The Journal of Immunology

Infection-Induced Myelopoiesis during Intracellular Bacterial Infection Is Critically Dependent upon IFN-g Signaling

Katherine C. MacNamara,* Kwadwo Oduro,† Olga Martin,* Derek D. Jones,* Maura McLaughlin,* Kyunghee Choi,† Dori L. Borjesson,‡ and Gary M. Winslow*

Although microbial infections can alter steady-state hematopoiesis, the mechanisms that drive such changes are not well under- stood. We addressed a role for IFN-g signaling in infection-induced bone marrow suppression and anemia in a murine model of human monocytic ehrlichiosis, an emerging tick-borne disease. Within the bone marrow of Ehrlichia muris-infected C57BL/6 mice, we observed a reduction in myeloid progenitor cells, as defined both phenotypically and functionally. Infected mice exhibited a concomitant increase in developing myeloid cells within the bone marrow, an increase in the frequency of circulating monocytes, and an increase in splenic myeloid cells. The infection-induced changes in progenitor cell phenotype were critically dependent on IFN-g, but not IFN-a, signaling. In mice deficient in the IFN-g signaling pathway, we observed an increase in myeloid progenitor cells and CDllbloGr1lo promyelocytic cells within the bone marrow, as well as reduced frequencies of mature granulocytes and monocytes. Furthermore, E. muris-infected IFN-gR–deficient mice did not exhibit anemia or an increase in circulating monocytes, and they succumbed to infection. Gene transcription studies revealed that IFN-gR–deficient CDllbloGr1lo promyelocytes from E. muris-infected mice exhibited significantly reduced expression of irf-1 and irf-8, both key transcription factors that regulate the differentiation of granulocytes and monocytes. Finally, using mixed bone marrow chimeric mice, we show that IFN-g–dependent infection-induced myelopoiesis occurs via the direct effect of the cytokine on developing myeloid cells. We propose that, in addition to its many other known roles, IFN-g acts to control infection by directly promoting the differentiation of myeloid cells that contribute to host defense. The Journal of Immunology, 2011, 186: 1032–1043.

s most mature blood cells have only a limited lifespan, result in the enhanced production of granulocytes (4–8). This re- hematopoiesis, under homeostatic conditions, acts to sponse, termed “emergency granulopoiesis” (4), functions to A maintain the continual production of new cells. However, increase neutrophil output in response to infection, thereby faci- during immunological stress, caused by factors such as chemo- litating innate host defense. Factors that promote emergency therapy or microbial infection, the demand for immune cells is granulopoiesis include the CSFs, IL-3, and IL-6 (9–11), and the met through enhanced production and mobilization of blood cells. transcription factor CCAAT/enhancer-binding protein b (12). Thus, This requirement for new blood cells can, in turn, result in the inflammation, mediated by soluble factors, promotes the pro- proliferation and differentiation of hematopoietic progenitor cells. duction of granulocytes and augments the innate immune response. Although much of our basic knowledge of hematopoiesis has been Despite knowledge on how these mature bone marrow cells can obtained from studies of steady-state hematopoiesis, a variety of respond to infection-induced hematopoietic stress, less is known viral, bacterial, and fungal infections can alter the cellular com- regarding how myeloid progenitor cells respond to inflammatory position and/or phenotype of bone marrow progenitor cells (1–3). conditions. The mechanisms underlying such changes are likely multifactorial IFN-g plays a critical role in innate immunity to intracellular and infection-dependent; however, the factors that regulate he- pathogens, including the ehrlichiae, and is essential for host de- matopoiesis during microbial infections have not been character- fense via its ability to activate macrophages and to promote Th1 ized in depth. responses. A lesser known function of IFN-g is its ability to act as Studies that have addressed the effects of infections and inflam- a moderator of hematopoiesis. A historical paradigm has been that mation on hematopoiesis have revealed that these processes can IFN-g suppresses hematopoiesis, as was shown in several in vitro culture models (13–15). A role for IFN-g in governing infection- *Wadsworth Center, New York State Department of Health, Albany, NY 