Skip to main content
NCBI home page
The EMBO Journal logoLink to The EMBO Journal
. 2017 Jul 3;36(16):2390–2403. doi: 10.15252/embj.201796771

Integrin αvβ3 enhances the suppressive effect of interferon‐γ on hematopoietic stem cells

Terumasa Umemoto 1,✉, Yu Matsuzaki 1, Yoshiko Shiratsuchi 2, Michihiro Hashimoto 1, Takayuki Yoshimoto 3, Ayako Nakamura‐Ishizu 4, Brian Petrich 5, Masayuki Yamato 2, Toshio Suda 1,4,✉
PMCID: PMC5556268  PMID: 28673932

Abstract

Hematopoietic homeostasis depends on the maintenance of hematopoietic stem cells (HSCs), which are regulated within a specialized bone marrow (BM) niche. When HSC sense external stimuli, their adhesion status may be critical for determining HSC cell fate. The cell surface molecule, integrin αvβ3, is activated through HSC adhesion to extracellular matrix and niche cells. Integrin β3 signaling maintains HSCs within the niche. Here, we showed the synergistic negative regulation of the pro‐inflammatory cytokine interferon‐γ (IFNγ) and β3 integrin signaling in murine HSC function by a novel definitive phenotyping of HSCs. Integrin αvβ3 suppressed HSC function in the presence of IFNγ and impaired integrin β3 signaling mitigated IFNγ‐dependent negative action on HSCs. During IFNγ stimulation, integrin β3 signaling enhanced STAT1‐mediated gene expression via serine phosphorylation. These findings show that integrin β3 signaling intensifies the suppressive effect of IFNγ on HSCs, which indicates that cell adhesion via integrin αvβ3 within the BM niche acts as a context‐dependent signal modulator to regulate the HSC function under both steady‐state and inflammatory conditions.

Keywords: hematopoietic stem cells, IFNγ, integrin αvβ3, STAT1

Subject Categories: Immunology, Signal Transduction, Stem Cells

Introduction

Hematopoietic stem cells (HSCs) maintain homeostasis within the hematopoietic system through their self‐renewal and multilineage differentiation. Adult HSCs reside within the bone marrow (BM) and are regulated by a specialized microenvironment, or “niche”. During steady‐state hematopoiesis, the BM niche maintains HSCs in a quiescent state through cytokine signaling, cell‐extracellular matrix (ECM) adhesion, and cell–cell contacts (Arai et al, 2004; Nilsson et al, 2005; Qian et al, 2007; Kovtonyuk et al, 2016).

Integrin receptors function as adhesion molecules and modulate growth factor and cytokine signaling cascades to regulate cell survival and proliferation (Hynes, 2002; Juliano et al, 2004). We previously highlighted that mouse HSCs express integrin αvβ3 (CD51/CD61) higher than progenitor cells (Umemoto et al, 2006). Integrin β3 signaling contributes to the maintenance of HSCs by stimulating long‐term repopulating (LTR) activity, collaborating with thrombopoietin (TPO) (Umemoto et al, 2012), a crucial cytokine for HSC maintenance, quiescence (Qian et al, 2007; Yoshihara et al, 2007), and proliferation (Kimura et al, 1998; Fox et al, 2002; Kovtonyuk et al, 2016). By contrast, in the presence of interferon‐γ (IFNγ), integrin β3 signaling diminished LTR activity of HSCs even in the presence of TPO (Ishihara et al, 2014). Taken together, these reports indicate that integrin αvβ3 exerts a dual effect on HSCs that depend on the context of the surrounding cytokine environment.

The pro‐inflammatory cytokine IFN is produced by T, natural killer, and natural killer T cells in response to intracellular pathogens (e.g., mycobacteria and viruses) and contributes to the regulation of immunologic responses. Importantly, excess IFN is known to impair the LTR activity of HSCs (Essers et al, 2009; Baldridge et al, 2010; de Bruin et al, 2013; Nakagawa et al, 2015), possibly through the activation of STAT1, a core signaling molecule downstream of IFNs (Darnell et al, 1994). Increased IFNγ signaling or persistent IFNγ expression is known to lead to aplastic anemia (AA) (Anderson, 2008; Lin et al, 2014; Nakagawa et al, 2015). Furthermore, patients with AA exhibit enhanced IFNγ levels in blood serum as well as BM serum (Zoumbos et al, 1985), produced possibly by circulating or bone marrow T cells (Sloand et al, 2002). Thus, these reports strongly indicate that IFNγ suppresses BM HSCs and may cause BM failure. While our previous studies suggest integrin αvβ3 is involved in the response of HSCs to IFNγ (Ishihara et al, 2014), the role of integrin αvβ3 in IFNγ response on HSCs remains unclear.

In this study, we indicate that integrin αvβ3 plays a key role in the suppressive effect of IFNγ on HSCs, whereas the signaling of this integrin alone contributes to the maintenance of HSCs. When IFNγ negatively affects HSC proliferation, integrin αvβ3 intensifies IFNγ‐dependent suppression by enhanced activation of STAT1 through the promotion of serine 727 phosphorylation. Thus, integrin αvβ3 on HSCs acts a modulator yielding appropriate responses to predominant effects induced by the surrounding environment.

Results

Integrin αvβ3 contributes to the suppressive effect of IFNγ on HSC pool

There may be some confusion about the IFN effect on HSCs. IFNα causes the increase in Sca‐1 expression, which is one of the most general positive markers of HSCs, even in cells that originally have low and little expression of Sca‐1 (Pietras et al, 2014). Indeed, this bias of Sca‐1 expression led a part of cells other than HSCs to show HSC‐like immunophenotype, which often complicated the interpretation of the effect of IFN on HSCs. Therefore, at first, we defined the phenotypes of HSCs more strictly without relying on Sca‐1 in BM cells.

IFNγ administration increased the frequency of phenotypic LT‐HSCs defined by CD150+CD34−c‐kit+Sca‐1+lineage− (CD150+CD34−KSL), accompanied with changed expression pattern of Sca‐1 within CD150+CD34−c‐kit+lineage− (CD150+CD34−KL) fraction (Fig 1A). Although most of CD150+CD34−KSL cells in untreated mice show expression of integrin β3 or endothelial protein C receptor (EPCR), known as HSC markers (Balazs et al, 2006; Umemoto et al, 2006; Iwasaki et al, 2010; Gur‐Cohen et al, 2015), we found that the so‐called LT‐HSC fraction of IFNγ‐treated mice was mainly composed of integrin β3Low or EPCR− cells that were hardly observed in the fraction of PBS‐treated mouse HSCs (Fig 1B and C). Importantly, these integrin β3Low cells within the “LT‐HSCs fraction” showed no engraftment capacity after transplantation, even when transplanting 10‐fold input compared to integrin β3High cells (Fig 1D). Moreover, Sca‐1− cells within CD150+CD34−KL fraction in untreated mice showed low expression of integrin β3 (Fig 1E). Consistent with a previous report that IFNα stimulation leads cells to induce or enhance Sca‐1 expression (Pietras et al, 2014), these results strongly indicated that IFNγ treatment causes that Sca‐1− cells within CD150+CD34−KL fraction appear in CD150+CD34−KSL “LT‐HSC fraction” through the induction of Sca‐1 expression. Thus, Sca‐1 is not a good marker for the identification of HSCs from mice treated with IFNγ.

Figure 1. Expression of integrin β3 contributes to the accurate identification of HSCs after IFNγ treatment.

