Background: Sepsis causes acute kidney injury (AKI) in critically ill patients, although the mechanisms underlying the pathophysiology are not fully understood. SUMO-specific proteases 3 (SENP3), a member of the deSUMOylating enzyme family, is known as a redox sensor and could regulate multiple cellular signaling pathways. However, the role of SENP3 in septic AKI remains unclear. Objectives: The purpose of this study was to investigate the role of SENP3 in lipopolysaccharide (LPS)-induced AKI model. Methods: C57BL/6 mice were given intraperitoneal injection of LPS (10 mg/kg). NRK-52E cells were treated with LPS in vitro. The SENP3 protein expression was analyzed by Western blotting. The levels of reactive oxygen species (ROS) in cells were measured using DCFH-DA. SENP3-siRNA or SENP3-plasmid was, respectively, transfected into NRK-52E cells to knock down or overexpress the SENP3 expression. Western blotting was performed to analyze the protein expression of cleaved caspase 3, cytochrome c, and dynamin-related protein 1 (Drp1). The mitochondrial membrane potential was measured using JC-1 assay kit. Co-immunoprecipitation was used to determine the interaction of Drp1 and SMUO2/3. Results: SENP3 protein expression was obviously increased in renal tissues from the mouse model of LPS-induced AKI. Accordingly, SENP3 expression was upregulated in NRK-52E cells treated with LPS in a ROS-dependent manner in vitro. Knockdown of SENP3 dramatically ameliorated LPS-induced apoptosis of NRK-52E cells, whereas overexpression of SENP3 further aggravated LPS-induced apoptosis of NRK-52E cells. Mechanistically, SENP3 triggered Drp1 recruitment to mitochondria by increasing the deSUMOylation of Drp1. Conclusion: SENP3 aggravated renal tubular epithelial cell apoptosis in LPS-induced AKI via Drp1 deSUMOylation manner.
Sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection . Acute kidney injury (AKI) is one of the most common complications of sepsis, which significantly increases the risk of mortality . Patients with sepsis-induced AKI have a remarkably increased mortality compared to patients without AKI [3-5]. It has been reported that approximately 60% of patients with sepsis develop AKI and sepsis is the most common cause of AKI in the intensive care unit [6, 7]. However, there is no effective therapy for septic AKI, mainly due to the lack of complete understanding of kidney injury pathogenesis in sepsis, calling for a need to further elucidate its molecular mechanism.
The mechanisms underlying the pathophysiology of sepsis-induced AKI are complex and not fully understood, but they may include oxidative stress, mitochondrial dysfunction, and tubule epithelia cell apoptosis [8, 9]. Oxidative stress plays an important role in the development of sepsis-induced AKI and is the major contributor to mitochondrial dysfunction in renal cells after treatment with lipopolysaccharide (LPS) . Tubular epithelial cells are primary targets for acute injury  and are susceptible to apoptosis in AKI . Recent studies strongly suggest that mitochondrial fragmentation is a crucial mechanism contributing to tubular damage during AKI [12, 13]. Mitochondria are highly dynamic organelles that maintain their morphology by a delicate balance between two opposing processes, fusion and fission. Mitochondria fission is mainly regulated by dynamin-related protein 1 (Drp1), which is a cytosolic protein that will move to the outer mitochondrial membrane by activation .
Small ubiquitin-like modifier (SUMO) proteins express four SUMO paralogs in mammals, including SUMO1, SUMO2, SUMO3, and SUMO4. SUMO modification (SUMOylation), one of the post-translational modifications, has been discovered as a critical regulatory mechanism in multiple signaling pathways that alters the cellular localization, protein-protein interaction, and biological function of target proteins . SUMOylation is a dynamic process that can be reversed by SUMO-specific proteases (SENPs) to maintain the balance between the SUMOylation and deSUMOylation of target proteins . SENPs, composed of six members (SENP1, 2, 3, 5, 6, 7) in mammals, detach the specific SUMO paralogs from target proteins. Generally, SENP1 and SENP2 deconjugate SUMO1 or SUMO2/3, SENP3 and SENP5 have specific activity for SUMO2/3 deconjugation, and SENP6 and SENP7 can edit polySUMO2/3 chains [17, 18]. As a crucial member of the SENPs, SENP3 is a redox-sensitive enzyme and is unique in its rapid increases in protein level following a mild increase in reactive oxygen species (ROS) . Multiple studies have shown that SENP3 is associated with the progression of many diseases, including cardiovascular diseases, neurological diseases, nonalcoholic fatty liver disease and various cancers . However, it remains unknown whether SENP3 is involved in the pathogenesis of sepsis-induced AKI. Thus, the aim of the current study is to explore the role of SENP3 in the progression of LPS-induced AKI.
