All C57BL/6J mice used in this study were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The mice were male, 8 weeks, and housed under standard laboratory conditions with a 12-h light/dark cycle, temperature maintained at 22 °C–25 °C, and humidity at 40%–60%. They had free access to food and water. The mice were randomly divided into three groups: (I) Control group; (II) UIRI group; (III) UIRI + NM treatment group. The UIRI (unilateral ischemia-reperfusion injury) kidney model was established as follows: under general anesthesia, a midline abdominal incision was made, and the left renal pedicle was clamped for 35 min using a vascular clip. During ischemia, mice were kept in a thermostatic chamber at 38 °C. After 35 min, the clip was removed, and the abdomen was sutured layer by layer. NM (HY-B0190A, MCE) was dissolved in saline at a concentration of 10 mg/kg (Yan et al., 2022). Intraperitoneal administration began on day 4 post-UIRI and continued for 7 days. On day 10, the contralateral kidney was removed, and on day 11, the mice were euthanized for blood and kidney sample collection. The unilateral ureteral obstruction (UUO) mouse model was established with male C57BL/6J; 129 mice at 8 weeks of age. The left ureter of mice was exposed through the left incision and ligated with 4–0 silk thread. The sham operation group only received kidney exposure surgery and did not undergo ligation surgery. The left kidney was collected 7 days after the operation. The folic acid-induced nephropathy (FAN) mouse model was established with male C57BL/6J; 129 mice at 8 weeks of age. Renal injury was induced by intraperitoneal injection of 200 mg/kg folic acid dissolved in 0.3M NaHCO3. Mice in the control group only received intraperitoneal injection of 0.3M NaHCO3. The kidneys of the mice were collected 28 days after injection.

Human proximal tubular epithelial cells (HK-2) were cultured in DMEM/F12 medium supplemented with 10% fetal bovine serum (FBS) and 1% antibiotics (penicillin-streptomycin) at 37 °C under 5% CO₂. Three experimental treatments were performed: (I)TGF-β Stimulation Assay: HK-2 cells were inoculated in six-well plates and stimulated for 24 h in serum-free medium with TGF-β (2.5 ng/mL) (P01137, R&D Systems), and were divided into the control group and the TGF-β group. (II)NM Intervention Assay: The HK-2 cells cultured in vitro were co-incubated with TGF-β and NM (100 μM) in serum-free medium for 24 h and were divided into: the control group, the TGF-β group and the TGF-β + NM group. (III)GZMB Stimulation Assay: In serum-free medium, HK-2 cells were stimulated with GZMB cytokines (5 μg/mL) (Shen et al., 2016) (ab307481, Abcam) for 24 h and were divided into: the control group and the GZMB group. (IV) GZMB and NM Intervention Assay: The HK-2 cells cultured in vitro were co-incubated with GZMB and NM in serum-free medium for 24 h and were divided into: the control group, the GZMB group and the GZMB + NM group.

Protein lysates from HK-2 cells or renal tissues were extracted using RIPA buffer supplemented with PMSF, protease inhibitors, and phosphatase inhibitors. Protein concentrations were quantified using a BCA protein assay kit. Protein samples were separated by electrophoresis on 8%, 10%, or 12% SDS-PAGE gels and subsequently transferred to PVDF membranes. The membranes were blocked with 5% non-fat milk and incubated overnight at 4 °C with primary antibodies (see Supplementary Table S1). After washing, the membranes were incubated with horseradish peroxidase (HRP)-conjugated anti-mouse or anti-rabbit IgG secondary antibodies for 1 h at room temperature. Protein signals were detected using enhanced chemiluminescence (ECL) substrate. Western blot images were compiled using Adobe Photoshop, quantified with ImageJ software (v1.8.0), and statistically analyzed using GraphPad Prism 8.

Total RNA was extracted from HK-2 cells and mouse renal tissues using TRIzol reagent (Takara). cDNA was synthesized using the HiScript II Reverse Transcription System (Vazyme, Nanjing, China). Quantitative real-time PCR (qRT-PCR) was performed using the ChamQ SYBR Color qPCR Master Mix kit (Vazyme). Primer sequences are provided in Supplementary Table S2.

