in endotoxemia mouse model. We found LPS could induce mPGES-2 expression and up-regulation of mPGES-2 can protect renal tubular epithelial cells by promoting autophagy and inhibiting apoptosis.
The occurrence of sepsis-induced AKI is affected by many factors. In addition to hemodynamic changes, other factors such as renal cell apoptosis, endotoxin-induced complex inflammation and immune network response, endothelial dysfunction, glomerular embolization and necrosis-induced renal tubular obstruction are all involved in its pathophysiological changes [24][25][26] . Among them, apoptosis of renal tubular epithelial cells may play an important role in the development of sepsis induced AKI 27 . The use of caspase inhibitors can not only prevent renal cell apoptosis, but also inhibit the renal tissue inflammatory response, thereby protecting LPS-induced kidney damages 28,29 . Our study showed that inhibiting mPGES-2 expression significantly increased the apoptosis rate of renal tubular epithelial cells, suggesting that mPGES-2 may play a protective role by inhibiting renal tubular cell apoptosis. We also found that inhibiting mPGES-2 expression decreased renal autophagy. As a cellular stress response, autophagy plays important roles in many diseases [30][31][32] . Increased autophagy has been observed in AKI due to various causes (e.g., ischemia-reperfusion injury, cisplatin, LPS, etc.), while knockout of autophagy-related genes aggravated kidney damages 8,33,34 . Our results also showed that decrease in autophagy in LPS-induced AKI resulted in more severe renal injury in mice. The balance of autophagy and apoptosis in the body is affected by a variety of factors and has important impacts on the prognosis of the disease. For example, enhancing macroautophagy inhibits myocardial cell apoptosis and protects against ischemia/reperfusion injury in cardiac myocytes 35,36 . Autophagy induction protect against cisplatin-stimulated tubular cell apoptosis 37 . Enhanced autophagy by rapamycin exerts a renoprotective role via inhibiting tubular cell apoptosis during renal I/R injury 38 . Usually, the dynamic balance of autophagy and apoptosis, to a certain extent, determines the survival or death of cells. Therefore, it is important to find a common upstream factor that regulates the autophagy and apoptosis. Our in   vitro results showed that in LPS-induced AKI model, mPGES-2 is involved in the regulation of autophagy and apoptosis in renal tubular epithelial cell, and can be a common upstream factor of them. It is therefore reasonable to speculate that mPGES-2 decreases the apoptosis rate of LPS-induced renal tubular epithelial cell by regulating autophagy activity.
To further demonstrate this hypothesis, we treated HK-2 cells with autophagy inhibitor 3-methyladenine (3-MA) and autophagy agonist rapamycin. The results showed that 3-MA treatment blocked autophagy induced by mPGES-2, and increased apoptosis rate of HK-2 cells. Previous studies have shown that 3-MA can block autophagy by inhibiting class III phosphoinositide 3-phosphate kinase (PI3KC3) 39 . The activity of PI3KC3 is necessary for the nucleation and assembly of membrane pool formation in early autophagy 20,40 . Combined with our results, we believe that mPGES-2 can inhibit apoptosis by increasing autophagy and may play a role in the upstream of PI3KC3. In contrast, rapamycin treatment re-induced autophagy after interfering mPGES-2 expression, and reduced apoptosis rate. Rapamycin is an immunosuppressant and a negative regulator specific to mTOR signaling pathway 41,42 . A large number of studies have shown that rapamycin can induce autophagy 43 . Our results also showed that rapamycin reversed inhibition of autophagy due to interfering mPGES-2 expression, strongly suggesting that autophagy and apoptosis induced by mPGES-2 have an upstream and downstream relationship and both of them are involved in protection of LPS-induced injury in HK-2 cells. Meanwhile, our results showed that the PI3KC3 complex was regulated by mPGES-2, which further suggested that PI3KC3 complex and/or its upstream regulatory molecules may be the targets of mPGES-2. But how mPGES-2 affects PI3KC3 pathways need to be further investigated.
