Resistance to single-target therapies has spurred interest in multitarget strategies for acute myeloid leukemia (AML). Heat shock protein 90 (HSP90), a chaperone that stabilizes numerous oncogenic client proteins, represents an attractive therapeutic target for AML; however, the clinical development of early HSP90 inhibitors was limited by dose-limiting toxicities and an excessive heat-shock response (HSR). Through structural optimization of the marine aaptamine scaffold and target identification, ap-a48 was identified as a novel HSP90-targeting anti-AML lead that exhibits potent anti-AML activity and acceptable preliminary tolerability while inducing only a modest HSR. In rats, ap-a48 showed favorable pharmacokinetics with 65.3% oral bioavailability, and in HL-60 xenograft mouse models, it suppressed tumor growth (71.2% inhibition at intraperitoneal 40 mg/kg; 67.3% at oral 60 mg/kg) without significant hepatotoxicity or major organ abnormalities. These findings identify ap-a48 as a promising marine-natural-product-derived HSP90-targeting lead for AML therapy.

Interferon-induced transmembrane protein 3 (IFITM3) is overexpressed in acute myeloid leukemia (AML) and is associated with poor prognosis. Previous studies implicate IFITM3 in oncogenic processes and immune modulation; however, the mechanisms by which IFITM3 contributes to leukemogenesis and shapes the AML immune microenvironment remain unclear. Here, we integrated single-cell RNA sequencing (scRNA-seq) and bulk gene expression analyses with macrophage polarization assays using IFITM3high AML cell lines and their corresponding IFITM3 knockdown models, together with gain-of-function experiments in IFITM3low HL-60 cells, alongside functional assays and pharmacologic perturbations, to define the immunological function of IFITM3 in AML. Our results revealed for the first time that IFITM3high AML cell lines polarize M0 macrophages toward a M2-like phenotype. IFITM3 knockdown significantly reduced the M2 marker CD163 and increased the M1 marker CD80. Consistently, IFITM3-deficient AML cells decreased M2-associated gene expression and diminished secretion of TGF-β and IL-10, together with lower viability and G1/G0 cell-cycle arrest. Conversely, enforced IFITM3 expression in HL-60 cells biased macrophages toward a predominantly M2-like state, whereas vector controls promoted a mixed M1/M2 pattern. Mechanistically, loss of IFITM3 reduced total PI3K and AKT protein levels and markedly decreased PI3K/AKT phosphorylation. Pharmacologic PI3K inhibition with LY294002 suppressed PI3K/AKT phosphorylation and specifically attenuated IFITM3-dependent M2 polarization and associated cytokine programs, while also reducing leukemic cell proliferation in co-culture. Clinically, we extended the adverse prognostic effect of IFITM3 to AML patients with intermediate molecular risk. These findings provide comprehensive evidence that IFITM3 contributes to a tumor-supportive AML microenvironment by driving M2 macrophage polarization, highlighting IFITM3 as a potential therapeutic target in IFITM3high AML.

Acute myeloid leukemia (AML) remains a therapeutic challenge due to drug resistance and relapse, which are often driven by the protective bone marrow (BM) niche. Conventional xenograft models fail to adequately recapitulate this niche-specific pathophysiology. To overcome this limitation, a novel magnetically targeted intramedullary (MagIC-TI) xenograft model was developed. Magnetically labeled doxorubicin (DOX)-resistant HL60 cells (Mag-Re) were injected into the femurs of NSG (nonobese diabetic [NOD] Cg-Prkdc[scid]IL2rg[tm1Wjl]/SzJ) mice using a patented microinjection syringe under localized magnetic guidance. With the MagIC-TI model, rapid (day 1) and specific (100% by day 7) leukemic engraftment was achieved within the femoral BM, whereas intravenous (IV) injection led to delayed (mean 23.67 ± 10.26 days) and disseminated engraftment. Bioluminescence imaging, histopathological analysis, flow cytometry, and molecular assays confirmed that disease was localized in the MagIC-TI model. In contrast, extramedullary infiltration, predominantly in the lungs, spleen, liver, and kidneys, was observed early in progression in the IV model. The MagIC-TI model discriminated drug responses, showing effective tumor burden reduction with homoharringtonine (HHT) and unequivocal DOX resistance, a distinction that was obscured in heterogeneous IV models. Furthermore, employing a semisolid decalcification (SSD) system preserved green fluorescent protein (GFP) fluorescence, enabling high-resolution visualization of engrafted cells within bone tissue. The MagIC-TI model enables BM-targeted, rapid, and efficient leukemic engraftment and allows discrimination of drug sensitivity and resistance. This model provides a robust and reproducible platform for modeling the leukemia BM niche and for preclinical evaluation of niche-directed therapies.