The impact of GZMB and perforin co-treatment on cell viability was assessed using the CCK-8 assay. HK-2 cells were seeded in 96-well plates for 24 h, followed by 24-h co-incubation with GZMB (2 μg/mL) and perforin (160 ng/mL). According to the manufacturer’s protocol (CCK-8 kit, C0037, Beyotime), 100 μL of culture medium containing 10% CCK-8 reagent was added to each well. The plates were then incubated at 37 °C for 2 h. Absorbance was measured at 450 nm using a microplate reader.

Collecting 1.5 mL of whole blood from mouse and keeping it at 4 °C overnight. On the following day, separating the serum by centrifuging the sample at 4 °C, 3000rpm for 10 min to isolate the serum. Withdrawing 30 μL for Serum creatinine (Scr) measurement and storing the remaining serum at −80 °C for future use. Scr levels were measured using the DICT-500 system (Bioassay Systems, Hayward, CA, United States). Blood Urea Nitrogen (BUN) levels were measured using the corresponding kits (NO.C013-2-1) (Nanjing Jiancheng Company, China).

HK-2 cells were plated on coverslips and cultured for 24 h. The medium was then replaced with serum-free basal medium, and cells were treated with TGF-β (2.5 ng/mL) and NM (100 μM), categorized into control, TGF-β, and TGF-β + NM groups. Subsequent experiments were conducted after 24h of treatment. HK-2 cells were grown on coverslips and fixed with pre-chilled methanol for 15 min. The cell-covered coverslips were then incubated overnight at 4 °C with primary antibodies against fibronectin and α-SMA. The following day, samples were incubated for 1 h at room temperature with donkey anti-rabbit (H + L) secondary antibody (light-protected conditions), followed by nuclear staining with DAPI. Fluorescence images were acquired using a LSM780 confocal laser scanning microscope system (ZEISS, Oberkochen, Germany).

Renal tissue samples were fixed in 4% paraformaldehyde and embedded in paraffin. Sections (3 μm thick) were deparaffinized, rehydrated, and subjected to Masson’s trichrome staining to assess collagen deposition and fibrotic lesions. Quantitative analysis of fibrotic lesions was performed using ImageJ software (v1.8.0), with at least three randomly selected images analyzed per mouse.

The renal tissue sections were dewaxed and rehydrated into water through a series of graded alcohols. The sections were stained with hematoxylin for 10 min, then rinsed under running tap water to remove excessive staining agent. After differentiation and bluing, rinsed again. Then, re-stain with eosin staining solution for 50s. Seal the sheet after dehydration and cleaning. Finally, morphological observations were conducted using an optical microscope (Motic China Group Co., Ltd., China).

Total RNA was extracted from HK-2 cells and kidney tissues of mice subjected to ischemia-reperfusion and NM treatment using TRIzol^® Reagent (Thermo Fisher, Cat#15596026). The purity and concentration of RNA were measured using a NanoDrop 2000 spectrophotometer, while RNA integrity was assessed with an Agilent 2100 Bioanalyzer/LabChip. After confirming sample quality, library construction was performed as follows: eukaryotic mRNA was enriched using magnetic beads with Oligo (dT) primers, followed by fragmentation into 200–300 bp segments. First-strand cDNA and second-strand cDNA were synthesized using mRNA as a template, purified, and subjected to end repair, A-tailing, and adapter ligation. Finally, PCR amplification was performed to enrich the cDNA library. Quality control was conducted to ensure library integrity. Sequencing was then carried out on the Illumina NovaSeq 6,000 platform using a PE150 (Paired-End 150 bp) sequencing mode. Finally, data analysis was conducted, DEGs were identified using DESeq2 or edgeR, with a significance threshold of p < 0.05. Functional annotation of DEGs was performed through GO analysis, while pathway enrichment was assessed via KEGG analysis using Cluster Profiler.

All data are presented as Mean ± SD. Data analysis and visualization were performed using GraphPad Prism (GraphPad Software, San Diego, CA, United States). Two-tailed Student’s unpaired t-test analysis was utilized to compare between two groups. One-way ANOVA followed by LSD or Dunnett post-test was performed to compare more than three groups. A p-value <0.05 was considered statistically significant. All experiments were performed with at least three independent replicates.