To investigate whether mPGES-2 exerts protective role in other model, we stimulated HK-2 cells using H 2 O 2 . It is well known that H 2 O 2 induces cell apoptosis 44 . Our results revealed mPGES-2 also ameliorates H 2 O 2 induced apoptosis by regulating autophagy. These results support the nonspecific role of mPGES-2 in LPS or H 2 O 2 induced HK-2 cell injury models.
In conclusion, mPGES-2 down-regulation aggravates acute kidney injury by inhibiting autophagy and promoting apoptosis in mice. The protective effect of mPGES-2 is mediated, at least partially, through promoting autophagy and inhibiting apoptosis of renal tubular epithelial cells.

Materials and Methods
This study was approved by the Ethical Committee of the Animal Experimental Institute of Central South University, Changsha, China, and carried out in accordance with the approved protocol.
Animals and preparation of LPS-induced AKI mouse model. A total of 130 male C57BL/6 mice, weighing 20-25 g, from Hunan SJA Laboratory Animal Co., Ltd. were used in the experiments. They were maintained in a clean and quiet environment with relative humidity of 40~60% and light/dark cycle of 12 h/12 h at 25 °C. Among them, 30 mice were divided into 5 groups with 6 mice in each group (n = 6) and intraperitoneally injected LPS at 10 mg/kg body weight for 0 (control), 4, 8, 12 and 24 h, respectively. Under anesthesia, their eyeballs were removed for collection of blood in a clean centrifuge tube. Subsequently, mice were scarified and their left kidneys were collected for qRT-PCR, western blot and immunohistochemistry. The other 100 mice were used to establish mPGES-2 knockout mouse model as described in Section below.
Renal function testing and biochemical assays. Blood samples taken from the animals were placed still at room temperature for 2 h and sera were collected by centrifugation at 1000 g for 15 min. The renal function indexes such as urea nitrogen and creatinine were measured using automatic biochemical analyzer (Hitachi Modular System, Hitachi, Tokyo, Japan). Concentration of kidney PGE 2 was determined by enzyme immunoassay (Cusabio Life science, Hubei, China).
Observation of kidney damages by hematoxylin and eosin staining. The fresh mouse kidney tissues were fixed in 4% paraformaldehyde solution, stained hematoxylin and eosin and observed under a microscope. The degree of kidney damages was assessed using the blind method. Samples with normal kidney, renal tubular injury area <25%, 25-50%, 50-75% and 75-100% were scored 0, 1, 2, 3, and 4, respectively 45 . Immunohistochemistry. The expression of mPGES-2 in kidney was detected by immunohistochemistry Transmission Electron Microscopy. Fresh kidneys were quickly emerged into fixative for electron microscope at 4 °C for 2-4 h. After that, samples were transferred to Wuhan Goodbio technology Co., Ltd, prepared as specimens, examined and photographed under a transmission electron microscope.
Detection of renal tubular cell apoptosis using TUNEL method. The apoptotic level of kidney in paraffin sections was examined using TUNEL apoptosis detection kit (Roche) according to instructions provided by the manufacturer. The percentage of apoptosis cell was calculated as the number of apoptotic cells/total number of cells × 100% in 5-10 randomly selected fields.
Cell culture. The HK-2 cell line from ATCC was maintained in DMEM/F12 medium supplemented with 10% fetal bovine serum at 37 °C in an incubator supplemented with 5% CO 2 . Medium was renewed every 2 days.