Acute myeloid leukemia (AML) is a highly aggressive malignancy with frequent therapeutic resistance, necessitating the identification of novel molecular targets. This study aims to elucidate the role of the deubiquitinase OTUD5 in AML progression by regulating SLC7A11 to suppress ferroptosis. We analyzed OTUD5 expression in AML patient samples and cell lines using RNA sequencing and quantitative PCR. Functional roles were assessed through OTUD5 silencing and overexpression in AML cell lines (THP-1, HL-60), followed by proliferation, colony formation, and ferroptosis assays (ROS, labile Fe2 +, GSH, MDA). Co-immunoprecipitation and ubiquitination assays confirmed OTUD5-SLC7A11 interactions, while in vivo xenograft models validated findings. Molecular docking and transmission electron microscopy further elucidated mechanisms. OTUD5 was significantly upregulated in AML, correlating with ferroptosis suppression. OTUD5 directly interacted with and deubiquitinated SLC7A11, enhancing its stability and promoting AML cell survival. OTUD5 silencing induced ferroptosis, marked by increased labile iron, ROS, and mitochondrial damage, which was reversed by SLC7A11 overexpression or GSH supplementation. In vivo, OTUD5 knockdown reduced tumor growth, an effect mitigated by SLC7A11 overexpression or GSH. The OTUD5-SLC7A11 axis drives AML progression by suppressing ferroptosis, offering a novel therapeutic target to exploit ferroptosis sensitivity and overcome treatment resistance in AML.

Acute myelocytic leukemia (AML) is a common leukemia subtype in adults, necessitating novel, potent treatments. Microtubule-targeting agents (MTAs) are key anti-cancer drugs widely used in clinical practice. A series of novel furan-trimethoxyphenyl hybrids as tubulin polymerization inhibitors were designed using a molecular combination strategy and evaluated for their structure-activity relationships (SARs) in AML. SAR studies demonstrated that the trimethoxyphenyl unit and substituents on benzyl units play significant roles for the antiproliferative activity against human AML cells. Compound III-3 exerted the highest antiproliferative effects on human AML MV-4-11 (IC50 = 74.25 nM), HL-60 (IC50 = 87.99 nM), and THP-1 cells (IC50 = 92.20 nM). Compound III-3 could directly bind to β-tubulin at the colchicine-binding site, thereby disrupting the microtubule network structure. Compound III-3 activated the spindle assembly checkpoint (SAC), promoting M-phase arrest and mitotic catastrophe and activating the mitochondrial apoptotic pathway in AML cells. In the xenograft mouse model constructed with MV-4-11 cells, the tumor-inhibitory rate of compound III-3 (20 mg/kg body weight, 71.40%) was higher than that of the positive control Ara-C (50 mg/kg body weight, 58.41%). Thus, furan-trimethoxyphenyl hybrids are a promising class of tubulin inhibitors with potential therapeutic applications in AML.

Spatiotemporal gene expression is regulated by the SNF2 family of ATPases that remodel chromatin, among which helicase lymphoid-specific (HELLS) play an essential role in DNA-templated processes. Analysis from cancer databases revealed HELLS upregulation in several cancers, including acute myeloid leukemia (AML). Using an in vitro myeloid differentiation model, we found that HELLS deficiency promotes myeloid differentiation, apoptosis, cell-cycle arrest, and chromatin instability. Furthermore, HELLS deficiency enhanced myeloid differentiation and apoptosis in HL-60 cells when treated with chemotherapy drugs, including 5-azacytidine, cytarabine, and doxorubicin. Epigenetically, loss of HELLS reduced active histone mark (H3K4me3) and enhanced repressive promoter (H3K27me3), active enhancer (H3K27ac) marks, and chromatin accessibility. Transcriptomic analysis of BeatAML data and bioinformatics analysis revealed that HELLS upregulation is associated with adverse prognosis, as defined by the ELN-2017 classification, and significantly poorer survival. Classification of AML patients into HELLSHigh and HELLSLow gene-expression groups showed that lower HELLS expression was associated with a favorable prognosis and improved remission. HELLS upregulation was positively and negatively associated with leukemic stem/progenitor cell (LSPC) markers and myeloid lineage markers, respectively. Ex vivo validation in AML patient-derived LSPCs demonstrated that HELLS deficiency reduces leukemic marker burden (CD123 and TIM-3) and increases apoptosis. Overall, we show HELLS upregulation drives leukemogenesis with poor outcomes, highlighting its potential as a prognostic marker and therapeutic target in AML.

