Oncogenic KRAS^G12D in the Hematopoietic System Causes NLRP3 Inflammasome Activation Leading to Myeloproliferative Syndrome

Oncogenic Ras mutations occur frequently in myelodysplastic (MDS) and myeloproliferative syndromes (MPN). It was unclear if besides the direct transforming effect via constant RAS/MEK/ERK signaling, an inflammation related effect of KRAS contributes to the development of the disease.

To address this question we performed a microarray based analysis of bone marrow derived from Rosa26Cre^ERT2; LSL-Kras^G12D mice after induction of Kras^G12D with tamoxifen. We found that the NLRP3 inflammasome and related genes were upregulated. In agreement with increased NLRP3 sensitivity, Kras^G12D bone marrow derived dendritic cells (BMDCs) showed increased inflammasome activation upon stimulation with LPS and ATP as quantified by intracellular levels of IL-1β and caspase-1 p20 subunit.

To validate the functional role of the NLRP3 inflammasome for the MDS phenotype in vivo we generated Rosa26Cre^ERT2; LSL-Kras^G12D; NLRP3^-/- mice. We transferred BM (CD45.2) from WT mice, Rosa26Cre^ERT2; LSL-Kras^G12D or Rosa26Cre^ERT2; LSL-Kras^G12D; NLRP3^-/- mice into C57BL/6 WT recipients (CD45.1) after myeloablative conditioning and induced Kras^G12D by tamoxifen treatment.

Rosa26Cre^ERT2; LSL-Kras^G12D BM recipient mice developed MPN stigmata including anemia, leukopenia and thrombocytopenia. The mice also presented weight loss, splenomegaly, and an increase in myeloid cells in peripheral blood, spleen and BM, as well as increased CD11b⁺ckit⁺ immature myeloid progenitors. In contrast, Rosa26Cre^ERT2; LSL-Kras^G12D; NLRP3^-/- recipient mice developed no disease phenotype.

BMDCs isolated from Rosa26Cre^ERT2; LSL-Kras^G12D recipient mice exhibited strong NLRP3 inflammasome activation and high IL-1β levels which was not found in BMDCs of Rosa26Cre^ERT2; LSL-Kras^G12D; NLRP3^-/- mice.

CD11b⁺ cells of a Juvenile myelomonocytic leukemia (JMML) patient with an activating KRAS mutation showed increased inflammasome activation as compared to a non-mutant patient, which is in agreement with our observations in mouse models.

Our findings support the concept that oncogenic Kras^G12D does not only act via its oncogenic driver function, but also enhances activation of the NLRP3/IL-1β axis. This could lead to novel therapeutic approaches combining inhibition of oncogenic signaling and immune modulation via IL-1β blockade.