8-weeks-old male C57BLKS/J db/db mice and control db/m mice were purchased from Gem Pharmatech Co., Ltd. (Nanjing, China). They are type 2 diabetes model mice due to a spontaneous mutation and have obvious characteristics of type 2 diabetes (Li et al., 2023). The animal experiment protocol was approved by the Institutional Animal Use and Care Committee of Nanjing Medical University (Ethical No. IACUC-2204022) and was conducted in strict accordance with the Guidelines for the Use and Management of Laboratory Animals issued by the National Institutes of Health of China. All the mice were kept in an SPF room at a temperature of 22 °C ± 2 °C, a relative humidity of 50% ± 5%, a 12 h light-dark cycle, and provided food and water freely. After 2 weeks of adaptation, unilateral nephrectomy was performed on the mice. The mice were randomly divided into four groups (n = 6): the db/m group, db/m + PET group, db/db group and db/db + PET group. Eight weeks after the surgery, the PET treatment groups were given 10 mg/kg/d PET by gavage for 8 weeks according to previous studies (Nagaoka et al., 2019). Then, all the mice were anesthetized via intraperitoneal injection of pentobarbital sodium (30 mg/kg) and sacrificed via cervical dislocation. Blood samples were collected and stored at −80 °C for later use. The kidney tissues were partially fixed in 4% paraformaldehyde (PFA) for histological examination.

The human proximal renal tubule cell line (HK2) was supplemented with 10% bovine serum (ServiceBio, Wuhan, China), 100 U/mL penicillin and 100 μg/mL streptomycin (Gibco, China) and was cultured in 5.5 mmol/L glucose DMEM/F-12 (BasalMedia, Shanghai, China) at 37 °C, 95% humidity and 5% CO₂. HG intervention: HK2 cells received culture medium supplemented with 30 mmol/L glucose for 48 h. PET treatment: Petunidin chloride (purity ≥99%) was obtained from MCE (United States). PET was dissolved in DMSO, and then added to the cells for 48 h at different concentrations (25, 50, or 100 μM).

Body weight and glucose were assessed every 4 weeks. The mice were placed in a metabolic cage for 24 h, and urine was collected at week 0 and week 26. Mouse urine samples were centrifuged at 4 °C for 15 min, and the supernatant was collected. The concentrations of urinary albumin and creatinine were determined via a mouse urine protein ELISA kit (Shanghai Viao, China). Then, the urinary albumin/urinary creatinine ratio (ACR) was calculated. Mouse serum creatinine and blood urea nitrogen (BUN) were tested by ELISA kit (Shanghai Viao, China). A total of 10 mg of kidney tissue from each sample was ground to obtain the supernatant. The supernatant was centrifuged at 4 °C for 15 min, and the results were measured with an ELISA kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, Jiangsu, China) according to the manufacturer’s instructions.

The kidney tissues were fixed with 4% paraformaldehyde (PFA), embedded in paraffin, and cut into 4 μm thick slices. The kidney sections were stained with hematoxylin-eosin (HE), Schiff periodate (PAS), Masson and Sirius Red.

For immunofluorescence staining, the frozen sections were infiltrated with 0.3% Triton X-100 for 10 min, blocked with 3% BSA for 1 h, and incubated with the primary antibody at 4 °C overnight. On the second day, after the samples were washed with PBS, they were incubated with a fluorescent secondary antibody (1:500) at room temperature in the dark for 1 h. DAPI (1:500) was applied 8 min before fluorescence microscopy. The samples were observed via confocal microscopy, and the fluorescence intensity was analyzed via ImageJ software. The immunofluorescence procedure for cells was similar to that for kidney tissues. The primary antibodies used for immunofluorescence are shown in Supplementary Table S1.

Fresh kidney tissues and cultured HK2 cells were washed with PBS, mitochondria separation reagent (Beyotime, Shanghai, China) was added, they were homogenized in an ice bath approximately 10 times, and centrifuged at 4 °C for 5 min. The supernatant was transferred to another centrifuge tube and centrifuged at 11,000 × g at 4 °C for 10 min. The supernatant contained the isolated cytoplasm, and the precipitate contained the isolated mitochondria.

