AUTHORS
Daniel Ward Phillips, N/A, Caroline C Ennis, BS2, Alana M Franceski, BS, MSc3, Paran Goel, PhD4, Francesca Luca, PhD5, Ashley N Connelly, PhD6, Pritam Apitkatla7, Arthur W Meehan8, Christina R Larson9, Rob S Welner, PhD10
AFFILIATIONS
1 University of Alabama at Birmingham, Birmingham, Alabama, United States
2 University of Alabama at Birmingham, Birmingham, Alabama, United States
3 University of Alabama at Birmingham, Birmingham, Alabama, United States
4 University of Alabama at Birmingham, Birmingham, Alabama, United States
5 University of Alabama at Birmingham, Birmingham, Alabama, United States
6 University of Alabama at Birmingham, Birmingham, Alabama, United States
7 University of Alabama at Birmingham, Birmingham, Alabama, United States
8 University of Alabama at Birmingham, Birmingham, Alabama, United States
9 University of Alabama at Birmingham, Birmingham, Alabama, United States
10 University of Alabama at Birmingham, Birmingham, Alabama, United States
ABSTRACT
Introduction: During acute myeloid leukemia (AML), expanding leukemic blasts extensively remodel the bone marrow niche and disrupt normal hematopoiesis. Most work attributes the resulting immune dysfunction to loss of healthy hematopoietic output. Whether the leukemic niche also functionally reprograms the genetically normal ("bystander") cells that persist within it remains unclear. We interrogated bystander monocytes, innate effectors bridging homeostasis, immune surveillance, and antigen presentation, whose abundance, transcriptional state, and functional capacity in the leukemic niche remain uncharacterized.
Methods: We used a non-conditioned chimeric murine model in which HoxA9/Meis1-driven AML establishes a leukemic environment alongside a genetically distinguishable normal compartment, allowing us to track both cell states across control, low-, and high-burden disease.
Results: By flow cytometry, bone marrow monocytes declined 6.3-fold with increasing disease burden (9.95×10⁶ to 1.59×10⁶ cells; p<0.0001; n=8, 11, 14 for control, low, high), while bystander monocytes held a stable proportion of the monocyte compartment (9.5% vs 8.8%, p=0.45), indicating that the niche does not selectively deplete them. In peripheral blood, the bystander monocyte pool shifted toward the non-classical subset (0.9% to 2.1% of CD45.1 cells, p=0.0098;⁺ n=4, 6), away from the inflammatory-responsive classical phenotype.
Furthermore, single-cell transcriptomics of bystander marrow (15,064 cells, n=3/group) showed that classical monocytes maintained lineage identity (Ly6c2, Ccr2, or Sell all p>0.7) yet markedly lost interferon responsiveness. Unbiased pathway hallmark analysis ranked interferon gamma and interferon alpha responses as the two most suppressed programs of 3,813 gene sets tested (NES=−3.07, −3.03; FDR<0.001), and classical monocytes significantly downregulated interferon-stimulated genes including Ifit2 (log2FC=−1.74, FDR=3.8e-3), Oas1a (−1.21, FDR=1.2e-2), Ifitm3 (−0.58, FDR=2.1e-4), and Ifitm2 (−0.53, FDR=5.4e-4).
Suppression extended across several marrow populations but peaked in classical monocytes, whereas T cells within the same marrow increased interferon signaling (NES=+1.95, FDR=9.3e-5), indicating cell-type selective reprogramming rather than uniform alteration of the leukemic marrow. Notably, Stat1 target gene activity was reduced with no overlap between control and leukemic animals, while Stat1 transcript itself was unchanged, indicating suppressed signaling output rather than loss of the transcription factor. Since the niche appears to silence Stat1-mediated target genes while sparing lineage identity, we hypothesize that extrinsic suppression, likely TGFβ or IL-10, drives this state.
Classical monocytes similarly reduced MHC class I transcripts H2-K1 (log2FC=−0.43, FDR=1.2e-2), H2-D1 (−0.67, FDR=9.8e-9), and B2m (−0.44, FDR=1.1e-4), along with the immunoproteasome subunit Psmb8 (−0.43, FDR=0.035) and the interferon-gamma receptor Ifngr1 (−0.72, FDR=6.3e-5), and unbiased analysis independently ranked antigen processing and presentation among the most affected pathways. These cells also decreased the maturation regulators Mafb (−1.38, p=0.025), Cebpb (−0.87, FDR=9.8e-9), and Cebpd (−0.75, FDR=0.029) along with mature macrophage markers. Proteomic analysis of bystander marrow monocytes displayed a 51% reduction in CSF1R (M-CSFR) surface protein (gMFI 5,484 vs 2,702; p=0.005), extending arrested maturation to the protein level. Bystander monocytes recapitulated no published signature of myeloid-derived suppressor cells, endotoxin-tolerant, tumor-infiltrating, or NRF2- activated monocytes, defining a unique AML-induced inactivation state.
Conclusions: The AML niche therefore reprograms genetically normal monocytes into an inactivated state that preserves lineage identity while coordinately suppressing interferon responsiveness, maturation, and MHC class I antigen presentation. Immune compromise in AML thus reflects not only the loss of healthy immune cells but the functional silencing of those that persist, identifying the bystander compartment as a potentially reversible target for restoring innate immunity and enhancing therapy response in AML.
DISCLOSURES
Daniel Ward Phillips, N/A: No relevant conflicts of interest to declare.