Ischemic tissue damage activates hematopoietic stem and progenitor cells (HSPCs) in the bone marrow (BM)-generating myeloid cells, and persistent HSPC activity may drive chronic inflammation and impair tissue recovery. Although increased reactive oxygen species in the BM regulate HSPC functions, their roles in myelopoiesis of activated HSPCs and subsequent tissue recovery during ischemic damage are not well understood. In this paper, we report that deletion of Nox2 NADPH oxidase in mice results in persistent elevations in BM HSPC activity and levels of inflammatory monocytes/macrophages in BM and ischemic tissue in a model of hindlimb ischemia. Ischemic tissue damage induces oxidants in BM such as elevations of hydrogen peroxide and oxidized phospholipids, which activate redox-sensitive Lyn kinase in a Nox2-dependent manner. Moreover, during tissue recovery after ischemic injury, this Nox2-ROS–Lyn kinase axis is induced by Nox2 in neutrophils that home to the BM, which inhibits HSPC activity and inflammatory monocyte generation and promotes tissue regeneration after ischemic damage. Thus, oxidant signaling in the BM mediated by Nox2 in neutrophils regulates myelopoiesis of HSPCs to promote regeneration of damaged tissue.

This article is featured in In This Issue, p.2191

Hematopoietic stem and progenitor cells (HSPCs) maintain blood cell levels in the steady state and are activated in response to acute inflammation following tissue injury (1, 2). Acute inflammation is normally followed by resolution of inflammation, and failure of this process can lead to chronic inflammation associated with persistent HSPC activation (3, 4). HSPCs can be directly activated by damage-associated molecular patterns (DAMPs) through TLRs, resulting in the generation of myeloid cells, especially monocytes (monopoiesis) (1, 5, 6). For example, NF-κB–dependent inflammatory cytokine production from HSPCs promotes monopoiesis from HSPCs under TLR2/4 stimulation in vitro through autocrine mechanism (5). However, factors that limit or resolve HSPC activation in this context have yet to be reported. The HSPC niche, which is the discrete microenvironment in the bone marrow (BM) where the HSPC reside, plays a critical role in regulating HSPC activity (7, 8). Recent studies suggest that HSPC progeny are indispensable components of the niche and thus may play a role in regulating HSPC activity.

Reactive oxygen species (ROS) are known to play important roles in regulating inflammatory pathways (9) and stem cell activity (10). Myeloid cells, such as neutrophils and monocytes/macrophages, are major producers of ROS during inflammation, and NADPH oxidase (Nox) enzymes, especially Nox2, are the dominant sources of myeloid ROS (11, 12). Nox2 is involved in killing bacteria and fungi and is a critical regulatory component of signaling in inflammatory cells (12). Importantly, in some inflammatory conditions such as rheumatoid arthritis, systemic lupus erythematosus, and inflammatory bowel disease, Nox2-derived ROS limit rather than promote the inflammatory response (1315). At the cellular level, lack of ROS causes proliferation of proinflammatory T cells, and restoring ROS corrected the arthritogenic phenotype of T cells (13, 16). In an animal model of rheumatoid arthritis, the accumulation of arthritogenic T cells was accompanied by the expansion of myeloid cells in the BM through an unknown mechanism (13).

We previously identified that Gr-1+ myeloid cells in the BM cavity produced ROS in a Nox2-dependent manner following ischemic tissue damage in mouse hindlimb (17), similar to in acute bacterial infection (18). Although Nox2 produces superoxide anion, it can be converted to more stable forms such as H2O2 or can induce oxidized molecules. Moreover, ROS in damaged tissue or wound environment are quickly sensed by redox-regulated Src family kinases, which regulate cellular functions in cells in oxidative microenvironment (19). This critical pathway activated in oxidative environment has not yet been examined in the BM.

We also previously showed that Nox2-dependent ROS in the BM is associated with expansion of lineage-negative HSPCs in the BM and their mobilization from the BM to the blood (17). ROS can regulate HSPC fate and function in both cell-intrinsic and cell-extrinsic manners (10). In addition, both intracellular and extracellular ROS promote granulopoiesis of myeloid progenitor cells (18, 20). However, it remains to be determined how monocyte generation (monopoiesis) is regulated by Nox2-derived ROS during tissue recovery after ischemic damage. Using a mouse hindlimb ischemia model and in vitro HSPC monopoiesis assay, we report that following ischemic injury, extracellular ROS derived from Nox2 limit monopoiesis from HSPCs. This process is mediated by oxidative signals by neutrophil homing to the BM, diffusible H2O2, and oxidized phospholipids and Lyn kinase activation and plays a critical role in the resolution of inflammation during tissue regeneration.

Nox2 knockout (KO) mice (B6.129S-Cybbtm1Din/J), Nox2y/− (male), and Nox2−/− (female) CD45.1 (B6.SJL-Ptprca Pepcb/BoyJ) and C57BL/6J wild-type (WT) control mice were obtained from The Jackson Laboratory. In experiments using a congenic system (CD45.1 versus CD45.2), CD45.1 mice were used as Nox2+/+ controls. Lyn−/− mice (21) were back-crossed to a C57BL/6 background as described (22). Akt1−/− mice were generated in the laboratory of Nissim Hay as described (23). For NF-κB reporter mice, we used a line of transgenic mice (HLL [HIV–long terminal repeat/luciferase] mice) that carries the proximal 5′ HIV (HIV-1) long terminal repeat, a well-characterized NF-κB promoter, driving the expression of Photinus luciferase (24). HLL mice were cross-bred with Nox2−/− mice to generate Nox2y/−-HLL mice. All the mice were bred and/or maintained in specific pathogen-free facilities at the University of Illinois at Chicago. For surgical procedures, anesthesia and perioperative analgesia were achieved with isoflurane and s.c. injection of buprenorphine SR-LAB, respectively. Hindlimb ischemia surgery was performed as described previously (17, 25). All procedures involving animals were approved by The Animal Care Committee at the University of Illinois at Chicago and conducted in accordance with institutional guidelines.

Blood was collected from the retro-orbital cavity by a heparinized glass tube under terminal anesthesia and kept in an EDTA 2K Tube. Mice were perfused with PBS through the left ventricle of heart. The tibialis anterior muscle was harvested from the ischemic side for flow cytometry, gene expression, and histological analysis. The superior half of muscles were digested with 2.5 mg/ml collagenase B (Roche) in Dulbecco’s PBS with Ca++ and Mg++ along with mechanical disruption by passing through needles. BM cells were obtained by flushing of the cavity of the right femur and tibia. Total cells were counted manually on a hemocytometer.

RBCs in blood samples were lysed using PharmLyse solution (BD Biosciences). All samples except for myeloid progenitor characterization were blocked with 5 μg/ml anti-CD16/32 in FACS buffer (2% FBS and 2 mM EDTA in Dulbecco’s PBS) before surface staining on ice with the following Abs: Sca-1 (D7), cKit (2B8), CD34 (RAM34), FcγRII/III (CD16/32:93), CD11b (M/70), Ly-6G (1A8), Ly-6C (AL-21), F4/80 (BM8), CD62L (MEL-14), CD45.1 (A20), and CD45.2 (104). For lineage (Lin), mouse lineage panel (Ter-119, M1/70, RB6-8C5, 145-2C11, and RA3-6B2) in addition to CD4 (GK1.5), CD8a (53-6.7), CD19 (605), NK-1.1 (PK136), and IL7Ra (A7R34) were used. Abs were obtained from BD Biosciences, eBioscience, or Biolegend.

To determine cell proliferation, BrdU (Sigma-Aldrich; 1 mg/mouse) was injected into mice through the retro-orbital cavity 24 h before mice were euthanized. Isolated cells were immunolabeled as described above and then fixed, permeabilized, and labeled using FITC BrdU Flow Kit (BD Biosciences). To measure ROS levels, cells were incubated with 5 μM CellROX Green (Thermo Fisher Scientific) for 15 min at 37°C following surface marker staining and then fixed with 0.5% paraformaldehyde overnight. Samples were analyzed by CyAn ADP (Beckman Coulter) or LSRFortessa (BD Biosciences).

LineageSca-1+c-Kit+ (LSK) cells were sorted from the BM of WT, Nox2−/−, Lyn−/−, or CD45.1 mice. Femurs, tibias, ilia, and humeri were flushed with PBS containing 2% FBS, 2 mM EDTA, 100 IU/ml penicillin, and 100 μg/ml streptomycin and filtered by a 70-μm mesh cell strainer. Total BM cells were magnetically enriched by EasySep Mouse Hematopoietic Progenitor Cell Enrichment Kit (STEMCELL Technologies). The enriched HSPCs were further stained with Abs for Sca-1, cKit, and Lin and sorted using MoFlo Cell Sorter (Beckman Coulter). Purity of sorted LSK cells was averaged 98.4%, revealed by subsequent flow cytometry analysis excluding dead cells. Isolated LSK cells were equally divided and cultured in a U-bottom 96-well plate (3000–7500 cells as indicated in figure legends) with serum-free culture medium StemPro-34 SFM (Invitrogen) containing 2 mM l-glutamine, 100 IU/ml penicillin, 100 μg/ml streptomycin, 20 ng/ml Stem Cell Factor (PeproTech), and 50 ng/ml Flt3-Ligand (PeproTech). To stimulate myeloid differentiation, LPS (derived from Escherichia coli 055:B5 from Sigma-Aldrich or InvivoGen) was added to the culture. For myeloid progenitor cells, the magnetically enriched HSPCs were further stained with Abs for Sca-1, cKit, CD34, FcγRII/III, and Lin and sorted and cultured similarly to LSK cells. For pharmacological Lyn activation, the Lyn activator MLR-1023 (Sigma-Aldrich) or DMSO only was used. Catalase from bovine liver (Millipore) was added to scavenge extracellular H2O2 (26). The yield per input was calculated based on the number of total cells in each well, which was determined by event number in live cell gate based on forward and side scatter.

For culture experiments, neutrophils were isolated from BM by FACS sorting of Gr-1+ and intermediate side scatter population. For adoptive transfer, neutrophils were obtained from either BM or blood as indicated in the figure legends. BM neutrophils were enriched from the interface of Histopaque 1119 and Histopaque 1077 (Sigma-Aldrich) in either subsequent (27) or simultaneous (28) separation. Isolated cells were 80 ± 5% purity for CD11b+Ly-6G+ and >95% viability. One million neutrophils were adoptively transferred to one recipient via retro-orbital cavity. For the DAMPs injection experiment (Fig. 6), isolated neutrophils were cultured in RPMI 1640 medium supplemented with 10% FBS for 24 h. DAMPs were isolated as tissue homogenate from skeletal muscles of hindlimb from WT mice, which were homogenized in 1 ml of Dulbecco’s PBS using a beads homogenizer (2 × 30 s). A total of 160 μg of protein was injected via retro-orbital cavity.

To track transferred cells, we stained cells with DiI (Molecular Probes) prior to the transfer. In other experiments, for in vivo labeling blood cells, we i.v. injected a single 1-mg EZ-Link Sulfo-NHS-LC-Biotin (Life Technologies) in PBS. After 2 h, blood and BM were collected, stained for Ly-6G and CD62L, and labeled with Alexa 450–conjugated streptavidin. Samples from mice that received the injection 30 min prior to euthanasia served as a control to determine the threshold of biotin-positive signal.