Protein samples were extracted from mouse kidney tissues, HK2 cells, or isolated mitochondria via lysis and centrifugation. In accordance with the manufacturer’s instructions, MDA (Nanjing Jiancheng Bioengineering Institute, Nanjing, Jiangsu, China) test kits were used to measure the contents of MDA. The glutathione (GSH) and catalase (CAT) activities were determined via GSH and CAT assay kits (Beyotime, China). Superoxide dismutase (SOD) and manganese-SOD (Mn-SOD) activities were measured via Cu/Zn-SOD and Mn-SOD assay kits with WST-8 (Beyotime, China) (Chen et al., 2025b).

To determine 12-hydroxyeicosatetraenoic acid (12-HETE) and 15-HETE levels, 12-HETE and 15-HETE ELISA kits (Abcam, United States) were used according to the product instructions. To detect 4-hydroxynonenal (4-HNE), the OxiSelectTM 4-HNE Assay Kit (Cell Biolabs Inc., United States) was utilized following the kit’s instruction manual.

Total iron content was measured with an iron assay kit (Abcam) according to the product instructions. Briefly, mouse kidney tissue or HK2 cells in iron assay lysis buffer were rapidly homogenized and centrifuged at 16,000 × g at 4 °C for 10 min. The supernatant was then incubated with an iron reducer at 25 °C for 30 min, followed by incubation with iron probe buffer for 60 min at 25 °C in the dark. A microplate reader (Molecular Devices) was used to read the absorbance at 593 nm.

The lipid peroxidation (LPO) content was measured via a lipid peroxidation assay kit according to the manufacturer’s instructions (Nanjing Jiancheng Bio, China). In brief, the chromogen reagent was combined with lipid peroxide and incubated at 45 °C for 60 min to stabilize the chromophore. Readings were subsequently obtained via a microplate reader at 586 nm. The lactate dehydrogenase (LDH) release assay was conducted using an LDH cytotoxicity assay kit (MedChemExpress, Monmouth Junction, China).

The protein from kidney tissues and HK2 cells was extracted from the lysate, and the concentration was determined via the bicinchoninic acid (BCA) method. The protein extracts were separated and electrophoresed on a 10% polyacrylamide gel (SDS‒PAGE) at voltages of 80 V, 30 min and 100 V for 90 min. Then, the gel was transferred to a PVDF membrane. The membrane was blocked with fast block blocking buffer for 15 min. Primary antibodies were added to the membrane overnight at 4 °C. The membrane was subsequently washed with TBS-T and incubated with secondary antibodies for an hour at RT. The bands were displayed on a Tanon-5200 Multi. β-actin was used as an internal reference. The primary antibodies used for Western blotting are shown in Supplementary Table S1.

Apoptosis was measured with a TUNEL apoptosis assay kit. Kidney tissues were fixed with 4% PFA and infiltrated with 0.1% Triton X-100. Then, the samples were incubated with 50 μL of biotin-dUTP buffer at 37 °C for 1 h, and the positive samples were observed via fluorescence microscopy (Olympus, Tokyo, Japan).

Cell viability was measured using a CCK-8 assay (Topscience, Shanghai, China) according to the manufacturer’s instructions. HK2 cells were cultured in 96-well plates, incubated at 37 °C for 12 h, and then cultured in 25, 50, or 100 μM PET containing 30 mM glucose for 48 h. 100 μL of CCK-8 working solution were added to each well. A microplate reader was used to measure the absorbance at 450 nm.

In accordance with the manufacturer’s instructions, the ROS levels in renal tissues and HK2 cells were measured using a dihydroethyl (DHE, Sigma) probe and a 2′,7′-dichlorodihydrofluorescein (DCFH-DA) fluorescent probe (Beyotime, Shanghai, China) and then observed under an inverted microscope.

