ENO1 Activity Inhibition Induce Ferroptosis Susceptibility in BCR-ABL TKIs Resistant CML Cells through GPX4 Autophagy Degradation

Background: CML accounting for approximately 15% of adult leukemia, with imatinib and other tyrosine kinase inhibitors(TKIs) significantly improved the clinical outcomes of CML. However, nearly 30% of CML patients would be ultimately resistant to TKIs therapy and lead to relapse, which significantly shorten the outcome of the patients. ENO1 is a potential oncogene identified by our research group through chemical-biology methods in myeloid leukemia cells. The purpose of this study is aimed to explore the impact and mechanism of ENO1 on CML carcinogenesis and TKIs resistance, so as to provide potential molecular targets and evidence for the clinical treatment of CML TKIs resistance.

Methods: GEO/TCGA cohort data and our center's patient sample data were undermined; single-cell RNA sequence data was also employed to explore the efficacy of ENO1 in CML development. Western blotting and qPCR techniques were used to detect the expression level of ENO1 in myeloid leukemia cells; The CCK-8 , enolase activity assay kit was used to detect the ENO1 enzyme activity of AP-III-a4 (ENOblocker, AP) acting on K562 and its imatinib-resistant counterpart K562/G cells; Stable ENO1 overexpression cell clone, K562.OE, was constructed on the basis of K562 to further study the effect and mechanism of ENO1. Glycolysis/OXPHOS activity were determined and global metabolomics as well as Bulk-RNA sequence were also performed. Colony formation assay, flow cytometry and electron microscope were also applied. In vivo study with NSG mice was applied to confirmed the anti-leukemia effects of ENO1 inhibition in resistant CML models.

Results: ENO1 knockdown in K562 could significantly reduce the colony forming ability of CML cells and the in vivo tumorigenic ability; TCGA/GEO (GSE37642) database analysis found that ENO1 expression was significantly elevated in accelerated phase(AP) CML. Under low-dose imatinib stimulation(IC20), the expression of ENO1 gradually increases,accompanied by a decrease in BCR-ABL copies, which mimics the development of imatinib resistance. Single-cell RNA sequencing data showed that in patients with worse TKIs treatment response, the expression of ENO1 was higher in CD34+ subpopulation, and there also found a moderate co-expression relationship between ENO1 and GPX4. Imatinib resistant K562/G cells exhibited higher expression level of ENO1 and cross-resistance to other BCR-ABL inhibitors, such as dasatinib and nilotinib. When inhibiting ENO1 enzyme activity with AP), K562/G cells regained sensitivity to the above BCR-ABL inhibitors. This situation results were also validated in K562.OE cells, which collectively suggest higher expression of ENO1 may predict worse BCR-ABL inhibitor responses. Besides, K562/G cells exhibit faster proliferation ability and higher glycolytic activity, all of which could be reversed by AP. Global metabolomics and transcriptome analysis revealed that ferroptosis susceptibility and cysteine/methionine pathways were enriched when K562/G treated with AP. GSH level were higher in K562/G cells and K562.OE cells compared with K562 cells. Typical mitochondrial condensation and autophagosomes were observed under the co-treatment of AP and imatinib by electron microscope. Inhibiting ENO1 increases the expression of p62 and LC3-II/I proteins in K562/G cells, and reduces the expression of GPX4 and SLC7A11 gene. Western-blotting analysis found that autophagy inhibition by chloroquine partially restored the downregulation of GPX4 induced by AP, and no direct interaction between ENO1 and GPX4 was found by co-IP. In vivo results using NSG mice showed that the combination of AP and imatinib significantly reduced the tumorigenic burden , with minor liver or kidney toxicity. The above results suggest that ENO1 block by AP may be related to the autophagic degradation of GPX4, and ultimately impact the metabolism of cysteine/methionine, which finally induce ferroptosis subceptibility.

Conclusion: ENO1 would be an important biomarker for CML treatment response, and elevated expression of ENO1 may be associated with worse response to TKIs such as imatinib. Inhibition of ENO1 activity may promote autophagic degradation of GPX4, affect cysteine/methionine metabolism, and ultimately enhance susceptibility of CML cells to ferroptosis. Inhibition of ENO1 may provide as a candidate option to improve the anti-tumor effect of BCR-ABL inhibitors in CML cells.