Advertisement

HIF prolyl hydroxylase inhibitor FG-4497 enhances mouse hematopoietic stem cell mobilization via VEGFR2/KDR

Kavita Bisht, Marion E. Brunck, Taichi Matsumoto, Crystal McGirr, Bianca Nowlan, Whitney Fleming, Thomas Keech, Graham Magor, Andrew C. Perkins, Julie Davies, Gail Walkinshaw, Lee Flippin, Ingrid G. Winkler and Jean-Pierre Levesque

Data supplements

Article Figures & Data

Figures

  • Figure 1.

    VEGFR tyrosine kinase inhibitor PTK787 inhibits synergistic effect of FG-4497 on G-CSF–induced HSPC mobilization. (A) Mice were administered G-CSF for 2 days, FG-4497 or vehicle for 3 days, and PTK787 or vehicle for 3 days. At harvest, the numbers of CFCs (B-C), LKS+ HSPCs (D-E), and LKS+CD48CD150+ HSCs (F-G) were measured in peripheral blood (B,D,F) and spleen (C,E,G). Data are from 2 pooled experiments performed several months apart. Each point represents an individual mouse. Bars are means ± standard deviation. *P < .05, **P < 10−2, ***P < 10−3, ****P < 10−4. G, G-CSF with vehicles; G+F, G-CSF plus FG-4497; G+F+P, G-CSF plus FG-4497 plus PTK787; G+P, G-CSF plus PTK787.

  • Figure 2.

    VEGF-A protein increased in BM fluids during G-CSF treatment, but VEGFR1 (Flt1), VEGFR2 (Kdr), and VEGFR3 (Flt4) mRNA are not detected in HSPCs. (A) VEGF-A concentration in BM fluids from mice mobilized with G-CSF. Data are mean ± standard deviation of 3 to 9 mice per time point. (B) RNA-seq analysis of LKS+ cells sorted from the BM of mice treated with saline or G-CSF for 2 days. Results in fragments per kilobase of transcript per million mapped reads (FKPM) are from 3 mice for each treatment group (mean ± standard deviation). (C) In silico single-cell clustering of expression of Flt1, Kdr, Flt4, and Vegfa mRNA on Single-Cell Gene Expression Atlas Web site (http://blood.stemcells.cam.ac.uk/single_cell_atlas.html). Distribution of individual HSPC subsets in blue (HSCs), yellow (LKS+ HSPCs), and red (LKS HPCs) (upper left panel). In all other panels, the intensity of the mRNA signal increases from gray (negative) to red. Slamf1 (CD150) and Procr (EPCR) mRNA identify long-term reconstituting HSCs. Among all single HSPCs sequenced (gray dots), the red dots represent single cells positive for the indicated mRNA. **P < .01.

  • Figure 3.

    VEGFR1 and VEGFR2 are expressed by Sca1+endothelial cells in the BM. BM cells were extracted from the endosteal region, enriched in stromal cells by magnetic depletion of CD45+ leukocytes and Ter119+ erythroid cells, and stained for hematopoietic, endothelial, and mesenchymal markers together with anti-VEGFR1 and anti-VEGFR2 fluorescent antibodies. (A) Gating strategy to identify endothelial and mesenchymal cells. Following the exclusion of residual CD45+ and Ter119+ leukocytes and erythroid cells, nonhematopoietic CD45Ter119 cells were gated as CD31+Sca1 endothelial cells and CD31+Sca1+ endothelial cells. Remaining CD45Ter119CD31 mesenchymal cells were gated as PDGFRαSca1, PDGFRα+Sca1, PDGFRα+Sca1+, and PDGFRαSca1+ cells. (B) Typical dot-plots showing VEGFR1 and VEGFR2 expression on previously defined endothelial and mesenchymal populations. (C) Percentage of endothelial and mesenchymal BM cells expressing VEGFR1 and/or VEGFR2. Bar graphs show mean ± standard deviation of 4 mice. (D) Kdr (VEGFR2) mRNA expression by qRT-PCR on CD11b+F4/80+VCAM1+CD169+ macrophages (CD11b+Mφ), CD11bF4/80+VCAM1+CD169+ macrophages (CD11b-Mφ), CD11b+F4/80+VCAM1CD169 monocytes (MO), CD11b+F4/80Ly6G+ granulocytes (Gran), CD45Ter119CD31+ endothelial cells (EC), CD45Ter119CD31PDGFRα+/−Sca1+ mesenchymal progenitor cells (P±S+), PDGFRα+Sca1 mesenchymal progenitor cells (P+S-), and CD45Ter119CD31PDGFRαSca1 stromal cells (P-S-) sorted from the BM of mice treated with saline (C), FG-4497 alone (F), G-CSF (G), or G-CSF plus FG-4497 (G+F) for 3 days. Each dot represents a separate mouse and separate sort. Data are relative to Hprt mRNA. The P values were calculated using ANOVA.