Materials and Methods
C57BL/6 mice (male, 8 weeks old, 20–24 g) were purchased from the Animal Laboratory Center of Tongji University. All efforts were made to ameliorate the suffering of animals, and this study was approved by the Ethics Committee of Tongji Hospital of Tongji University. Mice were randomly assigned to two groups (n = 4–6 animals per group): control (Con) group and LPS group. The Con group was given intraperitoneal injection of saline (0.5 mL), while the LPS group was given intraperitoneal injection of LPS (10 mg/kg; Sigma-Aldrich) in 0.5 mL of saline. Mice were euthanized at 24 h after LPS treatment, and then blood and kidney samples were collected.
Renal Function and Histology Analyses
Blood urea nitrogen (BUN) and serum creatinine (Scr) were measured by analytical kits (Huilin Biotech, Changchun, China) using automatic biochemistry analyzer (Rayto Life and Analytical Sciences, Shenzhen, China). Kidney samples were fixed with paraformaldehyde (4%), then embedded in paraffin, and cut into 4 μm thick sections, which were stained with hematoxylin and eosin. The pathological changes of the stained kidney samples were observed under an optical microscope. To evaluate the renal tubular injury, a semiquantitative scoring method was used. In brief, the score was calculated under ×400 magnification observing ten randomly fields for each mouse. Kidney injury score is scored as 0, normal; 1, <10%; 2, 10–25%; 3, 25–50%; 4, 50–75%; 5, 75–100%. This criterion includes lumen occlusion, loss of brush border, and tubular dilation.
Transmission Electron Microscopy
Renal cortex tissues were fixed with 2.5% glutaraldehyde for 2 h at room temperature and transferred to 4°C for storage. After dehydration in ethanol, the specimens were embedded in EPON. Ultrathin sections were photographed under a Hitachi 7700 transmission electron microscope (Tokyo, Japan).
Normal rat kidney tubular epithelial cells (NRK-52E) (Rattus norvegicus, cell strain) were cultured using high-glucose Dulbecco’s Modified Eagle’s Medium (HyClone) containing 5% fetal bovine serum (Beyotime, Shanghai, China) under 37°C and 5% CO2 condition. Then, cells were incubated with the optimum concentration of LPS of the indicated times.
Transfection of siRNA and Plasmid
The SENP3-siRNA, negative Con (NC)-siRNA, SENP3-plasmid, and NC-plasmid (empty vehicle pEX-3) were purchased from GenePharma (Shanghai, China) and were transfected into NRK-52E cells using the Lipo6000TM (Beyotime) according to the manufacturer’s instructions. Then, NRK-52E cells were subjected to LPS addition for the indicated times. Next, cells were lysed for protein extraction, and samples were analyzed by Western blotting. The sequences of SENP3-siRNA oligonucleotides were as follows: sense 5′-GGUACUACAGCUGAUCCAATT-3′ and antisense 5′-UUGGAUCAGCUGUAGUACCTT-3′. The full sequence of SENP3-Plasmid was described in the online supplementary information (see www.karger.com/doi/10.1159/000525308 for all online suppl. material).