12201; induced myelopoiesis was first demonstrated in mycobacterial in- †Washington University School of Medicine, St. Louis, MO 63110; and ‡Department fection (16), where it was proposed that IFN-g acts to suppress of Pathology, Microbiology, and Immunology, School of Veterinary Medicine, University of California, Davis, Davis, CA 95616 myelopoiesis. Other studies have shown that IFN-g–induced genes Received for publication June 8, 2010. Accepted for publication November 10, 2010. are critical for hematopoietic stem cell (HSC) survival/maintenance (17, 18), also suggesting that IFN-g plays a direct role in hema- This work was supported by U.S. Public Health Service Grants R01AI064678 (to G.M.W.) and F32AI077242 (to K.C.M.). topoiesis. IFN-a, a related and important antiviral cytokine, was Address correspondence and reprint requests to Dr. Gary M. Winslow, Wadsworth shown to activate dormant HSCs, demonstrating that IFNs can act Center, 120 New Scotland Avenue, Albany, NY 12201. E-mail address: gary.winslow@ directly on HSCs (19). In a model of malaria infection, it has been wadsworth.org shown that IFN-g is essential for the generation of a unique pro- Abbreviations used in this article: ADAR, adenosine deaminase acting on RNA; CLP, common lymphoid progenitor; CMP, common myeloid progenitor; EGFP, enhanced genitor population that gives rise to mostly myeloid cells (20). Thus, GFP; GMP, granulocyte monocyte progenitor; HSC, hematopoietic stem cell; HME, IFN-g is emerging as a factor in the regulation of infection- or human monocytic ehrlichiosis; IRF, IFN regulatory factor; Lin, lineage; MEP, stress-induced hematopoiesis. megakaryocyte-erythrocyte progenitor. To address a role for IFN-g in infection-induced hematopoiesis, Copyright Ó 2011 by The American Association of Immunologists, Inc. 0022-1767/11/$16.00 we have used a murine model of human monocytic ehrlichiosis www.jimmunol.org/cgi/doi/10.4049/jimmunol.1001893 The Journal of Immunology 1033

(HME), an emerging infectious disease caused by the tick-borne macrophage, megakaryocyte) progenitor cells were scored after 7 d incubation pathogen Ehrlichia chaffeensis. The ehrlichiae are obligate in- at 37˚C in 5% CO2. tracellular bacteria that infect mononuclear cells (21). HME is Generation of bone marrow chimeric mice a febrile illness and patients exhibit a severe loss in hemoglobin + and platelets (21, 22). Infection of C57BL/6 mice with a closely C57BL/6 and IFN-gR–deficient (CD45.2 ) mice were lethally irradiated (950 rad administered in two doses, 3 h apart). After irradiation, the he- related pathogen, Ehrlichia muris, causes hematological abnor- matopoietic compartment of the irradiated mice was reconstituted with malities, including anemia and thrombocytopenia (23), similar to 2 3 106 bone marrow cells obtained from CD45.1 congenic mice. To what is observed in HME patients. We have proposed that the generate the mixed chimeras, CD45.1+ mice were irradiated as above and 3 6 hematological changes that occur during ehrlichiosis are a con- reconstituted with a 1:1 ratio of bone marrow (2 10 cells) harvested from EGFP-expressing and IFN-gR–deficient mice (both CD45.2). After sequence of alterations in hematopoiesis (23). In this study, we irradiation and reconstitution, all mice were administered antibiotics in extend our previous studies by demonstrating that ehrlichial in- their drinking water for 1 wk, and 6 wk postreconstitution, the mice were fection induces changes in the phenotype and function of myeloid bled and screened to determine the degree of hematopoietic chimerism. progenitor cells and augments the production of granulocytes and The chimeric mice were infected with E. muris ∼8 wk postreconstitution. monocytes, and that these changes require direct IFN-g signaling. Flow cytometric cell sorting These studies highlight a novel and important role for this factor in host defense during bacterial infection. Promyelocytes were identified in bone marrow by low surface expression of CDllb and Gr-1, using CDllb-FITC and Gr-1-PE. Bone marrow cells were purified by flow cytometry from day 15 E. muris-infected C57BL/6 mice Materials and Methods and IFN-gR–deficient mice, using an Aria II equipped with FACSDiva Mice software (BD Biosciences). The mice used in these studies were obtained from The Jackson Laboratory Gene expression analyses (Bar Harbor, ME) or were bred in the Animal Care Facility at the Wads- RNA was extracted from flow