Figure 1

Open in a new tab
  • A
    After WT mice were intravenously injected with 1 μg IFNγ three times every 48 h, the frequency of LT‐HSC fraction (CD150+CD34−KSL) was examined. The graph depicts the frequency of LT‐HSC fraction. Data are presented as means ± SD, and were analyzed using Student's t‐test (n = 5, *P < 0.01).
  • B, C
    Expression of integrin β3 (B) and EPCR (C) within LT‐HSC fraction (CD150+CD34−KSL) was examined. The graph depicts the frequency of integrin β3High/EPCR+ cells within indicated fractions after serial administration. Data are presented as means ± SD, and were analyzed using Student's t‐test (n = 5, *P < 0.01).
  • D
    Twenty cells of integrin β3High LT‐HSC fraction (CD150+CD34−KSL) derived from PBS or IFNγ‐administrated mice or 200 cells of integrin β3Low LT‐HSC fraction (CD150+CD34−KSL) obtained from IFNγ‐administrated mice (Ly5.1) were competitively transplanted with 2 × 105 competitor cells (Ly5.2) into lethally irradiated mice (Ly5.1). The gate for each fraction represents within the dot plot of panel (B). Twenty weeks later, the percent donor cells (Ly5.1+) were determined in peripheral blood. Recipient mice with < 0.5% donor cell chimerism in any lineage were considered not to be reconstituted (negative mice). Each plot represents donor‐derived cells (% Ly5.1+ cells) in the peripheral blood of recipient mice. Bars indicate mean values. Data were analyzed using Student's t‐test (n = 9−10, *P < 0.01).
  • E
    Expression of integrin β3 in all Sca‐1− and Sca‐1+ cells within CD150+CD34−c‐kit+Lineage− (CD150+CD34−KL) of untreated mice was examined. The graph depicts the frequency of integrin β3High cells within this fraction after PBS or IFNγ administration. Data are presented as means ± SD, and were analyzed using Student's t‐test (n = 5, *P < 0.01, **P < 0.05).
  • F
    Expression of EPCR in integrin β3High CD150+CD34−KL fraction was examined. The graph depicts the frequency of EPCR+ cells after serial administration. Data are presented as means ± SD, and were analyzed using Student's t‐test (n = 5, N.S., not significant).
  • G
    Expression of Sca‐1 in EPCR+ integrin β3High CD150+CD34−KL fraction was examined after serial administration. Numbers within plots depict the frequency of Sca‐1+ cells. Data are presented as means ± SD (n = 5).
  • H
    Forty cells of EPCR+ integrin β3High CD150+CD34−KL derived from PBS or IFNγ‐administrated mice (Ly5.1) were competitively transplanted with 2 × 105 competitor cells (Ly5.2) into lethally irradiated mice (Ly5.1). Twenty weeks later, the percent donor cells (Ly5.1+) were determined in peripheral blood. Each plot represents donor‐derived cells (% Ly5.1+ cells) in the peripheral blood of recipient mice. Bars indicate mean values. Data were analyzed using Student's t‐test (n = 9−12, N.S., not significant).

To avoid this issue, CD150+CD34−KL fraction without Sca‐1 has been often used in the analysis for HSCs under IFN stimulation. However, in contrast to CD150+CD34−KSL LT‐HSC fraction, CD150+CD34−KL population was mainly composed of integrin β3Low cells (Fig 1E), which imply that non‐usage of Sca‐1 greatly abolished the purity of LT‐HSCs within the defined HSC fraction. Therefore, we suggested that integrin β3 is available as a substitute HSC marker instead for Sca‐1 to prevent from IFNγ‐induced bias of Sca‐1 expression without loss of HSC purity in the defined fraction, since expression of integrin β3 is greatly correlated with Sca‐1 expression in CD150+CD34−KL fraction (Fig 1E). Indeed, integrin β3High CD150+CD34−KL showed no change of EPCR expression pattern after IFNγ treatment (Fig 1F). Moreover, in both untreated and IFNγ‐treated mice, almost all EPCR+ integrin β3High CD150+CD34−KL cells were gated in Sca‐1+ fraction (Fig 1G), and these cells showed equal engraftment (Fig 1H). These data indicate that integrin β3High CD150+CD34−KL or EPCR+ integrin β3High CD150+CD34−KL fraction can identify LT‐HSCs in IFNγ‐treated mice as accurately as in untreated mice. Moreover, since integrin β3Low cells that were mainly composed of Sca‐1− cells were excluded from CD150+CD34−KL cells by gating integrin β3High fraction (Fig 1E), integrin β3High CD150+CD34−KL fraction suggest showing greater HSC purity, compared to CD150+CD34−KL cells. Thus, the usage of integrin β3 (i.e., integrin β3High CD150+CD34−KL fraction) as an alternative to Sca‐1 (i.e., CD150+CD34−KSL fraction) enables a more accurate analysis of phenotypic LT‐HSCs in IFNγ‐treated mice.

Next, to examine the effect of integrin β3 signaling on IFNγ response in HSCs, we used integrin β3Y747A‐(Y747A) mutant mice, which carry an alanine substitution at tyrosine 747 and exhibit impaired integrin β3 outside‐in signaling (Petrich et al, 2007; Takizawa et al, 2010). Consequently, IFNγ led to decreased frequency and absolute number of integrin β3High CD150+CD34−KL cells within BM in wild‐type (WT) mice, but not in Y747A mutant mice (Fig 2A and B). In addition, we previously confirmed that this mutation hardly affects expression of integrin β3 in HSCs (Umemoto et al, 2012). These data suggest that integrin β3 signaling plays a key role in the suppressive effect of IFNγ on HSC pool. Furthermore, to confirm the in vivo effect of integrin β3 signaling on IFNγ‐mediated suppression of HSCs, we prepared chimeric mice by co‐transplantation from both WT and integrin β3 mutant (Y747A) BM cells and treated them with or without serial administration of IFNγ (Fig 2C). In agreement with our previous result that Y747A‐derived HSCs showed decreased LTR activity than WT HSCs (Umemoto et al, 2012), WT BM cells showed higher engraftment than integrin β3 mutant BM (Fig 2D). Interestingly, IFNγ treatment enhanced the chimerism of Y747A‐derived cells within integrin β3High CD150+CD34−KL fraction in BM (Fig 2D). Although the absolute number of WT‐derived integrin β3 High CD150+CD34−KL cells decreased, IFNγ treatment did not affect the number of Y747A‐derived cells (Fig 2E). Therefore, enhanced chimerism of Y747A‐derived cells after IFNγ treatment was attributed to decreased absolute number of WT‐derived cells. Moreover, gene set enrichment analyses (GSEA) showed that IFNγ‐associated gene sets significantly enriched in WT cells obtained from IFNγ‐treated chimeric mice, respectively (P‐ or q‐value > 0.05; Fig 2F, G and H), which was not observed in Y747A mutation cells (Fig 2F, I and J). Hierarchal clustering assay revealed that the group of IFNγ‐treated integrin β3 mutant HSCs was largely divided into the clusters containing PBS‐ or IFNγ‐treated WT cells (Fig 2K). These data indicate that integrin β3 signaling on HSCs greatly contribute to the suppressive effect of IFNγ through supporting IFNγ‐mediated gene expression.

Figure 2. Integrin β3 signaling is involved in the suppressive effect of IFNγ on HSC pool.