Proteins were collected from kidney tissues or NRK-52E cells with RIPA lysis buffer (Beyotime). The protein concentration was determined using the BCA assay kit (Beyotime). The protein (30–50 mg) from each group was separated by using SDS-PAGE and electrophoretically transferred onto PVDF membranes (CAT. NO. ISEQ00010, Immobilon®-PSQ). After transfer, nonspecific binding was prevented using 5% skimmed milk TBS-Tween (BBI Life Sciences). The primary antibodies were used as follows: anti-SENP3 (ab124790), anti-SUMO2/3 (ab109005), anti-Drp1 (ab184247), anti-cytochrome C (ab13575) (Abcam); anti-cleaved caspase-3, anti-COX IV (Cell Signaling Technology); anti-GAPDH (Proteintech). Membranes were then incubated with HRP-labeled secondary antibodies (1:1,000 dilution) at room temperature for 1 h. Each step described above was followed by three washes for 10 min each. Finally, the signals were detected using the AIC. AIphaView imaging system.
NRK-52E cells were lysed, the supernatants obtained by centrifuging were incubated with anti-Drp1 (ab184247; Abcam) antibody, or IgG (NC) on a shaker overnight at 4°C, and then protein A + G agarose (Beyotime) were added at 4°C slowly shook for 1–3 h. The beads were then rinsed with PBS and subjected to SDS-PAGE. Protein bands were detected by Western blot using anti-SUMO2/3 (ab109005; Abcam) antibodies.
Detection of Intracellular ROS Levels by Fluorescence Microscopy
The ROS levels in cells were measured using 2′,7′-dichlorofluorescein diacetate (DCFH-DA) (Beyotime). N-acetylcysteine (NAC) (Beyotime), a ROS inhibitor, was pre-treated before LPS administration. Cells were treated with LPS (1 μg/mL) for 24 h. Then, the cells were washed twice with cold PBS, treated with DCFH-DA (10 μM final concentration), and incubated for 30 min at 37°C in a light-protected, humidified chamber. After probe treatment, the cells were washed at least twice with cold PBS. A fluorescence microscope (Nikon) was used at set excitation and emission wavelengths in these experiments.
Mitochondrial and Cytosolic Component Extraction
Mitochondrial and cytosolic protein of renal tissue or NRK-52E cells was extracted using a Tissue Mitochondria Isolation Kit or Cell Mitochondria Isolation Kit (Beyotime). First, renal tissue or NRK-52E cells were well distributed by Mitochondria Isolation Solution containing PMSF in an ice bath for 15 min. A glass homogenizer was applied to grind the tissue or cells followed by centrifugation with 600 g for 10 min at 4°C. The liquid supernatant was shifted to another tube and centrifuged again with 11,000 g for 10 min at 4°C. The sediment was blended with Mitochondrial Lysate Solution to obtain mitochondrial proteins. The supernatant was centrifuged with 12,000 g for 10 min at 4°C to obtain cytosolic proteins.
Mitochondrial Membrane Potential
The mitochondrial membrane potential (∆Ψm) was measured using mitochondrial membrane potential assay kit with JC-1 (Beyotime) according to the manufacturer’s instructions. Briefly, after indicated treatments, NRK-52E cells cultured in 6-well plates were washed by PBS and were incubated with JC-1 working solution for 20 min at 37°C. Then, cells were washed twice with JC-1 staining buffer. JC-1 stained cells were measured using a fluorescence microscope (Nikon).
Data are expressed as means ± SEM. Sample means were compared using t tests or one-way ANOVA. Statistical analyses were performed with GraphPad Prism 8 software. A value of p < 0.05 was considered statistically significant.
