JOSD2 alleviates acute kidney injury through deubiquitinating SIRT7 and negativity regulating SIRT7-NF-κB inflammatory pathway in renal tubular epithelial cells

Abstract

Acute kidney injury (AKI), triggered by various stimuli including ischemia-reperfusion, nephrotoxic insult, and sepsis, is characterized by an abrupt deterioration in kidney function. Ubiquitination is a post-translational modification of proteins that plays a critical role in the pathogenesis and progression of AKI. In this study, we aimed to investigate the role and underlying mechanism of the deubiquitinating enzyme Josephin Domain-containing protein 2 (JOSD2) in AKI. We found that deficiency of JOSD2 exacerbated renal tubular injury and inflammation in AKI mice induced by cisplatin or ischemia-reperfusion injury. Conversely, the specific overexpression of JOSD2 in renal tubular epithelial cells effectively prevented renal tubular injury and inflammation induced in AKI mice. Mechanistically, we identified Sirtuin 7 (SIRT7) as a potential substrate of JOSD2 through mass spectrometry combined with co-immunoprecipitation analysis. JOSD2 removes the K63-linked ubiquitination of SIRT7 via its active site C24 and promotes P62-mediated autophagic degradation of SIRT7, which subsequently prevents the phosphorylation and nuclear translocation of P65 and reduces inflammatory responses in renal tubular epithelial cells. Our study reveals the role of the JOSD2-SIRT7 axis in regulating AKI-induced renal inflammation and highlights the potential of JOSD2 as a promising therapeutic target for AKI.

Introduction

Acute kidney injury (AKI) is a prevalent clinical complication, affecting approximately 13.3 million individuals annually and resulting in 1.7 million mortalities worldwide each year [1]. AKI is characterized by a sudden and rapid decline in renal function, accompanied by tubular cell damage, especially in the proximal tubules. Cisplatin (CP) is a potent chemotherapy agent used to treat various solid tumors. Despite its therapeutic efficacy, its severe nephrotoxicity led to the occurrence of AKI [2, 3]. In addition, ischemia-reperfusion injury (IRI), which involves a period of diminished blood flow followed by re-establishment of circulation, is another frequent event that can lead to AKI [4]. Current clinical treatments for AKI primarily involve renal replacement, vasoactive drugs, and antibiotics, yet their therapeutic efficacy remains unsatisfactory. Accumulating evidence indicates that inflammation is a crucial feature of AKI, and strategies targeting inflammation help mitigate renal tubular injury induced by AKI [5, 6]. Therefore, identifying the inflammation-related proteins in regulating renal tubule damage may lead to developing new therapeutic strategies and targets for AKI treatment.

Ubiquitination is a pivotal post-translational modification process of numerous proteins, which can be reversed by deubiquitinating enzymes (DUBs) [7]. DUBs play a crucial role in regulating the stability and function of proteins and are involved in regulatory mechanisms in various diseases [8]. Emerging evidence suggests that DUBs may play a key role in AKI. Luo et al. have demonstrated that deubiquitinating enzyme BAP1 protects AKI mice by enhancing the stability of BRCA1 protein and inactivating the NF-κB signaling pathway [9]. Moreover, the USP25-PKM2-aerobic glycolysis axis positively regulates M1-like polarization and exacerbates AKI in mice [10]. Thus, DUBs may serve as a potential molecular library for identifying targets for the treatment of AKI.

Josephin Domain-containing protein 2 (JOSD2) belongs to the Machado-Josephin Domain-containing proteases (MJDs) subfamily of DUBs along with other three cysteine-proteases, Ataxin-3, Ataxin-3L, and JOSD1, which share a common catalytic Josephin domain [11]. JOSD2 preferentially recognizes substrates containing K11, K48, and K63 linkages, suggesting a potential role in maintaining protein quality control [12]. Our research group has recently demonstrated that JOSD2 attenuates cardiac hypertrophy and heart failure by deubiquitinating SERCA2a in cardiomyocytes [13]. In addition, JOSD2 is involved in the pathological process of lung cancer [14], cholangiocarcinoma [15], and acute myeloid leukemia [16]. However, the biological function and mechanism of JOSD2 in AKI are poorly understood yet.

In the present study, we explored the roles and underlying mechanisms of JOSD2 in AKI. We found that knockout of JOSD2 exacerbated renal tubule injury and inflammation induced by AKI, whereas JOSD2 overexpression in renal tubular epithelial cells mitigated AKI in mice. Mechanistically, we revealed that JOSD2 deubiquitinates Sirtuin 7 (SIRT7) and promotes the P62-mediated autophagic degradation of SIRT7, subsequently alleviating renal tubular epithelial cell inflammation. Our study suggests that JOSD2 serves a protective role in AKI and indicates JOSD2 as a potential therapeutic target for the treatment of AKI.

Materials and methods

Antibodies

Antibodies against SIRT7 (1:1000, 29729-1-AP), P-P65(1:1000, 82335-1-RR), P65(1:1000, 10745-1-AP), FLAG-Tag (1:1000, 20543-1-AP), HA-Tag (1:1000, 51064-2-AP), and His-Tag (1:1000, 66005-1-IG) were purchased from Proteintech. Antibody against Lamin B1 (1:1000, ab133741) was purchased from Abcam. Antibodies against GAPDH (1:1000, MB001) was purchased from Bioworld. Antibodies against P62(1:1000, 5114s) and ubiquitin (1:1000, 3936S) were purchased from Cell Signaling Technology. Antibody against KIM-1(1:1000, HA721535) was purchased from HUABIO. Antibody against JOSD2 (1:1000, SAB2103354) was purchased from Sigma-Aldrich.

Animal experiments

Josd2 gene-knockout mice (Josd2^−/−) with a C57BL/6J background and littermate wild-type mice (WT) were generously provided by Prof. Fu-ping You from Peking University Health Science Center. The care and handling of all mice strictly followed the guidelines of the National Institutes of Health (USA) and was approved by the Animal Policy and Welfare Committee of Wenzhou Medical University (Approval number: Wydw2021-0182). In all animal experiments, the mice were fed standard rodent chow and water and housed in an environment controlled by 22–25 °C, 50%–60% humidity with a 12-h light/dark cycle. Male mice are generally more susceptible to AKI modeling and exhibit more severe renal damage compared to females. Their physiological responses are also more consistent, avoiding the variability associated with hormonal cycles in female mice. This makes male mice a convenient and widely accepted model for studying AKI pathophysiology. Therefore, we used only male mice in this study.