  • Figure 4.

    Neutralization of VEGFR2 blocks the effect of FG-4497 on G-CSF–induced HSPC mobilization. (A) Mice were administered G-CSF for 2 days, FG-4497 or vehicle for 3 days, and neutralizing rat anti-mouse VEGFR2 mAb DC101 or control rat IgG1 every other day. At harvest, the numbers of CFCs (B-C), LKS+ HSPCs (D-E), and LKS+CD48CD150+ HSCs (F-G) were measured in peripheral blood (B,D,F) and spleen (C,E,G). Data are from 2 pooled experiments performed several months apart. Each point is an individual mouse. Bars are means ± standard deviation. *P < .05, **P < 10−2, ***P < 10−3. G+DC, G-CSF with vehicle DC101 mAb; G+F+DC, G-CSF plus FG-4497 plus DC101; G+F+IgG, G-CSF plus FG-4497 plus rat IgG1; G+IgG, G-CSF with vehicle with rat IgG1.

  • Figure 5.

    VEGF-A concentration is increased in vivo in response to DC101 and in vitro in response to FG-4497. (A-B) Mice were treated with G-CSF, FG-4497, neutralizing rat anti-mouse VEGFR2 mAb DC101, or control rat IgG1, as described in Figure 4A. Mouse VEGF-A protein concentration was measured by enzyme-linked immunosorbent assay in blood plasma (A) and femoral BM fluids (B). Data are from 2 pooled experiments performed at a 1-month interval. Each point represents an individual mouse. (C) FG-4497 stimulated VEGF-A secretion by BMDMs in vitro. BMDMs were cultured for 48 hours in the presence of 100 ng/mL rhuG-CSF (G), 40 µM FG-4497 (F), or both (G+F). Control cells were added vehicle. Supernatants were harvested at 24 and 48 hours, and mouse VEGF-A concentration was measured by enzyme-linked immunosorbent assay. Data are means ± SD of 3 replicates per conditions. (D) Western blot for HIF-1α on BMDMs cultured for 24 or 48 hours in the presence of 100 ng/mL rhuG-CSF or 40 µM FG-4497. (E) Vegfa, Vegfc, and Vegfd mRNA expression by qRT-PCR on CD11b+F4/80+VCAM1+CD169+ macrophages (CD11b+Mφ), CD11bF4/80+VCAM1+CD169+ macrophages (CD11b-Mφ), CD11b+F4/80+VCAM1CD169 monocytes (MO), CD11b+F4/80Ly6G+ granulocytes (Gran), CD45Ter119CD31+ endothelial cells (EC), CD45Ter119CD31PDGFRα+/−Sca1+ mesenchymal progenitor cells (P±S+), PDGFRα+Sca1 mesenchymal progenitor cells (P+S-), and CD45Ter119CD31PDGFRαSca1 stromal cells (P-S-) sorted from the BM of mice treated with saline (C), FG-4497 alone (F), G-CSF (G), or G-CSF plus FG-4497 (G+F) for 3 days. Each circle represents a separate mouse and separate sort (n = 5 mice per population per treatment condition). Data are relative to Hprt mRNA. *P < .05, **P < 10−2, ***P < 10−3, ****P < 10−4, ANOVA.

  • Figure 6.

    VEGF-A has no chemotactic effect on HSPCs in vitro. Chemotaxis assay with Kit+ enriched BM cells from naive C57BL/6 mice incubated for 4 hours at 37°C in the presence of 50 ng/mL CXCL12 in the bottom chamber or 100 ng/mL mouse VEGF-A in the bottom chamber or in the top insert. The percentage of LKS+ cells (A) and LKS+CD48CD150+ HSCs (B) that migrated to the bottom chamber was measured by flow cytometry. Data are from 2 pooled experiments performed several weeks apart. Each symbol is a well. Bars are means ± standard deviation. Each individual group with CXCL12 in the bottom well was significantly different from each group without CXCL12. ****P < 10−4, ANOVA.