The chemical reagents were purchased as below: Myricitrin (Myr): Selleck, S2327; Cisplatin (CDDP): MCE, HY-17394; 3-Methyladenine (3-MA): MCE, HY-19312; Chloroquine (CQ): MCE, HY-17589A; acetylcysteine (NAC): MCE, HY-B0215. All other chemicals were of high-grade purity available from commercial sources.

The antibodies were obtained as below: GPX4 (Abcam, ab125066), FTH (proteintech, 10727-1-AP), LC3A/B (CST, 4108), NCOA4 (abclonal, A5695), SQSTM1 (CST, 5,114), GAPDH (abclonal, AC001) and HRP-conjugated Goat Anti Rabbit IgG (proteintech, SA0001-2).

The relevant materials were provided as following: Cell Counting Kit-8 Kit (Beyotime, C0038), Annexin V/PI Apoptosis Analysis Kit (KeyGEN, KGA1013), Cell Ferrous Iron Colorimetric Assay Kit (Biosharp, BL1147A), Lipid Peroxidation MDA Assay Kit (Beyotime, S0131M), GSH Assay Kit (Jiancheng Bioengineering, A006-2-1), Bodipy 581/591 C11 (abclonal, RM02821), Reactive oxygen species Assay Kit (Beyotime, S0033S), MitoSox (ThermoFisher, M36008), and JC-1 applied to detect mitochondrial membrane potential (MCE, S0131M).

The SMILES format and structure of myricitrin were obtained from PubChem and we imported these information to SwissTargetPrediction database, comparative toxicogenomics database (CTD), and PharmMapper database (Daina et al., 2019; Davis et al., 2023; Wang et al., 2017). The gene interaction network was established by using STRING database (Szklarczyk et al., 2023). Furthermore, the identified genes were imported into Cytoscape to construct a protein-protein interaction (PPI) network for analysis by utilizing the CytoNCA method (Shannon et al., 2003; Tang et al., 2015). It is our aim to identify and evaluate the potential main targets of myricitrin in this study. In addition, the biological process of the key genes was enriched by using online Metascape database and visualized as a chordal graph by the application of bioinformatics (Zhou et al., 2019; Tang D. et al., 2023).

To explore the pharmacodynamic mechanism of myricitrin on acute kidney injury stimulated by cisplatin, the disease-related targets were collected at first from the following three databases: GenCards, Online Mendelian Inheritance in Man (OMIM) database, and DisGeNET (Stelzer et al., 2016; Amberger et al., 2015; Piñero et al., 2021). The two key phrases “cisplatin induced acute kidney injury” and “cisplatin nephrotoxicity” were used for searching, and only “Homo sapiens” proteins linked to the disease from results were selected. Finally, the targets were obtained from the overlaps between the myricitrin-related targets and the disease-related targets.

The “myricitrin on cisplatin acute kidney injury” perdition target network was constructed by using Cytoscape software. In the network, proteins were represented as nodes, and interactions between those molecular species were represented as edges. Functional annotation of target genes was analyzed by using the online Metascape database. The statistical significances were defined as p < 0.05 and the gene sets containing more than five genes were significant as well. Based on the online Metascape database, the GO and KEGG analysis projects were employed to explore the predicted action targets.

Human kidney epithelial tubular HK-2 cells were grown in F12 medium with 10% fetal bovine serum and 1% streptomycin/penicillin mixture. A humidified 5% CO2 atmosphere was applied to maintain all cell cultures at 37°C. After seeded in plates and cultured for 24 h, the HK-2 cells were incubated with fresh complete culture medium, and exposed to different treatments.

To establish cisplatin (CDDP) activated cells, HK-2 cells were exposed to different concentrations of cisplatin (2.5 or 5 ug/mL) for 24 h. To explore whether ferritinophagy occurs and is involved, we first employed the autophagy inhibitor, 3-MA (5, 10 uM) and CQ (10, 20 uM), to the HK-2 cells with cisplatin co-treatment for 24 h. To observe the effect of myricitrin on cisplatin activated cells, HK-2 cells were divided into four groups: the normal control, the CDDP group exposed to 5 ug/mL cisplatin, the Myr group cultured in medium containing 5 uM myricitrin for 24 h, and the CDDP + Myr group incubated with 5 uM myricitrin for 1 h prior to cisplatin stimulation for sequentially 24 h. Then, known as the autophagy inducer, 100 nM rapamycin (Rapa) was applied to the HK-2 cells 30 min before the administration of both myricitrin and cisplatin.

