C57BL/6J male mice aged 8–10 weeks (20–25 g) were purchased from Hunan Sleek Jingda Experimental Animal Co., Ltd. Mice were housed in a temperature-controlled room (25 ± 2°C) with relative humidity of 40∼60% on a 12 h light/dark cycle during the study. All animal experiments were approved by the Experimental Animal Ethics Committee of Central South University. The experimental mice were acclimatized for 7 days before administration of a 10 mg/kg dose of LPS or control saline.

For analysis of 15-PGDH expression in LPS-induced AKI, 12 mice were divided into four groups with three mice in each group (n = 3), and each mouse was administered intraperitoneally with 10 mg/kg body weight LPS (from Escherichia coli 0111:B4, Sigma-Aldrich, St. Louis, MO, United States) for 0, 6, 12, and 24 h, respectively. Under anesthesia, their eyeballs were removed for collection of blood, and tissues were collected for Western blot and immunohistochemistry.

For effect of SW033291 on inhibiting 15-PGDH, 30 mice were divided into five groups with six mice in each group (n = 6), and each mouse was administered intraperitoneally with 10 mg/kg body weight SW033291. Animals were sacrificed at 0, 3, 6, 12, and 24 h, followed by collection of kidneys for PGE₂ ELISA experiment. SW033291 was diluted according to Zhang et al. (2015).

For survival experiments, 100 mice were divided into four groups (n = 20 in control and SW033291 groups; n = 43 in LPS group; n = 37 in LPS + SW033291 group). Following LPS (10 mg/kg body weight), stimulated mice were treated with 10 mg/kg SW033291 or vehicle control twice daily for five doses. Then, mice survival was monitored and recorded daily. The other 70 mice were used to explore the role and mechanism of 15-PGDH on LPS-induced AKI, as described in the section below.

The mice were anesthetized, and the blood was collected for the assay of renal function. The kidneys were used for analysis of real-time PCR, Western blot, ELISA, HE staining, and immunohistochemistry.

Scr assay kit (C011-2-1) and BUN assay kit (C013-2-1) were acquired from Nanjing Jiancheng Bioengineering Institute. After the mice were anesthetized, blood was collected from the eyeballs of the mice for serum creatinine (Cr) and urea nitrogen (BUN) assay, according to the manufacturer’s instructions. After sacrificed, the kidneys of each mouse were quickly removed and stored at −80°C.

After being fixed in 4% paraformaldehyde solution, the kidney tissue was dehydrated, transparent, paraffin embedded, and sliced. Then, HE staining was performed and observed under a microscope. Double-blind method was used to assess the renal tubular injury by the magnitude of tubular epithelial swelling, loss of brush border, interstitial cell infiltration, tubular cell necrosis, vacuolation, and desquamation on the basis of the following scale: grade 0, no morphological deformities; grade 1, 1–25%; grade 2, 26–50%; grade 3, 51–75%; grade 4, 76–100% (Kojima et al., 2007; Havakhah et al., 2014).

The total RNA from kidney tissue was isolated by using TRIzol. The isolated mRNA was used as a template to synthesize cDNA by using PrimeScriptTM RT Master Mix (Takara, Japan). Real-time PCR was performed by using the One Step SYBR^® PrimeScript TM RT-PCR Kit in a Biosystems 7500 instrument. The amplification conditions were 95°C for 30 s, 95°C for 5 s, and 60°C for 34 s for 40 cycles. The primer sequences used were presented as follows (see Table 1).

The expressions of 15-PGDH in the kidneys were detected by immunohistochemistry. The antibody against 15-PGDH antibody (Cayman chemical, Item No.160615) was diluted at 1:100, and the experiment was manipulated according to the instructions of the EnVisionTM FLEX Mini Kit (Dako, K8024). Immunohistochemistry stained tissue sections were scored by two pathologists blinded to clinical and pathological data. According to the dyeing intensity: 0, no staining; 1+, weak; 2+, medium intensity; 3+, strong. Then, H-score was calculated according to the following formula: [1 × (% cells 1 +) + 2 × (% cells 2 +) + 3 × (% cells 3 +)] (Kim et al., 2019).