We performed bioluminescence imaging in HLL or Nox2y/−-HLL mice using The Xenogen IVIS Spectrum (Perkin Elmer) following i.v. injection of 120 mg/kg d-luciferin (Gold Biotechnology). Hair was shaved prior to imaging and images were captured from the dorsal side. The number of photons per second was measured from identically sized regions of interest (ROI) located at the anatomically same location between left and right hindlimb.

Plasma was extracted from femur and tibia of the nonischemic leg. The long bone was drilled with a 26G needle and placed in an engineered P200 pipette tip and was centrifuged by 400 × g to collect supernatant of fluid and cells from the BM cavity and prefilled 50 μl of phosphate buffer. Total protein was measured by Pierce 660 nm Protein Assay Reagent (Thermo Fisher Scientific). Concentration-adjusted samples were used for downstream analysis. For hydrogen peroxide, we used Amplex Red kit according to the manufacturer’s protocol (Thermo Fisher Scientific). Oxidized phospholipids were measured as described (29). Briefly, we plotted the BM plasma onto a nitrocellulose membrane and let it dry. Oxidized 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphorylcholine (OxPAPC) on the membrane was probed by TopFluor-conjugated anti-OxPAPC (E06) Ab. Dried membrane was observed by an upright fluorescent microscope (Nikon) with a band pass filter for FITC. Captured images by a CCD camera (Nikon) were imported to ImageJ software, on which particle structures were quantified as integrated density, which represents the product of area and mean gray value. Background subtraction was performed using an image taken from an area without samples.

Mice were perfusion-fixed with 4% paraformaldehyde. Collected femurs or tibialis anterior muscles were further fixed for 4 h at 4°C and cryoprotected with sucrose solution and were mounted in tissue-freezing medium. For BM, nondecalcified 7-μm cryosections were taken using the Kawamoto method, which uses an adhesive film and a tungsten-made blade (30). A longitudinally sectioned bone from the center of the femur was obtained. After blocking, sections were then incubated with antiphosphorylated Lyn Ab (pY396, Epitomics). DiI signals of ex vivo–labeled cells were analyzed in some experiments. Four representative images at ×20 magnification per long bone section were analyzed. For each field, the number of fluorescent positive cells was manually counted, and data were averaged over sections to provide a representative value across all BM. For muscle, a transverse section from the center of the tibialis anterior muscles was obtained using the regular cryosection, and H&E staining was performed by a standard procedure. All digital images were obtained using either a Nikon 80i microscope, a Zeiss LSM 510, or a Zeiss LSM 710.

Analysis was performed as we described previously (17). The 2−ΔΔCT method was used to calculate relative mRNA expression among different samples normalized to Rn18s and Hprt.

BM-derived macrophages were generated as described (31). Cells were lysed in RIPA buffer containing a proteinase inhibitor mixture (Roche). Thirty micrograms of proteins separated by SDS-PAGE were blotted with Abs for phosphor-Lyn (pY396; Epitomics) at 1:500 dilution or total Lyn (Cell Signaling Technology) at 1:1000 dilution and then probed with fluorescent conjugated Abs (LI-COR Biosciences) at 1:10,000. Signals were visualized using Odyssey CLx imager with image processing by Image Studio software (LI-COR Biosciences).

Two-way ANOVA with Bonferroni multiple comparisons were used for all statistical comparisons except for the comparison of two groups in which an unpaired t test was used. A p value <0.05 was considered to be statistically significant on GraphPad Prism 6.

To examine the role of Nox2 and ROS in HSPC activity and inflammatory responses following ischemic injury, we surgically induced ischemia in one hindlimb of each mouse, harvested BM cells from the nonischemic contralateral leg, and analyzed HSPC activity (Fig. 1A). We found that LSKs were increased in Nox2-KO mice both prior to injury and on day 3 after ischemia compared with those in WT mice, whereas the number of cKit+Sca1Lin myeloid progenitors only differed prior to injury (Fig. 1B). The expansion of LSKs in Nox2-KO mice was associated with a sustained increase in proliferation measured by in vivo BrdU incorporation, whereas proliferation was increased only transiently in WT mice (Fig. 1C). In contrast, the transient increase in proliferation of myeloid progenitors observed in WT mice was absent in Nox2-KO mice (Fig. 1C). We further found a sustained increase of Ly-6Chi monocytes in the BM of Nox2-KO mice after hindlimb ischemia, whereas Ly-6G+ granulocytes, which were increased prior to injury, did not increase at later phase of injury (Fig. 1D). Thus, increased LSK activity was associated with increased levels of monocytes in the BM in Nox2-KO mice.

FIGURE 1.

Increased HSPC activity and monopoiesis in Nox2 KO mice following hindlimb ischemia. (A) Hindlimb ischemia was induced by femoral artery ligation and excision, and BM from the nonischemic leg was analyzed in mice without operation (Pre) and in indicated time points after hindlimb ischemia. (B) Dot plots of lineage-negative (Lin) 3 d after ischemia and quantification (cell number per tibia) of LSK cells and LincKit+Sca1 myeloid progenitors (MP) (n = 6–12 mice in each group per time point). (C) Contour plots of LSK cells 3 d after ischemia and BrdU+ proliferative LSK or MP in the BM (n = 4–6 in each group per time point). (D) Dot plots of CD11b+ myeloid cells in the BM and quantification (percentage in total BM cell and cell number per tibia) of Ly-6Chi monocyte (Mo) and Ly-6G+ granulocyte (Gr) (n = 6–12 mice in each group per time point). Statistical significance was evaluated by two-way ANOVA with Bonferroni multiple comparisons test. Error bars indicate SEM. *p < 0.05, ***p < 0.001.

FIGURE 1.

Increased HSPC activity and monopoiesis in Nox2 KO mice following hindlimb ischemia. (A) Hindlimb ischemia was induced by femoral artery ligation and excision, and BM from the nonischemic leg was analyzed in mice without operation (Pre) and in indicated time points after hindlimb ischemia. (B) Dot plots of lineage-negative (Lin) 3 d after ischemia and quantification (cell number per tibia) of LSK cells and LincKit+Sca1 myeloid progenitors (MP) (n = 6–12 mice in each group per time point). (C) Contour plots of LSK cells 3 d after ischemia and BrdU+ proliferative LSK or MP in the BM (n = 4–6 in each group per time point). (D) Dot plots of CD11b+ myeloid cells in the BM and quantification (percentage in total BM cell and cell number per tibia) of Ly-6Chi monocyte (Mo) and Ly-6G+ granulocyte (Gr) (n = 6–12 mice in each group per time point). Statistical significance was evaluated by two-way ANOVA with Bonferroni multiple comparisons test. Error bars indicate SEM. *p < 0.05, ***p < 0.001.

Close modal

Consistent with the notion that blood Ly-6Chi monocytes are a main source of inflammatory monocytes/macrophages during active inflammation (32), the increase in BM Ly-6Chi Mo is associated with prolonged tissue accumulation of inflammatory Ly-6Chi monocytes/macrophages (CD11b+Ly-6GLy-6Chi) together with Ly-6G+ granulocytes in injured muscle in Nox2-KO mice (Fig. 2A). In contrast, WT mice showed a transient increase in Ly-6Chi monocytes/macrophages population with a rapid increase on day 3 and return to baseline on day 7 (Fig. 2A). This parallels a prolonged increase in blood Ly-6G granulocytes in Nox2-KO mice (Supplemental Fig. 1A). We used NF-κB reporter mice and in vivo bioluminescence imaging to analyze inflammatory NF-κB in vivo and found that the increase in inflammatory cells in tissue resulted in persistent NF-κB activation in the ischemic hindlimb in Nox2-KO mice crossed with the reporter mice compared with WT reporter mice (Fig. 2B). The increased tissue NF-κB activity was correlated with persistent increases in TNF-α and IL-6 gene expressions (Supplemental Fig. 1B) and reduced regenerating skeletal muscle fibers (Fig. 2C) on day 7 after hindlimb ischemia in Nox2-KO mice. Thus, downregulation of LSK activity on day 3 may be required for the reduction in inflammatory Ly-6Chi monocytes/macrophages after this time point in ischemic muscles and for the induction of inflammation resolution, and failure to reduce LSK activity may result in monocytosis and a prolonged increase in tissue Ly-6Chi monocytes/macrophages, leading to persistent inflammation.

FIGURE 2.

Increased inflammatory monocytes/macrophages and impaired muscle regeneration in Nox2 KO mice following hindlimb ischemia. (A) Contour plots of CD11b+ myeloid cells and quantification of Ly-6Chi monocytes/macrophages (Mo/Mφ), Ly-6G+ granulocytes (Gr), and Ly-6Clo macrophages (Mφ) in ischemic tibialis anterior muscles 3 and 7 d after hindlimb ischemia. Nonoperated mice were used as healthy control (Pre). (n = 3 mice for day 0 and n = 6 mice for day 3 and 7 per group per time point) (B) Bioluminescent images of NF-κB reporter (HLL) mice. Red circles indicate ROI symmetrically chosen from ischemic and nonischemic hindlimbs. Ratio of radiance (photon per second) from ROI of ischemic leg to that of nonischemic leg is shown (n = 6 mice per group). (C) Histological analysis by H&E staining of tibialis anterior muscles on day 7 (n = 6). Scale bars, 100 μm. Statistical significance was evaluated by two-way ANOVA with Bonferroni multiple comparisons test. Error bars indicate SEM. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 2.

Increased inflammatory monocytes/macrophages and impaired muscle regeneration in Nox2 KO mice following hindlimb ischemia. (A) Contour plots of CD11b+ myeloid cells and quantification of Ly-6Chi monocytes/macrophages (Mo/Mφ), Ly-6G+ granulocytes (Gr), and Ly-6Clo macrophages (Mφ) in ischemic tibialis anterior muscles 3 and 7 d after hindlimb ischemia. Nonoperated mice were used as healthy control (Pre). (n = 3 mice for day 0 and n = 6 mice for day 3 and 7 per group per time point) (B) Bioluminescent images of NF-κB reporter (HLL) mice. Red circles indicate ROI symmetrically chosen from ischemic and nonischemic hindlimbs. Ratio of radiance (photon per second) from ROI of ischemic leg to that of nonischemic leg is shown (n = 6 mice per group). (C) Histological analysis by H&E staining of tibialis anterior muscles on day 7 (n = 6). Scale bars, 100 μm. Statistical significance was evaluated by two-way ANOVA with Bonferroni multiple comparisons test. Error bars indicate SEM. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

To test whether in vitro monopoiesis from HSPCs is altered by genetic loss of Nox2, we isolated LSKs from the BM of either WT or Nox2-KO mice and stimulated those cells with LPS, a TLR2/4 ligand (Fig. 3A). Consistent with previous studies (5, 6), TLR2/4 stimulation of LSKs enhanced monopoiesis both by increasing total cells (Supplemental Fig. 2A) and skewing differentiation toward CD11b+Ly-6C+ cells (Fig. 2B). We further discovered that Nox2-deficient LSKs exhibit enhanced monopoiesis in vitro (Fig. 2B). Interestingly, this enhanced monopoiesis exhibited by Nox2-deficient LSK cells was not observed in lineageSca-1c-Kit+ myeloid progenitor populations, CD34loFcγRint common myeloid progenitor, and CD34+FcγRhi granulocyte-monocyte progenitor (Fig. 3C). Thus, the effect of Nox2 that limits monopoiesis is specific for TLR2/4-stimulated LSKs.

