Dimethyl fumarate prevents ferroptosis to attenuate acute kidney injury by acting on NRF2

Epidemiological studies indicate that acute kidney injury (AKI) is still rapidly increasing every year.1 Currently, there is no satisfactory clinical treatment for AKI. Ferroptosis is an iron-dependent form of regulated necrosis.2, 3 Iron metabolism and lipid peroxidation were the main inductors of ferroptosis.2, 3 This process can be inhibited by several endogenous negative regulators of ferroptosis, including GPX4 and NRF2.3, 4 Dysregulated ferroptosis has been observed in AKI induced by ischemia/reperfusion, cisplatin, and folic acid.5 However, there is a lack of agents targeting ferroptosis for AKI treatment. Dimethyl fumarate (DMF), an oral therapeutic small-molecule drug has been approved by the US FDA in 2013 for therapy of multiple sclerosis and psoriasis.6 DMF is considered as a prodrug because it is rapidly cleaved into monomethyl fumarate (MMF) following oral administration in vivo.7, 8 Both DMF and MMF can induce NRF2 gene transcription and inhibit the degradation of NRF2, resulting in the activation of NRF2.8 Several studies have shown that DMF protects against tissue fibrosis and diabetic nephropathy possibly by improving inflammation and redox balance.9-11 However, the effect of DMF on AKI still needs further investigation. In a cisplatin-induced AKI mouse model (Supporting Methods), our results showed that 10 mg/kg DMF pretreatment remarkably decreased the levels of blood urea nitrogen (BUN) and serum creatinine (SCr) (Figure S1A,B; Figure 1A). Meanwhile, the substantial renal histological damage induced by cisplatin was also significantly improved after DMF treatment (Figure 1B). The Western blotting showed that the elevated kidney injury markers including KIM-1 and NGAL were markedly decreased after DMF treatment (Figure 1C). Furthermore, the enhanced TUNEL-positive cells induced by cisplatin were significantly decreased after DMF treatment (Figure 1D). Accordingly, the upregulation of Bax and cleaved caspase-3 induced by cisplatin was also suppressed by DMF (Figure 1E). The results of qRT-PCR (primer sequences are listed in Table S1) confirmed increased expression of MCP-1, TNFα, IL-1β, IL-6, and COX-2 in the kidneys of cisplatin-treated mice, which was significantly attenuated by DMF therapy (Figure 1F). Furthermore, the ELISA assay showed that circulatory inflammatory factors including TNFα (Figure 1G) and IL-6 (Figure 1H) were significantly reduced after DMF treatment. Next, genome-wide transcriptome analysis was performed to determine possible mechanisms of DMF therapy. The RNA-seq analysis identified more than 9000 differentially expressed genes in the vehicle versus vehicle + cisplatin groups (Supporting File S1). KEGG pathway enrichment analysis detected enriched genes associated with cisplatin-induced cell death, mainly including apoptosis, necroptosis, and ferroptosis (Figure 2A), with higher enrichment of ferroptosis-associated pathways than apoptosis and necroptosis (Figure 2B). A heat map of ferroptosis-related genes showed that the expression of GPX4 was significantly decreased after cisplatin treatment and was remarkably restored by DMF therapy (Figure 2C). The results of qRT-PCR showed that DMF treatment restored the mRNA levels of GPX4 (Figure 2D). Furthermore, DMF markedly suppressed the enhanced mRNA levels of positive ferroptosis regulators including ACSL4 (Figure 2E) and PTGS2 (Figure 2F) along with the restoration of the reduced GPX4 (Figure 2G). Interestingly, the protein levels of the NRF2 were efficiently increased after DMF treatment (Figure 2G,J; Figure S2A), which was accompanied by reduced lipid peroxidation (Figure 2I,K,M; Figure S2B), oxidative stress (Figure 2H,N), and mitochondrial damage (Figure 2L; Figure S2C). As renal tubular epithelial cells are the most vulnerable cells in AKI,1 the human renal tubular epithelial cells (HK2) were used to verify the effects of DMF in vitro. DMF is rapidly cleaved into MMF in vivo; hence, we used MMF for