LPS-Induced Mitochondrial Fragmentation and Kidney Injury in Mice
In the current study, sepsis-induced AKI mice model was established by intraperitoneal injecting LPS (10 mg/kg) for 24 h. Saline-injected mice were used as Con group. We observed the morphology of mitochondria in kidney cortices of the two groups’ mice by transmission electron microscopy. As shown in Figure 1a, mitochondria in tubular cells of mice in the LPS group had significantly increased fragmentation compared to the Con group. At the molecular level, we evaluated the mainly mitochondrial fission regulating proteins of Drp1 with extracted mitochondrial fractions and whole kidney tissue lysate. We found that mitochondrial Drp1 was increased significantly in the LPS group compared with the Con group (Fig. 1b). No difference of Drp1 was observed in the whole tissue lysates between the two groups (Fig. 1c). Renal tubular cell apoptosis evaluated by Western blotting analysis of cleaved caspase-3 was increased significantly in the LPS group (Fig. 1c). Increased cytochrome c release into the cytosol was observed in the LPS group (Fig. 1d). As shown in Figure 1e, LPS-treated mice exhibited a significant increase in BUN (86.47 ± 1.88 mg/dL vs. 29.07 ± 1.81 mg/dL; p < 0.05) and Scr (66.58 ± 6.35 μmol/L vs. 35.34 ± 2.75 μmol/L; p < 0.05) compared with the Con group. As shown in Figure 1f, renal histology by H&E staining revealed that LPS-treated mice showed more severe renal tubular damage including lumen occlusion, loss of brush border, and tubular dilation. The tubular injury score of the LPS group was significantly higher than that in the Con group (2.83 ± 0.31 vs. 0.67 ± 0.33; p < 0.05; Fig. 1g). Taken together, these data demonstrated that LPS could induce mitochondrial fragmentation, cellular apoptosis, and AKI in vivo.
SENP3 Protein Expression Was Upregulated in LPS-Induced AKI Model in vivo and in vitro
To investigate the potential roles of SENP3 in LPS-induced AKI, we first examined SENP3 expression in the kidney tissue of mice with or without LPS injection. As shown in Figure 1c, Western blotting showed that the renal SENP3 protein expression was significantly increased at 24 h after LPS injection.
And then, to determine the SENP3 expression in LPS-induced AKI model in vitro, SENP3 protein levels were examined using Western blotting in NRK-52E cells stimulated with different concentrations or time of LPS. Figure 2a shows the SENP3 levels in NRK-52E cells stimulated with different concentrations of LPS 0, 0.2, 1, 2, 5, or 10 μg/mL. Figure 2b shows the levels of SENP3 in cells treated with LPS (1 μg/mL) for 6, 12, and 24 h. Thus, LPS dramatically increased the levels of the SENP3 protein partly in a dose- and time-dependent manner.
Since SENP3 was known as a redox sensor responding to ROS accumulation, we investigated whether ROS was responsible for the upregulated SENP3 in LPS-induced AKI. To this end, we suppressed ROS accumulation in LPS-treated NRK-52E cells by pre-treatment of ROS inhibitors NAC (Fig. 2c). We observed that the upregulation of SENP3 was significantly suppressed by NAC in LPS-induced NRK-52E cells (Fig. 2d). These data suggested that SENP3 protein was upregulated in a ROS-dependent manner in LPS-treated NRK-52E cells.
Knockdown of SENP3 Alleviated LPS-Induced Apoptosis of NRK-52E Cells
To further investigate the direct roles of SENP3 in LPS-induced tubular cell apoptosis, we transfected NRK-52E cells with SENP3-siRNA or NC-siRNA and then incubated in the absence or presence of LPS (1 μg/mL) for 24 h. Figure 3a shows that SENP3-siRNA markedly decreased SENP3 protein compared with NC-siRNA, and the levels of SENP3 were lower in SENP3-deficient NRK-52E cells following LPS addition. The apoptotic marker-cleaved caspase-3 was determined. As shown in Figure 3b, cleaved caspase-3 was induced by LPS and reversed by SENP3-siRNA.
Cytochrome c release from mitochondria to cytoplasm is a hallmark of the intrinsic pathway of cell apoptosis . After LPS administration, the amount of cytochrome c in cytoplasm was significantly increased compared to Con group (Fig. 3c). However, the cytochrome c in the cytoplasm was significantly decreased in SENP3-siRNA + LPS group than that in the LPS group (Fig. 3c).
