Myocardial Infarction Activates CCR2(+) Hematopoietic Stem and Progenitor Cells

Abstract

Following myocardial infarction (MI), myeloid cells derived from the hematopoietic system drive a sharp increase in systemic leukocyte levels that correlates closely with mortality. The origin of these myeloid cells, and the response of hematopoietic stem and progenitor cells (HSPCs) to MI, however, is unclear. Here, we identify a CCR2(+)CD150(+)CD48(-) LSK hematopoietic subset as the most upstream contributor to emergency myelopoiesis after ischemic organ injury. This subset has 4-fold higher proliferation rates than CCR2(-)CD150(+)CD48(-) LSK cells, displays a myeloid differentiation bias, and dominates the migratory HSPC population. We further demonstrate that the myeloid translocation gene 16 (Mtg16) regulates CCR2(+) HSPC emergence. Mtg16(-/-) mice have decreased levels of systemic monocytes and infarct-associated macrophages and display compromised tissue healing and post-MI heart failure. Together, these data provide insights into regulation of emergency hematopoiesis after ischemic injury and identify potential therapeutic targets to modulate leukocyte output after MI.

Figure 1. Activation of HSPC after MI

(A) CD48⁻CD150⁺ HSC proliferation measured by BrdU incorporation with flow cytometry in steady state (No MI) and at different time points after MI (permanent coronary ligation, n=8–29 per group). Hours refer to the time of BrdU injection after coronary ligation. (B) Proliferation imaged by ¹⁸F-FLT PET/CT in bone marrow 2 days after MI. Arrow heads indicate increased PET signal in the vertebral marrow (n=5–6 per group). (C) Flow cytometric gating strategy for CCR2⁺ CD150⁺ CD48⁻ LSK, LRP and MPP. (D) Proliferation of CD48⁻ CD150⁺ CCR2⁺ and CCR2⁻ LSK in steady state and 48 hours after MI (n=5–17 per group) in mice. Data are shown as mean ± SEM, * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001. See also Figure S1.

Figure 2. Human CCR2⁺ expression in HSC

(A) CCR2 expression on human non-classical monocytes and classical monocytes in the blood, and CMP, GMP and HSC in the bone marrow harvested from the sternum during open heart surgery. The gating strategy for human HSC is shown. The upper most right panel shows isotype control staining for the CCR2 antibody. (B) Proliferation of human CCR2⁺ and CCR2⁻ HSPC (n=5 per group). Data are shown as mean ± SEM, ** P < 0.01. Patient characteristics are listed in Table S1.

Figure 3. CCR2⁺ HSPC phenotype, location and function

(A) Electron microscopy images of cell subsets (left panel) and magnified cytosol showing distribution of ribosomes (right panel). The bar graph shows ribosome aggregation in subsets (one tail Wilcoxon rank sum test, p=0.036) (n=6 ROIs in 3 CCR2⁺ and n=13 ROIs in 5 CCR2⁻ HSPC). (B) DiO-labelled CCR2⁻ and DiD-labelled CCR2⁺ CD150⁺ CD48⁻ LSK were imaged in the skull with intravital microscopy 1 day after transfer (n=13–27 cells in 2 mice per group). DiO-labelled CCR2⁻ HSC are green, DiD-labelled CCR2⁺ HSPC are white, blood pool is red and osteoblasts are blue. (C) Hierarchical clustering dendrogram based on whole-transcriptome microarray data of CCR2⁺ and CCR2⁻ CD150⁺ CD48⁻ LSK sorted from steady state bone marrow (3 replicates per group). (D) Survival curve of lethally irradiated mice reconstituted with either CCR2⁻ (n=8), CCR2⁺ (n=10) CD150⁺ CD48⁻ LSK or MPP (n=4). (E) Limiting dilution assay determined functional HSC frequency among CCR2⁻ and CCR2⁺ CD150⁺ CD48⁻ LSK in wild type bone marrow. Multi-lineage blood chimerism of 0.1% or higher served as cut off value to determine responders (p < 0.001 for difference in frequency). Table lists dilution steps, numbers of mice analyzed 4 months after HSC transfer, and number of responders. Data are shown as mean ± SEM, * P < 0.05, *** P < 0.001. See also Figure S2 and Table S2.