Cell viability was measured by using the Cell Counting Kit-8 (CCK8) assay. Cells were inoculated in a 96-well plate with a density of 5,000 cells per well. After overnight incubation the cells were treated by cisplatin and myricitrin for 24 h described above. The cells were incubated for 4 h after we applied 10 ul CCK8 solution in each well. Finally, the absorbance value was measured at 450 nm by a microreader.

To explore the apoptosis level, the HK-2 cells from treatment groups and controls were harvested for different experimental needs, and then incubated by Annexin V-FITC and PI in the dark for at least 10 min according to the manufacturer’s protocols. Afterward, flow cytometry was utilized to examine the cells. Early apoptotic cells were determined by counting the percentage of Annexin V+/PI− cells; progressed apoptotic cells were obtained by counting the percentage of Annexin V+/PI + cells; necrotic cells were detected by counting the percentage of Annexin V-/PI + cells, and Annexin V−/PI− cells were considered as surviving cells.

According to the manufacturer’s guidelines, an iron assay kit was employed to quantify the iron content in the HK-2 cells from treatment groups and controls. Briefly, cells were harvested and further lysed in iron lysis buffer by ultrasound. After centrifuged at 12,000 rpm for 10 min, the supernatant mixed with the working reagent was incubated for 15 min at room temperature. A microreader was employed to detect the absorbance at a wavelength of 562 nm.

To investigate the MDA concentration, the HK-2 cells from treatment groups and controls were subjected to lysis by using Western & IP lysis buffer. In accordance with manufacturer’s instructions, the resulting lysate was centrifuged at 12,000 rpm for 10 min and then measured by MDA Assay Kit which implemented TBA approach. Thereafter, 532 nm absorbance was recorded by using a microplate reader.

The HK-2 cells from treatment groups and controls were homogenized by using an ultrasonic crusher and centrifuged at 12,000 rpm for 10 min after they were gathered. Then, the supernatant was collected for the following examination of GSH level. The experiment involved the process of the addition of samples mixed with GSH working mixture in the 96-well plate, which was incubated for 5 min at room temperature. The absorbance was determined at the wavelength of 405 nm.

The evaluation of lipid peroxidation was conducted by incubating different samples with 10uM Bodipy 581/591 for 1 h at 37 C in the dark. After the incubation was completed, the cells were washed twice with PBS to eliminate any surplus dye. The phenomenon of oxidized Bodipy was noticed by using a fluorescence microscope.

The intracellular ROS production was observed by using reactive species oxygen Assay Kit. The HK-2 cells from treatment groups and controls were incubated with 1:2000 DCFH-DA at 37°C for 15 min in the dark. Then, after being washed twice by PBS, the HK-2 cells were resuspended in stain buffer and measured immediately by flow cytometry. Mitochondrial ROS level was detected by Mitosox stained. After incubated with 5 uM MitoSox for 60 min, the HK-2 cells were observated in fluorescence microscope.

The Mito Probe JC‐1 Assay Kit was used to detect the changes in mitochondrial membrane potential (MMP). The HK-2 cells from treatment groups and controls were collected and incubated in 10 uM JC-1 working solution for 30 min in the dark. Then, the cells were washed twice by staining buffer and analyzed by flow cytometry within 1 h. The red aggregation showed normal MMP, while the green monomer indicated a reduction in MMP due to mitochondrial dysfunction. Results were presented as the percentage of green monomer stained by JC-1.

Western blotting was employed to detect the protein expression of GPX4, FTH, NCOA4, LC3, and SQSTM1 in the cells from treatment groups and controls. The proteins obtained from the lysate of HK-2 cells were employed in RIPA buffer on ice and then the concentration was quantified by adopting the BCA protein assay kit. After the addition of 15 µg of distinct samples into each individual well, the proteins were separated by using 10%–12% SDS-PAGE gel and transferred onto a PVDF membrane. The membrane was blocked for 2 h at room temperature in the 5% non-fatty milk. The membranes were sectioned into different portions based on the molecular weight of the target protein. Subsequently, the pieces were subjected to overnight incubation with primary antibody under the temperature condition of 4°C. After using a suitable secondary antibody for incubation, the immunoblots were eventually detected by super ECL reagent. All protein bands’ intensity was measured by ImageJ.