Figure 2

Open in a new tab
  • A, B
    After WT and integrin β3Y747A (Y747A) mice were intravenously injected with 1 μg IFNγ three times every 48 h, the frequency or absolute number of integrin β3High CD150+CD34−c‐kit+Lineage− (integrin β3High CD150+CD34−KL) cells was examined. The graph depicts the frequency (A) or absolute number (B) of integrin β3High CD150+CD34−KL cells in WT or Y747A mice after PBS or IFNγ administration. Data are presented as means ± SD, and were analyzed using Student's t‐test (n = 5, **P < 0.05).
  • C
    1 × 106 whole BM cells were obtained from interginY747A (Y747A) mice (Ly5.2) and transplanted with 1 × 106 WT cells (Ly5.1) into lethally irradiated mice (Ly5.1). After more than 12 weeks from the transplantation, BM chimera mice were serially administrated with PBS or IFNγ five times every 48 h.
  • D, E
    The graph depicts the frequency (D) or absolute number (E) of WT‐ or Y747A‐derived integrin β3High CD150+CD34−KL cells in BM chimera mice after PBS or IFNγ administration. Data are presented as means ± SD, and were analyzed using Student's t‐test (n = 10, **P < 0.05).
  • F
    After BM chimera mice were serially administrated with PBS or IFNγ five times every 48 h, sorted Y747A (Ly5.2)‐ or WT (Ly5.1)‐derived integrin β3High CD150+CD34−KL cells were subjected to RNA‐Seq. Prior to gene set enrichment analysis (GSEA) using these transcriptome data, IFNγ‐dependently changed genes were extracted by comparison between WT integrin β3High CD150+CD34−KL cells derived from PBS‐ or IFNγ‐treated mice (P < 0.05, fold change < −2 or > 2).
  • G–J
    The plots depict the enrichment of IFNγ‐dependently up‐ (G and I) or down‐regulated (H and J) gene sets within up‐regulated genes in WT (G and H) or Y747A (I and J) integrin β3High CD150+CD34−KL cells derived from PBS or IFNγ‐treated mice. p‐ and q‐value mean nominal P‐value and false data rate, respectively.
  • K
    Using thesse transcriptome data, hierarchical clustering analysis was performed after filtering based on IFNγ‐dependent changes of gene expression in WT HSCs (Student's t‐test, P < 0.05, fold change < −2 or > 2).

Next, to test whether the induction of integrin β3 signaling enhances this suppressive effect of IFNγ on LT‐HSCs, we cultured HSCs with or without vitronectin (VN), a known ligand for integrin αvβ3, in serum‐free medium containing few ligands of this integrin receptor, and conducted transplantation with the cultured cells. Interestingly, the addition of VN further decreased the engraftment of WT HSCs when IFNγ negatively affected LTR activity (Fig 3A), but VN conversely augmented the engraftment of HSCs under the conditions without IFNγ (Fig 3B). Importantly, these effects of VN were completely canceled in integrin β3‐deficient HSCs, while similar effect of IFNγ was observed between WT and integrin β3‐deficient HSCs under conditions with little induction of integrin β3 signaling (e.g., without VN) (Fig 3C). These data indicate that the induction of integrin β3 signaling negatively affected the engraftment of HSCs in response to IFNγ. Moreover, the decrease in HSC numbers in VN plus IFNγ‐treated cells after the culture was supported through a limiting‐dilution assay (Fig 3D–F). These data suggest that the induction of integrin β3 signaling promotes the suppressive effect of IFNγ on LT‐HSCs.

Figure 3. Integrin β3 signaling promotes IFNγ‐dependent suppression of HSC function.

Figure 3

Open in a new tab
  1. After 50 CD150+CD34−KSL LT‐HSCs derived from WT mice (Ly5.1) were sorted and cultured for 5 days on plates with or without of vitronectin (VN) coating, in the presence of both SCF and TPO, with or without IFNγ, whole cultured cells were transplanted into lethally irradiated mice (Ly5.2) along with 5 × 105 BM competitor cells (Ly5.2). Twenty weeks later, the % donor cells (Ly5.1+) were determined in peripheral blood (1st). Then, 12 weeks after secondary transplantation with chimeric BM cells from the primary recipients, peripheral blood from the secondary recipients was analyzed (2nd). Recipient mice with < 0.5% donor cell chimerism in any lineage were considered not to be reconstituted (negative mice). Each plot represents donor‐derived cells (% Ly5.1+ cells) in the peripheral blood of recipient mice. Bars indicate mean values. Data were analyzed using Student's t‐test (n = 9−12, *P < 0.01, **P < 0.05).
  2. A total of 50 sorted CD150+CD34−KSL LT‐HSCs derived from WT mice (Ly5.1) were cultured on plates with or without vitronectin (VN) coating in the presence of SCF and TPO. After 5 days of culture, serial transplantation assays were performed using whole BM cells along with 5 × 106 competitor cells (Ly5.2), as described above. Each plot represents donor‐derived cells (% Ly5.1+ cells) in the peripheral blood of recipient mice. Bars indicate mean values. Data were analyzed using Student's t‐test (n = 11−12, *P < 0.01, **P < 0.05).
  3. CD150+CD34−KSL LT‐HSCs derived from integrin β3−/− mice (Ly5.2) were also subjected to competitive transplantation assays using competitors (Ly5.1) and recipients (Ly5.1) following culture under indicated conditions, as described above. At 20 weeks after transplantation, the chimerism of donor cells (Ly5.2+) was determined in peripheral blood. Each plot represents donor‐derived cells in the peripheral blood of recipient mice. Bars indicate mean values. Data were analyzed using Student's t‐test (n = 10−12, **P < 0.05, N.S.; not significant).
  4. Functional HSC frequencies after culture in the presence of IFNγ were determined with limiting‐dilution assays. After 5 days culture, 30, 50, 100, or 150 cultured whole cells (Ly5.1) were transplanted into lethally irradiated recipient mice along with 2 × 105 BM supporting cells from Ly5.2 mice. Twenty weeks after transplantation, recipient mice with donor cell chimerism (> 0.5% for myeloid and B‐ and T‐lymphoid lineages) were considered to be multilineage‐reconstituted mice (positive mice). The % unreconstructed mice (% negative mice on y‐axis) was plotted vs. the number of input cells, leading to a theoretical HSC frequency based on a Poisson distribution. This plot was generated by gathering the results of three independent experiments. Each point was calculated from more than 12 recipient mice. In the case of control group, the point of 150 input cells could not be plotted due to no unreconstructed mice.
  5. After culturing 50 WT LT‐HSCs under each different condition, total cell numbers were quantified using microscopy. The graph depicts the total cell numbers after 5 days of culture. Data are presented as means ± SD, and were analyzed using Student's t‐test (n = 5, *P < 0.01).
  6. Functional HSC number after culture for 5 days was estimated using theoretical HSC frequencies (D) and total cell numbers (E). The graph depicts estimated HSC numbers after culture under the indicated conditions. Data are presented as means ± SD, and were analyzed using Student's t‐test (*P < 0.01, **P < 0.05).