FIGURE 3.

Nox2 limits Ly-6Chi monocyte generation from HSPCs in vitro. (A) LSK cells were sorted from BM of WT and Nox2 KO mice and cultured with serum-free StemPro 34 medium supplemented with 20 ng/ml stem cell factor and 50 ng/ml flt3 ligand (3000 cells per well) with 2 or 10 μg/ml LPS for 84 h. (B) Dot plots and quantifications of CD11b+Ly-6Chi monocytes derived from LSK cells. Frequency and number shown as yield per input of CD11b+Ly-6Chi monocytes from LSKs. The results are from three biological replicates of three independent sortings. (C) Common myeloid progenitors (CMP: LinSca1cKit+CD34intFcγRint) and granulocyte-macrophage progenitors (GMP: LinSca1cKit+CD34+FcγRhi) were cultured and stimulated with 10 μg/ml LPS for 84 h. The results are from three biological replicates of three independent sortings. (D) LSK cells were sorted from BM of CD45.1 (Nox2+/+) and Nox2 KO (Nox2−/−) mice. LSK cells of different genotype were cocultured with various mixing ratios (total 6000 cells) for 72 h under 10 μg/ml LPS stimulation. Number of differentiated Ly-6Chi monocytes (CD11b+Ly-6C+) from LSKs of each genotype was analyzed by flow cytometry on CD45.1+ (for Nox2+/+) or CD45.2+ (for Nox2−/−) gated population, separately. (E) Representative contour plots of CD45.1+ (upper) and CD45.2+ (for Nox2−/−) gated population. (F) Yield of CD11b+Ly-6C+ cells from input LSK cells was calculated (triplicated samples from a representative experiment of two independent experiments). Statistical significance was evaluated by two-way ANOVA with Bonferroni multiple comparisons test (B and F) and by two-tailed unpaired Student t test (C). Error bars indicate SEM. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 3.

Nox2 limits Ly-6Chi monocyte generation from HSPCs in vitro. (A) LSK cells were sorted from BM of WT and Nox2 KO mice and cultured with serum-free StemPro 34 medium supplemented with 20 ng/ml stem cell factor and 50 ng/ml flt3 ligand (3000 cells per well) with 2 or 10 μg/ml LPS for 84 h. (B) Dot plots and quantifications of CD11b+Ly-6Chi monocytes derived from LSK cells. Frequency and number shown as yield per input of CD11b+Ly-6Chi monocytes from LSKs. The results are from three biological replicates of three independent sortings. (C) Common myeloid progenitors (CMP: LinSca1cKit+CD34intFcγRint) and granulocyte-macrophage progenitors (GMP: LinSca1cKit+CD34+FcγRhi) were cultured and stimulated with 10 μg/ml LPS for 84 h. The results are from three biological replicates of three independent sortings. (D) LSK cells were sorted from BM of CD45.1 (Nox2+/+) and Nox2 KO (Nox2−/−) mice. LSK cells of different genotype were cocultured with various mixing ratios (total 6000 cells) for 72 h under 10 μg/ml LPS stimulation. Number of differentiated Ly-6Chi monocytes (CD11b+Ly-6C+) from LSKs of each genotype was analyzed by flow cytometry on CD45.1+ (for Nox2+/+) or CD45.2+ (for Nox2−/−) gated population, separately. (E) Representative contour plots of CD45.1+ (upper) and CD45.2+ (for Nox2−/−) gated population. (F) Yield of CD11b+Ly-6C+ cells from input LSK cells was calculated (triplicated samples from a representative experiment of two independent experiments). Statistical significance was evaluated by two-way ANOVA with Bonferroni multiple comparisons test (B and F) and by two-tailed unpaired Student t test (C). Error bars indicate SEM. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

To investigate the mechanism underlying enhanced monopoiesis from Nox2-KO HSPCs, we used a coculture system in which LSKs from WT and Nox2-KO mice are mixed at different ratios and analyzed separately based on CD45.1 (Nox2-intact) or CD45.2 (Nox2-KO) expression (Fig. 3D) so that we could determine whether extracellular factor(s) released from Nox2-intact mice influence Nox2-KO cells. Monopoiesis of both WT and Nox2-KO LSKs was significantly reduced as the proportion of WT LSKs was increased (Fig. 3E, 3F), suggesting that extracellular factor(s) released from Nox2-intact LSKs or their progeny limits in vitro monopoiesis, potentially through a paracrine mechanism (5).

We showed previously that Nox2 expression is significantly higher in differentiated Gr-1+ myeloid cells compared with progenitor cells in the BM (25). Thus, we sought to determine whether a coculture of Nox2-abundant Gr-1+ cells can increase ROS in the HSPC environment and reduce in vitro monopoiesis from LSKs (Fig. 4A). LSKs cocultured with Nox2-intact Gr-1+ cells but not Nox2-KO Gr-1+ cells exhibited significantly lower monocyte output (Fig. 4B). Importantly, this limiting effect on HSPC monopoiesis was blocked by catalase, an extracellular H2O2 scavenger, implicating H2O2 as the key regulator of monopoiesis (Fig. 4B). We added H2O2 in Nox2-KO LSK culture; however, extracellular H2O2 marginally reduced percentage of monocytes, whereas total yield of monocytes from LSKs was substantially reduced (Fig. 4C). This striking contrast of Nox2-limiting effect on LSK monopoiesis between Gr-1+ cell coculture and extracellular H2O2 suggests that additional factors derived from cellular components are involved in the extracellular Nox2-mediated limitation. During coculture with LSKs in serum-free medium, Gr-1+ cells reduced their number, likely due to apoptosis. An oxidative environment and apoptotic cells can produce oxidized phospholipids such as OxPAPC (33). Indeed, we found that OxPAPC significantly reduced both frequency and yield of differentiated monocytes from cultured LSKs stimulated with TLR2/4 ligand (Fig. 4D). Interestingly, OxPAPC itself modestly induced monocyte differentiation in LSKs (Fig. 4D). In contrast, in myeloid progenitors, this OxPAPC-mediated limitation of differentiation was not evident in terms of frequency (Supplemental Fig. 2B), which is partly similar to what we observed with those cells between WT and Nox2-KO mice (Fig. 3C), although OxPAPC inhibited the number of outputs from LPS-stimulated myeloid progenitors (Supplemental Fig. 2B). This indicates that OxPAPC may be a key factor in the extracellular Nox2-mediated inhibition of LSK differentiation.

FIGURE 4.

Extracellular Nox2-dependent ROS and oxidized phospholipids limit Ly-6Chi monocyte generation from HSPCs in vitro. (A) LSK cells were sorted from BM of CD45.1 (Nox2+/+) mice. Neutrophil-rich Gr-1+ myeloid cells were sorted from BM of WT and Nox2 KO mice. The LSK cells were cocultured with Gr-1+ cells at 1:10 ratio (7500 LSKs and 75,000 Gr-1+ cells) with serum-free StemPro 34 medium supplemented with 20 ng/ml stem cell factor and 50 ng/ml flt3 ligand and with 10 μg/ml LPS stimulation for 72 h. (B) Contour plots of CD45.1+ (LSK-derived cells) gated population. Catalase at 100 μg/ml was added to scavenge extracellular ROS, H2O2. Frequency and yield of CD11b+Ly-6Chi cells from LSK cells were compared among groups (triplicated samples from a representative experiment of three independent experiments). (C) H2O2 was added to Nox2 KO LSK cells cultured with serum-free StemPro 34 medium supplemented with 20 ng/ml stem cell factor and 50 ng/ml flt3 ligand and with 10 μg/ml LPS stimulation for 84 h (triplicated samples from a representative experiment of three independent experiments). (D) LSK cells from WT or Nox2 KO mice were stimulated with either or both 30 μg/ml oxidized phospholipid (OxPAPC) or/and 10 μg/ml LPS for 96 h (triplicated cultures of pooled LSK cells from two independent experiments). Statistical significance was evaluated by one-way (C) or two-way (B and D) ANOVA with Bonferroni multiple comparisons test. Error bars indicate SEM. *p < 0.05, **p < 0.01.

FIGURE 4.

Extracellular Nox2-dependent ROS and oxidized phospholipids limit Ly-6Chi monocyte generation from HSPCs in vitro. (A) LSK cells were sorted from BM of CD45.1 (Nox2+/+) mice. Neutrophil-rich Gr-1+ myeloid cells were sorted from BM of WT and Nox2 KO mice. The LSK cells were cocultured with Gr-1+ cells at 1:10 ratio (7500 LSKs and 75,000 Gr-1+ cells) with serum-free StemPro 34 medium supplemented with 20 ng/ml stem cell factor and 50 ng/ml flt3 ligand and with 10 μg/ml LPS stimulation for 72 h. (B) Contour plots of CD45.1+ (LSK-derived cells) gated population. Catalase at 100 μg/ml was added to scavenge extracellular ROS, H2O2. Frequency and yield of CD11b+Ly-6Chi cells from LSK cells were compared among groups (triplicated samples from a representative experiment of three independent experiments). (C) H2O2 was added to Nox2 KO LSK cells cultured with serum-free StemPro 34 medium supplemented with 20 ng/ml stem cell factor and 50 ng/ml flt3 ligand and with 10 μg/ml LPS stimulation for 84 h (triplicated samples from a representative experiment of three independent experiments). (D) LSK cells from WT or Nox2 KO mice were stimulated with either or both 30 μg/ml oxidized phospholipid (OxPAPC) or/and 10 μg/ml LPS for 96 h (triplicated cultures of pooled LSK cells from two independent experiments). Statistical significance was evaluated by one-way (C) or two-way (B and D) ANOVA with Bonferroni multiple comparisons test. Error bars indicate SEM. *p < 0.05, **p < 0.01.

Close modal

To determine whether extracellular Nox2 and its downstream events play a role in regulating HSPC activation in response to tissue ischemia, we analyzed BM plasma or supernatant in Nox2-KO mice after hindlimb ischemia. Paralleling ROS production reported in our previous study (17), BM plasma collected from nonischemic legs revealed that diffusible H2O2 is increased in the early phase of ischemic injury, and this ischemia-induced H2O2 in BM plasma is lacking in Nox2-KO mice (Fig. 5A). BM plasma blotted on nitrocellulose membrane showed that OxPAPC was increased by hindlimb ischemia in WT mice, but not in Nox2-KO mice, as a form of small particles (Fig. 5B). Importantly, these increases in H2O2 and OxPAPC in BM plasma in WT mice were transient, as both levels return to the homeostatic level or lower at day 6 and day 10 in WT after hindlimb ischemia. As a target of extracellular H2O2, we examined Lyn kinase, an Src family kinase that is activated through oxidation (34). We confirmed in cultured macrophages that extracellular H2O2 is sufficient to increase Lyn activation (Fig. 5C). LPS stimulation in those cells induced Lyn activation, which was blocked by Nox inhibitor (Fig. 5D) and in Nox2-KO cells (Fig. 5E). Furthermore, in situ Lyn activation was detected in BM sections from WT mice subjected to hindlimb ischemia, which were fixed by perfusion immediately after euthanasia, and this response was not induced in the BM of Nox2-KO mice (Fig. 5F).

