Male 6-week-old C57BL/6 mice (weighing 20 ± 2 g) were provided by the Guangdong Medical Laboratory Animal Center (SCXK (YUE) 2018-0002, Foshan, China). The mice were housed under controlled conditions of light (12 h light/dark cycle), temperature (24°C ± 2°C) and humidity (50%–60%) and had adequate food and tap water to ad libitum. The mice were divided into four groups randomly after 1 week of acclimatization: C, control group (n = 6); L, low-dose AAI group (n = 6); M, medium-dose AAI group (n = 6); H, high-dose AAI group (n = 6). Mice in the AAI groups were administered by intraperitoneal injection of AAI (A5512, Sigma-Aldrich, St Louis, MO, United States) at the dose of 1.25 (L), 2.5 (M), and 5 mg/kg/d (H), respectively, for 5 days. Control mice were injected intraperitoneally with PBS solution containing 5% DMSO as vehicle for 5 days. Twenty-4 hours after the last dose of intraperitoneal injection of AAI or vehicle, all mice were euthanized. Blood and kidney samples were collected immediately for further experiments. All animal experiments were carried out in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals and approved by the Ethics Committee of Shenzhen Top Biotech Co., Ltd (approved ID: TOP-IACUC-2021-0137).

The mice serum was prepared by centrifuging blood for 10 min at 2,000 rpm. Then, the supernatant was transferred to a new clean EP tube. The levels of serum creatinine (SCR) and blood urea nitrogen (BUN) were measured using commercially available kits (SKT-217 and SKT-213, StressMarq Biosciences, Victoria, British Columbia, Canada), according to the manufacturer’s protocols.

The upper poles of mice kidneys were fixed in 4% paraformaldehyde overnight, dehydrated and embedded in paraffin. For periodic acid-Schiff (PAS) staining, kidney wax blocks were cut into 4 μm sections, dewaxed, and rehydrated. Then the sections were oxidized in 0.5% periodic acid solution for 5 min. After rinsing in distill water, the sections were placed in Schiff reagent for 15 min and washed in lukewarm tap water for 5 min. The nuclei were counterstained with hematoxylin. Kidney lesions were scored by estimating the percentage of tubules in the cortex exhibiting tubular atrophy, tubular dilatation, vacuolar degeneration, loss of brush border, and cell necrosis and shedding with the following manner (0-5 points): 0 = none; 1 = ≤ 10%; 2 = 11–25%; 3 = 26–45%; 4 = 46–75%; 5 = ≥ 76% (Cortes et al., 2018). Three mice from each group were blindly assessed for tubular injury score using three microscopic fields (×200) per mouse.

The frozen kidney cortex (approximately 20 mg) was pulverized in liquid nitrogen and resuspended and homogenized in 400 μL RIPA lysis buffer (#9806, Cell Signaling Technology, Beverly, MA, United States) containing a protease-phosphatase inhibitor cocktail (A32961, Pierce, Rockford, IL, United States). The homogenate was centrifuged at 12,000 rpm for 10 min at 4°C, and the supernatant was collected for quantification by Bradford protein assay kit (C503031, Sangon Biotech, Shanghai, China). Equal amounts of protein were separated on 10% SDS-PAGE and subsequently transferred onto polyvinylidene difluoride membranes by electroblotting. Electrophoresis conditions: Tris-Glycine SDS Running Buffer, 90–120 V, 2 h. Electrotransfer conditions: Tris-Glycine Transfer Buffer, 90 V, 1.5 h. After being blocked with non-fat milk for 1 h at room temperature, these membranes were incubated with primary antibodies overnight at 4°C. The primary antibodies against fetuin B (Fetub, 67002-1-ig), cytochrome P450 family 2 subfamily E member 1 (Cyp2e1, 19937-1-AP), acyl-CoA synthetase medium-chain family member 3 (Acsm3, 10168-2-AP) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, 60004-1-IG) were purchased from Proteintech (Wuhan, China). Anti-microsomal glutathione S-transferase 1(Mgst1, ab131059) was purchased from Abcam (Cambridge, MA, United States) and anti-β-actin (A5441) was from Sigma-Aldrich (St Louis, MO, United States). Horseradish peroxidase (HRP)-conjugated anti-mouse IgG (SA00001-1) obtained from Proteintech (Wuhan, China) and HRP-conjugated anti-rabbit IgG (#7074) obtained from Cell Signaling Technology (Beverly, MA, United States) were used for secondary antibody incubation at room temperature for 1 h. Immobilon^® ECL Ultra Western HRP Substrate (WBULS0500, Merck, Billerica, MA, United States) working solution was prepare by gently mixing solutions A and B in a 1:1 ratio and was used to incubate the blots for 2 min at room temperature. Finally, immunoblotting was observed, and image densitometry analysis was performed by Image Lab™ software (Bio-Rad Laboratories, Hercules, CA, United States).

Statistical analysis and graphing were performed using GraphPad Prism 9 software (GraphPad Software, Inc., San Diego, CA, United States). The significance of the differences among groups was examined by one-way analysis of variance (ANOVA) followed by post hoc analysis with Dunnett’s test. Data are presented as mean ± standard error of the mean (SEM). p-value less than 0.05 was considered statistically significant.