Glycosylation is essential for hematopoietic cell homeostasis and malignant transformation. Dysregulated expression of glycosylation genes in leukemia cells accelerates disease progression and fosters drug resistance. Therefore, targeting these genes offers a promising avenue for anti-leukemic therapy. In this study, we explore the roles of N-glycans in acute promyelocytic leukemia (APL) differentiation using the ATRA-induced wild-type NB4 (WT/ATRA) or HL-60 cell model. We found that expression of N-acetylglucosaminyltransferase IVa (GnT-IVa, encoded by the MGAT4A gene) and its product (β1,4-GlcNAc-branched N-glycan) increased significantly during differentiation, as evaluated by lectin blot, real-time PCR, and flow cytometry. Interestingly, analysis of the Gene Expression Omnibus (GEO) public data showed that MGAT4A expression is significantly lower in APL patients, and higher MGAT4A expression was associated with favorable survival in AML cohorts. To address the role of GnT-IVa in differentiation, we established MGAT4A- and MGAT4B-knockout (KO) NB4 cell lines using CRISPR/Cas9. Compared to WT/ATRA cells, MGAT4A KO, but not MGAT4B KO, markedly suppressed ATRA-induced differentiation, as evidenced by reduced expression of CD11b and CD11c. We found that CD11b is a major glycoprotein carrying β1,4-GlcNAc-branched N-glycans. This modification enhanced CD11b stability, as CD11b expression declined more rapidly in MGAT4A KO cells in the presence of cycloheximide. In addition, MGAT4A KO suppressed ERK/MAPK signaling, which contributed to differentiation. Our study highlights the critical role of GnT-IVa in regulating APL differentiation, which may provide a basis for developing new differentiation therapies for APL.

Acute myeloid leukemia (AML) is a complex blood cancer that primarily affects relapsing or refractory patients receiving conventional chemotherapy. Nonsteroidal anti-inflammatory drugs (NSAIDs) have anticancer properties with restricted clinical efficacy attributable to cyclooxygenase (COX)-induced toxicities. To address this issue, a group of benzylamide analogs of the classical NSAIDs (NSI-1-NSI-9) were developed and synthesized to mask the carboxylic acid moiety and minimize COX-induced adverse effects while maintaining anticancer activity. The cytotoxic effect of such substances has been demonstrated in some leukemia cell lines (HL-60, MV4-11, KG1a, and K562). NSI-5 exerted the highest anti-leukemic activity among these sulindac analogs, as determined at a sub-micromolar level in all cell lines studied, by IC50. This mechanistic data also demonstrated that NSI-5 induced apoptosis that was dose-dependent, especially in HL-60 cell lines, and increased the sub-G1 cell fraction. This apoptotic process was also accompanied by a significant decrease in mitochondrial membrane potential, which is characteristic of the induction of the intrinsic apoptotic process. Interestingly, NSI-5 decreased the intracellular reactive oxygen species (ROS) and the expression of most antioxidants (catalase and glutathione synthetase), as well as the redox balance. Gene characterization in vitro also suggested activation of apoptotic pathways, where expression of Bax, Bak1, and Caspase-3 increased, suggesting a potential p53-independent apoptotic pathway, in contrast to control for Bcl-2 expression. Collectively, these findings indicate that NSI-5 is a promising in vitro anti-leukemic lead compound, with activity associated with mitochondrial dysfunction and altered redox regulation. The observed effects are consistent with previously reported COX-independent activity of structurally related NSAID derivatives, and support further investigation of NSI-5 in preclinical models.

The present study was undertaken to provide complementary information on the cell cycle of leukaemic lymphoblasts and myeloblasts based on the computer-assisted maximal nuclear diameter measurements. The measurements were carried out in "one-size cell layer" loci of bone marrow smears of patients suffering from B-cell acute lymphoblastic leukaemia (B-ALL), acute myeloblastic leukaemia with minimal differentiation (M0 AML), acute myeloblastic leukaemia without maturation (M1 AML), acute myeloblastic leukaemia with maturation (M2 AML) and chronic myeloid leukaemia (CML). The maximal nuclear diameter was also measured in peripheral blood mature lymphocytes of B-cell chronic lymphocytic leukaemia (CLL), which are known to be in the G0/G1 phase of the cell cycle. In contrast, the maximal nuclear diameter in lymphoblasts of patients with B-ALL was larger and reflected the S and G2 cell cycle phase. The largest incidence of myeloblasts with the smallest maximal nuclear diameter was in M0 AML. The smaller incidence of such cells was also noted in patients with M1 AML. Myeloblasts with larger nuclear bodies were present in M2 AML and CML. Thus, post-mitotic myeloblasts in G0 and G1 phase were characteristic of M0 and to a smaller extent of M1 AML. Myeloblasts with larger maximal nuclear diameter reflecting the S and G2 phase were dominant in M2 AML and CML. In summary, the nuclear size heterogeneity of leukaemic lymphoblasts and myeloblasts in bone marrow smears depends on various phases of the cell cycle classified according to the maximal nuclear diameter.