FIGURE 5.

Hindlimb ischemia increases ROS and oxidized phospholipids and activates Lyn kinase in BM hematopoietic niche in a Nox2-dependent manner. (A) H2O2 levels were measured in BM plasma or supernatant by Amplex Red assay (n = 4 each group and time points). We tested twice for the Pre (no ischemia) versus day 2 comparison using a different set of animals (n = 2–4) and found a similar result. BM plasma was obtained from the nonischemic leg. (B) Oxidized phospholipid (OxPAPC) was identified as microparticles on the nitrocellulose membrane that BM plasma was plotted, and the membrane was labeled by TopFluor-conjugated anti-OxPAPC Ab (images were taken with 20 × objective, n = 9–10 for Pre (no ischemia) for day 2 and n = 3–4 for day 6 and 10. (C) Extracellular H2O2 activates Lyn kinase (p-Lyn at Y396) in vitro (macrophages from BM). (D) Nox inhibitor Diphenyleneiodonium (DPI) blocks Lyn activation induced by 100 ng/ml LPS. (E) Lyn activation by LPS is reduced in Nox2 KO macrophages (n = 3 independent experiments). (F) In situ Lyn activation in the BM after hindlimb ischemia. Activating Lyn phosphorylation (p-Lyn at Y396) shown as green was detected by immunofluorescence of BM sections. Quantification of p-Lyn density in the BM (n = 4 in each group per time point). Images were taken by Zeiss LSM 510 Meta with 20× objective and processed by LSM Browser. Bars indicate 50 μm. (G) LSK cells were sorted from BM of WT and Lyn−/− mice, and 5000 LSK cells were cultured with serum-free StemPro 34 medium supplemented with cytokines shown in Fig. 4A for 72 h under 1 or 10 μg/ml LPS stimulation. Dot plots show CD11b+Ly-6Chi monocytes in samples stimulated with 10 μg/ml LPS. Frequency (%) and quantification of Ly-6Chi monocytes. (H) LincKit+Sca1 myeloid progenitors (MP) were cultured and stimulated with 10 μg/ml LPS for 84 h. (I) LSK cells were sorted from BM of WT and Nox2−/− mice, and 5000 LSK cells were cultured with serum-free StemPro 34 medium supplemented with cytokines shown in Fig. 4A for 72 h under 1 or 10 μg/ml LPS stimulation. A Lyn activator, MLR1023, at 20 μM was added at the beginning of culture. (Triplicated culture in each two independent experiments). Statistical significance was evaluated by two-way ANOVA with Bonferroni multiple comparisons test. Error bars indicate SEM. *p < 0.05, ***p < 0.001, #p < 0.05 between genotypes.

FIGURE 5.

Hindlimb ischemia increases ROS and oxidized phospholipids and activates Lyn kinase in BM hematopoietic niche in a Nox2-dependent manner. (A) H2O2 levels were measured in BM plasma or supernatant by Amplex Red assay (n = 4 each group and time points). We tested twice for the Pre (no ischemia) versus day 2 comparison using a different set of animals (n = 2–4) and found a similar result. BM plasma was obtained from the nonischemic leg. (B) Oxidized phospholipid (OxPAPC) was identified as microparticles on the nitrocellulose membrane that BM plasma was plotted, and the membrane was labeled by TopFluor-conjugated anti-OxPAPC Ab (images were taken with 20 × objective, n = 9–10 for Pre (no ischemia) for day 2 and n = 3–4 for day 6 and 10. (C) Extracellular H2O2 activates Lyn kinase (p-Lyn at Y396) in vitro (macrophages from BM). (D) Nox inhibitor Diphenyleneiodonium (DPI) blocks Lyn activation induced by 100 ng/ml LPS. (E) Lyn activation by LPS is reduced in Nox2 KO macrophages (n = 3 independent experiments). (F) In situ Lyn activation in the BM after hindlimb ischemia. Activating Lyn phosphorylation (p-Lyn at Y396) shown as green was detected by immunofluorescence of BM sections. Quantification of p-Lyn density in the BM (n = 4 in each group per time point). Images were taken by Zeiss LSM 510 Meta with 20× objective and processed by LSM Browser. Bars indicate 50 μm. (G) LSK cells were sorted from BM of WT and Lyn−/− mice, and 5000 LSK cells were cultured with serum-free StemPro 34 medium supplemented with cytokines shown in Fig. 4A for 72 h under 1 or 10 μg/ml LPS stimulation. Dot plots show CD11b+Ly-6Chi monocytes in samples stimulated with 10 μg/ml LPS. Frequency (%) and quantification of Ly-6Chi monocytes. (H) LincKit+Sca1 myeloid progenitors (MP) were cultured and stimulated with 10 μg/ml LPS for 84 h. (I) LSK cells were sorted from BM of WT and Nox2−/− mice, and 5000 LSK cells were cultured with serum-free StemPro 34 medium supplemented with cytokines shown in Fig. 4A for 72 h under 1 or 10 μg/ml LPS stimulation. A Lyn activator, MLR1023, at 20 μM was added at the beginning of culture. (Triplicated culture in each two independent experiments). Statistical significance was evaluated by two-way ANOVA with Bonferroni multiple comparisons test. Error bars indicate SEM. *p < 0.05, ***p < 0.001, #p < 0.05 between genotypes.

Close modal

To examine the role of Lyn activation in HSPCs, we analyzed in vitro monopoiesis from HSPCs using Lyn−/− mice. LSK cells isolated from Lyn−/− mice exhibited enhanced monopoiesis in vitro compared with those isolated WT mice at both low and high doses of LPS (Fig. 5G). Moreover, pharmacologic Lyn activation in Nox2-KO LSKs was sufficient to inhibit their enhanced LPS-stimulated monopoiesis (Fig. 5H). Akt, a downstream target of Lyn that is activated in the BM following hindlimb ischemia in a Nox2-dependent manner (17), likely plays an important role in inhibition of monopoiesis from HSPCs, as HSPCs from Akt1−/− mice also exhibited enhanced monopoiesis in vitro (Supplemental Fig. 3A). In contrast to LSK cells, Lyn-mediated inhibition of monopoiesis was not observed in myeloid progenitors (Fig. 5I). However, Lyn protein expression levels measured by intracellular cell staining did not explain the Nox2- and Lyn-dependent inhibitions of monocytosis specifically observed in LSK cells but not in myeloid progenitors (Supplemental Fig. 3B).

To introduce Nox2 in extracellular space for HSPCs in vivo, we used adoptive transfer of Nox2-intact neutrophils to Nox2-KO mice after hindlimb ischemia; previous studies have shown that i.v.-transferred neutrophils preferentially home to the BM under homeostasis (8, 3537) and following infection (37). We found that DiI-labeled neutrophils transferred on day 1 of hindlimb ischemia homed to the BM within 18 h and that homing ability was not different between Nox2-intact (WT) neutrophils and Nox2-KO neutrophils (Fig. 6A). Adoptive transfer of WT neutrophils but not of Nox2-KO neutrophils is sufficient to rescue Lyn activation after hindlimb ischemia in recipient Nox2-KO mice (Fig. 6B). Under these conditions, transfer of Nox2-intact but not Nox2-KO neutrophils decreased LSKs in the BM of recipient Nox2-KO mice (Fig. 6C) and reduced monocytosis in the BM (Fig. 6D) and in the blood (Fig. 6E) of Nox2-KO mice during hindlimb ischemia.

FIGURE 6.

Nox2 in transferred neutrophils inhibit persistent inflammatory response in Nox2 KO mice following hindlimb ischemia. (A and B) Blood neutrophils (NE) were isolated from WT and Nox2 KO mice, labeled with fluorescent DiI dye, and then adoptively transferred to Nox2 KO mice on day 1 after hindlimb ischemia. After 24 h, mice were euthanized to analyze BM. DiI signals indicate homing of transferred NE in the BM (A). Immunofluorescence and quantification of activating p-Lyn in the BM (n = 4). Images were taken by Zeiss LSM 510 Meta with 20× objective and processed by LSM Browser. Scale bars, 50 μm (B). (C) One million neutrophils were isolated from BM of WT and female Nox2 KO mice and transferred to Nox2 KO mice on day 1 of hindlimb ischemia. Dot plots show LSK cells in the BM on day 3 (n = 5). (D) On day 7, Ly-6Chi monocytes (Mo) in the BM and (E) in the blood were analyzed by flow cytometry (n = 6). (F) One million neutrophils were isolated from blood of WT and Nox2 KO mice and transferred to NF-κB reporter HLL-Nox2y/− mice on day 1 and day 3 of hindlimb ischemia. At indicated time points, NF-κB activity of hindlimb ischemia was measured by bioluminescent imaging (n = 5). (G) In the mice used in (F), histological analysis using H&E staining was used for quantification of regenerating fibers in tibialis anterior muscles on day 7 (n = 5). Scale bars, 100 μm. (H) Schematic illustration of DAMPs injection and subsequent transfer of NE from BM, which were cultured in RPMI 1640 medium for 24 h. BrdU was i.v. injected 4 h prior to the tissue harvest for flow cytometry 24 h after DAMPs injection. (I) LSK cell activity (BrdU+ levels) in femur (n = 4, each) after DAMPs injection with/without NE transfer. Images were taken by Nikon 80i with 20× objective and processed on ImageJ. Scale bars, 100 μm. Statistical significance was evaluated by two-tailed unpaired Student t test (A–C and G), one-way ANOVA with Bonferroni multiple comparisons test (D and E), and two-way ANOVA with Bonferroni multiple comparisons test (F). Error bars indicate SEM. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 6.

Nox2 in transferred neutrophils inhibit persistent inflammatory response in Nox2 KO mice following hindlimb ischemia. (A and B) Blood neutrophils (NE) were isolated from WT and Nox2 KO mice, labeled with fluorescent DiI dye, and then adoptively transferred to Nox2 KO mice on day 1 after hindlimb ischemia. After 24 h, mice were euthanized to analyze BM. DiI signals indicate homing of transferred NE in the BM (A). Immunofluorescence and quantification of activating p-Lyn in the BM (n = 4). Images were taken by Zeiss LSM 510 Meta with 20× objective and processed by LSM Browser. Scale bars, 50 μm (B). (C) One million neutrophils were isolated from BM of WT and female Nox2 KO mice and transferred to Nox2 KO mice on day 1 of hindlimb ischemia. Dot plots show LSK cells in the BM on day 3 (n = 5). (D) On day 7, Ly-6Chi monocytes (Mo) in the BM and (E) in the blood were analyzed by flow cytometry (n = 6). (F) One million neutrophils were isolated from blood of WT and Nox2 KO mice and transferred to NF-κB reporter HLL-Nox2y/− mice on day 1 and day 3 of hindlimb ischemia. At indicated time points, NF-κB activity of hindlimb ischemia was measured by bioluminescent imaging (n = 5). (G) In the mice used in (F), histological analysis using H&E staining was used for quantification of regenerating fibers in tibialis anterior muscles on day 7 (n = 5). Scale bars, 100 μm. (H) Schematic illustration of DAMPs injection and subsequent transfer of NE from BM, which were cultured in RPMI 1640 medium for 24 h. BrdU was i.v. injected 4 h prior to the tissue harvest for flow cytometry 24 h after DAMPs injection. (I) LSK cell activity (BrdU+ levels) in femur (n = 4, each) after DAMPs injection with/without NE transfer. Images were taken by Nikon 80i with 20× objective and processed on ImageJ. Scale bars, 100 μm. Statistical significance was evaluated by two-tailed unpaired Student t test (A–C and G), one-way ANOVA with Bonferroni multiple comparisons test (D and E), and two-way ANOVA with Bonferroni multiple comparisons test (F). Error bars indicate SEM. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

Finally, we generated data to demonstrate the impact of Nox2 in neutrophils on regulating inflammatory responses and tissue regeneration after hindlimb ischemia. Importantly, adoptive transfer of Nox2-intact neutrophils but not by Nox2-KO neutrophils was sufficient to normalize the NF-κB activity in ischemic tissue of Nox2-KO reporter mice by inducing NF-κB in early phase and reducing in later phase (Fig. 6F). Furthermore, this correction of NF-κB activity was associated with improved regeneration of ischemic muscles (Fig. 6G).