The mitochondrial membrane potential was detected with a JC-1 fluorescent probe (Beyotime, Shanghai, China). One milliliter of JC-1 dye solution was added, the mixture was incubated at 37 °C for 20 min, and then the mixture was washed with JC-1 dyeing buffer twice. An inverted microscope was used for image acquisition.

HK2 cells were collected and freeze-thawed under liquid nitrogen, and cell lysates were diluted with PBS and divided into two aliquots, with one aliquot being added to PET and the other being added to DMSO as control. After incubation at 37 °C for 1 h, all lysates were equally divided into nine PCR tubes, and each tube of cells was heated separately at different temperatures (37 °C, 40 °C,43 °C, 49 °C, 52 °C, 55 °C, 58 °C, 61 °C, and 64 °C) for 3 min. After that, samples were centrifuged at 4 °C, and the soluble fractions were subjected to Western blot analysis (Chen et al., 2025a; Fu et al., 2025).

All data are expressed as the mean ± standard error of the mean (SEM). Statistical analysis was performed with GraphPad Prism 9 software (San Diego, CA). Student’s t test, one-way ANOVA followed by Tukey’s test, and two-way ANOVA followed by Tukey’s test were performed for comparisons. P values <0.05 were considered significantly different.

Whether PET can alleviate diabetic kidney damage is not known. For this purpose, we treated db/db mice with PET by gavage. Compared with those of the db/m group, the blood glucose and body weight of the db/db group were significantly increased, whereas those of the PET group decreased (Figures 1A,B). The results suggested PET has an antidiabetic effect, which is consistent with previous studies (Strugała et al., 2019). Pet alleviated the indicators of diabetic kidney injury, including serum creatinine, BUN and urinary ACR in db/db mice (Figures 1C–E). The effects of PET on the morphological changes in db/db mice were observed via HE, PAS, Masson and Sirius Red staining. HE and PAS staining revealed obvious pathological changes in the kidney tissues of db/db mice, including glomerular hypertrophy, mesangial matrix dilatation, tubular lumen dilatation and glycogen deposition, which were alleviated by PET treatment (Figure 1F). The results from Masson and Sirius red staining and Western blot analysis of fibrosis-related proteins revealed that PET treatment mitigated renal fibrosis (Figures 1F–H). Thus, PET attenuated db/db mouse kidney injury.

Given that PET has previously been shown to have antioxidative stress effects, we investigated whether PET treatment could mitigate oxidative stress in db/db mice. PET treatment inhibited the increase in renal ROS levels in db/db mice (Figure 2A). Compared with those in the db/m group, the MDA levels were significantly increased, and the GSH, CAT and total SOD activities were significantly decreased in the db/db group (Figure 2B). In contrast, PET treatment reversed these oxidative stress injuries in db/db mice (Figure 2B).

Mn-SOD (SOD2) is an essential mitochondrial antioxidant regulatory enzyme that plays an important role in regulating mitochondrial homeostasis (Chen et al., 2020). We found that PET could increase the activity of Mn-SOD in kidney tissue (Figure 2C) and we further evaluated markers of mitochondrial oxidative stress in db/db mouse kidney tissue. The MDA content increased, and the GSH and CAT activities decreased in the mitochondria of db/db mice, which was alleviated by PET treatment (Figure 2C). TUNEL staining revealed that PET reduced apoptosis in diabetic renal tissue (Figure 2D). These results indicated that PET treatment could alleviate oxidative stress in renal tissue and mitochondria.

Previously, we confirmed that SOD2 levels were reduced in mitochondria in db/db mice and that PET reversed this trend. SOD2 inactivation is associated with its acetylation (Ac), and PGC-1α expression is a key indicator for evaluating mitochondrial biosynthesis. Western blot analysis revealed decreased levels of SOD2 and PGC-1α and increased levels of Ac-SOD2 in db/db mice, while PET treatment reversed these trends (Figure 3A), which suggested that PET could improve mitochondrial function in diabetes.