The decrease of ∆Ψm is a landmark event in the early stages of cell apoptosis. JC-1 staining was used to measure ∆Ψm of NRK-52E cells. JC-1 is a monomer present in the cytoplasm (which stains green), and it also accumulates as aggregates in normally polarized mitochondria (which stains red). A shift in fluorescence from red to green indicates the decreased membrane potential. As shown in Figure 3d, LPS induced mitochondrial depolarization of NRK-52E cells, as evidenced by diffuse cytoplasmic JC-1 green staining. Knockdown of SENP3 using SENP3-siRNA restored mitochondrial membrane potential in LPS-induced NRK-52E cells. These results demonstrated that knockdown of SENP3 alleviated LPS-induced apoptosis of NRK-52E cells.
Overexpression of SENP3 Further Aggravated LPS-Induced Apoptosis of NRK-52E Cells
NRK-52E cells were transfected with SENP3-plasmid or NC-plasmid and then treated with the absence or presence of LPS (1 μg/mL) for 24 h. Figure 4a shows that the SENP3 protein was remarkably increased in the SENP3-plasmid group compared with the NC-plasmid group, and the levels of SENP3 were further higher in SENP3-overexpression NRK-52E cells following LPS addition. As shown in Figure 4b, the apoptotic marker-cleaved caspase-3 was significantly increased in SENP3-plasmid + LPS groups compared with the NC-plasmid + LPS groups. Figure 4c shows that the amount of cytochrome c in cytoplasm in the SENP3-plasmid + LPS groups was further increased relative to the LPS groups. Overexpression of SENP3 further decreased mitochondrial membrane potential in LPS-induced NRK-52E cells (Fig. 4d). These data indicated that overexpression of SENP3 further aggravated LPS-induced apoptosis of NRK-52E cells.
SENP3 Promoted deSUMOylation of Drp1 and Facilitated Drp1 Translocation to Mitochondria
Our forementioned results demonstrated that SENP3 played a detrimental role of LPS-induced AKI. To further gain insight into the underlying molecular mechanism mediating the pathological effects of SENP3, we detected the effect of SENP3 on Drp1. Co-IP method was used to detect the interaction between Drp1 and SUMO2/3, and this result indicated that Drp1 can be SUMOylated by SUMO2/3 (Fig. 5a, b). LPS exposure decreased the conjugation between SUMO2/3 and Drp1, and this decrease was further enhanced after transfected with SENP3-plasmid (Fig. 5b) and however transfected with SENP3-siRNA restored the conjugation between SUMO2/3 and Drp1 (Fig. 5a). As shown in Figure 5c, mitochondrial Drp1 expression was significantly increased in LPS-treated NRK-52E cells, which was significantly suppressed by SENP3 silencing while further augmented by SENP3 overexpression. These results demonstrated that SENP3 promoted deSUMOylation of Drp1 and facilitated Drp1 translocation to mitochondria in LPS-treated NRK-52E cells.
Increased understanding of the mechanisms underlying the physiopathology of LPS-induced AKI is important for the identification of novel therapeutic targets. In the present study, we provided new insights into the role of SENP3 in LPS-induced AKI. First, we found that the expression of SENP3 was upregulated in LPS-induced AKI model in vivo and in vitro. Second, knockdown of SENP3 alleviated LPS-induced apoptosis of NRK-52E cells, whereas overexpression of SENP3 further aggravated LPS-induced apoptosis of NRK-52E cells, indicating a negative role of SENP3 in the setting of LPS-induced AKI. Mechanistic studies indicated that SENP3 promoted LPS-induced apoptosis mainly via deSUMOylation of Drp1-mediated mitochondrial fission (Fig. 6).
Previous studies have demonstrated that SENP3 is widely expressed in various organ and cell types and is upregulated in multiple disease conditions largely in a ROS-dependent manner. It has been reported that SENP3 was expressed and upregulated in the mouse heart depending on ROS production in response to myocardial ischemia-reperfusion (MIR) injury . Hepatic SENP3 was upregulated in nonalcoholic fatty liver disease patients and an animal model in vivo and after loading hepatocytes with free fatty acids in vitro . SENP3 expression was obvious upregulation in the brain after traumatic brain injury, especially in the neurons . SENP3 was highly expressed in vascular smooth muscle cells of remodeled arteries, accompanied by elevated ROS levels . SENP3 mRNA and protein levels were significantly increased in high-glucose cultured human aortic endothelial cells in a time-dependent manner . There was an increase in ROS level and SENP3 expression in Dex-induced osteoporotic BM-MSCs . However, there are few studies concerning the expression of SENP3 in kidney injury. In the current study, we demonstrated that SENP3 was expressed in the mouse kidney tissue and was remarkably upregulated in the mouse kidney tissue in LPS-induced AKI model. In addition, our in vitro findings also suggest that SENP3 was remarkably upregulated depending on ROS accumulation in NRK-52E cells following LPS treatment. It is shown that SENP3 elevation in pathological stress conditions is not organ-specific.