cytometrically sorted CDllbloGr-1lo cells by worth Center under microisolator conditions. C57BL/6 mice were used as a the RNeasy method (Qiagen, Valencia, CA). RNA integrity was analyzed congenic control strain. The following transgenic and gene-targeted strains with an Agilent bioanalyzer (Agilent Technologies, Foster City, CA), and were used: IFN-aR–deficient mice (a gift of Dr. Susan Swain, Trudeau only RNA preparations with an RNA integrity number of 8.0 or higher Institute, Saranac Lake, NY), IFN-gR1–deficient mice (B6.129S7- tm1Agt were used for gene expression studies. Genomic DNA was removed from Ifngr1 /J, referred to as IFN-gR–deficient mice), enhanced GFP RNA samples with TurboDNA-free DNAse (Ambion, Austin, TX), and (EGFP)-transgenic mice (C57BL/6-Tg(ACTB-EGFP)1Osb/J), and a C57BL/ a b cDNAs were generated with a first-strand synthesis kit (SABiosciences, 6 congenic strain of mice (B6.SJL-Ptprc Pepc /BoyJ). All animal studies Frederick, MD). Gene expression analyses were performed using an IFN-g were performed in accordance with Wadsworth Center Animal Care and gene array assay (SABiosciences; APMM-064). Use Committee guidelines. Statistical analyses Bacteria and infections Statistical analyses were performed with a Student t test, using GraphPad Mice were infected, via i.p. injection, between 6 and 12 wk of age, with Prism software (GraphPad Software, La Jolla, CA); a p value ,0.05 was 50,000 copies of the E. muris bacterium. The inoculum was generated from considered significant. spleen mononuclear cells that were harvested from infected mice and stored at 280˚C, as previously described (24). Bacterial copy number was determined by real-time quantitative probe-based PCR, as previously de- Results scribed (23, 25). Type II IFN-dependent alterations in bone marrow progenitor Flow cytometry and Abs cells during infection Mononuclear cells were harvested from spleens by homogenization of the E. muris infection of C57BL/6 mice provides a model in which to tissue between frosted glass slides in HBSS. Bone marrow mononuclear investigate the pathogenesis of HME (21). In C57BL/6 mice, cells were harvested by flushing of the marrow from femurs using a 26.5- gauge needle and 2 ml HBSS. Tissue homogenates were passed through bacterial burden is maximal between days 8 and 9 postinfection in a 70-mm filter. Erythrocytes were removed by lysis in a hypotonic buffer the spleen, and it declines to very low numbers by day 30 post- containing 0.84% ammonium chloride. The cells were washed with a infection (21, 26). E. muris causes marked anemia and thrombo- buffer containing 5% FCS and 0.01% sodium azide in 13 PBS (wash cytopenia, two hallmarks of HME (21, 22). We previously demon- buffer). Prior to staining, the cells were incubated in blocking buffer (wash strated that these blood abnormalities, which are maximal during buffer containing 1 mg/ml 2.4G2 [FcgRII/III; anti-CD16/CD32]) for 15 min at 4˚C. The Abs used for flow cytometry included the following: FITC- and the second week of infection, occur concomitantly with a decrease biotin-conjugated lineage markers specific for CD3 (clone 17A2), CD11b in bone marrow colony-forming cells (23). Additional studies re- (M1/70), Ly-6G (RB6-8C5), Ter119 (Ly-76), and CD45R (RA3-6B2). Other vealed that infection induced the mobilization and/or depletion of Abs used were PE-conjugated Sca-1 (D7), PE-cychrome-7–conjugated B220+ lymphocytes from the bone marrow, which was coincident Sca-1 (D7), PerCP-Cy5.5-conjugated CD45.2 (104), Alexa 700-CD45.2 (104), PerCP-Cy5.5-CD127 (A7R34), allophycocyanin-conjugated c-Kit with an increase in the frequency of immature and mature gran- (2B8), PE-Gr-1 (RB6-8C5), allophycocyanin-cychrome-7-streptavidin, and ulocytes at this site (23). Significant changes in the cellular com- eFluor 450-streptavidin (eBioscience, San Diego, CA), as well as Pacific position of the bone marrow begins by day 8 postinfection and Blue-CDllb (M1/70; BioLegend, San Diego, CA). Unstained cells were used persists through day 15 postinfection, which prompted us to use as negative controls to establish the flow cytometer voltage settings, and these time points for our analyses. In this study, we have extended single-color positive controls were used for adjustment of the compensation. The flow cytometric data were acquired using FACSCalibur or LSR II flow our previous observations by examining hematopoietic progenitor cytometers (BD Biosciences), and data analysis