1008 - XPO1 inhibitor Selinexor induces heme oxygenase-1 mediated p53-dependent ferroptosis in chronic lymphocytic leukemia

Chronic lymphocytic leukemia (CLL) is a chronic disease resulting from B-cells clonal proliferation. In recent years, CLL treatment has entered the era of non-chemotherapy, and exploring new treatment targets and strategies has important value. Exportin-1 (XPO1) is a critical nuclear-cytoplasmic transport protein in cells that is responsible for the unidirectional export of cargo proteins (including tumor suppressors) from the nucleus to the cytoplasm, and is closely related to the occurrence, development, and drug resistance of various tumors. In previous studies, we found high expression of XPO1 is associated with poor prognosis and showed the effect of Selinexor on proliferation inhibition, cell cycle arrest, and apoptosis, and observed effective inhibition of the NF-κB pathway and up-regulation of the FOXO pathway in CLL (Clin Exp Med. 2023 Feb 4.).
In our previous study, we found that the results of CCK8 were not consistent with the flow cytometry results of apoptosis when treating CLL cells with a relatively low concentration of Selinexor. This suggested that there may be other pathways of cell death activated. Therefore, we used a variety of cell death inhibitors, including Z-VAD-FMK (an apoptosis inhibitor), Necrosulfonamide (a necrosis inhibitor), 3-MA (an autophagy inhibitor), and the iron chelator DFO (a ferroptosis inhibitor), to treat CLL cell lines. We compared the effects of each inhibitor on the drug activity of Selinexor before and after treatment and found that only the ferroptosis inhibitor DFO could significantly reverse the inhibitory activity of Selinexor on p53 wild-type JVM3 cells (Fig 1A). This suggested that the drug activity of Selinexor can be exerted through the ferroptosis pathway.
After detecting the total ROS and lipid peroxidation levels in cells after Selinexor treatment, we found that the lipid ROS level increased significantly in JVM3 cells (Fig 1B-C). Overproduction of Fe2+ was also detected in JVM3 cells after treatment (Fig 1D). Then, we used RNA-Seq to perform high-throughput sequencing on JVM3 cells before and after Selinexor treatment and conducted GSEA enrichment analysis on the expression profile data (Fig 1E). The results showed that the ferroptosis-related gene pathway was activated in JVM3, suggesting that Selinexor could induce ferroptosis.
We observed the mitochondrial morphology changes in JVM3 cells treated with Selinexor at a concentration of 0.5 μmol/L for 24 hours using transmission electron microscopy. We found that the mitochondria were reduced in size, the mitochondrial cristae were reduced, the outer membrane was ruptured and shrunk, and the double-layer membrane density was increased (Fig 1F). This indicated that Selinexor can induce ferroptosis in p53 wild-type JVM3 cells. Selinexor can also cause p53 nuclear retention (Fig 1G). RNAseq analysis revealed that Selinexor activated the p53 pathway in JVM3 cells.
To further investigate whether Selinexor exerts its effects through the TP53 gene, we used CRISPR Cas9 technology to knock out TP53 in the JVM3 cell line. After treating p53 wild-type and knockout cells with Selinexor, we found that the total ROS and lipid ROS levels increased significantly in p53 wild-type JVM3 cell lines, while there was no significant change in the ROS levels in p53 knockout cells (Fig 1H). These results suggested that Selinexor-induced ferroptosis depends on the normal function of p53. GSEA enrichment analysis showed the ferroptosis-related pathway was also significantly enriched in p53 wild-type JVM3 cells (Fig 1I). We then combined the differentially expressed genes (DEGs) from the two groups with the ferroptosis database and identified 34 potential ferroptosis target genes (Fig 2 A-B).
After analyzing the expression and relationship with ferroptosis of the potential target genes, we speculated Heme Oxygenase 1 (HMOX1) as the bridge to connect p53 and ferroptosis. HMOX1 degrades heme and releases free iron to generate oxidized lipids in the mitochondria membrane to induce ferroptosis (Proc Natl Acad Sci U S A. 2019;116(7):2672-80). The expression of HMOX-1 is regulated by p53 (Nucleic Acids Res. 2007;35(20):6924-34). XPO1 inhibitor treatment increased HMOX1 expression in a dose-dependent manner in JVM3 cells (Fig 2C-D). It was found that the levels of p53 and HMOX1 increased significantly in wild-type cells. Silencing HMOX1 decreased lipid ROS levels in CLL cells after Selinexor treatment (Fig 2E). Our data thus indicated that targeting XPO1 by Selinexor increased HMOX1 expression in a p53-dependent pathway and upregulated the lipid ROS to trigger ferroptosis in CLL (Fig 2F).