Proteins from the kidney samples of mice were extracted by sonicating the tissues in 1% PMSF-containing RIPA buffer and were centrifuged at 4°C for 10 min at 12,000 rpm. The protein concentration was determined by BCA method. Based on the protein concentration, 80–100 μg proteins were mixed with one quarter volume of 5× SDS loading buffer, boiled at 95°C for 10 min to denature the proteins. The proteins were subjected to SDS-PAGE and transferred to membrane. The membranes were blocked at room temperature for 1 h, washed with TBS-T (contains 0.1% Tween-20), and incubated with primary antibodies against 15-PGDH (1:1000 dilution, Cayman chemical, Item No.160615), caspase-3 (1:1000 dilution, proteintech, Item No.66470-2-Ig), microtubule-associated protein light chain 3 (LC-3B) (1:1000 dilution, CST, Item No. 3868), p62 (1:1000 dilution, CST, Item No. 5114), and β-actin (1:2000 dilution, Sigma-Aldrich, Item No. A1978), respectively, for overnight at 4°C. The membranes were then washed with TBST and incubated with secondary antibody for 1 h at RT, and signal was developed by adding the ECL luminescence substrate. The signal intensity of each protein was scanned by intensity-scanning software, and relative expression level of protein was calculated.

The kidney homogenates were prepared. The MDA content and the activities of SOD and CAT were determined according to the manufacturer’s instructions (Nanjing Jiancheng Bioengineering Institute). The absorbance value (OD value) was measured with a microplate reader, and the MDA content and SOD and CAT activities were calculated according to the formula in the instructions.

Paraffin sections were deparaffinized and rehydrated, further performed for antigen retrieval and blocked with goat serum for 30 min. The section samples were incubated with LC-3B primary antibody (1:100, CST) overnight at 4°C and washed three times with PBS (pH 7.4) at 5 min for each wash. Sections were then incubated with FITC-labeled anti-rabbit secondary antibody (1:400 dilution, MultiSciences) in the dark for 1 h at room temperature and washed three times with PBS (pH 7.4). The samples were counterstained with DAPI, mounted, and imaged under a fluorescence microscope. The nucleus stained with DAPI was blue under UV excitation, and the positive expression was shown as fluorescein-labeled red light.

The apoptotic levels in the kidneys were measured according to the instruction of the TUNEL apoptosis detection kit (Roche). The percentage of apoptotic cells was calculated according to the number of apoptotic cells/total number of cells in randomly selected 5–10 fields.

Freshly isolated kidneys were quickly put into fixing buffer and fixed at 4°C for 2–4 h. After this, samples were transferred to Wuhan Goodbio Technology Co., Ltd., prepared as specimens, examined, and photographed under a transmission electron microscope.

Pre-cooled 1 × PBS was added at a ratio of 1:10 to prepare tissue homogenates. The PGE₂ concentration of the renal tissue homogenate was determined according to the instruction of PGE₂ ELISA kit (Huamei Biology Company, Wuhan, China).

GraphPad Prism version 5.02 (GraphPad Prism Software Inc, San Diego, CA, United States) was used for data analysis. All the experiments were repeated at least three times. Data were presented as mean ± SD. Statistical analysis was performed by ANOVA followed by Tukey’s test. For the purpose of survival analysis, Kaplan-Meier method followed by log-rank test was applied to compare the survival rate between groups. A value of P < 0.05 was considered to be statistically significant.

At first, we established an AKI mouse model using LPS. As shown in Figure 1A, the survival rate of mice in LPS group significantly decreased to 53% compared to the control group. Through HE staining, we observed marked edema, vacuolar degeneration, inflammatory cell infiltration, and the narrowing of tubular lumen at 12 and 24 h after LPS treatment, contrasting the normal renal morphology in vehicle-treated mice (Figure 1B). Accordingly, the serum levels of creatinine and urea nitrogen were significantly increased after LPS challenge compared to the control mice (Figures 1C,D). We determined the expression of 15-PGDH in kidney in response to LPS treatment. As shown in Figures 1E,F, LPS treatment significantly enhanced protein levels of 15-PGDH in kidneys; the highest level of 15-PGDH occurred at 12 h and lasted for 24 h. By immunohistochemistry, we found that 15-PGDH protein was mainly localized in the cytoplasm of renal tubular epithelial cells in renal cortex and outer medulla and that it had low expression levels in glomeruli and the inner medulla (Figures 1G,H).

As shown in Figure 2A, the PGE₂ content in kidney time-dependently increased with SW033291 treatment and peaked at 12 h. Strikingly, although the renal PGE₂ content in both control group and SW033291 group increased after 12 h treatment with LPS, the increase of PGE₂ content was higher in SW033291 group mice than that in control mice (Figure 2B). Furthermore, the mRNA and protein levels of 15-PGDH were significantly downregulated in kidney after SW033291 treatment (Figures 2C,D).