Integrin β3 signaling promotes IFNγ/STAT1‐dependent change of gene expression in HSCs

Next, to clarify the mechanism through which integrin β3 signaling affects the effect of IFNγ on HSCs, we compared the gene expression profiles of untreated‐ or VN‐treated HSCs that were cultured with or without IFNγ in the presence of stem cell factor (SCF) plus TPO (Fig 4A). We used GSEA with gene sets whose expression was significantly altered during the culture with IFNγ. IFNγ‐dependent up‐regulated gene sets were significantly enriched in HSCs treated with VN plus IFNγ (Fig 4B). Consistent with these results, IFNγ‐dependent down‐regulated gene sets were also enriched in IFNγ (Fig 4C). An increase in the expression of such IFNγ‐related genes such as Irf1, 7, 8, or 9 in VN plus IFNγ‐treated HSCs was confirmed using real‐time RT–PCR (Fig 4D). By contrast, VN without IFNγ in the presence of SCF plus TPO did not influence expression of IFNγ‐dependent genes (Fig 4E and F). These data indicate that integrin β3 signaling promotes expression of IFNγ‐dependent genes in HSCs only in the presence of IFNγ.

Figure 4. Integrin β3 signaling promotes IFNγ/STAT1‐dependent gene expression in HSCs.

Figure 4

Open in a new tab
  • A
    Wild‐type (WT) LT‐HSCs were cultured on plates with or without vitronectin (VN) coating, in the presence of SCF plus TPO, in the absence or presence of IFNγ. RNA‐Seq was then performed using the sorted CD48−KSL fraction, which is regarded as the cultured HSC fraction (Noda et al, 2008), obtained from cultured cells. The transcriptome data from HSCs cultured with IFNγ or with IFNγ + VN were compared using gene set enrichment analysis (GSEA). Gene sets exhibiting IFNγ‐dependent changes were detected by comparing HSCs cultured under control conditions and the presence of IFNγ (P < 0.05, fold change < −5 or > 5).
  • B, C
    Enrichment plots depict the IFNγ‐dependently changed gene sets that were enriched among up‐ (B) or down‐regulated (C) genes in the presence of VN coating.
  • D
    Real‐time RT–PCR analysis revealed IFNγ‐dependent up‐regulated expression of Irf1, ‐7, ‐8, and ‐9 genes in CD150+CD34−KSL LT‐HSCs cultured for 5 days with or without VN in the presence or absence of IFNγ. The graphs depict the mRNA expression of the indicated genes. Data are expressed as the mean ± SD, and were analyzed using Student's t‐test (n = 4, *P < 0.01, **P < 0.05). p‐ and q‐value mean nominal P‐value and false data rate, respectively.
  • E, F
    Enrichment plots depict the IFNγ‐dependently changed gene sets that were enriched among up (E)‐ or down‐regulated (F) genes in the absence of VN. p‐ and q‐value mean nominal P‐value and false data rate, respectively.
  • G
    Using the transcriptome data from WT and STAT1−/− HSCs cultured with or without IFNγ, hierarchical clustering analysis was performed after filtering based on IFNγ‐dependent changes in gene expression (Student's t‐test, P < 0.05, fold change < −2 or > 2) in WT HSCs.
  • H, I
    The plots depict the enrichment of STAT1‐dependently up‐regulated gene sets (IFNγ‐dependent genes extracted in panel A whose expression was inhibited by > 50% upon STAT1 deficiency) within changed genes upon exposing cells to VN coating in the presence (H) or absence (I) of IFNγ. p‐ and q‐value mean nominal P‐value and false data rate, respectively.
  • J, K
    The plots depict the enrichment of STAT1‐dependently up‐regulated gene sets within changed genes upon WT‐ (J) or Y747A‐derived (K) integrin β3High CD150+CD34−KL cells obtained from BM chimera mice in response to IFNγ treatment. p‐ and q‐value mean nominal P‐value and false data rate, respectively.

STAT1 is a core signaling molecule downstream of IFNγ (Darnell et al, 1994) that appears to play a key role in IFNγ‐dependent impairment of LTR activity (King et al, 2011). Indeed, hierarchical clustering analysis showed that IFNγ‐treated STAT1‐deficient HSCs were clustered in the same group as WT HSCs cultured in the absence of IFNγ, rather than in the group containing HSCs treated with IFNγ (Fig 4G). In addition, expression of Irf1, 7, 8, or 9 was greatly impaired by STAT1‐deficiency (Fig 4G) Moreover, STAT1‐dependent up‐regulated gene sets (IFNγ‐dependent genes which expression was inhibited by > 50% upon STAT1‐deficiency) were significantly enriched among genes whose expression was enhanced by VN in the presence of IFNγ (Fig 4H), but not in the absence of IFNγ (Fig 4I). Furthermore, in the chimeric mice described before (Fig 2C), STAT1‐up‐regulated genes were significantly enriched within WT cells derived from IFNγ‐treated chimera mice, but Y747A mutation showed no statistical significance (P‐ or q‐value > 0.05) and decreased enrichment (Fig 4J and K). In addition, the facilitation of integrin β3 signaling hardly affected expression of STATs or JAKs (Fig EV1). These data indicate that integrin β3 signaling in HSCs promotes STAT1‐related gene expression in response to IFNγ stimulation.

Figure EV1. Integrin β3 signaling hardly affects expression of STATs or Jaks in HSCs.

Figure EV1

Open in a new tab
  • Real‐time RT–PCR analysis revealed expression of Jak/Stat family in CD150+CD34−KSL LT‐HSCs cultured for 5 days with or without VN in the presence of IFNγ. The graphs depict the mRNA expression of the indicated genes. Data are expressed as the mean ± SD (n = 4, N.S., not significant).

Integrin β3 signaling promotes serine phosphorylation of STAT1 during IFNγ stimulation in HSCs

Because integrin β3 signaling promoted IFNγ/STAT1‐dependent gene expression, we suspected that integrin β3 signaling affects STAT1 activation. To examine this possibility, we assessed STAT1 phosphorylation after 48 h of IFNγ stimulation in HSCs cultured with or without VN. However, although IFNγ treatment induced phosphorylation of STAT1 at Tyr701, VN did not alter phosphorylation of STAT1 at Tyr701 in the absence or presence of IFNγ (Fig 5A). As an alternative, we focused on the phosphorylation of STAT1 at Ser727, as phosphorylation at this position is required for full STAT1 transcriptional activity (Wen et al, 1995). When phosphorylation of STAT1 at Ser727 was enhanced by IFNγ stimulation at the same time as tyrosine phosphorylation, VN increased serine phosphorylation level of STAT1 (Fig 5B). However, VN hardly affected serine phosphorylation in the absence of IFNγ (Fig 5B). Interestingly, although approximately 52% of HSCs cultured with IFNγ alone showed STAT1 serine phosphorylation, VN treatment led to approximately 88% of HSCs exhibiting STAT1 phosphorylation at Ser727 (Fig 5B). Importantly, although both tyrosine and serine phosphorylation of STAT1 were hardly affected by integrin β3‐deficiency under culture conditions without any external ligands of integrin αvβ3 (Fig 5A and B), integrin β3‐deficient HSCs did not exhibit the positive effect of VN on Ser727 phosphorylation in the presence of IFNγ (Fig 5B). Furthermore, VN in the presence of IFNγ did not affect expression of integrin β3, STAT1, or IFNγ receptor α chain in HSCs (Fig EV2). These data show that the induction of integrin β3 signaling increases the activation of STAT1 through enhanced serine 727 phosphorylation in HSCs, when IFNγ induces STAT1 activation.

Figure 5. Integrin β3 signaling promotes STAT1 serine 727 phosphorylation during IFNγ‐induced prolonged activation of STAT1.

Figure 5

Open in a new tab
  • A, B
    After culturing CD34−KSL cells under indicated conditions in the presence of TPO plus SCF for 48 h, STAT1 phosphorylation at Tyr701 (A) or Ser727 (B) in WT and integrin β3−/− HSCs was examined. Each graph depicts geometric mean fluorescence intensity (GeoMFI) relative to the value of cells treated with isotype control, or the percent of STAT1pS727‐positive cells. Data are expressed as the mean ± SD, and were analyzed using Student's t‐test (n = 3−4, *P < 0.01, **P < 0.05).

