Erythroid lineage Jak2V617F expression promotes atherosclerosis through erythrophagocytosis and macrophage ferroptosis
Wenli Liu, … , Alan R. Tall, Nan Wang
Wenli Liu, … , Alan R. Tall, Nan Wang
Published May 19, 2022
Citation Information: J Clin Invest. 2022;132(13):e155724. https://doi.org/10.1172/JCI155724.
View: Text | PDF
Research Article Vascular biology

Erythroid lineage Jak2V617F expression promotes atherosclerosis through erythrophagocytosis and macrophage ferroptosis

Abstract

Elevated hematocrit is associated with cardiovascular risk; however, the causality and mechanisms are unclear. The JAK2V617F (Jak2VF) mutation increases cardiovascular risk in myeloproliferative disorders and in clonal hematopoiesis. Jak2VF mice with elevated WBCs, platelets, and RBCs display accelerated atherosclerosis and macrophage erythrophagocytosis. To investigate whether selective erythroid Jak2VF expression promotes atherosclerosis, we developed hyperlipidemic erythropoietin receptor Cre mice that express Jak2VF in the erythroid lineage (VFEpoR mice). VFEpoR mice without elevated blood cell counts showed increased atherosclerotic plaque necrosis, erythrophagocytosis, and ferroptosis. Selective induction of erythrocytosis with low-dose erythropoietin further exacerbated atherosclerosis with prominent ferroptosis, lipid peroxidation, and endothelial damage. VFEpoR RBCs had reduced antioxidant defenses and increased lipid hydroperoxides. Phagocytosis of human or murine WT or JAK2VF RBCs by WT macrophages induced ferroptosis, which was prevented by the ferroptosis inhibitor liproxstatin-1. Liproxstatin-1 reversed increased atherosclerosis, lipid peroxidation, ferroptosis, and endothelial damage in VFEpoR mice and in Jak2VF chimeric mice simulating clonal hematopoiesis, but had no impact in controls. Erythroid lineage Jak2VF expression led to qualitative and quantitative defects in RBCs that exacerbated atherosclerosis. Phagocytosis of RBCs by plaque macrophages promoted ferroptosis, suggesting a therapeutic target for reducing RBC-mediated cardiovascular risk.

Authors

Wenli Liu, Nataliya Östberg, Mustafa Yalcinkaya, Huijuan Dou, Kaori Endo-Umeda, Yang Tang, Xintong Hou, Tong Xiao, Trevor P. Fidler, Sandra Abramowicz, Yong-Guang Yang, Oliver Soehnlein, Alan R. Tall, Nan Wang

×

Figure 5

Macrophage lipid peroxidation and ferroptosis induced by erythrophagocytosis are reversed by Liprox-1.

Options: View larger image (or click on image) Download as PowerPoint
Macrophage lipid peroxidation and ferroptosis induced by erythrophagocyt...
WT BM-derived macrophages were treated with vehicle (M0), LPS plus IFN-γ (M1), or IL-4 (M2) for 24 hours and then incubated with an equal number of control or VFEpoR RBCs for another 6 hours. (A) Representative C11 BODIPY histogram and statistics of C11 BODIPY+ macrophage percentage by flow cytometry. (B) Immunoblot of glutathione peroxidase 4 (GPX4), malondialdehyde modified proteins (MDA adduct), ferritin, arginase I, and inducible NOS (iNOS) and β-actin of M0, M1, and M2 cell lysates. The bar graph shows quantification of immunoblots normalized to β-actin. C denotes control and VF denotes VFEpoR RBCs. (C) M0, M1, and M2 macrophage viability were quantified as the percentage of propidium iodide–negative macrophage (live cell) versus total macrophages by flow cytometry (n = 3 replicates). (D) Macrophage LDH release in culture medium was measured after incubation with control or VFEpoR RBCs for 6 hours. (E) C11 BODIPY+ macrophage percentage and cell death ratio were tested by flow cytometry after 6 hours of erythrophagocytosis assay with or without Liprox-1 (200 nM) treatment. (F) M0 or M2 BM-derived macrophages from WT or GPX4 transgenic mice were treated with control and VFEpoR RBCs for 6 hours. Lipid peroxidation and cell death were assessed by C11-BODIPY and propidium iodide staining and analyzed by flow cytometry. (G) Human peripheral monocyte-derived macrophages generated from healthy donors were treated overnight with RBCs from JAK2VF-positive patients with MPN or matched healthy controls in the presence or absence of Liprox-1 (200 nM). Lipid peroxidation and cell death were assessed by C11-BODIPY and propidium iodide staining and analyzed by flow cytometry; unpaired 2-tailed t test (A, C, and D) or 1-way ANOVA (B and E–G). *P < 0.05, **P < 0.01, ***P < 0.001.