was performed using FlowJo cell populations in the bone marrow. software (Tree Star, Ashland, OR). In uninfected mice, expression of IL-7Ra (CD127), Sca-1, and c-Kit on lineage-negative (Lin2) bone marrow cells can distinguish In vitro hematopoietic progenitor cell assays common lymphoid progenitors (CLPs; IL-7Ra+c-KitloSca-1+)and Bone marrow cells were plated at 2.5 3104 cells per 35-mm tissue culture dish, myeloid progenitors (IL-7Ra2c-KithiSca-12) (27, 28). Using this in duplicate, and cultured in methylcellulose media (MethoCult GF M3434; criteria, we determined that E. muris infection caused a modest StemCell Technologies, Vancouver, BC, Canada) along with recombinant cytokines for colony assays of murine cells, following the manufacturer’s reduction in phenotypically defined CLPs in C57BL/6 mice: a small instructions. Total colonies derived from granulocyte-macrophage (CFU decrease in frequency was observed on day 8 postinfection (Fig. granulocyte, macrophage) and multipotential (CFU granulocyte, erythrocyte, 1A), although the number of CLPs was unaffected (data not shown). 1034 IFN-g–DEPENDENT MYELOPOIESIS

In contrast, there was almost a complete loss of the IL-7Ra2c-Kithi cells in both C57BL/6 and IFN-aR–deficient mice postinfection, Sca-12 myeloid progenitor cells by day 8 postinfection, and a popu- but no significant decreases in this population in IFN-gR–deficient lation of Sca-1+ cells emerged. mice. Thus, the alterations in bone marrow progenitor cells ob- Sca-1 expression is known to be regulated by IFN-g signaling served during infection were dependent on IFN-g, but not IFN-a, (29), and recent studies have implicated a role for IFNs in driving signaling. the differentiation of hematopoietic progenitor and stem cells (19, 30, 31). To determine whether IFN signaling was responsible for Infection-induced alterations in bone marrow progenitor cells the loss of IL-7Ra2c-KithiSca-12 cells and the emergence of are correlated with anemia and changes in leukocyte the Sca-1+ population, we examined the bone marrow of infected frequencies in the peripheral blood mice deficient for either the IFN-a or IFN-gR. In the absence of To further investigate the role of IFN-g signaling on infection- IFN-a signaling, little effect was noted on the phenotype of Lin2 induced hematopoiesis, we next examined the bone marrow pro- IL-7Ra2 progenitor cells during E. muris infection, as compared genitor cell populations in greater detail. During infection there 2 with infected C57BL/6 mice (Fig. 1B). However, in the absence of was a paucity of Sca-1 cells in C57BL/6 mice. As Sca-1 is known IFN-g signaling, reduced frequencies and numbers of IL-7Ra2c- to be induced by IFN signaling, we used two additional markers, KithiSca-1+ cells were observed postinfection, relative to infected CD34 and FcgRII/III (CD16/32), to characterize myeloid pro- 2 2 2 C57BL/6 mice (Fig. 1C,1D). Moreover, we observed an ∼8-fold genitor subsets within the Lin IL-7Ra c-Kithi Sca-1 and Sca-1+ decrease in the frequency and numbers of IL-7Ra2c-KithiSca-12 fractions of cells (as shown in Fig. 1B). In uninfected mice, cells

FIGURE 1. E. muris infection causes a loss of CMPs in the bone marrow. A, Lin2 IL-7Ra+ and IL-7Ra2 bone marrow cells from mock-infected or E. muris-infected mice (left panels) were analyzed for c-Kit and Sca-1 expression. The regions in the right panels demarcate the CLPs (Lin2IL-7Ra+c-Kitlo Sca-1+), CMPs (Lin2IL-7Ra2c-Kithi Sca-12), and the infection-induced Lin2IL-7Ra2c-KithiSca-1+ progenitor cells. B, The Lin2IL-7Ra2 cells were analyzed for expression of c-Kit and Sca-1 in C57BL/6, IFN-aR–deficient, and IFN-gR–deficient mice. C, The frequencies of the CLPs, CMPs, and Lin2IL-7Ra2c-KithiSca-1+ progenitor cells in the bone marrow were determined in C57BL/6, IFN-aR–deficient, and IFN-gR–deficient mice. The error bars represent the means and SD. D, The number of Sca-12 and Sca-1+ cells within the Lin2IL-7Ra2c-Kithi population is indicated for each strain of mouse. The lineage markers used were the following: CD3, B220, CDllb, Gr-1, and Ter119. The error bars represent the averages and SEM. The bone marrow was harvested from both femurs. The data are representative of at least three separate experiments, where each experiment included at least three mice per group. A Student t test was used to evaluate statistically significant differences between the indicated groups, and significant differences are indicated as follows: *p , 