Figure EV2. Integrin β3 signaling hardly affects expression of integrin β3, STAT1 or IFNγR α chain at protein level in HSCs.

Figure EV2

Open in a new tab
  • A–C
    Expression of integrin β3 (A), STAT1 (B), and IFNγR α chain (C) on WT HSCs after IFNγ treatment with or without VN for 48 h in the presence of SCF plus TPO was determined using flow cytometry: white, isotype control; gray, indicated antibody. Each graph depicts the relative geometric mean fluorescence intensity (GeoMFI) of the indicated antigens; cells treated with isotype control served as the control. Data are expressed as the mean ± SD (n = 4, N.S., not significant).

Integrin β3 signaling supports the suppressive effect of IFNγ on HSC function through STAT1

In order to confirm whether integrin β3 signaling contributes to the suppressive effect of IFNγ on LT‐HSCs through STAT1, we performed transplantation assays using STAT1‐deficient HSCs cultured with or without VN in the presence of IFNγ (Fig 6A). In agreement with in vitro data, STAT1 deficiency completely reverses the effect of VN that was observed in HSCs cultured with IFNγ (Fig 6A compared to Fig 3A). Limited dilution of whole cultured cells exhibited that VN increased the number of STAT1‐deficient HSCs in the context that this cytokine led to increased number of STAT1‐deficient HSCs (Fig 6B–D). Our data underline that STAT1 deficiency eliminated the IFNγ‐dependent suppressive effect of integrin β3 signaling on HSC function, and indicate that integrin β3 signaling in the presence of IFNγ suppresses LT‐HSCs through the predominant effect of STAT1.

Figure 6. Integrin β3 signaling supports the effect of IFNγ through STAT1.

Figure 6

Open in a new tab
  1. STAT1−/− CD150+CD34−KSL HSCs (Ly5.2) were cultured for 5 days in the presence of SCF and TPO, with or without vitronectin (VN), in the absence or presence of IFNγ, after which they were transplanted into lethally irradiated mice (Ly5.1) along with 5 × 105 BM competitor cells (Ly5.1). Twenty weeks later, the percent donor cells (Ly5.2+) were determined in peripheral blood. Each plot depicts the chimerism of donor‐derived cells (% Ly5.2+ cells) in the peripheral blood of recipient mice. Bars indicate mean values. Data were analyzed using Student's t‐test (n = 11−14, *P < 0.01, **P < 0.05).
  2. After culturing 50 STAT1−/− LT‐HSCs as described in (A), total cell numbers were quantified using microscopy. The graph depicts the total cell numbers after 5 days of culture. Data are expressed as the mean ± SD, and were analyzed using Student's t‐test (n = 4, *P < 0.01).
  3. Functional STAT1−/− HSC frequencies among whole cultured cells were determined using limiting‐dilution assays. After 5 days of culture, 50 or 100 whole cultured cells (Ly5.2) were transplanted into lethally irradiated recipient mice along with 2 × 105 BM supporting cells from Ly5.1 mice. After determining the % multilineage reconstructed mice, theoretical HSC frequencies were estimated based on a Poisson distribution.
  4. Functional STAT1−/− HSC number after 5 days of culture was estimated based on total cell numbers (B) and HSC frequencies (C). The bars depict estimated numbers of functional STAT1−/− HSCs after culture. Data are presented as means ± SD, and were analyzed using Student's t‐test (*P < 0.01).

Discussion

Hematopoietic stem cells are regulated by intrinsic programs and extrinsic stimulation through cytokines and adhesion molecules. When HSC sense external stimuli, their adhesion status (i.e., adhesion or non‐adhesion) might be critical for the regulation of HSC function. Integrin αvβ3 is activated through HSC adhesion to the ECM and niche cells. Here, we clarified the synergistic mechanism of IFNγ and integrin β3 signaling. We observed that integrin β3 signaling enhanced STAT1‐mediated responses of HSCs (i.e., impaired engraftment, gene expression and serine phosphorylation) only in the presence of IFNγ, but not in the absence of IFNγ (Figs 3, 4, 5, 6). Moreover, that integrin β3 mutation mitigates IFNγ‐dependent negative action on HSCs in vivo (Figs 1 and 2). Therefore, our finding strongly suggests that this synergistic effect is attributed to a mechanistic link between IFNγ and integrin β3 signaling via STAT1. On the one hand, the deletion of integrin β3 signaling hardly affected the effect of IFNγ on HSCs in vitro (Fig 3C), unlike in vivo (Fig 2). This may be due to our “serum‐free” culture system that contains few ligands of integrin αvβ3. Indeed, unless external ligand of integrin αvβ3, this integrin signaling is hardly induced even in WT HSCs under our serum‐free culture conditions, resulting in similar response to IFNγ between WT and integrin β3‐deficient HSCs. In contrast, our previous study suggests that ligands of integrin αvβ3 are presented in HSC niche in vivo (Umemoto et al, 2012). Moreover, integrin β3‐deficiency mitigated the response of HSC to IFNγ in the presence of VN (Fig 3C), similarly to in vivo (Fig 2C–E). Thus, the effect of integrin β3‐deficiency on IFNγ appears to be dependent on the presence of their ligands around HSCs. Therefore, our results also suggest that integrin β3 signaling constantly affects HSC regulation via this mechanistic link during IFNγ stimulation in vivo.

Compared to stimulation of a soluble ligand alone, immobilization of cells via cell adhesion molecules may enable effective and prolonged cell signaling (Tohyama et al, 1998; Saik et al, 2011). The membrane‐associated Kit‐ligand has been shown to act differently from the secreted Kit‐ligand in T‐cell differentiation (Buono et al, 2016). Similarly, when TPO does affect the maintenance of LT‐HSCs, integrin β3 signaling supports this effect of TPO on LT‐HSCs (Umemoto et al, 2012).

Here, we have shown that IFNγ negatively affects the function of LT‐HSCs. At that time, integrin β3 signaling enhances the inhibitory effect of IFNγ on HSCs through enhanced serine 727 phosphorylation of STAT1 (Figs 3, 4, 5, 6). Moreover, IFNγ treatment augmented the engraftment of STAT1‐deficient HSCs by increasing numbers of functional HSCs (Fig 6), probably through STAT1‐masked potential of IFNγ to enhance tyrosine phosphorylation of both STAT5 and 3 (Fig EV3). Under these conditions, integrin β3 signaling again appeared to promote the predominant effect of this context in LT‐HSCs (Fig 6A and D). These results thus suggest that integrin β3 signaling regulates the response of HSCs to external stimulation. HSC immobilization through cell attachment within the stem cell niche enables HSCs to modify their behavior in response to external signaling, and the appropriate regulation of this cell adhesion within niches contributes to the maintenance of hematopoietic homeostasis. Consequently, dysregulation of HSC adhesion via integrin αvβ3 during chronic infection or inflammation with increased IFNγ levels may cause hematopoietic disorders.

Figure EV3. IFNγ stimulation leads to enhanced phosphorylation of STATs in STAT1‐deficient HSCs.