cell culture experiments. The MMF concentrations used in this study were determined by a CCK-8 assay (Figure S3). Interestingly, MMF markedly induced nuclear accumulation of NRF2 according to the data of Western blotting (Figure 3A) and IF staining (Figure 3B). The results of flow cytometry (Figure 3D) and LDH assay (Figure S3D) showed that the cell death induced by cisplatin was reduced after MMF treatment. Meanwhile, the MMF treatment enhanced GPX4 and decreased cleaved caspase3 determined by Western blotting (Figure 3C; Figure S3C). Accordingly, DMF inhibited cisplatin-induced ferroptosis in HK2 cells as evidenced by lower levels of lipid peroxidation (Figure 3E; Figure S3H) and MDA (Figure S3F), and restored GSH (Figure S3E) and mitochondrial membrane potential (Figure S3I). Additionally, we also verified the protective effect of MMF against cisplatin-induced cell death in renal mesangial cells (Figure S3L). To determine whether NRF2 mediated the protective effect of DMF, NRF2-knockout HK2 cells (NRF2 KO) were constructed by CRISPR/Cas9 (Figure 3G; Figure S3G), and the results showed knockout of NRF2 greatly blunted the protective effects of MMF against cisplatin-induced ferroptosis in HK2 cells (Figure 3F–K; Figure S3J,K). Finally, the effects of DMF on AKI were also examined in folic acid (FA)- and ischemia/reperfusion (IR)-induced AKI. As shown in Figure 4A,D,G, DMF treatment markedly improved renal pathology and decreased the levels of BUN, SCr, KIM-1, and NGAL after FA treatment for 72 h. Moreover, DMF activated NRF2 (Figure 4F) and ameliorated FA-induced ferroptosis as evidenced by the upregulation of GPX4 (Figure 4F) and the improvement of lipid peroxidation (Figure 4B,E; Figure S4C), oxidative stress (Figure 4C), mitochondrial damage (Figure S4D), and cell death (Figure 4H). Strikingly, similar protective effects of DMF were also observed in an IR-induced model as shown by the markedly improved renal pathology (Figure 4L) and decreased the levels of BUN, SCr, KIM-1, and NGAL after IR for 24 h (Figure 4I,M; Figure S4E). Furthermore, IR-induced ferroptosis was similarly inhibited by DMF (Figure 4J,K,N,O; Figure S4F). Apoptosis, necroptosis, and ferroptosis are the main pathological features of cisplatin- and IR-induced AKI.5 However, a study reported that ferroptosis not necroptosis was the primary cause of FA-induced AKI.12 Our results demonstrated that DMF prevented ferroptosis and ameliorated AKI possibly by acting on NRF2 and anti-peroxidation, suggesting the clinical potential of DMF for AKI therapy. This work was supported by grants from the National Key Research and Development Program (2016YFC0906103), the National Natural Science Foundation of China (81700642), the Natural Science Foundation of Jiangsu Province (no. BK20170150), the China Postdoctoral Science Foundation (2018M640505), and Jiangsu Postdoctoral Science Foundation (2018K042A). National Key Research and Development Program, Grant number: 2016YFC0906103; National Natural Science Foundation of China, Grant number: 81700642; Natural Science Foundation of Jiangsu Province, Grant number: BK20170150; China Postdoctoral Science Foundation, Grant number: 2018M640505; Jiangsu Postdoctoral Science Foundation, Grant number: 2018K042A The authors declare that there is no conflict of interest. Yunwen Yang, Fangfang Cai, Suwen Liu, and Ning Zhou performed the experiments and prepared the figures. Songming Huang, Zhanjun Jia, and Yunwen Yang designed the experiments, analyzed the data, and wrote the main text of the manuscript. Peipei Wang, Shengnan Zhang, Aihua Zhang, and Yue Zhang offered the assistance with the manuscript preparation, and all the authors reviewed the manuscript. The data that support the findings of this study are available from the corresponding authors upon reasonable request. All animal procedures of this study were approved by the Institutional Animal Care and Use Committee of Nanjing Medical University (registration number: IACUC 14030112-2). Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.