To further evaluate the specific molecular mechanism under PET treatment, we detected the expression levels of mitochondrial oxidative stress-related proteins. The Keap1/Nrf2 pathway plays an important role in the regulation of oxidative stress. The results of renal tissue immunofluorescence revealed that Keap1 expression increased and Nrf2 expression decreased in db/db mice, whereas PET treatment reversed these trends (Figure 3B). In view of the significant improvement in mitochondrial function by PET, the distributions of proteins related to the Keap1/Nrf2 pathway in the cytoplasm and mitochondria were further clarified. The results revealed that only the mitochondrial distribution of NQO1 were significantly decreased (Figures 3C,D). However, PET treatment significantly facilitated the mitochondrial translocation of NQO1 (Figures 3C,D). Moreover, among the proteins associated with oxidative stress, significant changes in Keap1 expression were observed (Figure 3C). Therefore, it was reasonable to hypothesize that PET might reduce oxidative stress damage by targeting Keap1 and NQO1 mitochondrial translocation (Figures 3C,D).

Given the significant changes in Keap1 protein levels, we speculated that PET might influence Keap1 protein expression and predicted the effect of PET on the Keap1 protein via molecular docking methods. The results revealed that 8 hydrogen bonds and 2 salt bonds formed between PET and Keap1, with a predicted docking score of −11.48. This finding suggested a possible direct interaction between PET and Keap1 (Figure 3E). To study the relationship between PET and Keap1, db/db mice were treated with different concentrations of PET (2.5, 5, or 10 mg/kg), and the results revealed that the Keap1 protein expression level was negatively correlated with the PET concentration (Figure 3F). Furthermore, the cellular thermal shift assay (CETSA) was used to demonstrate the biological interaction between Keap1 and PET, which showed that PET incubation led to the stabilization of Keap1 (Figure 3G). These data suggest that Keap1 might be a potential target for PET to ameliorate oxidative stress.

NQO1 is a target gene of NRF2 and plays a role in ferroptosis (Zuo et al., 2023). We next explored whether the oxidative stress injury caused by diabetes was associated with ferroptosis. Ferroptosis is related to iron overload, lipid peroxidation and weakened antioxidant capacity. The results revealed that the levels of total iron, LPO, 12-HETE and 4-HNE in the cytoplasm and mitochondria were significantly increased in db/db mice, whereas treatment with PET resulted in the opposite trend (Figures 4A,B). GPX4, SLC7A11, ACSL4, FTH1, and TFR1 are the key indicators of ferroptosis. Immunofluorescence staining and Western blot analysis were used to assess the levels of ferroptosis-associated proteins. Immunofluorescence staining revealed reduced GPX4 expression in db/db mice, and PET treatment increased GPX4 expression (Figure 4C). Furthermore, Western blot analysis revealed that the expression of GPX4, SLC7A11, and FTH1 decreased in db/db mice and that the expression of ACSL4 and TFR1 increased, the effects of which were reversed by PET treatment (Figure 4D). These findings suggested that PET could improve ferroptosis in db/db mice.

Furthermore, we confirmed the protective effect of PET on oxidative stress in HK2 cells in vitro. HK2 cells were cultured in medium supplemented with 30 mM glucose and treated with different PET concentrations (0, 25, 50, and 100 μM). PET treatment increased cell viability, reduced LDH release, and suppressed Keap1 expression in a dose-dependent manner (Figures 5A–C). The benefits of PET reached peak at 50 μM, therefore, 50 μM was chosen as the dose for subsequent in vitro experiments.

To further investigate the role of Keap1 in oxidative stress, we overexpressed the Keap1 gene in HK2 cells via viral transfection (Figure 5D). In accordance with the animal experiments, PET treatment alleviated cellular and mitochondrial oxidative stress injuries in HK2 cells (Figures 5E–G). Additionally, Pet alleviated mitochondrial injury as evidenced by improved mitochondrial membrane potential (Figure 5H). However, after the overexpression of Keap1, the beneficial effects of PET were compromised (Figures 5E–H). The above experimental results indicated that Pet alleviated HG-induced oxidative stress in HK2 cells by inhibiting Keap1.