Figure EV3

Open in a new tab
  • After culturing STAT1‐deficient CD34‐KSL cells with or without IFNγ in the presence of TPO plus SCF for 48 h, tyrosine phosphorylation of STAT5 and STAT3 in WT HSCs was examined. Each graph depicts geometric mean fluorescence intensity (GeoMFI) relative to cells treated with isotype control. Data are expressed as the mean ± SD (n = 3, *P < 0.01; **P < 0.05).

There may be some confusion about the IFN effect on HSCs. It was reported that IFN induces the expansion of HSCs, which may have been influenced by the increase in Sca‐1 expression by IFNγ. In this manuscript, we employed novel markers (EPCR and integrin β3) and functional assays to stringently define HSCs rather than relying on the expression of Sca‐1. We found that cell fraction with integrin β3Low cells are increased with the gate of KSL by the stimulation with IFNγ (Fig 1C), and that LT‐HSCs are not in this population, but in β3 integrin High cells (Fig 1C). Thus, to examine the effect of IFNγ on HSCs in vivo, we focused on integrin β3+CD150+CD34−KL cells and showed that IFNγ decreased HSC number (Fig 2A and B).

Although several mechanisms through which IFNs suppress HSC function (decrease quiescence, enhanced differentiation, suppression of HSC proliferation or apoptosis) have been reported (Essers et al, 2009; Baldridge et al, 2010; de Bruin et al, 2013; Matatall et al, 2014; Pietras et al, 2014), it remains unclear by which mechanism integrin β3 signaling is involved. However, our findings indicate that the suppressive effect of IFNγ is completely dependent on STAT1, and that the contribution of integrin β3 signaling to the action of IFNγ was also mediated through STAT1. Therefore, integrin β3 signaling in the presence of IFNγ appears to be involved in STAT1‐dependent mechanisms for HSC regulation.

In conclusion, this study unveils a synergistic effect of integrin αvβ3 in the response of HSCs to IFNγ. Our finding indicated that integrin αvβ3 acts as an amplifier of the predominant effects induced by external stimuli, enabling HSCs to effectively respond to their surrounding environment. Hence, the appropriate regulation of cell adhesion within BM niches may play a key role in the maintenance of hematopoietic homeostasis under steady‐state conditions as well as inflammatory conditions. Our observations may provide insight into the pathogenesis and treatment of hematological disorders secondary to chronic inflammation.

Materials and Methods

Animals

C57BL/6‐Ly5.2 and C57BL/6‐Ly5.1 mice were obtained from Sankyo Labo Service Corporation (Tokyo, Japan) or Japan SLC Inc. (Shizuoka, Japan), and β3 integrin‐deficient mice were obtained from The Jackson Laboratory (Bar Harbor, ME) unless otherwise noted. Integrin β3 Y747A knock‐in mutant mice and STAT1‐deficient mice were described previously (Petrich et al, 2007; Shimizu et al, 2013). Each strain was used between 8 and 12 weeks of age.

Antibodies for flow cytometry

The following monoclonal antibodies were used for cell sorting and flow cytometric analysis of surface markers: anti‐c‐Kit (2B8), anti‐CD34 (RAM34, eBioscience, San Diego, CA), anti‐CD150 (TC15‐12F12.2), anti‐CD48 (HM48‐1, BioLegend), anti‐integrin β3 (2C9.G2), anti‐IFNγ receptor α chain (2E2), anti‐Sca‐1 (E13‐161.7), anti‐CD45.2 (104), anti‐CD45.1 (A20), anti‐B220/CD45R (RA3‐6B2), anti‐Mac‐1 (M1/70), anti‐Gr‐1 (RB6‐8C5), anti‐CD4 (RM4‐5), and anti‐CD8 (53‐6.72) antibodies. All antibodies were obtained from BioLegend (San Diego, CA) unless otherwise noted.

Cell preparation

Suspensions of BM cells were prepared from mice as described previously (Umemoto et al, 2006, 2012).

IFNγ administration

To stimulate HSCs with IFNγ in vivo, mice were intravenously injected with 1 μg IFNγ (Shenandoah Biotechnology) three or five times every 48 h. Twenty‐four hours after the last administration, BM cells were removed for transplantation assays. PBS‐administered mice served as a control.

Cell sorting and flow cytometric analysis

We used a MoFlo™ XDP (Beckman Coulter, Fullerton, CA), FACS Aria II (BD Bioscience, San Jose, CA), FACS CANTO II (BD Bioscience), or FACS VERSE (BD Bioscience) for cell sorting and flow cytometric analyses, as described previously (Umemoto et al, 2006, 2012).

Long‐term competitive repopulation assays

Long‐term competitive repopulation assays were performed by transplanting the indicated cells into lethally irradiated (10 Gy) C57BL/6‐Ly5.2 or C57BL/6‐Ly5.1 congenic mice through i.v., as described previously (Umemoto et al, 2012). Twenty weeks after transplantation, recipient mice with donor cell chimerism (> 0.5% for myeloid and B‐ and T‐lymphoid lineages) were considered to be multilineage‐reconstituted mice (positive mice). For serial transplantation, 1 × 107 whole BM cells were obtained from primary transplanted mice and transplanted into secondary irradiated recipient mice.

Limiting‐dilution assay

After culture, exactly indicated number of indicated fractions or whole cultured cells were sorted using a cell sorter, followed by transplantation along with 2 × 105 BM supporting cells derived from Ly5.2 mice through i.v., as described previously (Umemoto et al, 2012). Twenty weeks after transplantation, recipient mice with donor cell chimerism (> 0.5% for either myeloid or B‐ or T‐lymphoid lineages) were considered to be multilineage‐reconstituted mice (positive mice). In addition, functional HSC frequencies among fresh uncultured CD150+CD34−KSL cells were assessed after single cell transplantation along with 2 × 105 BM supporting cells. HSC numbers were estimated based on total cell numbers counted and the HSC frequency determined from limiting‐dilution assays, as described previously (Umemoto et al, 2012).

Analysis for BM chimera mice

WT (Ly5.1)‐ and Y747A (Ly5.2)‐derived whole BM cells were transplanted into lethally irradiated mice (Ly5.1) at the ratio of 1:1. After > 12 weeks from the transplantation, BM chimera mice were intravenously injected with IFNγ five times every 48 h and subsequently used for indicated used for the analyses.

RNA sequencing

After culture under the indicated conditions, indicated cells were sorted, and first‐strand cDNA was synthesized as described previously (Umemoto et al, 2012). Synthesized cDNA was then amplified by 18 cycles of PCR, followed by shearing into small fragments (75–200 bp) using a Covaris S2 system (Covaris, Inc., Woburn MA). The sheared cDNA was subjected to end‐repair and subsequent ligation of IonXpress™ P1 and A adaptors to both ends. To amplify the cDNA fragment libraries, the ligated fractions were subjected to limited rounds of PCR using primers complementary to the P1 and A adaptor sequences. After purification of the cDNA fragments with both adaptor sequences, emulsion PCR was run using beads with P1 primers covalently attached to their surfaces. Thereafter, ~100,000,000 beads were sequenced on an Ion Proton™ sequencer (Life Technologies, Carlsbad, CA). All of these procedures were carried out according to the manufacturer's instructions. After sequencing, the raw data were analyzed using Torrent Suite v3.6 (life Technologies). In addition, each read was mapped to the reference sequence “GRCm38/mm10” using CLC genomic workbench v7.0.4 (Qiagen, Aarhus, Denmark), and expression signals were normalized and subjected to the statistical analyses based on EdgeR. All RNA‐seq data were deposited in the Gene Expression Omnibus under accession number GSE81559.