To examine further the role of Nox2 in neutrophils in LSK activity in the BM, we injected DAMPs isolated from skeletal muscles in recipient mice, in which we found that DAMPs injection sufficiently induced LSK proliferation in the BM without hindlimb ischemia 24 h after injection (Fig. 6H) but did not induce significant changes in levels of progenitors and monocytes. In this condition, we found that Nox2-intact (WT) neutrophils but not Nox2-KO neutrophils, which were injected just after DAMPs injection, inhibit DAMPs-induced LSK activation (Fig. 6I). These data suggest that Nox2 in neutrophils directly influence the BM stimulated with DAMPs in vivo.

We also examined the underlying mechanism of why WT neutrophils increased NF-κB activity in early phase (Fig. 6F) and found that inflammatory Ly-6Chi monocytes were increased in ischemic tissue at day 3 after WT neutrophil transfer compared with Nox2-KO neutrophil transfer (Supplemental Fig. 4A). This suggests that Nox2 in neutrophils may promote early (day 3) and inhibit later (day 7) monocyte generation/migration.

Because adoptive transfer of neutrophils in Nox2-KO mice resulted in their homing to the BM, we next determined how BM homing of neutrophils is regulated during hindlimb ischemia. To do this, we labeled blood leukocytes in vivo with biotin and analyzed biotin-bound cells in the BM 2 h later (Fig. 7A). Consistent with a previous study (8), we observed neutrophil homing to the BM in uninjured mice (Fig. 7B). Following hindlimb ischemia, neutrophils labeled with biotin were increased in the BM on days 2 and 5 after injury compared to baseline levels (Fig. 7C). Biotin-labeled CD62Llo neutrophils, which have been shown to represent aged neutrophils that preferentially cleared in the BM (8, 38) were significantly increased in the BM on day 2 of hindlimb ischemia, whereas biotin-labeled CD62Lhi neutrophils were not significantly increased (Fig. 7D). Moreover, biotin-labeled BM neutrophils exhibit significantly higher ROS levels compared with nonlabeled neutrophils (Fig. 7E), indicating that neutrophils returning to the BM likely contribute to ROS production in the BM.

FIGURE 7.

Neutrophil homing and residency in the BM are increased after hindlimb ischemia. (A) Blood cells were labeled with biotin in vivo on day 2 or day 5 of hindlimb ischemia. After 2 h, blood and BM were harvested following perfusion with PBS. (B) Dot plots of biotin-labeled Ly-6G+ neutrophils in the BM on day 0 (Pre) and day 2 after hindlimb ischemia. (C) Quantification of biotin-labeled Ly-6G+ neutrophils in the BM (n = 4–5). (D) CD62Llo aged-type neutrophils and CD62Lhi counterparts labeled with biotin were quantified (n = 4–5). (E) ROS levels in biotin-labeled or nonlabeled BM neutrophils were measured by median fluorescent intensity (MFI) of CellROX dye staining. (F) Neutrophil phagocytosis by BM resident macrophages was quantified by intracellular Ly-6G expression in F4/80+CD115int-hi cells. After labeling with Abs for F4/80 and CD115, Ly-6G was further labeled following fixation/permeabilization. Resident macrophages in the BM were defined as F4/80+CD115int-hi cells (upper panels). Neutrophils phagocytosed by BM macrophages were identified by Ly-6G expression in the gate of resident macrophages (lower panels). Quantification of neutrophils phagocytosed by BM macrophages after hindlimb ischemia (n = 4–5). Statistical significance was evaluated by one-way ANOVA with Bonferroni multiple comparisons test (B, D, and F) and two-tailed unpaired Student t test (E). Error bars indicate SEM. *p < 0.05, **p < 0.01.

FIGURE 7.

Neutrophil homing and residency in the BM are increased after hindlimb ischemia. (A) Blood cells were labeled with biotin in vivo on day 2 or day 5 of hindlimb ischemia. After 2 h, blood and BM were harvested following perfusion with PBS. (B) Dot plots of biotin-labeled Ly-6G+ neutrophils in the BM on day 0 (Pre) and day 2 after hindlimb ischemia. (C) Quantification of biotin-labeled Ly-6G+ neutrophils in the BM (n = 4–5). (D) CD62Llo aged-type neutrophils and CD62Lhi counterparts labeled with biotin were quantified (n = 4–5). (E) ROS levels in biotin-labeled or nonlabeled BM neutrophils were measured by median fluorescent intensity (MFI) of CellROX dye staining. (F) Neutrophil phagocytosis by BM resident macrophages was quantified by intracellular Ly-6G expression in F4/80+CD115int-hi cells. After labeling with Abs for F4/80 and CD115, Ly-6G was further labeled following fixation/permeabilization. Resident macrophages in the BM were defined as F4/80+CD115int-hi cells (upper panels). Neutrophils phagocytosed by BM macrophages were identified by Ly-6G expression in the gate of resident macrophages (lower panels). Quantification of neutrophils phagocytosed by BM macrophages after hindlimb ischemia (n = 4–5). Statistical significance was evaluated by one-way ANOVA with Bonferroni multiple comparisons test (B, D, and F) and two-tailed unpaired Student t test (E). Error bars indicate SEM. *p < 0.05, **p < 0.01.

Close modal

Because neutrophils that home to the BM are cleared by resident macrophages (8), we examined macrophage engulfment of neutrophils in the BM after hindlimb ischemia using intracellular staining of the neutrophil Ag Ly-6G (39). BM macrophages defined as F4/80+/CD115int-hi cells that contained engulfed neutrophils were decreased following hindlimb ischemia (Fig. 7F) in parallel with the increased number of newly homed neutrophils (Fig. 7C). Importantly, CD62Lhi neutrophils migrating into the BM are higher in number than CD62Llo neutrophils after hindlimb ischemia. A previous study (7) also showed both CD62Lhi and CD62Llo neutrophils similarly migrating into the BM, indicating that the capacity of migrating into the BM is similar between these neutrophil subsets at least in the steady state. However, CD62Lhi and CD62Llo neutrophils behave differently after migrating into the BM; the same study showed that CD62Llo cells were engulfed 73% more efficiently by macrophages than CD62Lhi neutrophils in BM HSPC niche (7), and our data show that increased residency of BM-homed CD62Llo neutrophils was paralleled with the reduced engulfment by macrophages (Fig. 7F). These indicate that CD62lo neutrophils are more influenced by BM resident macrophages after migrating into the BM. Because of the reduced macrophage engulfment, neutrophils that home to the BM during ischemic injury likely maintain residence in the BM for a longer period compared with the noninjured condition, contributing to the increased ROS production in the BM microenvironment (17).

Our data indicate critical roles of Nox2-derived ROS in limiting HSPC activity, which may be required for timely induction of the resolution of inflammation to promote tissue regeneration following ischemic injury. In response to tissue ischemic injury, Nox2 in the BM contributes to increases in diffusible H2O2 and OxPAPC in BM plasma, which limits monocyte generation by LSK HSPCs induced by damage-associated signals and potentially inflammatory cytokines. This inhibition of LSK cells mediated by Lyn and Akt occurs in the acute phase of inflammation after tissue damage, leading to reduced output of monocytes at later time points, thus contributing to the resolution of inflammation. Our data also indicate a critical role of neutrophils and their homing and residency in the BM microenvironment in the regulation of HSPC activity during inflammation.

Previous studies have shown that primitive HSPCs expand in response to acute inflammation during organ injuries (1, 2, 17, 40), contributing to the systemic increase in inflammatory cells such as neutrophils and inflammatory Ly-6Chi Mo in the acute phase of inflammation (1). This acute inflammatory myeloid response is followed by its resolution during normal wound healing (41). Our results show that failure of the Nox2 regulatory system during inflammation results in persistent inflammation and impaired tissue regeneration following ischemic tissue injury. This protective role of Nox2 in tissue injury response is consistent with previous studies (25, 42), whereas others showed conflicting results—no effect in heart ischemia-reperfusion and worsening role of Nox2 in brain, kidney, and heart ischemia-reperfusions (4345), indicating that Nox2-derived ROS can contribute to both tissue damage and protection. In our Nox2-KO mice, which have basally activated myelopoiesis (inconsistent with the previous study) (Fig. 1D) (13), adoptive transfer of Nox2-intact (WT) neutrophils was sufficient to induce a protective effect of Nox2 in Nox2-KO mice, in which the majority of neutrophils are Nox2 deficient. Therefore, we argue that low amounts of Nox2 in neutrophils play a crucial role in protective, tissue regenerative, and resolution signals during ischemic tissue recovery associated with reduced monopoiesis.

The present data extend our understanding of the critical role of Nox2-derived ROS in regulating HSPCs following tissue ischemia (17). Our data show that diffusible oxidative products H2O2 and OxPAPC are increased in the BM plasma after hindlimb ischemia in a Nox2-dependent manner (Fig. 5) and that these oxidative products inhibit activation and monocyte generation from LSKs (Fig. 4). In mouse acute infection models, it has been shown that Nox2 in myeloid cells increases H2O2 in the BM, which in turn acts on myeloid progenitor cells to drive granulopoiesis (18, 46). Consistent with these studies, we observed impaired granulopoiesis at later time points as determined by BM Ly-6G+ cell levels in our Nox2-deficient model during ischemic tissue recovery (Fig. 1). Thus, Nox2-derived extracellular oxidative signals may drive granulopoiesis by limiting monopoiesis at myeloid-committed progenitor level. In addition, we found population-specific responses to ischemia between LSK cells and myeloid progenitor cells; LSK cells but not myeloid progenitor cells are increased in Nox2-KO after ischemia (Fig. 1B), and myeloid progenitor cells have increased levels of proliferation in WT mice at day 1 postischemia but not in Nox2-KO mice (Fig. 1C), suggesting that Nox2 signals may stimulate myeloid progenitor cells and inhibit LSK cells after ischemia. The adoptive transfer experiment also shows that Nox2 in neutrophils stimulate myeloid progenitor cells in early (Supplemental Fig. 4A) and inhibit LSK cells in later phase of ischemia (Fig. 6C). The possible mechanism behind this may be explained by the following: 1) the fates of hematopoietic progenitors or how they respond to each stimulus may be predetermined (47, 48); 2) timing of generation of monocytes: 3–4 d for LSK and 1–2 d for myeloid progenitor cells, based on observation in in vitro differentiation (6); and 3) difference in signaling: M-CSF (49) and IL-1 (4) have LSK-specific roles, whereas IL-3 and GM-CSF are more specific for myeloid progenitor cells (50). Also, possible influence of cell death, mobilization, or alternative differentiation pathways may explain the inconsistency of response to ischemia between LSK and myeloid progenitors. Our study and previous studies (18, 46) argue that extracellular oxidative signals induce tissue-protective hematopoiesis in BM, likely through regulating HSPCs in cell population– and timing-specific manners.

