Back

Myeloid IRF5 is required for TLR7-driven inflammatory hemophagocyte differentiation and Macrophage Activation Syndrome

Thulin, N. K.; Lu, A.; Orozco, S. L.; Huang, A. Y. Y.; Nguyen, L. P.; Mishra, G.; Savan, R.; Clapp, W.; Ray, J.; Hamerman, J.; Barnes, B. J.

2026-08-23 immunology
10.64898/2026.08.18.745337 bioRxiv
Show abstract

In TLR7-driven macrophage activation syndrome (MAS), inflammatory hemophagocytes (iHPCs) differentiate from Ly6CHI monocytes, phagocytose red blood cells and promote disease, including anemia and thrombocytopenia. We demonstrate here that IRF5 is required for iHPC differentiation and MAS in TLR7-overexpressing (TLR7.1) mice. Both constitutive and myeloid-specific Irf5 deletion reduced iHPCs and improved anemia, thrombocytopenia and survival. Furthermore, therapeutic inhibition of IRF5 ameliorated MAS features and reduced splenic and circulating iHPCs. While cell-intrinsic IRF5 expression was required for iHPC differentiation, it was not required for TLR7.1 Ly6CHI monocyte differentiation and monocyte transcriptional programs. We further show that the transcriptome and chromatin landscape changed dramatically as iHPCs differentiated from TLR7.1 Ly6CHI monocytes. Many transcriptional programs gained in iHPCs were enriched in genes associated with IRF5-binding accessible chromatin regions, including those associated with NF-kB signaling, cytokine and chemokine production, and complement activation. Our data suggest that IRF5 collaborates with other transcription factor families, including NF-kB, ETS and AP1 members, to regulate iHPC gene programs. Together, our findings demonstrate that expression of IRF5 in myeloid cells is critical for MAS, for iHPC differentiation, and acts broadly across iHPC-specific gene programs in TLR7-driven inflammation.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.