To further demonstrate that PET affects oxidative stress and mitochondrial function by targeting Keap1, we examined the expression levels of a series of oxidative stress-related proteins. The results showed that PET treatment increased the protein expression levels of Nrf2 in HK2 cells induced by high glucose but decreased the protein expression level of Keap1 (Figure 6A). However, the overexpression of Keap1 almost completely blocked these PET-mediated changes (Figure 6A). PET increased the protein levels of SOD2 and PGC-1α and alleviated HG-induced acetylation of SOD2, which was abolished by Keap1 overexpression (Figure 6B). These data further suggested that PET improved oxidative stress damage and improved mitochondrial function by inhibiting Keap1.

Animal experiments revealed that PET inhibited Keap1 expression and promoted the mitochondrial translocation of NQO1 in the kidney tissue of db/db mice. Therefore, in subsequent experiments, we further investigated the role of NQO1 in high glucose-stimulated HK2 cells and its relationship with Keap1. PET treatment significantly increased the expression level of NQO1, and promoted the translocation of NQO1 to mitochondria, the effects of which were obviously inhibited by Keap1 overexpression (Figure 6C), indicating that PET promoted NQO1 upregulation and mitochondrial translocation by inhibiting Keap1.

Similarly, we investigated the effects of PET on ferroptosis in HK2 cells. The total iron and LPO contents and the levels of 12-HETE and 4-HNE increased in the high-glucose intervention group, which was reversed by PET treatment (Figure 6D). In addition, Western blot analysis revealed that PET treatment effectively attenuated high glucose-induced ferroptosis, but was abolished by the overexpression of Keap1 (Figure 6E). Taken together, these results suggested that Pet alleviated high glucose-induced oxidative stress, impaired mitochondrial function and promoted ferroptosis by regulating Keap1 in HK2 cells.

To further confirm the role of mitochondrial NQO1 in alleviating oxidative stress, NQO1 with a mitochondrial translocation peptide (mitoNQO1) was overexpressed in HK2 cells (Figure 7A). Western blotting was used to detect the transfection efficiency and specificity of mitoNQO1 (Figure 7A). As previously shown, overexpression of Keap1 significantly inhibited cell viability and increased cell damage under high-glucose conditions, and these effects were eliminated by overexpression of mitoNQO1 (Figures 7B,C). In addition, when Keap1 was overexpressed, cellular and mitochondrial MDA and ROS contents increased, and cellular and mitochondrial GSH, CAT, total SOD and Mn-SOD activities and the mitochondrial membrane potential decreased (Figures 7D–G). However, these effects of Keap1 overexpression were largely eliminated in the case of mitoNQO1 overexpression (Figures 7D–G). These findings confirmed the key roles of the Keap1/mitoNQO1 pathway in PET-mediated renal protection against oxidative stress and mitochondrial function under high-glucose conditions.

To further study the influence of the Keap1/mitoNQO1 pathway on oxidative stress, Western blotting was also performed, and the results revealed that Keap1 overexpression suppressed Nrf2 expression under high glucose conditions, whereas mitoNQO1 overexpression promoted its expression (Figure 8A). The Western blotting results revealed decrease protein levels of SOD2 and PGC-1α, along with an increase in the protein level of Ac-SOD2 caused by Keap1 overexpression, whereas mitoNQO1 overexpression counteracted these harmful effects (Figure 8B).

In addition, overexpressing mitoNQO1 alleviated ferroptosis under Keap1 overexpression, primarily evidenced by reduced the levels of 12-HETE and 4-HNE, as well as the total iron and LPO contents (Figure 8C). Moreover, the expression of ferroptosis-related proteins could be improved by mitoNQO1 overexpression (Figure 8D). Together, these findings suggest a permissive role of the Keap1/mitoNQO1 pathway in PET-mediated renal protection against ferroptosis under high-glucose conditions.