Our data also show that Nox2, H2O2, and OxPAPC inhibit myeloid differentiation from LSKs but not from myeloid-committed progenitors (Figs. 3, 5) in vitro. Although we did not consider subpopulations of LSKs that play differently during inflammatory hematopoiesis (4, 5, 51) in this study, our findings suggest that oxidative signals in the BM microenvironment have an inhibitory effect on myeloid commitment of undifferentiated HSPCs, at least in vitro. We show that Nox2-inhibited LSK differentiation into monocytes was promoted by catalase, a scavenger for extracellular H2O2 (Fig. 3). Consistent with our data, a previous study reported that catalase increased myeloid differentiation of HSPCs in short-term culture (26). Thus, we propose that extracellular H2O2 acts as an endogenously and exogenously produced inhibitory factor for inflammatory myelopoiesis for LSK cells but likely not for myeloid-committed progenitors. Note that as discussed previously (10), the effects of intracellular and extracellular ROS on HSPC function may differ. Intracellular ROS, typically induced by growth factors such as G-CSF and stem cell factor, are known to positively regulate myeloid cell growth (20, 52). In addition, higher intracellular ROS levels are associated with myeloid skewing of HSPCs (5355). These data on intracellular ROS differ from the findings of the current study, which indicate that extracellular oxidative events specifically limit HSPC monopoiesis. In vitro culture of HSPCs has been used to show cytokine signals and damage associated signals that drive the myeloid commitments (46, 49), and a model of endogenously activated inhibitory signals during acute hematopoietic proliferation is proposed (7). Further investigation is needed to determine the exact role of extracellular ROS in the hematopoietic niche in vivo, although our neutrophil adoptive transfer experiment (Fig. 7) supports our claim.

Because of its diffusible nature, H2O2 is a candidate messenger that could travel through the extracellular space and activate oxidative signals in the HSPC microenvironment. However, H2O2 can pass through cellular membrane (56) and promote intracellular ROS production (57). This may explain why addition of H2O2 in cultured LSKs failed to reproduce the strong myeloid suppression of LSKs that we observed in coculture experiments (Figs. 3, 4) and suggests that an oxidation-inducible cell-derived factor may be increased in the BM in addition to H2O2. Oxidized phospholipids, such as OxPAPC, inhibit TLR4 activation by binding CD14 (58, 59), and surface expression of CD14 level is significantly higher in LSK cells than in myeloid progenitors (6). Thus, the diverging responses to the oxidative signals between LSKs and myeloid progenitors may be due to the difference in CD14 levels. We indeed demonstrate that in BM plasma OxPAPC levels were correlated with H2O2 levels [increased 2 d after ischemic injury in WT but not in Nox2-KO mice (Fig. 5)]. Although further studies are needed to determine how OxPAPC is increased in the BM plasma in response to tissue injury and whether OxPAPC is derived from neutrophils that return to the BM, we provide evidence that Nox2-dependent H2O2 in the BM microenvironment promotes OxPAPC accumulation, which limits TLR-dependent monopoiesis from LSK cells.

As intracellular sensors of LSK cells that respond to extracellular oxidative signals, we present evidence that Lyn and Akt mediate the inhibitory signals of myeloid differentiation. Lyn kinase has been shown as an oxidation-sensitive kinase that senses extracellular H2O2 derived from Nox (34). OxPAPC can also activate Lyn kinase (60). In our in vivo system, Lyn activation is Nox2 dependent (Fig. 5F), and adoptive transfer of Nox2-intact neutrophils is sufficient to induce Lyn activation in Nox2-KO mice (Fig. 6B). Lyn interacts with TLR4 and other proteins in TLR4-mediated signaling and can regulate downstream events positively or negatively in a context-dependent manner (61). In addition, we previously showed Akt activation in the BM after hindlimb ischemia is also Nox2 dependent (17). Akt kinase has been shown to mediate the inhibitory action of Lyn in TLR4 signaling in murine macrophages (62) and dendritic cells (63). Indeed, our data demonstrate that Lyn protein is expressed on HSPCs and their myeloid progeny (Supplemental Fig. 3B) and that lack of Akt1 inhibits monopoiesis from LSKs (Supplemental Fig. 3A). Thus, Lyn-Akt activation appears to act as a negative regulator for TLR4 signaling in this context. Lyn activation also negatively regulates thrombopoietin-induced proliferation in megakaryocyte progenitors (64) and negatively regulates erythroid progenitor proliferation (65), whereas G-CSF increases myeloid growth via Lyn activation (20). Thus, Lyn may be a master regulator of hematopoiesis in the BM.

We demonstrate that neutrophil homing to the BM may regulate the injury-induced oxidative signals in BM. Previous studies reported that neutrophil homing to the BM modulates the inflammatory response during different inflammatory states; in hyperglycemia, neutrophils in the BM stimulate myeloid progenitors to promote myelopoiesis (27, 66), suggesting neutrophils promote inflammatory response in the BM. During skin infection, neutrophils homing to the BM interact with T cells to prime their memory function (37). Thus, neutrophil homing to the BM may play an important role in transmitting signals from periphery to the BM that modulate inflammatory responses. Together with our data showing neutrophil homing transmits regulatory signals in the BM, HSPCs potentially sense inflammatory or resolution signals or both from neutrophils. Further studies are needed to elucidate those signals in the BM. In addition, neutrophils in BM can influence hematopoietic niche through acting on other niche cells; macrophage uptake of aged neutrophils induces a release of hematopoietic cytokines and thus regulates hematopoiesis of HSPCs (8, 67, 68). Our result indicates this clearance of aged-type neutrophils may be suppressed in the acute phase of ischemic tissue repair (Fig. 7). Therefore, macrophage function can influence immunomodulation by neutrophil homing. However, given that CD62Lhi neutrophils consist of ∼80% blood neutrophils, it is still possible that some CD62Lhi neutrophils migrating into the BM also contribute to the Nox2-dependent signals in the BM. These complex relationships between CD62Lhi and CD62Llo migrating into the BM remain to be further determined. Finally, myeloid populations homing back to the BM during ischemic injury are not exclusively neutrophils, as macrophages derived from tissue were also found in the BM after myocardial infarction (69). We also found that circulating monocytes, but not lymphocytes, homed back to the BM after hindlimb ischemia (Supplemental Fig. 4B). We plan to elucidate whether recirculated neutrophils and monocytes differently modulate immune responses in BM in future studies.

Our findings not only provide insight into the regulation of HSPC during inflammation, they also have implications for a number of different pathological conditions that involve chronic inflammation in which resolution of inflammation is impaired. Both increased myelopoiesis and nonresolving monocyte/macrophage responses are associated with conditions such as obesity, diabetes, atherosclerosis, and autoimmune diseases. Improved understanding of the mechanisms of cell-intrinsic and cell-extrinsic regulation of HSPCs in chronic inflammatory conditions may lead to novel therapeutic approaches for each of these conditions. For example, targeting oxidant signaling or neutrophils may be an attractive strategy to modulate HSPC function in BM and its downstream events.

We thank the Flow Cytometry Service and the Core Imaging Facility of the Research Resource Center at the University of Illinois at Chicago for technical assistance. We also thank Dr. Jaehyung Cho for helpful discussion and advice and Dr. Ben Gantner for reagents and advice.

This work was supported by grants from the American Heart Association (12SDG12060100 to N.U.), the National Institutes of Health (R01DK111489 to N.U., R01HL062350 and R01HL080264 to X.D., R01HL075557 to J.W.C., R01HL070187 and R01HL116976 to T.F., R01GM092850 to T.J.K., and R01HL077524, R01HL116976, and R21HL112293 to M.U.-F.), and by Department of Veterans Affairs Merit Review Grant 2I01BX001232 (to T.F.).

The online version of this article contains supplemental material.

Abbreviations used in this article:

BM

bone marrow

DAMP

damage-associated molecular pattern

HSPC

hematopoietic stem and progenitor cell

KO

knockout

LSK

lineageSca-1+c-Kit+

Nox

NADPH oxidase

OxPAPC

oxidized 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphorylcholine

ROI

region of interest

ROS

reactive oxygen species

WT

wild-type.