Cisplatin-induced acute kidney injury model

8–12 weeks old male Josd2^−/− mice and WT mice were randomly divided into 4 groups (n  =  6 for each group): WT + Ctrl group, WT + CP group, Josd2^−/− + Ctrl group, and Josd2^−/− + CP group. Cisplatin was purchased from Sigma-Aldrich (S1166, MO, USA). Cisplatin was dissolved in PBS solution and injected intraperitoneally into mice at a dose of 25 mg/kg [17].

Bilateral renal ischemia-reperfusion injury (IRI) induced acute kidney injury model

8–12 weeks old male Josd2^−/− mice and WT mice were randomly divided into 4 groups (n  =  6 for each group): WT + sham group, WT + IRI group, Josd2^−/− + sham group, and Josd2^−/− + IRI group. The mice were anesthetized with isoflurane. The left and right kidneys were found through the abdominal midline incision, and the renal pedicles were blocked with a non-invasive arteriolar clamp for 28 min [17]. Completely ischemic kidneys changed from red to dark purple. Then, the clamp is released and the kidney color gradually returns to red. 4-0 Vicryl sutures are used to close the muscle layer of the incision. The sham group received the same procedure without clamping. During the procedure, the mice were placed on a 37 °C thermostatic blanket and covered with sterile gauze.

Renal tubular epithelial specific overexpression of JOSD2

JOSD2 overexpression was achieved by recombinant adeno-associated virus serotype 9 (AAV9). AAV9s carrying Josd2 cDNA and a renal tubular epithelial specific promoter Ksp-cadherin (AAV-JOSD2, Bioqure Genetech, Hangzhou, China) were applied. 6-8 weeks male WT mice were injected with AAV9 via the tail vein (2E+ 11vg). The control group was injected with an AAV9-empty vector (EV). After 4 weeks, acute renal injury models were given according to the above method, which was specifically divided into Josd2^−/−-CP + AAV-EV group and Josd2^−/−-CP + AAV-JOSD2 group, or Josd2^−/−-IRI + AAV-EV group and Josd2^−/−-IRI + AAV-JOSD2 group (n  =  6 for each group). The overexpression efficiency was detected by immunoblot.

Three days after the administration of cisplatin or the induction of IRI, mice were humanely euthanized under pentobarbital sodium anesthesia. Blood samples were collected and serum creatinine and urea nitrogen levels were measured using an automated chemistry analyzer. Kidney tissues were collected for further analysis.

Histological analysis

Mouse kidney tissue was fixed using 4% polyformaldehyde and embedded in paraffin, followed by sectioning at a thickness of 5 µm. The histopathological damage in the kidney sections was assessed using hematoxylin and eosin (H&E, G1120, Solarbio, Beijing, China) staining and periodic acid-Schiff (PAS, G1285, Solarbio, Beijing, China) staining. The extent of tubulointerstitial damage was evaluated as previously described [18]. According to the percentage of the area affected by hyaline casts, tubular atrophy, and tubular lumen dilation, the sections were scored semi-quantitatively on a scale of 0 to 5+ (0, normal; 1, <10%; 2, 10%–25%; 3, 26%–50%; 4, 51%–75%; and 5, >75%). Five to eight independent fields in the kidneys were analyzed, and the average value was graphed.

Single-cell RNA sequencing (scRNA-seq)

The kidney was dissociated into single cells by dissociation solution. The single-cell nuclear suspension was added to the 10x Chromium chip according to the instructions for the 10x Genomics Chromium Single-Cell 3’ kit (V3). cDNA amplification and library construction were performed according to standard protocols. Libraries were sequenced by LC-Bio Technology (Hangzhou, China) on an Illumina NovaSeq 6000 sequencing system. Results from Illumina sequencing offline were converted to FASTQ format using bcl2fastq software (version 5.0.1).

Immunofluorescence

Immunofluorescence (IF) staining of kidney tissues was performed on 5 μm thick frozen sections. Briefly, the sections were immersed in methanol for a fixation period of 10 min to preserve tissue integrity. To prevent non-specific binding, the sections were incubated with 5% Bovine Serum Albumin (BSA) for 1 h to block any remaining unspecific binding sites. The sections were then incubated overnight with the appropriate primary antibodies. After washing to remove any unbound primary antibodies, a fluorescence-conjugated secondary antibody from YEASEN (Shanghai, China) was applied to the sections and incubated at 37 °C for 1 h. Sections were stained with 4’,6-diamidino-2-phenylindole (DAPI) and imaged by Nikon microscope imaging system (Nikon Instruments, Japan).

Cell culture and transfection

Mice renal tubular epithelial cell line (TCMK1) was obtained from Shanghai Tongpai Biological Technology (Shanghai, China). NIH/3T3 cell was obtained from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). TCMK1 and NIH/3T3 were maintained in the 1640 medium and DMEM medium (Gibco, Germany), respectively. The culture medium was supplemented with 10% FBS (Meilunbio, PWL001), 100 U/mL penicillin, and streptomycin (Sbjbio, Nan Jing, BC-CE-007).

Josd2 genes were silenced in TCMK1 by small interfering RNA (siRNA, GenePharma, Shanghai, China). The silenced sequences were transfected by LipofectAMINE™ 2000 (11668019, Thermo Fisher, Germany).

Mouse plasmids (Flag-JOSD2-WT, Flag-JOSD2-C24A, Flag-JOSD2-H125A, HA-Ub, HA-K63 Ub, and HA-K48 Ub) were synthesized by Genechem (Shanghai, China). Mouse plasmid (His-SIRT7) was synthesized by Tsingke Biotechnology Co., Ltd. (Beijing, China). These plasmids were transfected with LipofectAMINE™ 3000 (L3000015, Thermo Fisher, Germany).