Real time PCR. The total RNA was extracted from kidneys and HK-2 cells with TRIzol. mRNA were reverse transcribed to cDNA using PrimeScript ™ RT Master Mix (Takara, Japan). Real-time PCR was performed using cDNA as template and One Step SYBR ® PrimeScript ™ RT-PCR Kit on the Biosystems 7500 Real Time PCR System. The amplification was carried out for 40 cycles at conditions of 95 °C for 30 s, 95 °C for 5 s and 60 °C for 34 s. The primer set 5′-CAT TGC TGA CAG GAT GCA GAA GG-3′ and 5′-TGC TGG AAG GTG GAC AGT GAG G-3′ as well as the primer set 5′-GGT AGA CCT CTA TGA AGC AGC C-3′ and 5′-CAT CAC TCG CAG CAC ACC ATA C-3′ were used for amplification of mouse β-actin and mPGES-2, respectively. The primer set 5′-CAC CAT TGG CAA TGA GCG GTT C-3′ and 5′-AGG TCT TTG CGG ATG TCC ACG T-3′ as well as the primer set 5′-CCT CTA TGA GGC TGC TGA CAA G-3′ and 5′-ATC ACA CGC AGC ACG CCA TAC A-3′ were used for amplification of human β-actin and mPGES-2, respectively. Western blot. Mouse kidney and HK-2 cells were lysed using RIPA buffer plus PMSF. Proteins in the extractions were determined using BCA method. The same amount of proteins were mixed with the same volume of The membranes were blocked at room temperature for 1 h, and then incubated overnight with primary antibodies against mPGES-2 (1:1,000 dilution, cayman chemical), microtubule-associated protein light chain 3 (LC-3B) (1:1,000 dilution, CST), p62 (1:1,000 dilution, CST), PI3KC3 (1:1000 dilution, CST) and β-actin (1:2,000 dilution, Sigma-Aldrich), respectively, at 4 °C. After washed with PBS, the membranes were then incubated with corresponding secondary antibodies for 1 h at room temperature. Signals were visualized using ECL substrate and quantified using ImageJ software.
Immunofluorescence analyses. Paraffin sections were subjected to antigen repair, blocked using sheep serum for 30 min, and incubated with primary antibody against LC-3B (1:100, CST) overnight at 4 °C. After washed with PBS, the sections were incubated with FITC-labeled anti-rabbit secondary antibody (1:400 dilution, MultiSciences) at room temperature in the dark for 1 h, counterstained with DAPI, and photographed under a fluorescence microscope.
HK-2 cells were seeded on coverslips. After transfection and stimulation according to the experimental requirements, cells were washed with PBS three times, fixed with 100% methanol for 5 min, washed three times with PBS again, and blocked using sheep serum for 30 min. After that cells were stained with LC-3B antibody, FITC-labeled anti-rabbit secondary antibody and subjected to Immunofluorescence analyses as described above.
Transfection of plasmids and siRNA into HK-2 cells. pENTER eukaryotic plasmids expressing mPGES-2 and small interfering RNA (siRNA) of mPGES-2 were purchased from ViGene Biosciences and Shanghai GenePharma Co., Ltd, respectively. The plasmid was extracted using Plasmid Maxprep Kit (Vigorous Biotechnology) according to the manufacturer's instruction. Plasmid and siRNA transfections into HK-2 cells were carried out using Lipofectamine 3000 ™ (Invitrogen) according to the manufacturer's instruction.
Analysis of cell proliferation and detection of LDH level in culture supernatant. The viability of cells cultured in 96-well plates in each group was measured using CCK-8 kit (Dojindo, Japan) following the manufacturer's instruction. The culture supernatants were used to measure LDH level using LDH cytotoxicity test kit (Nanjing Jiancheng Bioengineering Institute) following the manufacturer's instructions.
Apoptosis detection using flow cytometry. The cells in each group were washed, digested and subjected to apoptosis analysis using Annexin V-FITC/PI Apoptosis Detection Kit (BD Pharmingen, CA, USA) according to the manufacturer's instructions.
Statistical analysis. GraphPad Prism version 5.02 (GraphPad Prism Software Inc, San Diego, CA) was used for data analysis. All the experiments were repeated at least three times. The quantitative data were presented as Mean ± SD. Multiple comparisons were analyzed for significant differences by using the one-way ANOVA with Tukey post hoc test. Kaplane-Meier plots were used to illustrate survival of mice between different groups, and statistical assessment was performed by the log-rank test. A value of P < 0.05 was considered statistically significant.