SW033291 treatment enhanced the survival rate of LPS-treated mice from 30 to 49% (Figure 3A). Meanwhile, renal function assay showed that BUN and Cr levels were significantly increased after 6 h of LPS treatment compared with the control group and peaked at 12 h, whereas the BUN and Cr levels were significantly decreased after SW033291 treatment (Figures 3B–E). The decline in the BUN and Cr levels of mice with 10 mg/kg SW033291 treatment is more obvious than 5 mg/kg dose at 12 h (Figures 3D,E). HE staining showed that there were no significant pathological lesions in the kidneys of the control and SW033291 group mice. In LPS-treated group, the tubular epithelial cells of the mice were edematous with larger cellular volume, vacuolar degeneration and narrowed lumen of the renal tubules. It was also observed that the glomerular structure was disordered, accompanied by infiltration of scattered inflammatory cells and narrowing of the renal capsule. These pathological lesions were evidently alleviated after SW033291 treatment (Figures 4I–Q). PAS staining confirmed above findings in renal morphology (Figures 4A–H). These data indicated a protective effect of SW033291 against LPS-induced AKI.

TUNEL assay showed enhanced renal cell apoptosis after intraperitoneal injection of LPS for 12 h in mice, but the apoptosis level was decreased significantly after SW033291 therapy (Figures 5A,B). After LPS stimulation, apoptosis-related protein cleaved-caspase-3 levels increased evidently, whereas SW033291 treatment showed a significant reduction of cleaved-caspase-3 protein expression levels (Figures 5C,D). At the same time, the elevated mRNA levels of Fas, caspase-3, and caspase-8 in kidney after LPS stimulation were all blunted by the pretreatment of SW033291 in line with the restoration of reduced mRNA expression of Bcl-2 (Figures 5E–H). All these data demonstrated an anti-apoptotic role of SW033291 in this experimental setting.

As shown in Figures 6A,B, the expression of 15-PGDH protein was significantly upregulated by LPS treatment. In contrast, SW033291 treatment significantly suppressed 15-PGDH expression in LPS-treated mice. During the development of autophagy, cytosolic soluble LC3 (i.e., LC3-I) is converted to membrane-type LC3 (i.e., LC3-II) and localized on the autophagosome membrane. After LPS stimulation, the LC3-II/LC3-I ratio was significantly increased along with the enhanced P62 protein compared to the control group. Interestingly, SW033291 treatment resulted in a greater increment of the LC3-II/LC3-I ratio after stimulation with LPS (Figures 6A–D). By immunofluorescence, we further confirmed the regulatory role of SW033291 for LC3B in kidneys of AKI mice (Figure 6E). Transmission electron microscopy showed that SW033291 treatment evidently enhanced the number of autophagic vacuoles in kidney cells after LPS stimulation, which further confirmed a promoted autophagic response (Figure 6F).

As shown in Figures 7A–C, compared with the control group, the MDA level in the kidney tissue of the LPS group was significantly increased, while the activity of antioxidant enzymes of SOD and CAT was evidently suppressed. Importantly, SW033291 treatment significantly decreased the MDA level and enhanced the activity of SOD and CAT in the kidneys of AKI mice (Figures 7A–C). In agreement with above findings, the mRNA levels of kidney NADPH oxidase subunits gp91phox, p40phox, and p47phox were strikingly upregulated after 12 h LPS challenge compared with the control group, which was significantly blunted after SW033291 treatment (Figures 7D–F). Similarly, the upregulated mRNA expression of iNOS was also reduced by the pretreatment of SW033291 (Figure 7G). Furthermore, we confirmed that the reduction of antioxidant genes of SOD1 and catalase in kidneys after a 12 h injection of LPS was significantly reversed by SW033291 (Figures 7H,I).

It has been shown that inflammation is a pathogenic factor of AKI. Meanwhile, prostaglandin E2 is a well-established inflammatory mediator. We evaluated whether inflammation was involved in the role of SW033291 in LPS-induced AKI. As shown in Figure 8, LPS treatment significantly enhanced the release of IL-6, IL-1β, TNF-α, and MCP-1 in the kidneys, while SW033291 treatment had no effect on the release of these inflammatory factors. These data suggested that the beneficial role of SW033291 in LPS-induced AKI could be independent of the inflammation.