Western blots and co-immunoprecipitation

Proteins were separated from renal tissues and cultured cells by RIPA lysate (P0013C, Beyotime Biological, Shanghai, China) containing a protease and phosphatase inhibitor (P1051, Beyotime Biological, Shanghai, China). Equal amounts of protein were separated by 10% SDS-PAGE gel and transferred to polyvinylidene difluoride membranes (Bio-Rad). The membranes were blocked with 5% skim milk for 1.5 h at room temperature and subsequently incubated with specific primary antibodies overnight at 4 °C. After washing with TBST and incubating with HRP-linked secondary antibodies (Beyotime Biological, Shanghai, China), images were detected using an enhanced chemiluminescence (ECL) system. Protein density quantification was analyzed using ImageJ software (Version 1.38e, NIH, Bethesda, MD).

For co-immunoprecipitation (co-IP) assays, the specific primary antibody or control IgG was added to cell and renal tissue lysates and incubated at 4 °C overnight. Then, protein G-sepharose beads (P2012, Beyotime Biological, Shanghai, China) were added into the lysate and incubated at 4 °C for immunoprecipitated. The supernatant was discarded after centrifugation at 2500 rpm, and the precipitate was washed 5 times with PBS. The protein was eluted with an SDS loading buffer for Western blot.

Liquid chromatography-tandem mass spectroscopy (LC-MS/MS) analysis

NIH/3T3 cells were transfected with either Flag-vector or Flag-JOSD2 plasmids. Subsequently, cell lysates were prepared and incubated with anti-Flag antibodies and protein G-sepharose beads. The immunoprecipitated complexes were then subjected to LC-MS/MS analysis by PTM Bio Co., Ltd. (Zhejiang, China). Candidate substrate proteins were identified and selected based on their scores and detected masses from the analysis.

RNA isolation and quantitative real-time PCR

Total RNA was isolated from renal tissues (10 mg) and TCMK1 cells with TRIzol (#9109, Takara, Japan). RNA (1 μg) was reverse transcribed to cDNA by HiScript III All-in-one RT SuperMix Reagent Kit (R333-01, Vazyme, Nanjing, China). The quantitative real-time PCR (qPCR) was performed using the SYBR Green reagent Master Mix (Q711-02, Vazyme, Nanjing, China) in QuantStudio^TM 3 Real-Time PCR System (Thermo Fisher, CA, USA). β-actin was used as an internal reference gene. Primer sequences used are listed in Supplementary Table S1.

Statistical analysis

All statistical analyses were conducted using GraphPad Prism version 8.0 (San Diego, CA, USA). Data are presented as the mean ± SD. Comparisons between the two groups were made using a two-tailed Student’s t test. For multiple comparisons, one-way analysis of variance (ANOVA) was employed, followed by Bonferroni’s correction. Two-way ANOVA with Bonferroni’s correction was applied when analyzing more than two variables. P value of less than 0.05 was considered indicative of a statistically significant difference.

Results

JOSD2 deletion exacerbates cisplatin-induced renal tubule injury and inflammation

To analyze whether JOSD2 is involved in AKI, the expression of JOSD2 in the kidney was measured. Interestingly, we observed that both the protein (Fig. 1a and Supplementary Fig. S1a) and mRNA (Fig. 1b) levels of JOSD2 were downregulated in cisplatin-challenged mice renal tissues. However, the renal mRNA levels of Josd1 and Atxn3, which are other two members of the MJD family, were not altered in response to cisplatin-induced AKI modeling (Supplementary Fig. 1b). To further evaluate the potential role of JOSD2 in AKI, we generated whole-body Josd2 knockout mice (Josd2^−/−), which were confirmed by Western blot analysis (Supplementary Fig. S1c). Cisplatin (CP) was administered intraperitoneally to induce AKI in wild-type (WT) and Josd2^−/− mice, respectively (Fig. 1c). Examinations on serum creatinine and urea nitrogen revealed that the renal functional injury was significantly aggravated in Josd2^−/− mice following cisplatin treatment (Fig. 1d, e). Furthermore, histopathological examination on Josd2^−/−-CP mice showed more severe renal tubule injury than WT-CP mice, characterized by obvious tubular dilation and brush margin loss (Fig. 1f). Quantitative analysis of renal tubular damage score confirmed the changing trends across four groups (Fig. 1g). Meanwhile, the levels of kidney injury molecule 1 (KIM-1), a biomarker of AKI [19], were further increased in Josd2^−/−-CP kidney tissues compared with WT-CP kidney tissues (Fig. 1h and Supplementary Fig. S1d, e). Deficiency of JOSD2 also increased the mRNA levels of Il6 and Tnfa in renal tissues of cisplatin-induced AKI mice (Fig. 1i, j). Together, JOSD2 deficiency aggravates renal tubule injury and inflammation in cisplatin-induced AKI mice.

Fig. 1. JOSD2 deletion exacerbates cisplatin-induced renal tubule injury and inflammation.

a Representative immunoblots of JOSD2 in renal cortex from wild-type control and CP-induced AKI mice. b Relative mRNA levels of Josd2 in renal cortex from wild-type control and CP-induced AKI mice. (n  =  6; P values were determined by two-tailed unpaired t test and data are presented as mean ± SD; **P < 0.01). c Schematic diagram depicting the procedure of CP-induced AKI mice. d, e The levels of serum creatinine and urea nitrogen were analyzed in mice. f Representative images of hematoxylin and eosin staining (H&E) and periodic acid-Schiff (PAS) in mice. Scale bar: 50 μm. g Tubular injury of the kidneys quantified by tubular injury scores obtained from the sections. h Representative immunofluorescence (IF) images of KIM-1 expression from mice. Scale bar, 50 μm. i, j Real-time qPCR showing mRNA levels of Il6 and Tnfα in kidney tissues of each group. (n  =  6; For (d, e, g, i, j), P values were determined by one-way ANOVA with Bonferroni’s correction and data are presented as mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001).

