Abstract
Introduction: Acute myeloid leukemia (AML) accounts for ~25% of pediatric leukemia cases and is associated with poor prognosis and high relapse rates compared to other childhood cancers. Arising from malignant myeloid blasts, AML is highly heterogeneous and treatment resistant. Macrophage migration inhibitory factor (MIF), a cytokine signaling through CD74 and co-receptors (CD44, CXCR2, CXCR4, CXCR7), promotes proliferation and migration while inhibiting apoptosis and has been linked to tumor progression and poor outcomes in solid cancers (Barthelmess et al., Cancers, 2023). In AML, MIF promotes blast proliferation and survival (Spertini et al., Cell Death Discov, 2024), yet its relevance in pediatric conditions and potential as a therapeutic target remain underexplored. In this study, we integrated single cell RNA-seq reanalysis of publicly available data sets from bone marrow (Mumme et al.,Nat Commun, 2023) and blast cells (Naldini et al, Nat Commun, 2021) of pediatric AML patients with functional validation in AML cell lines.
Methods: Publicly available single cell RNA-seq datasets from bone marrow of 15 pediatric AML patients (GEO: GSE235923) and blast cells from 13 pediatric AML patients (GEO: GSE185993) were reanalyzed using Seurat (v5.2) for quality control, clustering, and differential gene expression. After quality control filtering, 57,295 bone marrow cells (3820 ± 1759 cells per patient) and 65,361 blast cells (5028 ± 2714 cells per patient) were used for downstream analysis. Ligand–receptor interactions were inferred using CellChat (v2.1) to identify enriched signaling pathways.
AML cell lines (KG-1a, MOLM-13, HEL) were treated with 0, 25, 50, 75, and 100 μM of 4-IPP, a selective MIF inhibitor, for 72 hours. Cell growth and viability (7-AAD) as well as CD74 and CD44 expression, were assessed by flow cytometry. MIF secretion was quantified by ELISA. Statistical analyses included one-way ANOVA with Tukey's post hoc test and Pearson correlation.
Results: CellChat analysis of pediatric AML bone marrow samples revealed the MIF-CD74/CD44 axis as a dominant signaling pathway among malignant blast cells. Monocytes, T cells, and B cells also expressed high levels of MIF, CD74, and CD44, suggesting cross-talk between blasts and other immune cells. Co-receptors CXCR2 and CXCR7 (ACKR3) were minimally expressed, whereas CXCR4 showed robust expression in monocytes, T cells, and B cells, implicating this receptor in additional MIF-mediated interactions. Across all 28 pediatric AML samples (bone marrow and blast cell datasets) MIF, CD74, and CD44 were consistently expressed at equal or higher levels than established therapeutic targets (FLT3, CD33, BCL2), supporting the broad relevance of this signaling axis in pediatric AML.
Inhibition of MIF with 4-IPP reduced cell growth (KG-1a: F = 63.0, p < 0.0001; MOLM-13: F = 100.5, p < 0.0001; HEL: F = 3.4, p = 0.057) and viability (KG-1a: F = 399.8, p < 0.0001; MOLM-13: F = 390.6, p < 0.0001; HEL: F = 47.6, p<0.0001) across all AML cell lines, with the greatest reductions observed at 50-100 μM. Sensitivity to 4-IPP varied between cell lines: after 72 hours at 100 μM, KG-1a and MOLM-13 viability declined from ~90% to <20%, whereas HEL only fell to ~70%. CD74 and MIF expressions were significantly lower in KG-1a compared to MOLM-13 and HEL (p < 0.01 and p < 0.001, respectively), while CD44 expression was significantly lower in HEL compared with both MOLM-13 and KG-1a (p < 0.05). Viability reduction following 4-IPP treatment correlated strongly with CD44 expression (R = 0.81, p < 0.01) but not with MIF (R = -0.58, p = 0.062) or CD74 (R = -0.53, p = 0.095), implicating CD44 as a key determinant of 4-IPP sensitivity.Conclusions: Integrated reanalysis of pediatric AML single cell datasets and functional validation in cell lines reveal the MIF signaling axis as a prominent signaling pathway promoting blast survival and proliferation. These data further suggest that MIF signaling may influence the tumor microenvironment by engaging local immune cells. CD44 expression strongly correlated with sensitivity to MIF inhibitor 4-IPP, identifying it as a potential biomarker. Heterogeneous responses among AML cell lines highlight subtype-specific signaling dependencies and support further exploration of MIF–CD74/CD44 interactions as a therapeutic target, including potential combination strategies with existing AML treatments.
