Mitochondrial RNA helicase DDX28 promotes cell cycle and DNA repair programs and shapes an immunosuppressive microenvironment in acute myeloid leukemia

• Identifies DDX28 as a n prognostic biomarker linked to immune evasion in AML. • Reveals DDX28 remodels an immunosuppressive TIME via M2 macrophage and Treg enrichment. • Links DDX28-driven metabolic reprogramming to DNA repair and chemoresistance. • Maps DDX28 to malignant blasts and exhausted T cells using single-cell transcriptomics. • Demonstrates targeting DDX28 suppresses proliferation, suggesting a new therapeutic strategy. Acute myeloid leukemia (AML) is a heterogeneous malignancy with frequent relapse, driven by intertwined alterations in mitochondrial function, cell cycle control, genome maintenance, and immune evasion. DDX28 is a mitochondrial DEAD-box RNA helicase required for mitoribosome assembly and mitochondrial translation, and has been implicated in bioenergetic regulation in other tumor contexts. Here, we profiled DDX28 expression across AML cohorts using integrated multi-omics resources and evaluated its associations with prognosis, immune microenvironment features, and predicted drug response. Functional annotation, pathway analysis, GSEA, and targeted in vitro assays were used to explore potential mechanisms. High DDX28 expression was associated with inferior survival and higher blast burden. Mechanistically, elevated DDX28 expression was linked to promoter hypomethylation and was positively correlated with the transcription factor THAP11. Transcriptomic signatures in the high DDX28 group were enriched for cell cycle progression and DNA damage repair programs, together with an immune-suppressive landscape characterized by increased regulatory T cells and M2 macrophage signatures. Single-cell RNA-seq analyses further showed DDX28 enrichment in malignant blasts and exhausted or proliferative T-cell states. Consistently, DDX28 knockdown by siRNA in HEL cells reduced proliferation and impaired migration and invasion. Although direct metabolic flux measurements were not performed, the mitochondrial localization of DDX28 and the enrichment of proliferation and repair programs support a model in which DDX28 couples mitochondrial translation with the biosynthetic and genome maintenance demands of rapidly cycling AML cells. Collectively, our findings identify DDX28 as a prognostic indicator and a candidate regulator of malignant and immune states in AML, with potential relevance to therapeutic response.