The antibiotic linezolid targets the Jak2V617F mutation in polycythemia vera.

Abstract Current treatment options for polycythemia vera (PV) patients suppress excessive hematopoiesis, but lack specificity against JAK2V617F+ cells. Our group initially aimed to study crosstalk between the microbiome and inflammation in PV, and unexpectedly identified Linezolid (LZD)–an FDA-approved antibiotic used to treat patients with gram-positive bacterial infections– as a JAK2V617F selective inhibitor. In JAK2V617F mice, which recapitulate the PV phenotype (Li J., Blood, 2014), LZD treatment effectively normalized spleen sizes and disease associated hematopoietic parameters. Specifically, after 6 weeks of LZD treatment by way of drinking water (1g/L), hematocrit levels in JAK2V617F mice were reduced from 77% to 51% (P=0.002), and hemoglobin, reticulocyte counts, and white blood cell counts were normalized to levels seen in healthy wild-type (WT) mice. Treatment of WT mice with LZD at 1 g/L for 6 weeks did not lead to any changes in hematological parameters, supporting the conclusion that LZD exerts a specific inhibitory effect on JAK2V617F mice. Molecular docking studies revealed that LZD specifically interacted with the pseudokinase domain of JAK2V617F protein but not with WT JAK2 protein–a key feature that distinguishes LZD from existing JAK2 inhibitors. In vitro treatment of homozygous JAK2V617F+HEL cells with LZD suppressed cell proliferation, downregulated STAT5 signaling, altered cell cycle status, and increased apoptosis. By contrast, LZD did not inhibit cell proliferation of RS4:11 cells, which express WT JAK2, further confirming the specificity of LZD against JAK2V617F. As a positive control, ruxolitinib–a JAK2 inhibitor that is currently widely used–significantly reduced cell proliferation of both HEL and RS4:11 cells, highlighting the unique selectivity profile of LZD. PV patients contain healthy normal hematopoietic cells mixed with JAK2V617F+ cells. To replicate this situation, we utilized a competitive bone marrow (BM) transplantation approach where BM cells from WT (CD45.1) and JAK2V617F (CD45.2) mice were mixed and transplanted into lethally irradiated WT recipients. LZD treatment consistently rescued the PV-like phenotypes induced by JAK2V617F+ cells with normalization of hematopoiesis and spleen size. Splenic fibrosis was significantly reduced in the LZD group compared to the untreated group (P=0.001). LZD significantly reduced the amount of JAK2V617F mutant RBCs present in the peripheral blood circulation and consequently led to more WT RBCs in circulation (P=0.01), indicating a decrease in variant allele frequency(VAF). Most importantly, our data also demonstrated that LZD treatment in vivo lowered the JAK2V617F+ hematopoietic stem cell (HSC) numbers and increased apoptosis of JAK2V617F+ Megakaryocyte-Erythroid Progenitor (MEP) cells. LZD treatment of PV patient (N=6) mononuclear cells inhibited hematopoietic colony formation in a dose-dependent manner(10ug/ml, P=0.001; 50ug/ml, P<0.0001), with more primitive progenitor cells exhibiting greater sensitivity to this inhibition at the lowest dose tested (5ug/ml, P=0.002). This observation aligned with our animal data, which demonstrated that more primitive stem/progenitor cells were more susceptible to the effects of LZD. Additionally, LZD significantly reduced JAK2V617F colony counts (P=0.03) with no change detected in WT colony counts(P=0.4), again confirming the specificity of LZD against JAK2V617F mutant cells. Reduction of the JAK2V617F–driven disease phenotypes across multiple model systems provided a firm foundation for the future evaluation of compounds structurally similar to LZD as a promising new therapeutic agent selectively targeting JAK2V617F+ cells in PV patients.