To investigate the role of GZMB in renal fibrosis, we established multiple murine models of kidney fibrosis. In the UIRI-induced renal fibrosis model, Western blot and qRT-PCR analyses revealed significant upregulation of GzmB in fibrotic kidneys compared to controls (Figures 1A–C). Meanwhile, UUO operation was performed on mice, and it was found that the expression levels of GzmB protein and mRNA in the kidneys of this model increased (Figures 1D–F). We also further analyzed through Western blot and found that the FAN model enhanced the expression of GZMB in the kidneys of mice (Figures 1G,H). To further explore the effect of fibrosis on the expression of GZMB in vitro, we stimulated human proximal renal tubular (HK-2) cells with TGF-β, a key mediator of renal fibrosis, to induce the occurrence of fibrosis. The Western blot results indicated that TGF-β stimulation increased the protein level of GZMB (Figures 1I,J). These findings suggest that GZMB is a potential participant in the pathogenesis of renal fibrosis.

To explore the effects of the human recombinant protein GZMB on cells, we added GZMB to HK-2 cells for cultivation. After 24 h, we evaluated the damage, proliferation, inflammation, and p-EMT of HK-2 cells (Figure 2A). Dual-field microscopy revealed that compared to the control group, the number of HK-2 cells treated with GZMB was significantly reduced (Figure 2B). The decline ratio can reach 56%. To further confirm the cytotoxic effects of GZMB, we analyzed the expression of Ki-67, a cell proliferation marker, and Ngal, a kidney injury marker, via Western blot, as shown in Figures 2C–E, the protein expression levels of Ki-67 and Ngal in GZMB-treated cells significantly decreased, indicating that GZMB treatment inhibited cell proliferation and exacerbated cell damage. Meanwhile, we also detected the mRNA expression of LCN2 and HAVCR1, both of which consistent results were obtained (Figures 2F,G). Next, we examined the effect of GZMB on the expression of pro-inflammatory factors in HK-2 cells. qRT-PCR results demonstrated that the mRNA levels of IL1B, IL6 and CXCL8 were significantly upregulated, suggesting that GZMB triggers an inflammatory response in human renal tubular epithelial cells (Figures 2H–J). Additionally, to assess the potential role of GZMB in p-EMT, we analyzed related genes. GZMB treatment significantly increased the mRNA levels of FN1, ACTA2, TGFB1 and COL1A1, indicating that GZMB promotes fibrotic responses in HK-2 cells (Figures 2K–N). In summary, these findings demonstrate that GZMB exacerbates cell damage, induces the release of inflammatory factors, and contributes to the progression of p-EMT.

Since GZMB exerts its pro-apoptotic function in a perforin-dependent way, we examined the expression of perforin in fibrotic kidneys. In two classical kidney fibrosis models, we observed a significant increase in the expression of Prf gene (Figures 3A,B). To verify whether GZMB interacts with perforin, we co-cultured HK-2 cells with both GZMB and perforin in the medium and assessed cell viability using the CCK-8 assay. The results showed that the combined treatment led to a significant decrease in cell viability (Figures 3C,D), indicating that GZMB-mediated cytotoxicity in a perforin-dependent way.

To further reveal the gene expression changes induced by the combined action of GZMB and perforin through transcriptomic analysis, we performed RNA-seq on HK-2 cells co-incubated with GZMB and perforin compared to control cells. The Pearson correlation heatmap showed that the correlation coefficients between intra-group samples were close to 1, confirming good experimental reproducibility. While the transcriptomic profile in negative control (NC) group significantly differed from the GZMB and PRF treated (GZMB + PRF) group (Figure 3E). Principal component analysis (PCA) further supported these findings, revealing low intra-group variation and high reproducibility in both GZMB + PRF and NC groups, while a significant difference was observed between NC and GZMB + PRF groups along PC1 (51.21%) (Figure 3F). Volcano plot analysis identified 2,294 differentially expressed genes (DEGs) meeting the threshold criteria (FDR<0.05, |log2(FC)|>1), including 848 upregulated and 1,446 downregulated genes (Figure 3G). Additionally, a heatmap of DEGs expression further demonstrated that samples from the same treatment group exhibited similar gene expression patterns, while significant differences were observed between NC and GZMB + PRF groups (Figure 3H).