JOSD2 deletion aggravates renal tubule injury and inflammation induced by IRI

We then utilized Josd2^−/−mice to explore the role of JOSD2 in another AKI mouse model performed by ischemia-reperfusion injury (IRI). Specifically, Josd2^−/− mice and WT mice were subjected to 28 min of bilateral renal ischemia or sham surgery, followed by reperfusion for 3 days (Fig. 2a). Similarly, Western blot (Fig. 2b and Supplementary Fig. S2a) and qPCR analyses revealed a reduction in JOSD2 levels in the renal tissue of IRI-induced mice compared to normal controls (Fig. 2c), while the mRNA levels of Josd1 and Atxn3 were not altered (Supplementary Fig. S2b). In addition, JOSD2 deficiency exhibited more severe renal injury in IRI mice, as indicated by serum creatinine and urea nitrogen examinations (Fig. 2d, e). Histological examinations by H&E and PAS staining showed that renal tubule injury was elevated in response to IRI, which was further increased in the Josd2^−/− mice (Fig. 2f). Correspondingly, higher tubular injury scores were graded in Josd2^−/−-IRI mice than in WT-IRI mice (Fig. 2g). Immunofluorescence staining and Western blot of KIM-1 further validated the enlarged degree of renal tubular injury in Josd2^−/−-IRI mice (Fig. 2h and Supplementary Fig. S2c, d). In addition, the mRNA levels of Il-6 and Tnfα in renal tissues were significantly increased in Josd2^−/−-IRI mice compared to WT-IRI mice (Fig. 2i, j). These results suggest that JOSD2 deficiency also exacerbates IRI-induced AKI in mice.

Fig. 2. JOSD2 deletion aggravates renal tubule injury and inflammation induced by IRI.

a Schematic diagram depicting the procedure of IRI-induced AKI mice. b, c Representative immunoblots and relative mRNA levels of JOSD2 in renal cortex from wild-type control and IRI-induced AKI mice. (n  =  6; P values were determined by two-tailed unpaired t test and data are presented as mean ± SD; ***P < 0.001). d, e The levels of serum creatinine and urea nitrogen were analyzed in mice. f Representative images of H&E and PAS in mice. Scale bar: 50 μm. g Tubular injury of the kidneys quantified by tubular injury scores obtained from the sections. h Representative IF images of KIM-1 expression from mice. Scale bar, 50 μm. i, j Real-time qPCR showing mRNA levels of Il6 and Tnfα in kidney tissues of each group. (n  =  6; For (d, e, g, i, j), P values were determined by one-way ANOVA with Bonferroni’s correction and data are presented as mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001).

JOSD2 negatively regulates inflammation and P65 activation in renal tubular epithelial cells

To elucidate the cellular source of JOSD2 in the kidney, we performed single-cell RNA sequencing (scRNA-seq) analysis of kidney cells from wild-type mice. Based on the expression of specific marker genes (Supplementary Fig. S3a), 10 main cell types were classified, including podocyte (Podo), mesangial cell (MC), endothelial cell (EC), macrophage (MФ), loop of Henle (LH), connecting tubule (CNT), distal convoluted tubule (DCT), principal cell (PC), intercalated cell (IC-A, IC-B), and proximal tubule segment 1-3 (PT S1-S3) (Fig. 3a). Notably, Josd2 mRNA expression was mainly distributed in proximal tubule cells (Fig. 3a and Supplementary Fig. S3b). Furthermore, double immunofluorescence staining confirmed that JOSD2 was expressed in AQP1-positive tubular epithelial cells (Fig. 3b) and its expression was reduced in both cisplatin-challenged and IRI-operated mice.

Fig. 3. JOSD2 negatively regulates inflammation and P65 activation in renal tubular epithelial cells.

a Clustering and cell-type identification of nuclei using wild-type mice. UMAP plot of the combined dataset is shown by splitting conditions. We identified all major cell types in the kidney, including podocyte (Podo), mesangial cell (MC), endothelial cell (EC), macrophage (MФ), loop of Henle (LH), connecting tubule (CNT), distal convoluted tubule (DCT), principal cell (PC), intercalated cell (IC-A, IC-B) and proximal tubule segment 1-3 (PT S1-S3). Biaxial scatter plot showing the expression pattern of JOSD2 (purple) in these main cell types. b Representative immunofluorescence staining images of kidney tissues from wild-type control and CP- or IRI-induced AKI mice, showing immunoreactivity to JOSD2 (green) and AQP1 (red). Tissues were counterstained with DAPI (blue). Scale bar, 50 μm. c, d Volcano plot analysis and GO enrichments of DEGs in CP-treated TCMK1 cells transfected with Flag or Flag-JOSD2. e, f TCMK1 cells transfected with Flag-JOSD2 were stimulated with CP for 12 h. Real-time qPCR showing mRNA levels of Il6 and Tnfα. (n  =  3 independent experiments; P values were determined by one-way ANOVA with Bonferroni’s correction and data are presented as mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001). g, h TCMK1 cells transfected with si-JOSD2 were stimulated with CP for 12 h. Real-time qPCR showing mRNA levels of Il6 and Tnfα. (n  =  3 independent experiments; P values were determined by one-way ANOVA with Bonferroni’s correction and data are presented as mean ± SD; **P < 0.01, ***P < 0.001). i, j TCMK1 cells transfected with Flag-JOSD2 or si-JOSD2 were stimulated with CP. Representative immunoblot of KIM-1 expression. k The enrichment plot of NF-kappa B signaling pathway. l TCMK1 cells transfected with Flag-JOSD2 were stimulated with CP for 30 min. Representative immunoblot of phosphorylated and total P65 levels. m Representative immunoblot of phosphorylated and total P65 levels in kidney tissues of each group.

To investigate the regulatory mechanism of JOSD2 in renal tubular epithelial cells, we performed RNA-seq on cisplatin-challenged TCMK1 cells overexpressing JOSD2 (Fig. 3c). Gene Ontology (GO) analyses revealed that JOSD2 regulates immune and inflammatory responses (Fig. 3d). Then, we observed that overexpression of JOSD2 significantly suppressed the increases in Il6 and Tnfa mRNA levels in TCMK1 cells challenged by cisplatin (Fig. 3e, f and Supplementary Fig. S3c). Conversely, silencing JOSD2 expression via siRNA exacerbated CP-induced inflammation in TCMK1 (Fig. 3g, h and Supplementary Fig. S3d). The regulation of renal tubule injury by JOSD2 followed a similar trend, as evidenced by the KIM-1 examination (Fig. 3i, j). KEGG enrichment analyses of the RNA-seq gene expression profiling further showed that JOSD2 negatively regulates the NF-κB signaling pathway (Fig. 3k). NF-κB is a central transcription factor for inflammatory gene expression, and p65 phosphorylation is the key step and landmark in NF-κB activation [20]. Subsequent Western blot analysis confirmed that JOSD2 overexpression suppressed cisplatin-induced P65 phosphorylation in TCMK1 cells (Fig. 3l). In contrast, JOSD2 knockdown aggravated NF-κB activation (Supplementary Fig. S3e). Similar results were observed in the kidney tissues of Josd2^−/− mice (Fig. 3m and Supplementary Fig. S3f, g). Collectively, these findings suggest that JOSD2 in renal tubular epithelial cells mitigates NF-κB-mediated inflammation, thereby alleviating the renal tubule injury.