1
Dutta
,
P.
,
H. B.
Sager
,
K. R.
Stengel
,
K.
Naxerova
,
G.
Courties
,
B.
Saez
,
L.
Silberstein
,
T.
Heidt
,
M.
Sebas
,
Y.
Sun
, et al
.
2015
.
Myocardial infarction activates CCR2(+) hematopoietic stem and progenitor cells.
Cell Stem Cell
16
:
477
487
.
2
Rodgers
,
J. T.
,
K. Y.
King
,
J. O.
Brett
,
M. J.
Cromie
,
G. W.
Charville
,
K. K.
Maguire
,
C.
Brunson
,
N.
Mastey
,
L.
Liu
,
C. R.
Tsai
, et al
.
2014
.
mTORC1 controls the adaptive transition of quiescent stem cells from G0 to G(Alert).
Nature
510
:
393
396
.
3
Baldridge
,
M. T.
,
K. Y.
King
,
N. C.
Boles
,
D. C.
Weksberg
,
M. A.
Goodell
.
2010
.
Quiescent haematopoietic stem cells are activated by IFN-gamma in response to chronic infection.
Nature
465
:
793
797
.
4
Pietras
,
E. M.
,
C.
Mirantes-Barbeito
,
S.
Fong
,
D.
Loeffler
,
L. V.
Kovtonyuk
,
S.
Zhang
,
R.
Lakshminarasimhan
,
C. P.
Chin
,
J. M.
Techner
,
B.
Will
, et al
.
2016
.
Chronic interleukin-1 exposure drives haematopoietic stem cells towards precocious myeloid differentiation at the expense of self-renewal.
Nat. Cell Biol.
18
:
607
618
.
5
Zhao
,
J. L.
,
C.
Ma
,
R. M.
O’Connell
,
A.
Mehta
,
R.
DiLoreto
,
J. R.
Heath
,
D.
Baltimore
.
2014
.
Conversion of danger signals into cytokine signals by hematopoietic stem and progenitor cells for regulation of stress-induced hematopoiesis.
Cell Stem Cell
14
:
445
459
.
6
Nagai
,
Y.
,
K. P.
Garrett
,
S.
Ohta
,
U.
Bahrun
,
T.
Kouro
,
S.
Akira
,
K.
Takatsu
,
P. W.
Kincade
.
2006
.
Toll-like receptors on hematopoietic progenitor cells stimulate innate immune system replenishment.
Immunity
24
:
801
812
.
7
Csaszar
,
E.
,
D. C.
Kirouac
,
M.
Yu
,
W.
Wang
,
W.
Qiao
,
M. P.
Cooke
,
A. E.
Boitano
,
C.
Ito
,
P. W.
Zandstra
.
2012
.
Rapid expansion of human hematopoietic stem cells by automated control of inhibitory feedback signaling.
Cell Stem Cell
10
:
218
229
.
8
Casanova-Acebes
,
M.
,
C.
Pitaval
,
L. A.
Weiss
,
C.
Nombela-Arrieta
,
R.
Chèvre
,
N.
A-González
,
Y.
Kunisaki
,
D.
Zhang
,
N.
van Rooijen
,
L. E.
Silberstein
, et al
.
2013
.
Rhythmic modulation of the hematopoietic niche through neutrophil clearance.
Cell
153
:
1025
1035
.
9
West
,
A. P.
,
I. E.
Brodsky
,
C.
Rahner
,
D. K.
Woo
,
H.
Erdjument-Bromage
,
P.
Tempst
,
M. C.
Walsh
,
Y.
Choi
,
G. S.
Shadel
,
S.
Ghosh
.
2011
.
TLR signalling augments macrophage bactericidal activity through mitochondrial ROS.
Nature
472
:
476
480
.
10
Urao
,
N.
,
M.
Ushio-Fukai
.
2013
.
Redox regulation of stem/progenitor cells and bone marrow niche.
Free Radic. Biol. Med.
54
:
26
39
.
11
Weinberg
,
S. E.
,
L. A.
Sena
,
N. S.
Chandel
.
2015
.
Mitochondria in the regulation of innate and adaptive immunity.
Immunity
42
:
406
417
.
12
Holmström
,
K. M.
,
T.
Finkel
.
2014
.
Cellular mechanisms and physiological consequences of redox-dependent signalling.
Nat. Rev. Mol. Cell Biol.
15
:
411
421
.
13
Lee
,
K.
,
H. Y.
Won
,
M. A.
Bae
,
J. H.
Hong
,
E. S.
Hwang
.
2011
.
Spontaneous and aging-dependent development of arthritis in NADPH oxidase 2 deficiency through altered differentiation of CD11b+ and Th/Treg cells.
Proc. Natl. Acad. Sci. USA
108
:
9548
9553
.
14
Campbell
,
A. M.
,
M.
Kashgarian
,
M. J.
Shlomchik
.
2012
.
NADPH oxidase inhibits the pathogenesis of systemic lupus erythematosus.
Sci. Transl. Med.
4
:
157ra141
.
15
Campbell
,
E. L.
,
W. J.
Bruyninckx
,
C. J.
Kelly
,
L. E.
Glover
,
E. N.
McNamee
,
B. E.
Bowers
,
A. J.
Bayless
,
M.
Scully
,
B. J.
Saeedi
,
L.
Golden-Mason
, et al
.
2014
.
Transmigrating neutrophils shape the mucosal microenvironment through localized oxygen depletion to influence resolution of inflammation.
Immunity
40
:
66
77
.
16
Yang
,
Z.
,
Y.
Shen
,
H.
Oishi
,
E. L.
Matteson
,
L.
Tian
,
J. J.
Goronzy
,
C. M.
Weyand
.
2016
.
Restoring oxidant signaling suppresses proarthritogenic T cell effector functions in rheumatoid arthritis.
Sci. Transl. Med.
8
:
331ra38
.
17
Urao
,
N.
,
R. D.
McKinney
,
T.
Fukai
,
M.
Ushio-Fukai
.
2012
.
NADPH oxidase 2 regulates bone marrow microenvironment following hindlimb ischemia: role in reparative mobilization of progenitor cells.
Stem Cells
30
:
923
934
.
18
Kwak
,
H. J.
,
P.
Liu
,
B.
Bajrami
,
Y.
Xu
,
S. Y.
Park
,
C.
Nombela-Arrieta
,
S.
Mondal
,
Y.
Sun
,
H.
Zhu
,
L.
Chai
, et al
.
2015
.
Myeloid cell-derived reactive oxygen species externally regulate the proliferation of myeloid progenitors in emergency granulopoiesis.
Immunity
42
:
159
171
.
19
Tauzin
,
S.
,
T. W.
Starnes
,
F. B.
Becker
,
P. Y.
Lam
,
A.
Huttenlocher
.
2014
.
Redox and Src family kinase signaling control leukocyte wound attraction and neutrophil reverse migration.
J. Cell Biol.
207
:
589
598
.
20
Zhu
,
Q. S.
,
L.
Xia
,
G. B.
Mills
,
C. A.
Lowell
,
I. P.
Touw
,
S. J.
Corey
.
2006
.
G-CSF induced reactive oxygen species involves Lyn-PI3-kinase-Akt and contributes to myeloid cell growth.
Blood
107
:
1847
1856
.
21
Chan
,
V. W.
,
F.
Meng
,
P.
Soriano
,
A. L.
DeFranco
,
C. A.
Lowell
.
1997
.
Characterization of the B lymphocyte populations in Lyn-deficient mice and the role of Lyn in signal initiation and down-regulation.
Immunity
7
:
69
81
.
22
Yin
,
H.
,
J.
Liu
,
Z.
Li
,
M. C.
Berndt
,
C. A.
Lowell
,
X.
Du
.
2008
.
Src family tyrosine kinase Lyn mediates VWF/GPIb-IX-induced platelet activation via the cGMP signaling pathway.
Blood
112
:
1139
1146
.
23
Chen
,
W. S.
,
P. Z.
Xu
,
K.
Gottlob
,
M. L.
Chen
,
K.
Sokol
,
T.
Shiyanova
,
I.
Roninson
,
W.
Weng
,
R.
Suzuki
,
K.
Tobe
, et al
.
2001
.
Growth retardation and increased apoptosis in mice with homozygous disruption of the Akt1 gene.
Genes Dev.
15
:
2203
2208
.
24
Blackwell
,
T. S.
,
F. E.
Yull
,
C. L.
Chen
,
A.
Venkatakrishnan
,
T. R.
Blackwell
,
D. J.
Hicks
,
L. H.
Lancaster
,
J. W.
Christman
,
L. D.
Kerr
.
2000
.
Multiorgan nuclear factor kappa B activation in a transgenic mouse model of systemic inflammation.
Am. J. Respir. Crit. Care Med.
162
:
1095
1101
.
25
Urao
,
N.
,
H.
Inomata
,
M.
Razvi
,
H. W.
Kim
,
K.
Wary
,
R.
McKinney
,
T.
Fukai
,
M.
Ushio-Fukai
.
2008
.
Role of nox2-based NADPH oxidase in bone marrow and progenitor cell function involved in neovascularization induced by hindlimb ischemia.
Circ. Res.
103
:
212
220
.
26
Gupta
,
R.
,
S.
Karpatkin
,
R. S.
Basch
.
2006
.
Hematopoiesis and stem cell renewal in long-term bone marrow cultures containing catalase.
Blood
107
:
1837
1846
.
27
Denny
,
M. F.
,
S.
Yalavarthi
,
W.
Zhao
,
S. G.
Thacker
,
M.
Anderson
,
A. R.
Sandy
,
W. J.
McCune
,
M. J.
Kaplan
.
2010
.
A distinct subset of proinflammatory neutrophils isolated from patients with systemic lupus erythematosus induces vascular damage and synthesizes type I IFNs.
J. Immunol.
184
:
3284
3297
.
28
Swamydas
,
M.
,
M. S.
Lionakis
.
2013
.
Isolation, purification and labeling of mouse bone marrow neutrophils for functional studies and adoptive transfer experiments.
J. Vis. Exp.
77
:
e50586
. Available at: https://www.jove.com/video/50586/isolation-purification-labeling-mouse-bone-marrow-neutrophils-for.
29
Baldan
,
A.
,
A.
Gonen
,
C.
Choung
,
X.
Que
,
T. J.
Marquart
,
I.
Hernandez
,
I.
Bjorkhem
,
D. A.
Ford
,
J. L.
Witztum
,
E. J.
Tarling
.
2014
.
ABCG1 is required for pulmonary B-1 B cell and natural antibody homeostasis.
J. Immunol.
193
:
5637
5648
.
30
Kawamoto
,
T.
2003
.
Use of a new adhesive film for the preparation of multi-purpose fresh-frozen sections from hard tissues, whole-animals, insects and plants.
Arch. Histol. Cytol.
66
:
123
143
.
31
Bryer
,
S. C.
,
G.
Fantuzzi
,
N.
Van Rooijen
,
T. J.
Koh
.
2008
.
Urokinase-type plasminogen activator plays essential roles in macrophage chemotaxis and skeletal muscle regeneration.
J. Immunol.
180
:
1179
1188
.
32
Leuschner
,
F.
,
P. J.
Rauch
,
T.
Ueno
,
R.
Gorbatov
,
B.
Marinelli
,
W. W.
Lee
,
P.
Dutta
,
Y.
Wei
,
C.
Robbins
,
Y.
Iwamoto
, et al
.
2012
.
Rapid monocyte kinetics in acute myocardial infarction are sustained by extramedullary monocytopoiesis.
J. Exp. Med.
209
:
123
137
.
33
Huber
,
J.
,
A.
Vales
,
G.
Mitulovic
,
M.
Blumer
,
R.
Schmid
,
J. L.
Witztum
,
B. R.
Binder
,
N.
Leitinger
.
2002
.
Oxidized membrane vesicles and blebs from apoptotic cells contain biologically active oxidized phospholipids that induce monocyte-endothelial interactions.
Arterioscler. Thromb. Vasc. Biol.
22
:
101
107
.
34
Yoo
,
S. K.
,
T. W.
Starnes
,
Q.
Deng
,
A.
Huttenlocher
.
2011
.
Lyn is a redox sensor that mediates leukocyte wound attraction in vivo.
Nature
480
:
109
112
.
35
Suratt
,
B. T.
,
S. K.
Young
,
J.
Lieber
,
J. A.
Nick
,
P. M.
Henson
,
G. S.