We performed functional annotation and enrichment analysis on the identified differentially expressed genes (DEGs). For the differentially expressed genes (DEGs) enriched in biological processes, we further constructed a network diagram, which revealed that the DEGs were primarily clustered in regulation of smooth muscle cell proliferation, transforming growth factor beta receptor signaling pathway, indicating p-EMT (Figure 4A). Furthermore, KEGG pathway analysis categorized the enriched DEG pathways into four major branches: Cellular Processes, Environmental Information Processing, Genetic Information Processing, and Human Diseases. Within Cellular Processes, apoptosis-related DEGs constituted the largest subset (33 genes), followed by endocytosis, phagosome, and tight junction pathways. In Environmental Information Processing, the PI3K-Akt signaling pathway contained the highest number of associated genes (50 genes), with the MAPK signaling pathway ranking second. For Genetic Information Processing, protein processing in the endoplasmic reticulum was the dominant pathway (22 genes). In Human Diseases, cancer-related DEGs were most abundant (89 genes), followed by Herpes simplex virus 1 infection. These findings collectively indicate that most GZMB-activated pathways are implicated in regulating cellular damage, apoptosis, and inflammatory responses (Figure 4B). The KEGG enrichment network analysis of differentially expressed genes (DEGs) revealed their predominant enrichment in the Pathways in cancer (Figure 4C), these findings strengthen the association between GZMB/PRF and the regulation of apoptosis and cell cycle. We performed Gene Set Enrichment Analysis (GSEA)-based GO biological pathway analysis were primarily enriched in collagen trimer (Figure 4D) and extracellular (Figure 4E). GSEA-based KEGG pathway analysis revealed significant downregulation of DNA replication-related genes (Figure 4F), while demonstrating marked upregulation of ferroptosis-associated gene sets (Figure 4G). These results demonstrate that the differentially expressed genes are predominantly associated with apoptosis, inflammation, and fibrotic signaling, along with a distinct disruption of extracellular matrix homeostasis. These findings cell damage characterized by apoptosis, p-EMT, and ferroptosis in PRF abundant.

To further investigate whether NM could ameliorate GZMB-induced p-EMT, cell damage, and inflammatory responses, HK-2 cells were co-cultured with NM and GZMB. Western blot analysis demonstrated that the expression levels of p-EMT-associated markers, including Fibronectin and α-SMA, as well as the renal injury biomarker Ngal, were significantly reduced in the NM-treated group (Figures 5A–D). These findings demonstrated that NM effectively mitigates the p-EMT process in tubular cells and alleviates cell damage. Additionally, to assess the potential anti-inflammatory effects of NM, we examined the expression of inflammation-related genes such as IL1B, IL6, and CXCL8. The results showed a notable decrease in the expression levels of these genes following NM treatment (Figures 5E–G), indicating that NM can effectively suppress the production of inflammatory cytokines. In summary, these results demonstrate that NM can alleviate GZMB-induced cellular injury, inhibit inflammatory cytokine release, and slow the progression of p-EMT.

We have previously demonstrated that GZMB, a serine protease, induces renal tubular injury. To further explore regulatory mechanisms, we analyzed the impact of the serine protease inhibitor NM on renal injury progression. To determine the potential protective effect of NM in renal p-EMT process, we co-cultured HK-2 cells with TGF-β in the presence or absence of NM. CCK8 found that compared with the control group, NM with a concentration of 100 μM had no significant effect on cell activity (Supplementary Figure S1A). Western blot demonstrated that NM treatment can improved the expression of Fibronectin, E-cadherin and Vimentin induced by TGF-β (Figures 6A–E). Meanwhile, qRT-PCR, and IF staining consistently demonstrated that NM treatment significantly suppressed TGF-β-induced fibronectin expression (Figures 6F,G). TGF-β, a key cytokine implicated in EMT (Ram et al., 2022b), significantly upregulated fibronectin expression in HK-2 cells. Importantly, administration of the serine protease inhibitor NM attenuated TGF-β-induced p-EMT. During EMT, some transformed mesenchymal cells differentiate in α-SMA-positive myofibroblasts, which excessively produce ECM (Yuan et al., 2022b). As a core marker of myofibroblasts, the expression level of α-SMA indirectly reflects the progression of EMT. Furthermore, NM treatment markedly downregulated α-SMA expression in TGF-β-stimulated HK-2 cells (Figure 6H), suggesting that NM may downregulate fibronectin expression by inhibiting myofibroblasts activation. In summary, NM exerts potential anti-fibrotic effects by inhibiting the TGF-β signaling pathway and suppressing the expression of fibrotic markers in renal tubular epithelial cells.