All experiments were conducted three to five times, with each repetition carried out independently. Data was expressed as mean + standard deviation and analyzed by using GraphPad Prism 8. Data analysis and comparison between groups used one-way ANOVA and two-tailed Student’s t-test. Values of p < 0.05 were considered statistically significant.

As shown in Figure 1A, a total of 242 target genes of myricitrin were obtained by searching three online databases (PharmMapper database, SwissTargetPrediction database, and superpre database). With the help of the STRING database, a pharmacological target of the myricitrin network was constructed. According to Cytoscape by applying CytoNCA calculation, Figure 1B showed that a network of top 27 interactional genes is obtained by setting parameters (Degree ≥18, Betweenness Centrality ≥370, Closeness Centrality ≥0.248). Then, with analysis results of the 27 genes via Metascape database, it was surprisingly noted that the biological process of the top gene was mainly enriched in cellular response to cytokines stimulus, regulation of reactive oxygen species metabolic, cellular response to lipid, and so on. In addition, it was fascinating to mention that when conducting a search in the ferroptosis database and the related literature, a significant majority of the top 27 genes were identified as hallmark indicators of ferroptosis. These genes contain PTGS2, NFE2L2, KEAP1, SLC2A1, and so on (Figure 1C). To sum up, it can be speculated that administration of myricitrin was intricately countering the incidence and progression of ferroptosis.

HK-2 cells were treated with different doses and by different times of cisplatin to determine the exact concentrations that inhibiting the cell growth. As shown in Figure 2A, cisplatin significantly inhibited the cell viability compared with the control group. Furthermore, confirmed with the above results, Figure 2B depicted that cisplatin increased the cell death-rate, which was detected by Annexin-V flow cytometry based on dose dependence. Known as specific markers for ferroptosis, Western blot was used to examine the GPX4 protein expression. The results showed that GPX4, a crucial enzyme against lipid peroxidation and ferroptosis, was also suppressed in the cisplatin group (Figure 2C). Furthermore, our results showed that cisplatin exposure led to enhanced accumulation of intracellular iron content. Lipid peroxidation was considered as a key component involved in the cell death cascade driven by ferroptosis. The amount of MDA could reflect the degree of intracellular lipid peroxidation. It was observed that cisplatin could significantly increase MDA levels in HK-2 cells (Figure 2D). C11-BOPIDY staining showed that the level of oxidation C11-BODIPY was significantly increased in the cisplatin group compared with the control group (Figure 2E).

It has been confirmed in an increasing number of studies that autophagy participates in the occurrence and development of cisplatin-related renal tubular cell injury. Moreover, ferritinophagy is an autophagy mode of ferritin degradation mediated by NCOA4, which can regulate ferroptosis. Therefore, it was speculated that ferritinophagy might be involved in HK-2 cell injury caused by cisplatin. Meanwhile, it was speculated that NCOA4, FTH1, SQTM1, and LC3II/LC3I protein-related expression were obtained by applying Western blot. Compared with the CDDP group, the expressions of NCOA4 and LC3II/LC3I were downregulated in myricitrin pretreatment group, while the FTH1 and SQTMQ expression were increasing (Figures 3A,B). However, the classical autophagy inhibitors, specifically 3-methyladenine (3-MA) and chloroquine (CQ), were all employed to suppress the ferritinophagy activity. As illustrated in Figures 3C,D, high concentrations of 3-MA and CQ pretreatment could enhance the cell viability and diminish the cell damage. These results indicated that ferritinophagy might have participated in the progress of the injury of renal tubular cells with cisplatin administration.

First of all, according to the results of cell viability (Figure 4A), myricitrin at a concentration between 0.5 and 10 uM stimulation of 24 h showed no cytotoxicity effect on HK-2 cells. Therefore, 5 uM myricitrin incubation was selected for subsequent cell experiments. Secondly, we found that the decreased cell viability induced by cisplatin could be alleviated by myricitrin pretreatment (Figure 4B). In addition, it was observed that myricitrin could promote cell survival whereas cisplatin stimulated the apparent apoptosis of HK-2 cells (Figure 4C).

