Il6- and Tnf-mediated inflammation drive HSC dysfunction and hematological malignancy in miR-146a^−/−mice. (A) EPCR^hi/CD150^hi ESLAM subpopulation in WT, miR-146a^−/−, Il6^−/−miR-146a^−/−, Tnf^−/−miR-146a^−/−, and Nfkb1 miR-146a^−/− ESLAMs (sample of 200 cells shown per genotype). (B) Percentage of EPCR^hi/CD150^hi ESLAMs calculated from panel A (n = 660, 716, 394, 360, and 285 cells, respectively). P by binomial logistic regression. (C) Percentage of single ESLAMs divided at 24 hours in culture, expressed as mean (±SEM) fold change relative to WT (n = 6, 5, 3, 3, 4 replicates). Data for WT and miR-146a^−/− are adapted from Figure 3C. P by linear regression analysis. (D) Survival curve of WT, miR-146a^−/−, Il6^−/−miR-146a^−/−, and Tnf^−/−miR-146a^−/− mice to 1.5 years. Sample sizes (n) and log-rank test Ps shown. (E) Strategy for investigating the sensitivity of WT and miR-146a^−/− ESLAMs to IL6 stimulation. (F) Cell division timing and index data of WT and Il6^−/−miR-146a^−/− (IDKO) ESLAMs treated with 50 ng/mL IL6 or control (n = 69, 69, 68, and 70 cells). P by ordinal logistic regression, for interaction between genotype and cytokine treatment. (G) Model showing how aging promotes loss of miR-146a expression in HSCs. Loss of miR-146a promotes activation of NF-κB, an effector of the miR-146a target TRAF6,⁴¹  driving IL6 signaling, which in turn drives TNF signaling. IL6 and TNF promote expansion of an HSC subpopulation with inflammaging phenotypes. Reducing inflammation mediated by IL6 or TNF restores HSC subpopulation composition and reduces the malignant potential of miR-146a^−/− HSCs, demonstrating that loss of miR-146a is a genetic mechanism linking HSC inflammaging and myeloid malignancy. SEM, standard error of the mean.