To further investigate the role of SENP3 in LPS-induced AKI, we developed a model in vitro using NRK-52E cells exposed to LPS. Our study found that knockdown of SENP3 reversed LPS-induced increases in cleaved caspase 3, cytochrome c in cytosol, and lower ∆Ψm of NRK-52E cells. By contrast, overexpression of SENP3 further aggravated LPS-induced apoptosis of NRK-52E cells, indicating a negative role of SENP3 in the setting of LPS-induced AKI. In other words, our study suggested that SENP3 exerted deleterious effects on cells, which are consistent with previous studies. It has been reported that SENP3 upregulation pivotally contributes to MIR injury . SENP3 overexpression significantly promoted, and sh-RNA-mediated knockdown markedly inhibited vascular smooth muscle cells proliferation and migration . SENP3 was involved in high-glucose-induced endothelial dysfunction . Enhanced adipogenesis and weakened osteogenesis in osteoporotic BM-MSCs might be caused by upregulated SENP3 . However, whether SENP3 promotes cell death or cell survival remains controversial. Some studies have reported a protective effect. For example, knockdown of SENP3 promoted apoptosis after ischemia-reperfusion, indicating the protective role of the SENP3 in MIR injury .
Mitochondria are abundant in the kidney and represent the powerhouses of the renal tubular epithelial cells. Mitochondrial dysfunction has been increasingly recognized as a key contributor to the pathogenesis of kidney injure. Drp1 is a GTPase recruited over the mitochondrial surface upon activation to initiate fission and pivotally mediates mitochondria-dependent apoptosis . Here, our results demonstrated that knockdown of SENP3 reversed Drp1 recruited over mitochondrial in NRK-52E cells treated by LPS, whereas overexpression of SENP3 significantly augmented Drp1 localization at mitochondria. Subsequently, our findings demonstrate that Drp1 was the deSUMOylation target of SENP3. These data suggest that Drp1 might mediate the detrimental effects of SENP3 overexpression on mitochondrial dysfunction and subsequent injury. Consistent with our findings, it has been reported that SENP3 promoted Drp1 mitochondrial translocation in reoxygenated neurons, leading to augmented neural apoptosis . It has also been shown that Drp1 is the major SENP3 target mediating its effects in cardiomyocytes . Collectively, our results provide clues that SENP3-mediating Drp1 deSUMOylation signaling may represent a potential pathway in the mitochondrial damage and kidney injury in LPS-induced AKI.
In summary, SENP3 was upregulated in LPS-induced AKI model in vivo and in vitro. SENP3-mediated deSUMOylation of Drp1 promoted mitochondrial fission and aggravated renal tubular epithelial cell apoptosis in LPS-induced AKI. Such data might shed new light on the pathogenesis of LPS-induced AKI, which may be used as a potential therapeutic target for treatment of sepsis-associated AKI. The precise role of SENP3 in the development of septic AKI will continue to be studied.
Statement of Ethics
All animal experiments were performed under animal study protocols approved by the Ethics Committee of Tongji Hospital of Tongji University (No. 2019-DW-001).
Conflict of Interest Statement
The authors declare that they have no competing interests.
This study was supported in part by Grants of the National Natural Science Foundation of China (NSFC No. 82170696, 81800631).
Lirui Wang: study design, experiments, data analysis, manuscript preparation, and drafting the article. Jiangtao Li and Chen Yu: study design and revising the article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author on reasonable request.