In Figure 4D, the HK-2 cells with myricitrin incubation showed a dramatic increase in GPX4 expression compared with the cisplatin induction group, which showed the depletion of GPX4. Meanwhile, the contents of intracellular iron in myricitrin treatment group were obviously lower than the cisplatin group (Figure 4E). As illustrated in Figure 4F, after myricitrin intervention, the increase of MDA content stimulated by cisplatin was partially reserved, whereas the levels of GSH inhibited by cisplatin were effectively restored. At the same time, the oxidation of C11-BODIPY fluorescent intensity was suppressed by myricitrin administration, indicating a decrease in lipid peroxidation and inhibition of ferroptosis, which was in alignment with the results mentioned before. In short, these results demonstrated that myricitrin ameliorated lipid peroxidation and ferroptosis against cisplatin-induced HK-2 cell injury.

Previous research suggested the acetylcysteine, the oxidatve stress inhibitor, could attenuated the ferroptosis in various disease. The CCK-8 assays showed that does independed acetylcysteine could improved the cell viability of cisplatin induced cells (Figure 5A). Furthermore, intracellular iron content and MDA were lower in acetylcysteine group compare to the cisplatin group. Acetylcysteine could also increased the anti-oxidant GSH level (Figure 5B). In summary, we validated the positive role of ROS in the occurrence of ferrpotosis. To determine the effect of myricitrin on oxidative stress in renal tubular cells, the intracellular ROS levels were measured by using flow cytometry to detect the HK-2 cells stained with DCFDA. As shown in Figure 5C, ROS levels were significantly higher in cisplatin group, whereas both myricitrin and acetylcysteine could reverse oxidative stress via reducing intracellular ROS production. In addition, myricitrin could reduce the increased level of mitochondrial ROS, which induced by cisplatin (Figure 5D).

As the main organelle of energy metabolism and ROS production, mitochondria plays an important role in renal proximal tubular epithelial cells. In our study, it was found that cisplatin intervention in HK-2 cells resulted in a decrease in the percentage of JC-1 monomers, indicating MMP reduction and mitochondrial damage. However, lower percentage of green monomers was detected in the cisplatin-induced cells with myricitrin treatment (Figure 5E). These results manifested that myricitrin had an antioxidant effect as well as a protective function of mitochondria.

It has been demonstrated in our earlier research that the autophagic degradation of ferritin, a process referred to as ferritinophagy, can promote ferroptosis in activated HK-2 cells with cisplatin stimulation. Additionally, NCOA4 is a specific cargo receptor that facilitates ferritin degradation in lysosomes. Herein, we investigated whether the NCOA4 mediated degradation of ferritin was associated with myricitrin’s protective mechanism against ferroptosis induced by cisplatin in renal tubular cells. Western blotting revealed that the expression of vital ferritinophagy makers LC3II/LC3I and NCOA4 were downregulated, while FTH as well as SQTM1 expression were upregulated by myricitrin treatment (Figure 6A).

To further investigate the role of ferritinophagy in cisplatin-induced renal tubular cell injury, activated HK-2 cells were pretreated with the ferritinophagy inducer and rapamycin (Rapa), and co-treated with myricitrin for 24 h. Our results demonstrated that myricitrin alleviated ferroptosis by promoting the cell viability and survival (Figures 6B,C), decreasing intracellular iron level and MDA content, and increasing GSH content, which was reversed by Rapa co-treatment (Figure 6D). In addition, results from C11-BODIPY immunofluorescence showed that myricitrin significantly decreased the level of lipid peroxidation, while co-incubation with Rapa deteriorated this influence (Figure 6E). Overall, these findings revealed that myricitrin suppressed the activated HK-2 cells ferroptosis by inhibiting NCOA4-mediated ferritinophagy.

From GeneCards and OMIM, 2005 related targets for cisplatin-induced acute kidney injury were retrieved. Figure 7A displayed 94 putative targets for myricitrin acting on cisplatin induced acute kidney injury that derived from gene mapping. After removing the irrelevant genes from STRING database, the protein interaction network was created and enhanced by Cytoscape software. Figure 7B illustrated the construction of the targets of “myricitrin—cisplatin induced AKI” network. As shown in Figure 7C, the GO and KEGG pathway analysis results obtained from Metascape database, showed the enrichment in the process of response to lipid and reactive oxygen species metabolism, which was closely related to ferroptosis mechanism. Subsequently, the CytoNCA algorithm was utilized to identify the key proteins, including AKT1, SRC, EGFR, NFKB1, STAT1, PPARG, HIF1A, MTOR, ESR1, TNF, and so on (Figure 7D). Hence, it seemed sensible to assume that the AKT and NF-κB pathway might be implicated in myricitrin’s anti-ferroptosis effect on cisplatin-induced acute kidney injury.