Renal tubular epithelial cell-specific restoration of JOSD2 alleviates cisplatin- and IRI-induced renal tubule injury and inflammation

Then, we tested the protective effect of JOSD2 in renal tubular epithelial cells against AKI. We constructed AAV9s carrying Ksp-cadherin promoter to overexpress JOSD2 (AAV-JOSD2) specifically in renal tubular epithelial cells in Josd2^−/− mice. Josd2^−/− mice were injected with AAV-JOSD2 or empty vector (AAV-EV) for 4 weeks and harvested at 3 days after the cisplatin injection (Fig. 4a). Western blot and immunofluorescence analysis confirmed the successful restoration of JOSD2 in renal tubular epithelial cells (Supplementary Fig. S4a, b). Notably, the renal injury induced by CP in Josd2^−/− mice was substantially ameliorated by the restoration of JOSD2 (Fig. 4b, c). Histological evaluations (Fig. 4d) and tubular injury scores (Supplementary Fig. S4c) indicated that AAV-JOSD2 preserved CP-induced renal tubule structural integrity. The protective effect was further confirmed by the reduced levels of KIM-1 (Fig. 4e). Additionally, qPCR and Western blot analyses revealed a decrease in the expression of inflammation cytokines (Fig. 4f, g) and P65 phosphorylation (Fig. 4h) in JOSD2-restoring mice. Subsequently, we administered AAV9-JOSD2 or AAV9-EV to Josd2^−/− mice for 4 weeks and harvested at 3 days after the IRI operation (Fig. 4i). As expected, AAV-mediated overexpression of JOSD2 in renal tubular epithelial cells effectively attenuated renal injury (Fig. 4j, k), tubular injuries (Fig. 4l, m and Supplementary Fig. S4d), and inflammation (Fig. 4n–p) in the kidneys of IRI mice. These findings demonstrated the potential of JOSD2 restoration in renal tubular epithelial cells to protect against AKI in mice.

Fig. 4. Renal tubular epithelial cell-specific overexpression of JOSD2 alleviates cisplatin- and IRI-induced renal tubule injury and inflammation.

A schematic diagram illustrating the procedure of CP-induced (a) or IRI-induced (i) Josd2^−/− mice injected with AAV-EV or AAV-JOSD2. Serum creatinine (b, j) and blood urea nitrogen (c, k) levels in the indicated mouse groups. d, l Representative images of H&E and PAS in mice. Scale bar: 50 μm. e, m Representative immunoblot of KIM-1 expression in the kidneys of indicated groups. f, g, n, o Real-time qPCR showing mRNA levels of Il6 (f, n) and Tnfα (g, o) in the kidneys of indicated groups. (h, p) Representative immunoblot of phosphorylated and total P65 levels in the kidneys of indicated groups. (n  =  6; For (b, c, f, g, j, k, n, o), P values were determined by two-tailed unpaired t test and data are presented as mean ± SD; **P < 0.01, ***P < 0.001).

JOSD2 interacts with SIRT7 and promotes autophagy-lysosomal degradation of SIRT7

DUBs exert their biological functions by modifying substrate proteins. To identify potential substrates of JOSD2, we performed an LC-MS/MS analysis using cells overexpressing JOSD2 (Fig. 5a). We were intrigued to discover that SIRT7 emerged as a putative substrate of JOSD2 (Fig. 5b). SIRT7 has been recognized to positively regulate NF-κB activation and AKI pathogenesis, and its deficiency has been shown to inhibit NF-κB and mitigate inflammation and tubular damage in AKI [21]. We confirmed the endogenous interaction between JOSD2 and SIRT7 proteins in TCMK1 cells and renal tissues using co-immunoprecipitation assays (Fig. 5c, d). In addition, the interaction between JOSD2 and SIRT7 was reduced in cisplatin-stimulated tubular epithelial cells, possibly due to the cisplatin-reduced JOSD2 expression in these cells (Supplementary Fig. S5a). Then, we co-transfected Flag-JOSD2 and His-SIRT7 plasmids into NIH/3T3 cells, demonstrating that SIRT7 interacted with JOSD2 exogenously (Fig. 5e). These findings validate a direct interaction between JOSD2 and SIRT7.

Fig. 5. JOSD2 interacts with SIRT7 and promotes autophagy-lysosomal degradation of SIRT7.