Gene set enrichment analysis

Whole transcriptomes were subjected to gene set enrichment analysis (GSEA) using GSEA v2.2.0 software, available from the Broad Institute (http://www.broad.mit.edu/gsea) (Subramanian et al, 2005). IFNγ‐dependent changes gene sets (in vivo) were extracted by filtering genes whose response to IFNγ was defined by a > twofold change (P < 0.05) between WT HSCs derived from PBS‐ and IFNγ‐treated BM chimera mice. Gene sets showing IFNγ‐dependent changes in vitro were extracted by filtering genes whose response to IFNγ was defined by a > fivefold change (P < 0.05) between WT HSCs cultured in the presence and absence of IFNγ. STAT1‐dependently changed gene sets were extracted from the IFNv‐dependently changed sets (in vitro) by filtering genes suppressed more than 50% by STAT1 deficiency. Enriched gene sets were selected based on a threshold set at P < 0.05 and FDR (q‐value) < 0.05.

HSC cultures

CD150+CD34−KSL cells were sorted and cultured for 5 days in S‐Clone SF‐03 medium (Sanko‐Junyaku Co., Tokyo, Japan) supplemented with 0.5% bovine serum albumin (Sigma, St. Louis, MO), 50 ng/ml mouse stem cell factor (SCF), and 50 ng/ml mouse TPO (all from R&D Systems, Minneapolis, MN). To induce IFNγ signaling, IFNγ (5 ng/ml) (Shenandoah Biotechnology, Warwick, PA) was added to the medium. Signaling via integrin αvβ3 was induced by incubating cells in 96‐well plates coated with 5 μg/ml vitronectin (VN: Molecular Innovations, Novi, MI) overnight at 4°C, as described previously (Umemoto et al, 2012). Total cell numbers were then quantified using light microscopy.

Real‐time quantitative RT–PCR

mRNA expression was assessed using real‐time quantitative RT–PCR as described previously (Umemoto et al, 2006).

Hierarchical analysis

Following filtration based on the indicated threshold, the selected genes were subjected to hierarchical clustering analysis using MeV v4.9.0 software, available from the J. Craig Venter Institute (http://www.tm4.org).

Analysis for phosphorylated STAT1

Intracellular staining was performed using a PerFix EXPOSE kit (Beckman Coulter) according to the manufacturer's instructions. Briefly, sorted CD34−KSL cells were fixed, permeabilized, and staining first with antibodies against STAT1pY701 (BD Bioscience) and STAT1pS727 (Cell Signaling Technology, Danvers, MA) and then with a PE‐conjugated secondary antibody against rabbit IgG (BioLegend). After staining, the cells were analyzed using flow cytometry.

Author contributions

TU designed the study and wrote the manuscript. TU performed most of the experiments. AN‐I and TS edited the manuscript. YM and TY bred and maintained the mice. YS helped with the transplantation assays. YM helped with RNA‐seq. MH helped with the analyses for phosphorylation of STATs. BP generated the integrin knock‐in mice. MY and TS supervised all aspects of the project and helped with manuscript preparation.

Conflict of interest

The authors declare that they have no conflict of interest.

Supporting information

Expanded View Figures PDF

Click here for additional data file. (264.7KB, pdf)

Review Process File

Click here for additional data file. (373KB, pdf)

Acknowledgements

We thank Dr. M. H. Ginsberg (University of California, San Diego, La Jolla, CA) for providing Y747A KI mice. We thank all of the members of our research group for discussion. We are extremely grateful to Ms. M. Fujita and Dr. J. Berlier for her help in sequencing the whole transcriptome and careful reading, respectively. This study was supported by Grant‐in‐Aid for Young Scientists (Grant Numbers 15K19560, 17K16190) (to T.U.), Grant‐in‐Aid for Scientific Research (Grant Number 26221309) (to T.S.) from Japan Society for the Promotion of Science (JSPS), and the SENSHIN Medical Research Foundation (to T.U.).

The EMBO Journal (2017) 36: 2390–2403

Contributor Information

Terumasa Umemoto, Email: umemoto@kumamoto-u.ac.jp.

Toshio Suda, Email: csits@nus.edu.sg.