The results of kidney function-related indexes in mice are shown in Figures 1A, B. Both SCR and BUN levels were considerably higher in the high-dose AAI group than in the control group (p < 0.001). The medium-dose AAI group showed a significant increase in SCR (p < 0.05) compared with control, while no significant difference was observed in the low-dose AAI group. Furthermore, no significant difference was found in BUN levels in low- and medium-dose AAI groups in contrast to control. In PAS staining, the control group showed clear structure of renal cortex and tight arrangement of renal tubules (Figure 2A). On the contrary, in the AAI groups, necrosis and exfoliation of proximal convoluted tubular epithelial cells were observed along with dilation of tubular lumen, vacuolar degeneration, and infiltration of inflammatory cells in certain areas (Figure 2A). Pathological injuries were scored according to the scoring criteria mentioned above (Figure 2B). These data indicated that the degree of kidney injury in mice was positively correlated with the intake of AAI.

To improve the quality of proteomics analysis results and reduce the false positive rate, the quality control of the screening results was carried out. Peptides with too short sequences (less than 7 amino acids) are mostly filtered out due to their simple composition, while peptides with too long sequences (generally higher than 40 amino acids) are difficult to be identified by mass spectrometry due to their high molecular weight. Most peptides in this study were in the range of 7–20 amino acids in length (Supplementary Figure S1A). The distribution of the number of missed cut sites of peptides reflects the completeness of enzymatic cleavage. The results showed that the peptide with 0 missed cleavage number was the most abundant, which indicated that the enzymatic cleavage was complete and favorable for identification (Supplementary Figure S1B). The mass deviation of the peptide m/z was normally distributed in the range of −10 to 10 ppm, indicating the ideal mass accuracy of the mass spectra (Supplementary Figure S1C). Moreover, no contamination (marked red) was found in the analysis (Supplementary Figure S1D). Collectively, these quality control data ensured the accuracy of the results of this study.

After the screening of the above conditions, we obtained a total of 4,141 proteins. To investigate the function of proteins, functional database annotation was made on the identified proteins. The proteins mainly localized in the cytoplasm (30.99%), nucleus (20.56%), and mitochondria (15.53%) (Supplementary Figure S2A). Domains are the basic units of protein structure, function and evolution. The most annotated domain was P-loop_NTPase (Supplementary Figure S2B). In GO annotation, integral component of membrane, protein transport, and adenosine triphosphate (ATP) binding was the most annotated entry in cellular component (CC), biological process (BP), and molecular function (MF), respectively (Supplementary Figure S2C). In KEGG annotation, the metabolic pathways were the most enriched pathway (Supplementary Figure S2D). In the KOG database, the top 3 protein functions were general function prediction only, signal transduction mechanisms, and posttranslational modification, protein turnover, chaperones (Supplementary Figure S2E).

A total of 3,802 proteins with relative quantitative values were further subjected to expression analysis. In Figure 3A, correlation analysis revealed a high correlation of protein expression between the three samples within each group, which indicated good biological replication between the samples within the group. Compared with the control group, the correlation of protein expression between groups became smaller as the AAI dose increased. This suggested that the more severe the AAN model the greater the difference in renal protein expression, with the most pronounced difference between the control and high-dose AAI (5 mg/kg/d) group (Figure 3A). Principal component analysis confirmed that protein expression differences between groups were amplified with increasing AAI dose (Figure 3B).

The DEPs were identified according to the criteria of Fold change >2 or <0.5 and p-value <0.05. Compared to the control group, 31 proteins were upregulated, and 27 proteins were downregulated in the AAI low dose group (Figure 4A). Doubling of the AAI dose increased the number of upregulated proteins to 91 and the number of downregulated proteins to 119 (Figure 4B). The highest number of DEPs reached 271 (145 upregulated and 126 downregulated) in the AAI high dose group compared to the control group (Figure 4C). Intersection of these 3 sets of comparisons yielded 37 common DEPs (Figure 4D). The expression trends of these 37 proteins in each sample were presented in Figure 4E and Supplementary Table S1. These results helped to clarify the proteins that could be affected by all three doses of AAI. The expression of these proteins correlated with the occurrence and severity of AAN.

GO annotation analysis of DEPs was performed to describe the role of genes and proteins in cells, to comprehensively describe the characteristics of genes and gene products in organisms. In CC, 37 DEPs were mainly concentrated in cytoplasm, extracellular region, and endoplasmic reticulum (Figure 5A). These DEPs were involved in some biological processes of BP, among which lipid metabolic process was the most important (Figure 5B). As for MF, it was mainly related to metal ion binding, identical protein binding, and hydrolase activity (Figure 5C). The diagram of KEGG pathway analysis showed seven pathways, and the most important one was metabolic pathways (Figure 5D). PPI network analysis showed that 23 DEPs were interconnected, while the other 14 DEPs showed no association. Cytoscape software was used to present the overall perspective of the relationship within 23 DEPs (Figure 5E). According to the value of betweenness, Cyp2e1, Acsm3, Fetub, Mgst1 were screened as the most important DEPs (Figure 5F).

The relative abundance of Cyp2e1, Acsm3, Fetub, and Mgst1 in the proteomics analysis was summarized in Figures 6A–D. Exposure to AAI suppressed the expression of Cyp2e1 and Acsm3 but enhanced the expression of Fetub and Mgst1. This result was further validated by Western blot analysis. As shown in Figures 6E–I, compared with the control group, the expression of Acsm3 and Cyp2e1 in the AAI groups were significantly downregulated, while the expression of Fetub and Mgst1 were upregulated in the AAI groups. It is worth noting that AAI modulated the expression levels of these proteins in a dose-dependent manner.