a Schematic illustration of interactome analysis for JOSD2 substrate screening. NIH/3T3 cells were transfected with Flag or Flag-JOSD2 plasmids. Anti-Flag and protein G-Sepharose beads were added to the cell samples for co-IP. The binding proteins were extracted, digested to peptide, and then subjected to LC-MS/MS analysis. b Mass spectrometry/mass spectrometry (MS/MS) spectrum of the peptide WQDPIFSLATTPLR from SIRT7. c, d Co-IP of JOSD2 and SIRT7 in TCMK1 cells and kidney tissues. Endogenous OTUD5 was immunoprecipitated. e Co-IP of exogenous Flag-JOSD2 and His-SIRT7 in lysates from NIH/3T3 expressing Flag-JOSD2 and SIRT7. Exogenous JOSD2 was immunoprecipitated using an anti-Flag antibody. f Representative immunoblot of SIRT7 in NIH/3T3 cells transfected with Flag-JOSD2 in a gradient. g–j Representative immunoblot of SIRT7 in kidney tissues of each group. k Representative immunoblot of SIRT7 in NIH/3T3 cells transfected with Flag-JOSD2, followed by proteasome inhibitor (MG132, 10 μM) or autophagy inhibitors (3-MA, 10 mM; CQ, 50 μM and Baf A1, 200 nM) for 6 h respectively. l, m Representative immunoblot of SIRT7 in NIH/3T3 cells transfected with Flag-JOSD2, followed by autophagy inducer Earle’s balanced salt solution (EBSS) treatment for the indicated time points and quantification of SIRT7. (n  =  3 independent experiments; P values were determined by two-way ANOVA with Bonferroni’s correction; *P < 0.05, **P < 0.01). n Schematic illustration of the construct for mutating the ubiquitinated lysine residue of JOSD2. o NIH/3T3 cells were transfected with Flag-JOSD2 (WT, C24A or H125A). Lysates were probed for SIRT7. p–q Representative immunoblot of SIRT7 in NIH/3T3 cells transfected with Flag-JOSD2 (WT or C24A), followed by autophagy inducer Earle’s balanced salt solution (EBSS) treatment for the indicated time points and quantification of SIRT7. (n  =  3 independent experiments; P values were determined by two-way ANOVA with Bonferroni’s correction; *P < 0.05, **P < 0.01).

Next, we investigated the impact of JOSD2 on the stability of SIRT7. As shown in Fig. 5f and Supplementary Fig. S5b, c, the protein level of SIRT7 progressively decreased with increasing levels of JOSD2 expression, while the mRNA level of SIRT7 remained unaffected by JOSD2 expression. We also observed decreased SIRT7 levels in mouse renal tissues with tubular epithelial cells-specific overexpression of JOSD2 (Fig. 5g, h). Conversely, JOSD2 deficiency increased the SIRT7 protein level in mouse kidneys (Fig. 5i, j). These results suggested that JOSD2 promotes the degradation of SIRT7. It is well-established that proteasome and lysosome-involved autophagy pathways are the primary mechanisms of protein degradation [22, 23]. To determine which degradation system is involved in JOSD2-mediated SIRT7 degradation, we utilized proteasome inhibitor and lysosomal inhibitors. As depicted in Fig. 5k, the degradation of SIRT7 mediated by JOSD2 was blocked by chloroquine (CQ), 3-methyladenine (3 MA) and bafilomycin A₁ (Baf A₁), but not by the proteasome inhibitor MG132. Furthermore, JOSD2 significantly enhanced the degradation of SIRT7 in cells incubated with the autophagy inducer EBSS (Fig. 5l, m).

The active site cysteine-24 (C24) and histidine-125 (H125) have previously been reported to be involved in JOSD2 catalytic activity and are evolutionarily conserved across species [14]. Therefore, we constructed C24A (mutation of cysteine to alanine at C24) and H125A (mutation of histidine to alanine at H125) mutants of JOSD2 (Fig. 5n). Compared to wild-type JOSD2 (JOSD2 WT), the catalytically inactive JOSD2 C24A mutant was unable to promote the degradation of SIRT7 (Fig. 5o–q). Collectively, these findings indicated that JOSD2 promotes the degradation of SIRT7 through the autophagy-lysosomal pathway via the C24 site.

JOSD2 protects renal tubular epithelial cells against cisplatin-induced injuries via reducing SIRT7

As expected, JOSD2 overexpression significantly inhibited CP-induced P65 phosphorylation in TCMK1 cells, while the C24A mutant of JOSD2 failed (Fig. 6a). Moreover, JOSD2 inhibited the nuclear translocation of the P65 in CP-challenged TCMK1 cells but C24A mutant failed (Fig. 6b, c). These results indicate that JOSD2 negatively regulates NF-κB activation in TCMK1 cells through its deubiquitinating active site C24. To validate the JOSD2-SIRT7 axis in p65 activation and inflammation, we silenced SIRT7 using siRNA in TCMK1 cells and then overexpressed JOSD2. As expected, either JOSD2 overexpression or SIRT7 silencing protected renal tubular epithelial cells from CP-induced P65 phosphorylation (Fig. 6d) and nuclear translocation (Fig. 6e), and inflammatory gene expression (Fig. 6f, g). Notably, JOSD2 overexpression failed to show more protective effects in SIRT7-deficient TCMK1 cells exposed to cisplatin (Fig. 6d–g), indicating that JOSD2 protects renal tubular epithelial cells against CP-induced inflammation via SIRT7.

Fig. 6. JOSD2 protects renal tubular epithelial cells against cisplatin-induced injuries via reducing SIRT7.

a TCMK1 cells transfected with or without Flag-JOSD2 C24A were stimulated with CP for 30 min. Representative immunoblot of P-P65 and P65. b TCMK1 cells transfected with Flag-JOSD2 (WT or C24A) were stimulated with CP for 1 h. Cytoplasmic and nuclear P65 levels were determined. GAPDH and Lamin B were used as internal controls. c Immunofluorescence staining images of TCMK1 cells stimulated with CP for 1 h with or without Flag-JOSD2 (WT or C24A). Effects on nuclear translocation of P65 (green) are shown and counterstained with DAPI (blue). Scale bar: 25 μm. d, e TCMK1 cells transfected with si-SIRT7 were pretreated transfected with Flag-JOSD2 before exposure to CP. Levels of SIRT7, P-P65 and P65 were detected by Western blot. Cytoplasmic and nuclear P65 levels were also determined. f, g Real-time qPCR showing mRNA levels of Il6 and Tnfα in TCMK1 cells of each group. (n  =  3 independent experiments; P values were determined by one-way ANOVA with Bonferroni’s correction and data are presented as mean ± SD; **P < 0.01).

JOSD2 modulates the K63-linked deubiquitination of SIRT7 and promotes the degradation of SIRT7 via P62-mediated autophagy

We investigated how JOSD2 regulates the deubiquitination and stability of SIRT7. As shown in Fig. 7a, JOSD2 overexpression remarkably decreased the ubiquitination of SIRT7. In vivo, SIRT7 ubiquitination levels were also diminished in mice injected with AAV-JOSD2 compared to the control group (Fig. 7b, c). Immunoprecipitation assays further showed that JOSD2 reduced the total and K63-linked ubiquitination of SIRT7, rather than K48, K11, K27, and K33-linked ubiquitination (Fig. 7d and Supplementary Fig. S6a). As expected, the JOSD2 C24A mutant lost the ability to remove K63-linked ubiquitin molecules from SIRT7 (Fig. 7e), indicating that JOSD2 removes K63-linked ubiquitination of SIRT7 via its active site C24.