Given NM’s demonstrated efficacy in mitigating renal fibrosis by blocking TGF-β signaling in vitro, we employed a murine UIRI model to evaluate its potential in vivo effects. In the UIRI model, daily intravenous NM administration was initiated at D4 post-surgery, followed by contralateral nephrectomy at D10 and sample collection at D11 (Figure 7A). Renal function assessment revealed that while UIRI induced elevated Scr and BUN levels, NM treatment significantly attenuated this increase (Figures 7B,C). Next, we evaluated the effect of NM treatment on renal fibrosis in mice. As shown in Figures 7D,E, NM ameliorated UIRI-induced upregulation of Fibronectin and Collagen I expression, with quantitative analysis of their protein levels presented in Figures 7F,G mRNA expression analysis of fibrotic markers yielded consistent results with the protein expression patterns (Figures 7H,I). Masson’s trichrome staining revealed substantial collagen deposition in renal interstitium following UIRI, which was significantly attenuated by NM treatment (Figures 7J,K). Collectively, these in vivo findings demonstrate that NM exerts potent renoprotective and anti-fibrotic effects by modulating ECM remodeling and suppressing key fibrotic factors, thereby mitigating the progression of ischemic injury-induced renal fibrosis.

To further confirm the protective effect of NM treatment on renal injury in UIRI mice, we analyzed the expression levels of E-cadherin, a surface marker of healthy tubular cells, and Caspase-9, a common marker of cellular injury and apoptosis, by Western blot (Figures 8A,B). Compared with the control group, the UIRI group showed significantly reduced E-cadherin protein levels, which were restored by NM treatment (Figure 8C), indicating that NM preserves renal epithelial integrity. Additionally, the UIRI group exhibited markedly increased Caspase-9 protein levels, while the NM-treated group showed a downward trend (Figure 8D), suggesting that NM inhibits apoptosis. The Western blot analysis revealed that the expression of TGF-β was significantly upregulates in the UIRI group, and the protein level recovered after NM treatment (Supplementary Figures S2A,B). These findings further confirm that the promotion of renal fibrosis is partially mediated through the activation of the TGF-β signaling pathway. In addition, we assessed the impact of ischemic injury on mitochondrial function and observed a marked downregulation of PGC-1α expression in the UIRI group. Notably, NM treatment restored the expression of PGC-1α, suggesting that NM can protect the kidneys by improving mitochondrial function (Supplementary Figures S2A–C). H&E staining showed that the structure of renal tubular was significantly damaged after UIRI, and NM treatment could improve this phenomenon (Supplementary Figures S2D,E). In summary, UIRI leads to significant apoptosis, loss of epithelial characteristics, pro-fibrotic process, and mitochondrial function impairment. NM treatment can effectively improve these pathological changes and demonstrate a good protective effect on the kidneys.

To elucidate the molecular mechanisms underlying NM’s therapeutic effects, we performed RNA-seq analysis on renal tissues. As shown in Figure 9A, high correlation coefficients between replicate samples confirmed experimental reproducibility. We identified 155 upregulated and 40 downregulated DEGs in NM-treated mice compared to UIRI controls (Figure 9B). Heat map revealed significant gene expression differences between UIRI and NM groups (Figure 9C). Functional enrichment analysis of DEGs identified the top 10 most significantly enriched GO terms, suggesting a prominent enrichment of genes associated with inflammatory cell chemotaxis (Figure 9D). KEGG pathway analysis of the 15 most prominent DEGs highlighted the IL-17 signaling pathway for further investigation (Figure 9E). The IL-17 signaling pathway activates the downstream Act1-TRAF6 complex via IL-17A binding to IL-17RA, leading to the induction of c-Fos (Figure 9F). In GZMB-stimulated HK-2 cells, IL17B mRNA expression was significantly elevated after 24 h (Figure 9G). Correspondingly, in vivo studies demonstrated altered expression of both Il17a (Figure 9H) and Il17b (Figure 9I). Western blot analysis confirmed that NM treatment reduced c-Fos protein expression in UIRI kidneys (Figures 9J–L). These results demonstrate that NM exerts renal protection by inhibiting the IL-17/c-Fos signaling pathway.