0.05; **p , 0.001. dpi, day postinfection; n.s., not significant. The Journal of Immunology 1035 that are Lin2IL-7Ra2c-KithiSca-12 and CD342FcgRII/III2 are absence of IFN-g–dependent signaling. E. muris infection sup- megakaryocyte-erythrocyte progenitors (MEPs) and the CD34lo presses bone marrow colony formation (23), which is consistent FcgRII/III2 fraction are common myeloid progenitors (CMPs); with the reduced numbers of classically defined myeloid progen- the CD34+FcgRII/III+ population has granulocyte and monocyte itors that we have observed. In contrast to C57BL/6 mice, E. muris potential and are thus referred to as granulocyte monocyte pro- infection did not inhibit colony forming activity in IFN-gR– genitors (GMPs) (28). We observed that, in mock-infected mice, deficient mice; in fact, bone marrow harvested from E. muris expression of FcgRII/III is very low on the Lin2IL-7Ra2c-Kithi infected IFN-gR–deficient mice exhibited increased colony for- Sca-1+ cells but is increased during infection in both C57BL/6 and mation in vitro (Fig. 3A). These data suggest that the IFN-g– IFN-gR–deficient mice. We gated on total Lin2IL-7Ra2c-Kithi dependent changes in the phenotype of myeloid progenitor cells Sca-12 cells and quantified the numbers of classically defined observed during infection are associated with functional differ- MEPs, CMPs, and GMPs (Fig. 2A,2B). We observed that the ences in these bone marrow cells. frequency and number of each of these myeloid populations was We next addressed whether the alterations in progenitor pop- reduced during E. muris infection, with significant reductions in ulations within the bone marrow correlated with changes in the the frequency of CMPs and GMPs in C57BL/6 mice. A significant peripheral blood during infection. Hemoglobin levels began to reduction in the number of CMPs was also noted in IFN-gR– diminish on day 8 postinfection, and they were significantly re- deficient mice. We next examined the frequency and numbers of duced by day 15 postinfection in C57BL/6 mice but not in IFN-gR– infection-induced Sca-1+ MEPs, CMPs, and GMPs (Fig. 2C,2D). deficient mice (Fig. 3B). Platelet numbers were also reduced on As Sca-1 is not normally expressed on MEPs, CMPs, and GMPs, days 8 and 15 postinfection in C57BL/6 mice, but platelets did not each population in this study is referred to as “infection-induced.” decrease substantially in the IFN-gR–deficient mice until day 15 The numbers of infection-induced MEPs, CMPs, and GMPs were postinfection (Fig. 3C). In C57BL/6 mice, hemoglobin concen- significantly increased in C57BL/6 mice during infection, whereas tration and platelet numbers were restored to nearly normal levels IFN-gR–deficient mice exhibited an increase in the CMP pop- by day 30 postinfection (23). Thus, in IFN-gR–deficient mice the ulation only. The increase in Sca-1+ myeloid progenitor cells onset of anemia and thrombocytopenia occurred around the time mirrored the loss of Sca-12 cells, which suggested that the surface of death in infected mice. These data reveal that thrombocytopenia expression of Sca-1 was induced on the classically defined mye- was associated with reduction or loss of classically defined MEPs loid progenitors, as was previously documented during bacterial in C57BL/6 mice but not in IFN-gR–deficient mice, and they infection (2). Thus, IFN-g signaling contributed to the infection- suggest that the loss of the progenitor cell population is at least in induced loss of classically defined myeloid progenitors and the part responsible for the decline in platelet number in the periphery. emergence of a Sca-1+ population of progenitor cells. We also observed changes in the frequency of WBC populations Because the identification of myeloid progenitor cell subsets in the peripheral blood of infected mice. Most striking was the relies on Sca-1 expression, which can be altered during inflam- observation that infection induced an apparent expansion of blood mation, we next evaluated bone marrow colony formation to de- monocytes and neutrophils, as well as a contraction of lympho- termine whether myeloid progenitor cells were altered in the cytes, in the C57BL/6 mice (Fig. 3D). In contrast, IFN-gR–