Figure 4. Role of Mtg16 in myelopoiesis

(A) Gene-set enrichment analysis showing that CCR2⁺ HSPC express genes that are downregulated in hematopoietic stem and progenitor cells in Mtg16^−/− mice, while CCR2⁻ HSC express genes are upregulated in Mtg16^−/− (false discovery rate q-value <0.0001 in both cases). (B) Mtg16 mRNA expression in HSC normalized to Gapdh (n=6 per group). (C) Percentage of CCR2⁺ and CCR2⁻ CD150⁺ CD48⁻ LSK in wild type and Mtg16^−/− mice in steady state (n=7–8 per group). (D and E) FACS quantification of myeloid cells 3 days after MI in the blood (D) and infarct (E) of wild type and Mtg16^−/− mice (n=4–6 per group). (F) Seven thousand CCR2⁺ or CCR2⁻ CD150⁺ CD48⁻ LSK were sorted from GFP⁺ bone marrow and adoptively transferred into CD45.1 C57B/6 on day 4 after MI. Four days after transfer, progeny of GFP⁺ cells were analyzed in the blood and infarct (n=4 per group). (G) In vivo protease imaging in wild type and Mtg16^−/− mice on day 7 after MI (infarct indicated by yellow circle) (n=10–12 per group). (H) Change in end-diastolic volume (Δ EDV) between day 1 and 21 after MI in wild type and Mtg16^−/− mice measured with cardiac MRI. Original MR images show representative enddiastolic mid-ventricular short axis view 21 days after MI (n=6–7 per group). Data are shown as mean ± SEM, * P < 0.05, ** P < 0.01, *** P < 0.001. See also Figure S3.

Figure 5. Lineage relationship between CCR2⁺ and CCR2⁻ CD150⁺ CD48⁻ LSK

(A) Gene-set enrichment analysis for HSC genes shows enrichment in CCR2⁻ HSC (FDR qvalue= 0.001). (B) Principal components analysis. Projection of CCR2⁺ and CCR2⁻ CD48⁻ CD150⁺ LSK into the same space as mouse HSC, leukocytes and erythrocytes shows differentiation of CCR2⁺ CD150⁺ CD48⁻ LSK along the myeloid lineage. (C) Upper panel: Experimental design for competitive bone marrow reconstitution with 100 CCR2⁺ and CCR2⁻ CD150⁺ CD48⁻ LSK. Lower panel: plots showing HSC chimerism in the bone marrow and percentage of CCR2⁺ CD150⁺ CD48⁻ LSK derived from transferred cells. The bar graphs depict quantified flow cytometric data (n=4 per group). (D) Blood chimerism in myeloid cells derived from transferred HSC 4.5 months after bone marrow reconstitution (n=3 per group). (E) mRNA levels of myeloid (PU.1 and Cebpa) and lymphoid (Ikaros) transcription factors in FACS-isolated CCR2⁺ and CCR2⁻ CD150⁺ CD48⁻ LSK, normalized to Gapdh (n=3–10 per group). Data are shown as mean ± SEM, * P < 0.05, ** P < 0.01. See also Figure S4.

Figure 6. Migration of CCR2⁺ CD150⁺ CD48⁻ LSK from the bone marrow in steady state and after MI

(A) Gene-set enrichment analysis for chemokines and chemokine receptors shows enrichment in CCR2⁺ CD150⁺ CD48⁻ LSK (FDR q-value=0.01). (B) Percentage of CCR2⁺ and CCR2⁻ CD150⁺ CD48⁻ LSK in the blood during steady state at Zeitgeber 17 (midnight) and Zeitgeber 5 (noon) (n=4 per group). (C) CD45.1 and CD45.2 mice were analyzed after 5 weeks of parabiosis. Only CCR2⁺ CD150⁺ CD48⁻ LSK migrated to the other bone marrow of the other parabiont. The plots and the bar graph depict % of CCR2⁺ host and migratory CD150⁺ CD48⁻ LSK in the bone marrow (n=5 per group). (D) LSK in blood in CCR2^−/− and wild type mice 4 days after inducing myocardial infarction (MI) (n=5–12 per group). (E) Upper panel: Experimental design to investigate CCR2-dependent emigration from the bone marrow after MI. Lower panel: % of LSK chimerism in the blood (normalized to bone marrow chimera) 4 days after MI (n=5 per group). (F) Post-MI HSC release into blood is dominated by CCR2⁺ CD150⁺ CD48⁻ LSK. FACS plots are gated on HSC. The bar graph depicts the increase of absolute HSC numbers in blood from steady state to day 3 after MI (n=8–9 per group). Data are shown as mean ± SEM, * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001. See also Figure S5.

Figure 7. CCR2⁺ HSPC after LPS treatment

(A) Fold change of CCR2⁺ and CCR2⁻ CD150⁺ CD48⁻ LSK number in bone marrow of LPSinjected mice compared to PBS-injected controls at 4 hours after injection (n=4 per group). (B) Fold change of proliferation 4 hours after injecting LPS (n=6 per group). (C) Steady state TLR4 and TLR2 expression levels on HSC (n=9 per group). Data are shown as mean ± SEM, * P < 0.05, ** P < 0.01, *** P < 0.001. See also Figure S6.