Fig. 7. JOSD2 modulates the K63-linked deubiquitination of SIRT7 and promotes the degradation of SIRT7 via P62-mediated autophagy.

a NIH/3T3 cells were transfected with Flag-JOSD2 and then subjected to BafA₁ for 6 h. Ubiquitinated SIRT7 was enriched with anti-SIRT7 and then was detected with ubiquitin. b, c Ubiquitinated SIRT7 was detected by immunoblotting in the kidneys of indicated groups. d His-SIRT7, HA-WT Ub, HA-K63 Ub, and HA-K48 Ub were transfected into NIH/3T3 cells together with Flag-JOSD2 and then subjected to BafA₁ for 6 h. Ubiquitinated SIRT7 was detected by immunoblotting using an anti-HA antibody. e His-SIRT7 and HA-K63 Ub were transfected into NIH/3T3 together with Flag-JOSD2 (WT, C24A or H125A) and then subjected to BafA₁ for 6 h. Ubiquitinated SIRT7 was detected by immunoblotting using an anti-HA antibody. f, g NIH/3T3 cells were transfected with His-SIRT7 and HA-P62 or HA-LC3B. Anti-HA antibody was used for immunoprecipitation. h NIH/3T3 cells were transfected with HA-P62 and His-SIRT7, together with Flag-JOSD2. Protein lysates were harvested after BafA₁ treatment (6 h) for immunoprecipitation with anti-His antibody. i Representative immunoblot of SIRT7 in P62-silenced NIH/3T3 cells with or without JOSD2.

Accumulating evidence supports the idea that cargo receptors such as P62 and LC3B deliver direct specific target proteins after deubiquitination to lysosomes for autophagic degradation [24, 25]. We examined the interactions of SIRT7 with cargo receptors and found that SIRT7 specifically interacted with P62 but not LC3B (Fig. 7f, g). We further observed that JOSD2 overexpression increased the interaction between SIRT7 and P62, indicating that P62 prefers to bind to SIRT7 without ubiquitin chains (Fig. 7h). To determine whether JOSD2 controls SIRT7 degradation via P62, we silenced P62 expression using siRNA and found that JOSD2 was unable to promote SIRT7 degradation in the absence of P62 (Fig. 7i), although it still reduced SIRT7 ubiquitination (Supplementary Fig. S6b). These data suggested that JOSD2 promotes the deubiquitination of SIRT7 and subsequently increases the binding of SIRT7 to P62, thereby promoting the degradation of SIRT7 via the P62-mediated autophagic pathway.

Discussion

In this study, we observed the upregulation of JOSD2 in the renal tissue of mice with cisplatin- and IRI-induced AKI. We then showed that Josd2 gene knockout exacerbates renal tubule injury and inflammation in AKI mice. Through transcriptome analysis, we observed that JOSD2 is primarily expressed in renal tubular epithelial cells and negatively regulates inflammation and NF-κB activation. Furthermore, the overexpression of JOSD2 in renal tubular epithelial cells ameliorated renal tubule injury and inflammation in AKI mice. Mechanistically, we demonstrated that JOSD2 directly interacts with SIRT7 and removes the K63-linked ubiquitination of SIRT7 through its active site C24, subsequently promoting SIRT7-P62 interaction and P62-mediated degradation of SIRT7. JOSD2 failed to protect SIRT7-deficient TCMK1 cells exposed to cisplatin. Our study elucidates the role of JOSD2 and presents the JOSD2-SIRT7 axis in the pathogenesis of AKI, as schematically illustrated in Fig. 8.

Fig. 8. Schematic illustration of the main finding in this study.

JOSD2 interacts with SIRT7 and removes the K63-linked ubiquitination of SIRT7 through its active site C24, which subsequently prevents the phosphorylation and nuclear translocation of P65 and reduces inflammatory responses in renal tubular epithelial cells, leading to a renoprotective effect.

Increasing evidence has shown that inflammatory response in renal tubular epithelial cells is a critical mediator of AKI [26]. Upon exposure to injurious stimulation, renal tubular epithelial cells are activated and release various pro-inflammatory cytokines and chemokines, such as TNF-α, IL-6, and IL-1β. These cytokines contribute to the local inflammation in the kidneys, leading to a cascade of events that exacerbate AKI. Targeting renal tubular epithelial cell inflammation may be a strategy against AKI. Recent studies have highlighted the role of DUBs in regulating inflammation of tubular epithelial cells to intervene in the progression of AKI. Luo et al. showed that deubiquitinase BAP1 regulates the stability of BRCA1 protein and inactivates NF-κB signaling in renal tubule epithelial cells to protect mice from sepsis-induced AKI [9]. Moreover, Cezanne reduces renal inflammation and injury in response to ischemia-reperfusion by inhibiting RelA phosphorylation [27]. Here, we identified JOSD2 as a protective regulator that can reduce the expression of inflammation-related genes and the activation of P65 in tubular epithelial cells, thereby mitigating the progression of AKI induced by cisplatin or IRI. Given the small molecular mass of JOSD2 (~23 kDa), recombinant protein of JOSD2 or JOSD2-targeted gene therapy might be a promising and potential therapeutic strategy against AKI.

Sirtuins (SIRT1-7) are a family of histone deacetylases that regulate a wide variety of biological processes, including metabolism, stress responses, inflammation, tumorigenesis, and aging [28, 29]. SIRT7, the most recently identified member of SIRT, has garnered significant attention for its regulatory role in inflammation and its potential implications in various diseases, including kidney diseases. For instance, silencing SIRT7 in the pulmonary artery or microvascular endothelial cells attenuates inflammation, induces endo-mesenchymal transition, and increases vascular permeability [30]. In addition, SIRT7 deficiency exhibited diminished inflammation with a reduction in the level of multiple inflammatory factors such as TNF-α, IL-1β, and IL-6, and suppression of nuclear NF-κB p65 accumulation in cisplatin- or IRI-induced AKI [21, 31]. Therefore, SIRT7 emerges as a promising regulator in the context of inflammation and AKI. However, SIRT7 is ubiquitously expressed in almost all tissues, with the exception of skeletal muscle [32]. Here, we identified SIRT7 as a new substrate of JOSD2. In addition, we did not find SIRT1-6 as JOSD2-binding proteins in the LC-MS/MS analysis, indicating that JOSD2 specifically targets SIRT7 in the SIRT family members. JOSD2 deubiquitinates SIRT7 to promote autolysosome-related degradation of SIRT7. Compared with the direct inhibition of SIRT7, targeting JOSD2 has significant renal tubular epithelial cell specificity and may avoid potential side effects.

