Cytoplasmic retention of IRF3 binding Vimentin competitively with ERK1/2 mitigates acute myeloid leukemia through TFEB nuclear translocation

Acute myeloid leukemia (AML) is characterized by the accumulation of immature myeloid blasts. In this study, we investigated the role of interferon regulatory factor 3 (IRF3), a key transcription factor of type I interferon, in AML differentiation. Bone marrow mononuclear cell samples from cytogenetically normal AML patients were analyzed, revealing significantly elevated IRF3 protein levels. Knocking down IRF3 in THP-1 cells promoted PMA-induced cell differentiation while inhibiting proliferation, and subcutaneous implantation of these IRF3-knockdown cells into nude mice resulted in a significant reduction in both tumor volume and weight. Non-canonically, the differentiation inducer PMA inhibited the phosphorylation and nuclear translocation of IRF3 in THP-1 cells, and deletion of the nuclear localization sequence (NLS) of IRF3 significantly promoted cell differentiation marker expression. Furthermore, IRF3S386A, the phosphorylation site mutation, promoted PMA-induced AML cell differentiation, and relieved the AML leukemia burden in a disseminated intravenous xenograft leukemia model. Mechanistically, upon PMA stimulation, IRF3 was retained in the cytoplasm, where it competitively bound to Vimentin. This binding disrupted the interaction of Vimentin with p-ERK, a kinase that could phosphorylate and inhibit transcription factor EB (TFEB). Consequently, TFEB, whose phosphorylation was reduced due to the disrupted p-ERK activity, translocated to the nucleus, where it promoted the differentiation of AML cells. Translationally, the TFEB activator TA1 and the ERK inhibitor PD98059 both promoted cell differentiation, and mitigated the AML burden induced by IRF3 upregulation. In summary, IRF3 is highly expressed in AML, and its non-transcriptional cytoplasmic function is essential for regulating AML cell differentiation. Mechanistically, IRF3 binds to Vimentin in cytoplasm, which disrupts the Vimentin-p-ERK signaling axis and consequently promotes the nuclear translocation of TFEB. Our findings provide a mechanistic framework for exploring differentiation strategies in non‑M3 AML, and highlight its potential translational value for future experimental validation.