Worthen
.
2001
.
Neutrophil maturation and activation determine anatomic site of clearance from circulation.
Am. J. Physiol. Lung Cell. Mol. Physiol.
281
:
L913
L921
.
36
Furze
,
R. C.
,
S. M.
Rankin
.
2008
.
The role of the bone marrow in neutrophil clearance under homeostatic conditions in the mouse.
FASEB J.
22
:
3111
3119
.
37
Duffy
,
D.
,
H.
Perrin
,
V.
Abadie
,
N.
Benhabiles
,
A.
Boissonnas
,
C.
Liard
,
B.
Descours
,
D.
Reboulleau
,
O.
Bonduelle
,
B.
Verrier
, et al
.
2012
.
Neutrophils transport antigen from the dermis to the bone marrow, initiating a source of memory CD8+ T cells.
Immunity
37
:
917
929
.
38
Martin
,
C.
,
P. C.
Burdon
,
G.
Bridger
,
J. C.
Gutierrez-Ramos
,
T. J.
Williams
,
S. M.
Rankin
.
2003
.
Chemokines acting via CXCR2 and CXCR4 control the release of neutrophils from the bone marrow and their return following senescence.
Immunity
19
:
583
593
.
39
Dalli
,
J.
,
C. P.
Jones
,
D. M.
Cavalcanti
,
S. H.
Farsky
,
M.
Perretti
,
S. M.
Rankin
.
2012
.
Annexin A1 regulates neutrophil clearance by macrophages in the mouse bone marrow.
FASEB J.
26
:
387
396
.
40
Courties
,
G.
,
F.
Herisson
,
H. B.
Sager
,
T.
Heidt
,
Y.
Ye
,
Y.
Wei
,
Y.
Sun
,
N.
Severe
,
P.
Dutta
,
J.
Scharff
, et al
.
2015
.
Ischemic stroke activates hematopoietic bone marrow stem cells.
Circ. Res.
116
:
407
417
.
41
Novak
,
M. L.
,
T. J.
Koh
.
2013
.
Macrophage phenotypes during tissue repair.
J. Leukoc. Biol.
93
:
875
881
.
42
Matsushima
,
S.
,
J.
Kuroda
,
T.
Ago
,
P.
Zhai
,
Y.
Ikeda
,
S.
Oka
,
G. H.
Fong
,
R.
Tian
,
J.
Sadoshima
.
2013
.
Broad suppression of NADPH oxidase activity exacerbates ischemia/reperfusion injury through inadvertent downregulation of hypoxia-inducible factor-1α and upregulation of peroxisome proliferator-activated receptor-α.
Circ. Res.
112
:
1135
1149
.
43
Karim
,
A. S.
,
S. R.
Reese
,
N. A.
Wilson
,
L. M.
Jacobson
,
W.
Zhong
,
A.
Djamali
.
2015
.
Nox2 is a mediator of ischemia reperfusion injury.
Am. J. Transplant.
15
:
2888
2899
.
44
Chen
,
H.
,
Y. S.
Song
,
P. H.
Chan
.
2009
.
Inhibition of NADPH oxidase is neuroprotective after ischemia-reperfusion.
J. Cereb. Blood Flow Metab.
29
:
1262
1272
.
45
Braunersreuther
,
V.
,
F.
Montecucco
,
M.
Asrih
,
G.
Pelli
,
K.
Galan
,
M.
Frias
,
F.
Burger
,
A. L.
Quinderé
,
C.
Montessuit
,
K. H.
Krause
, et al
.
2013
.
Role of NADPH oxidase isoforms NOX1, NOX2 and NOX4 in myocardial ischemia/reperfusion injury. [Published erratum appears in 2014 J. Mol. Cell. Cardiol. 66: 189.]
J. Mol. Cell. Cardiol.
64
:
99
107
.
46
Zhu
,
H.
,
H. J.
Kwak
,
P.
Liu
,
B.
Bajrami
,
Y.
Xu
,
S. Y.
Park
,
C.
Nombela-Arrieta
,
S.
Mondal
,
H.
Kambara
,
H.
Yu
, et al
.
2017
.
Reactive oxygen species-producing myeloid cells act as a bone marrow niche for sterile inflammation-induced reactive granulopoiesis.
J. Immunol.
198
:
2854
2864
.
47
Yanez
,
A.
,
S. G.
Coetzee
,
A.
Olsson
,
D. E.
Muench
,
B. P.
Berman
,
D. J.
Hazelett
,
N.
Salomonis
,
H. L.
Grimes
,
H. S.
Goodridge
.
2017
.
Granulocyte-monocyte progenitors and monocyte-dendritic cell progenitors independently produce functionally distinct monocytes.
Immunity
47
:
890
902.e4
.
48
Paul
,
F.
,
Y.
Arkin
,
A.
Giladi
,
D. A.
Jaitin
,
E.
Kenigsberg
,
H.
Keren-Shaul
,
D.
Winter
,
D.
Lara-Astiaso
,
M.
Gury
,
A.
Weiner
, et al
.
2015
.
Transcriptional heterogeneity and lineage commitment in myeloid progenitors. [Published erratum appears in 2016 Cell 164: 325.]
Cell
163
:
1663
1677
.
49
Mossadegh-Keller
,
N.
,
S.
Sarrazin
,
P. K.
Kandalla
,
L.
Espinosa
,
E. R.
Stanley
,
S. L.
Nutt
,
J.
Moore
,
M. H.
Sieweke
.
2013
.
M-CSF instructs myeloid lineage fate in single haematopoietic stem cells.
Nature
497
:
239
243
.
50
Yvan-Charvet
,
L.
,
T.
Pagler
,
E. L.
Gautier
,
S.
Avagyan
,
R. L.
Siry
,
S.
Han
,
C. L.
Welch
,
N.
Wang
,
G. J.
Randolph
,
H. W.
Snoeck
,
A. R.
Tall
.
2010
.
ATP-binding cassette transporters and HDL suppress hematopoietic stem cell proliferation.
Science
328
:
1689
1693
.
51
Pietras
,
E. M.
,
D.
Reynaud
,
Y. A.
Kang
,
D.
Carlin
,
F. J.
Calero-Nieto
,
A. D.
Leavitt
,
J. M.
Stuart
,
B.
Göttgens
,
E.
Passegué
.
2015
.
Functionally distinct subsets of lineage-biased multipotent progenitors control blood production in normal and regenerative conditions.
Cell Stem Cell
17
:
35
46
.
52
O’Laughlin-Bunner
,
B.
,
N.
Radosevic
,
M. L.
Taylor
,
C.
Shivakrupa
,
C.
DeBerry
,
D. D.
Metcalfe
,
M.
Zhou
,
C.
Lowell
,
D.
Linnekin
.
2001
.
Lyn is required for normal stem cell factor-induced proliferation and chemotaxis of primary hematopoietic cells.
Blood
98
:
343
350
.
53
Jang
,
Y. Y.
,
S. J.
Sharkis
.
2007
.
A low level of reactive oxygen species selects for primitive hematopoietic stem cells that may reside in the low-oxygenic niche.
Blood
110
:
3056
3063
.
54
Tothova
,
Z.
,
R.
Kollipara
,
B. J.
Huntly
,
B. H.
Lee
,
D. H.
Castrillon
,
D. E.
Cullen
,
E. P.
McDowell
,
S.
Lazo-Kallanian
,
I. R.
Williams
,
C.
Sears
, et al
.
2007
.
FoxOs are critical mediators of hematopoietic stem cell resistance to physiologic oxidative stress.
Cell
128
:
325
339
.
55
Owusu-Ansah
,
E.
,
U.
Banerjee
.
2009
.
Reactive oxygen species prime Drosophila haematopoietic progenitors for differentiation.
Nature
461
:
537
541
.
56
Bienert
,
G. P.
,
J. K.
Schjoerring
,
T. P.
Jahn
.
2006
.
Membrane transport of hydrogen peroxide.
Biochim. Biophys. Acta
1758
:
994
1003
.
57
Zorov
,
D. B.
,
M.
Juhaszova
,
S. J.
Sollott
.
2014
.
Mitochondrial reactive oxygen species (ROS) and ROS-induced ROS release.
Physiol. Rev.
94
:
909
950
.
58
Erridge
,
C.
,
S.
Kennedy
,
C. M.
Spickett
,
D. J.
Webb
.
2008
.
Oxidized phospholipid inhibition of toll-like receptor (TLR) signaling is restricted to TLR2 and TLR4: roles for CD14, LPS-binding protein, and MD2 as targets for specificity of inhibition.
J. Biol. Chem.
283
:
24748
24759
.
59
Bochkov
,
V. N.
,
A.
Kadl
,
J.
Huber
,
F.
Gruber
,
B. R.
Binder
,
N.
Leitinger
.
2002
.
Protective role of phospholipid oxidation products in endotoxin-induced tissue damage.
Nature
419
:
77
81
.
60
Rahaman
,
S. O.
,
D. J.
Lennon
,
M.
Febbraio
,
E. A.
Podrez
,
S. L.
Hazen
,
R. L.
Silverstein
.
2006
.
A CD36-dependent signaling cascade is necessary for macrophage foam cell formation.
Cell Metab.
4
:
211
221
.
61
Xu
,
Y.
,
K. W.
Harder
,
N. D.
Huntington
,
M. L.
Hibbs
,
D. M.
Tarlinton
.
2005
.
Lyn tyrosine kinase: accentuating the positive and the negative.
Immunity
22
:
9
18
.
62
Keck
,
S.
,
M.
Freudenberg
,
M.
Huber
.
2010
.
Activation of murine macrophages via TLR2 and TLR4 is negatively regulated by a Lyn/PI3K module and promoted by SHIP1.
J. Immunol.
184
:
5809
5818
.
63
Brown
,
J.
,
H.
Wang
,
J.
Suttles
,
D. T.
Graves
,
M.
Martin
.
2011
.
Mammalian target of rapamycin complex 2 (mTORC2) negatively regulates Toll-like receptor 4-mediated inflammatory response via FoxO1.
J. Biol. Chem.
286
:
44295
44305
.
64
Murphy
,
A. J.
,
N.
Bijl
,
L.
Yvan-Charvet
,
C. B.
Welch
,
N.
Bhagwat
,
A.
Reheman
,
Y.
Wang
,
J. A.
Shaw
,
R. L.
Levine
,
H.
Ni
, et al
.
2013
.
Cholesterol efflux in megakaryocyte progenitors suppresses platelet production and thrombocytosis.
Nat. Med.
19
:
586
594
.
65
Kosmider
,
O.
,
D.
Buet
,
I.
Gallais
,
N.
Denis
,
F.
Moreau-Gachelin
.
2009
.
Erythropoietin down-regulates stem cell factor receptor (Kit) expression in the leukemic proerythroblast: role of Lyn kinase.
PLoS One
4
:
e5721
.
66
Nagareddy
,
P. R.
,
A. J.
Murphy
,
R. A.
Stirzaker
,
Y.
Hu
,
S.
Yu
,
R. G.
Miller
,
B.
Ramkhelawon
,
E.
Distel
,
M.
Westerterp
,
L. S.
Huang
, et al
.
2013
.
Hyperglycemia promotes myelopoiesis and impairs the resolution of atherosclerosis.
Cell Metab.
17
:
695
708
.
67
Stark
,
M. A.
,
Y.
Huo
,
T. L.
Burcin
,
M. A.
Morris
,
T. S.
Olson
,
K.
Ley
.
2005
.
Phagocytosis of apoptotic neutrophils regulates granulopoiesis via IL-23 and IL-17.
Immunity
22
:
285
294
.
68
Hong
,
C.
,
Y.
Kidani
,
N.
A-Gonzalez
,
T.
Phung
,
A.
Ito
,
X.
Rong
,
K.
Ericson
,
H.
Mikkola
,
S. W.
Beaven
,
L. S.
Miller
, et al
.
2012
.
Coordinate regulation of neutrophil homeostasis by liver X receptors in mice.
J. Clin. Invest.
122
:
337
347
.
69
Heidt
,
T.
,
G.
Courties
,
P.
Dutta
,
H. B.
Sager
,
M.
Sebas
,
Y.
Iwamoto
,
Y.
Sun
,
N.
Da Silva
,
P.
Panizzi
,
A. M.
van der Laan
, et al
.
2014
.
Differential contribution of monocytes to heart macrophages in steady-state and after myocardial infarction. [Published erratum appears in 2014 Circ. Res. 115: e95.]
Circ. Res.
115
:
284
295
.

The authors have no financial conflicts of interest.

Supplementary data