References

  1. Anderson P (2008) Post‐transcriptional control of cytokine production. Nat Immunol 9: 353–359 [DOI] [PubMed] [Google Scholar]
  2. Arai F, Hirao A, Ohmura M, Sato H, Matsuoka S, Takubo K, Ito K, Koh GY, Suda T (2004) Tie2/angiopoietin‐1 signaling regulates hematopoietic stem cell quiescence in the bone marrow niche. Cell 118: 149–161 [DOI] [PubMed] [Google Scholar]
  3. Balazs AB, Fabian AJ, Esmon CT, Mulligan RC (2006) Endothelial protein C receptor (CD201) explicitly identifies hematopoietic stem cells in murine bone marrow. Blood 107: 2317–2321 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Baldridge MT, King KY, Boles NC, Weksberg DC, Goodell MA (2010) Quiescent haematopoietic stem cells are activated by IFN‐gamma in response to chronic infection. Nature 465: 793–797 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. de Bruin AM, Demirel O, Hooibrink B, Brandts CH, Nolte MA (2013) Interferon‐gamma impairs proliferation of hematopoietic stem cells in mice. Blood 121: 3578–3585 [DOI] [PubMed] [Google Scholar]
  6. Buono M, Facchini R, Matsuoka S, Thongjuea S, Waithe D, Luis TC, Giustacchini A, Besmer P, Mead AJ, Jacobsen SE, Nerlov C (2016) A dynamic niche provides Kit ligand in a stage‐specific manner to the earliest thymocyte progenitors. Nat Cell Biol 18: 157–167 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Darnell JE Jr, Kerr IM, Stark GR (1994) Jak‐STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264: 1415–1421 [DOI] [PubMed] [Google Scholar]
  8. Essers MA, Offner S, Blanco‐Bose WE, Waibler Z, Kalinke U, Duchosal MA, Trumpp A (2009) IFNalpha activates dormant haematopoietic stem cells in vivo . Nature 458: 904–908 [DOI] [PubMed] [Google Scholar]
  9. Fox N, Priestley G, Papayannopoulou T, Kaushansky K (2002) Thrombopoietin expands hematopoietic stem cells after transplantation. J Clin Invest 110: 389–394 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Gur‐Cohen S, Itkin T, Chakrabarty S, Graf C, Kollet O, Ludin A, Golan K, Kalinkovich A, Ledergor G, Wong E, Niemeyer E, Porat Z, Erez A, Sagi I, Esmon CT, Ruf W, Lapidot T (2015) PAR1 signaling regulates the retention and recruitment of EPCR‐expressing bone marrow hematopoietic stem cells. Nat Med 21: 1307–1317 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Hynes RO (2002) Integrins: bidirectional, allosteric signaling machines. Cell 110: 673–687 [DOI] [PubMed] [Google Scholar]
  12. Ishihara J, Umemoto T, Yamato M, Shiratsuchi Y, Takaki S, Petrich BG, Nakauchi H, Eto K, Kitamura T, Okano T (2014) Nov/CCN3 regulates long‐term repopulating activity of murine hematopoietic stem cells via integrin alphavbeta3. Int J Hematol 99: 393–406 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Iwasaki H, Arai F, Kubota Y, Dahl M, Suda T (2010) Endothelial protein C receptor‐expressing hematopoietic stem cells reside in the perisinusoidal niche in fetal liver. Blood 116: 544–553 [DOI] [PubMed] [Google Scholar]
  14. Juliano RL, Reddig P, Alahari S, Edin M, Howe A, Aplin A (2004) Integrin regulation of cell signalling and motility. Biochem Soc Trans 32: 443–446 [DOI] [PubMed] [Google Scholar]
  15. Kimura S, Roberts AW, Metcalf D, Alexander WS (1998) Hematopoietic stem cell deficiencies in mice lacking c‐Mpl, the receptor for thrombopoietin. Proc Natl Acad Sci USA 95: 1195–1200 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. King KY, Baldridge MT, Weksberg DC, Chambers SM, Lukov GL, Wu S, Boles NC, Jung SY, Qin J, Liu D, Songyang Z, Eissa NT, Taylor GA, Goodell MA (2011) Irgm1 protects hematopoietic stem cells by negative regulation of IFN signaling. Blood 118: 1525–1533 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kovtonyuk LV, Manz MG, Takizawa H (2016) Enhanced thrombopoietin but not G‐CSF receptor stimulation induces self‐renewing hematopoietic stem cell divisions in vivo . Blood 127: 3175–3179 [DOI] [PubMed] [Google Scholar]
  18. Lin FC, Karwan M, Saleh B, Hodge DL, Chan T, Boelte KC, Keller JR, Young HA (2014) IFN‐gamma causes aplastic anemia by altering hematopoietic stem/progenitor cell composition and disrupting lineage differentiation. Blood 124: 3699–3708 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Matatall KA, Shen CC, Challen GA, King KY (2014) Type II interferon promotes differentiation of myeloid‐biased hematopoietic stem cells. Stem Cells 32: 3023–3030 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Nakagawa MM, Thummar K, Mandelbaum J, Pasqualucci L, Rathinam CV (2015) Lack of the ubiquitin‐editing enzyme A20 results in loss of hematopoietic stem cell quiescence. J Exp Med 212: 203–216 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Nilsson SK, Johnston HM, Whitty GA, Williams B, Webb RJ, Denhardt DT, Bertoncello I, Bendall LJ, Simmons PJ, Haylock DN (2005) Osteopontin, a key component of the hematopoietic stem cell niche and regulator of primitive hematopoietic progenitor cells. Blood 106: 1232–1239 [DOI] [PubMed] [Google Scholar]
  22. Noda S, Horiguchi K, Ichikawa H, Miyoshi H (2008) Repopulating activity of ex vivo‐expanded murine hematopoietic stem cells resides in the CD48‐c‐Kit+Sca‐1+lineage marker‐ cell population. Stem Cells 26: 646–655 [DOI] [PubMed] [Google Scholar]
  23. Petrich BG, Fogelstrand P, Partridge AW, Yousefi N, Ablooglu AJ, Shattil SJ, Ginsberg MH (2007) The antithrombotic potential of selective blockade of talin‐dependent integrin alpha IIb beta 3 (platelet GPIIb‐IIIa) activation. J Clin Invest 117: 2250–2259 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Pietras EM, Lakshminarasimhan R, Techner JM, Fong S, Flach J, Binnewies M, Passegue E (2014) Re‐entry into quiescence protects hematopoietic stem cells from the killing effect of chronic exposure to type I interferons. J Exp Med 211: 245–262 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Qian H, Buza‐Vidas N, Hyland CD, Jensen CT, Antonchuk J, Mansson R, Thoren LA, Ekblom M, Alexander WS, Jacobsen SE (2007) Critical role of thrombopoietin in maintaining adult quiescent hematopoietic stem cells. Cell Stem Cell 1: 671–684 [DOI] [PubMed] [Google Scholar]
  26. Saik JE, Gould DJ, Watkins EM, Dickinson ME, West JL (2011) Covalently immobilized platelet‐derived growth factor‐BB promotes angiogenesis in biomimetic poly(ethylene glycol) hydrogels. Acta Biomater 7: 133–143 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Shimizu M, Ogura K, Mizoguchi I, Chiba Y, Higuchi K, Ohtsuka H, Mizuguchi J, Yoshimoto T (2013) IL‐27 promotes nitric oxide production induced by LPS through STAT1, NF‐kappaB and MAPKs. Immunobiology 218: 628–634 [DOI] [PubMed] [Google Scholar]
  28. Sloand E, Kim S, Maciejewski JP, Tisdale J, Follmann D, Young NS (2002) Intracellular interferon‐gamma in circulating and marrow T cells detected by flow cytometry and the response to immunosuppressive therapy in patients with aplastic anemia. Blood 100: 1185–1191 [DOI] [PubMed] [Google Scholar]
  29. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, Mesirov JP (2005) Gene set enrichment analysis: a knowledge‐based approach for interpreting genome‐wide expression profiles. Proc Natl Acad Sci USA 102: 15545–15550 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Takizawa H, Nishimura S, Takayama N, Oda A, Nishikii H, Morita Y, Kakinuma S, Yamazaki S, Okamura S, Tamura N, Goto S, Sawaguchi A, Manabe I, Takatsu K, Nakauchi H, Takaki S, Eto K (2010) Lnk regulates integrin alphaIIbbeta3 outside‐in signaling in mouse platelets, leading to stabilization of thrombus development in vivo . J Clin Invest 120: 179–190 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Tohyama Y, Tohyama K, Tsubokawa M, Asahi M, Yoshida Y, Yamamura H (1998) Outside‐In signaling of soluble and solid‐phase fibrinogen through integrin alphaIIbbeta3 is different and cooperative with each other in a megakaryoblastic leukemia cell line, CMK. Blood 92: 1277–1286 [PubMed] [Google Scholar]
  32. Umemoto T, Yamato M, Shiratsuchi Y, Terasawa M, Yang J, Nishida K, Kobayashi Y, Okano T (2006) Expression of Integrin beta3 is correlated to the properties of quiescent hemopoietic stem cells possessing the side population phenotype. J Immunol 177: 7733–7739 [DOI] [PubMed] [Google Scholar]
  33. Umemoto T, Yamato M, Ishihara J, Shiratsuchi Y, Utsumi M, Morita Y, Tsukui H, Terasawa M, Shibata T, Nishida K, Kobayashi Y, Petrich BG, Nakauchi H, Eto K, Okano T (2012) Integrin‐alphavbeta3 regulates thrombopoietin‐mediated maintenance of hematopoietic stem cells. Blood 119: 83–94 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Wen Z, Zhong Z, Darnell JE Jr (1995) Maximal activation of transcription by Stat1 and Stat3 requires both tyrosine and serine phosphorylation. Cell 82: 241–250 [DOI] [PubMed] [Google Scholar]
  35. Yoshihara H, Arai F, Hosokawa K, Hagiwara T, Takubo K, Nakamura Y, Gomei Y, Iwasaki H, Matsuoka S, Miyamoto K, Miyazaki H, Takahashi T, Suda T (2007) Thrombopoietin/MPL signaling regulates hematopoietic stem cell quiescence and interaction with the osteoblastic niche. Cell Stem Cell 1: 685–697 [DOI] [PubMed] [Google Scholar]
  36. Zoumbos NC, Gascon P, Djeu JY, Young NS (1985) Interferon is a mediator of hematopoietic suppression in aplastic anemia in vitro and possibly in vivo . Proc Natl Acad Sci USA 82: 188–192 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Expanded View Figures PDF

Click here for additional data file. (264.7KB, pdf)

Review Process File

Click here for additional data file. (373KB, pdf)

Articles from The EMBO Journal are provided here courtesy of Nature Publishing Group

RESOURCES