Post-translational modifications (PTMs) on SIRT7 have been reported to regulate its protein stability or function. SIRT7 is phosphorylated by AMPK under energy stress, which plays a crucial role in the subcellular distribution and degradation of SIRT7 [33]. Furthermore, ketone body β-hydroxybutyrate (β-OHB), an HDAC2 inhibitor, has been shown to enhance acetylation levels at H3K9 and H3K14 in Sirt7 promoter to activate SIRT7 transcription [34]. Especially, SIRT7 protein can be modified by ubiquitination and deubiquitination. FBXO7, an adaptor protein in the SKP1-Cullin-1-F-box (SCF) E3 ligase complex, has been found to promote proteasomal degradation of SIRT7 through promoting K48-linked polyubiquitination of SIRT7 at four specific lysine sites (K292, K395, K397, and K398) [35]. Su et al. found that the deubiquitinating enzyme USP17L2 interacts with and deubiquitinates SIRT7, thereby increasing SIRT7 protein stability to regulate DNA damage repair [36]. USP7 can interact with SIRT7 to remove the K63-linked ubiquitination of SIRT7 and negatively regulate SIRT7 deacetylase activity [37]. Here, we showed that JOSD2 removes the K63-linked ubiquitination of SIRT7 and promotes the degradation of SIRT7 via the P62-mediated autophagy pathway. In our acknowledgement, this is the first time to find DUB-mediated SIRT7 degradation via the autophagy-lysosomal pathway. In general, DUBs inhibit proteasome-related degradation of substrate proteins by removing ubiquitin molecules from the target protein, while some DUBs can also promote the degradation of substrates in an autophagy-dependent manner [38]. The K63-linked ubiquitination predominantly affects the activation and protein-protein interaction of substrate proteins. In this study, we showed that JOSD2-mediated deubiquitination of SIRT7 increased SIRT7-P62 interaction, suggesting that K63-linked ubiquitination of SIRT7 hinders SIRT7-P62 interaction. We present a new way to regulate SIRT7 protein stability. These findings indicate that ubiquitinating modification of SIRT7 may have more multiple regulating roles on SIRT7 protein in different cell types and under different pathological conditions.

Transcriptional factor NF-κB is of central importance in inflammation and immunity [39]. In a canonical NF-κB activation process, the upstream IκB kinase (IKK) activation phosphorylates IκBα (inhibitor of nuclear factor κB type α), which leads to the ubiquitination and degradation of IκBα. IκBα degradation results in the release of free NF-κB subunit, such as P65 (RelA), which is subsequently phosphorylated and transferred into the nuclear to trigger target gene expression [40]. In our study, we chose to monitor P65 phosphorylation and nuclear translocation, rather than the upstream IKK phosphorylation and IKB degradation, based on the existing literature that suggests a direct regulatory role of SIRT7 in the phosphorylation of P65 [30]. Indeed, our results show that JOSD2 exerts an influence on P65 phosphorylation and subsequent nuclear translocation. Unfortunately, previous publications did not illustrate how SIRT7 positively regulates P65 phosphorylation. Although SIRT7 cannot directly deacetylate P65 [31], recent studies have shown that SIRT7 exhibits desuccinylase or defatted-acylase activity [41, 42]. Thus, SIRT7 might modulate P65 phosphorylation by de-succinylation modifications. Alternatively, we could not exclude the possibility that SIRT7 may directly regulate the upstream IKK and IκBα. Anyway, the mechanism by which SIRT7 regulates the activation of NF-κB signal is very interesting and warrants further investigations. Our data showed that silencing SIRT7 abolished the regulatory effect of JOSD2 on P65 phosphorylation, suggesting that JOSD2’s regulation on P65 phosphorylation and inflammation is SIRT7-dependent in renal tubular epithelial cells.

A limitation of this study may be that we evaluated the function of JOSD2 in AKI by using global JOSD2 knockout mice rather than renal tubular epithelial-specific JOSD2 knockout mice. Although subsequent experiments have confirmed that overexpression of JOSD2 specifically in tubular epithelial cells can mitigate renal tubule injury and inflammation induced by AKI in mice, studies using renal tubular epithelial cell-specific JOSD2 knockout mice may strengthen our conclusion. In addition, the mechanism of the reduction of JOSD2 in AKI is unclear. We made initial predictions about potential JOSD2 promoter-binding transcription factors using the Transcription factor prediction website (http://genome.ucsc.edu/). Several transcription factors were predicted to regulate Josd2 gene transcription, such as ZNF460, RREB1, and RFX4. It will be of great interest to further explore whether downregulation of JOSD2 in AKI is related to these transcription factors. Also, we do think that analyzing JOSD2 expression in renal biopsy specimens would provide more direct evidence of its role in clinical AKI. In summary, our findings extend our understanding of the role of DUBs in AKI. We demonstrate the protective role of JOSD2 against AKI and present a new JOSD2-SIRT7-NF-κB axis in renal tubular epithelial cells, indicating that JOSD2-targeted gene therapy might be a promising strategy for the treatment of AKI.

Author contributions

YZ: Investigation, Formal analysis, Writing – original draft. QQZ: Investigation, Methodology. SJF: Validation, Formal analysis. DYX: Investigation. LML: Software. WL: Software. BZY: Investigation. CPZ: Resources. HZ: Writing – review & editing, Resources. ZSZ: Formal analysis. YJZ: review & editing, Conceptualization, Supervision, GL: Writing – review & editing, Conceptualization, Project administration, Resources.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41401-025-01546-2.