FIGURE 2. Classically defined myeloid pro- genitor cells are diminished and infection-induced myeloid progenitors dominate the bone marrow compartment during E. muris infection. A, Classi- cally defined myeloid progenitor cells were iden- tified in the Lin2IL7Ra2c-KithiSca-12 bone mar- row cells by examining the expression of CD34 and FcgRII/III in C57BL/6 (top panels) and IFN- gR–deficient (bottom panels) mice. MEPs were identified as CD342FcgRII/III2, CMPs were iden- tified as CD34loFcgRII/III2, and GMPs were CD34+FcgRII/III+, as indicated in the text. B, The total number of each population was determined in mock-infected and E. muris-infected C57BL/6 and IFN-gR-deficient mice. C, Infection-induced myeloid progenitor cell subsets (identified within the Sca-1–expressing population of Lin2IL-7Ra2 c-Kithi cells) were identified based on the pattern of CD34 and FcgRII/III expression used above in C57BL/6 (top panels) and IFN-gR–deficient (bot- tom panels) mock and E. muris-infected mice. D, The numbers of cells within each population were determined in the bone marrow from the femurs. The error bars indicate the averages and SEM. *p , 0.05. 1036 IFN-g–DEPENDENT MYELOPOIESIS

FIGURE 3. IFN-g–dependent reduction of in vitro colony formation and alterations in pe- ripheral blood cell subsets. A, Bone marrow cells were harvested from C57BL/6 and IFN-gR– deficient mice and were analyzed by in vitro colony-forming assay. Total colonies derived from granulocyte-macrophage (CFU granulocyte, macrophage) and multipotential (CFU granulo- cyte, erythrocyte, macrophage, megakaryocyte) progenitor cells were enumerated on day 7 of culture. B–D, Peripheral blood was analyzed at the indicated times postinfection for hemoglobin concentration (B), platelet count (C), and the frequencies of leukocyte subsets (D).

deficient mice exhibited a much reduced expansion of the blood more than half of all cells within the bone marrow. The apparent monocytes, but increased neutrophils. These data are consistent expansion of the CDllbloGr-1lo cells was only noted during infection, with our hypothesis that the observed change in progenitor cell as uninfected C57BL/6 and IFN-gR–deficient mice exhibited very phenotype during infection is a consequence of IFN-g–dependent similar frequencies of granulocytes at each developmental stage. It is changes in hematopoietic progenitor cell function and altered my- possible that the infection-induced expansion of the CDllbloGr-1lo eloid differentiation. cells could have resulted from increased proliferation, decreased cell death, and/or a block in differentiation. Infected IFN-gR–deficient Altered infection-induced myelopoiesis in mice lacking IFN-g mice had fewer B220+ B cells in the bone marrow, as was observed signaling in C57BL/6 mice (Ref. 23 and data not shown), indicating that the During E. muris infection, the bone marrow becomes dedicated to infection-induced alterations in bone marrow B lymphocytes were the production of granulocytes (23). To determine whether the unaffected by the loss of IFN-g signaling. Thus, our data indicate IFN-g–dependent change in the number and phenotype of myeloid that IFN-g signaling is required for myeloid, but not lymphoid, progenitors had a direct impact on the differentiation of myeloid differentiation during infection. cells, we examined the bone marrow of C57BL/6 and IFN-gR– The infection-induced alterations in the bone marrow were deficient mice for its ability to support granulocyte production and reflected in the periphery, as infection increased the frequency and differentiation. To determine the maturation status of the bone mar- number of granulocytes in the spleens of both C57BL/6 and row granulocytes, we monitored cell-surface expression of CD11b IFN-gR–deficient mice (Fig. 4C,4D, Table I). Neutrophils nor- and Gr-1 (Fig. 4A). Promyelocytes express intermediate levels mally represent a small fraction (,2%) of the cells in the spleens of both cell surface molecules (i.e., are CDllbloGr-1lo), whereas of both C57BL/6 and IFN-gR–deficient mice. However, E. muris immature neutrophils are CDllbloGr-1hi and mature neutrophils infection induced a 1.5- to 2-fold increase in the frequency of are CDllbhiGr-1hi (7). The maturation status of these populations spleen granulocytes in C57BL/6 mice, but this corresponded to from mock and infected mice was confirmed cytologically (data a 10- to 12-fold increase in the numbers of CDllbloGr-1lo and not shown). CDllbloGr-1hi cells due to splenomegaly (Fig. 4D). This difference During the first 2 wk of infection, the frequency of granulocytes was more pronounced in infected IFN-gR–deficient mice, which in the bone marrow increased from ∼40%tonearly50%in exhibited an ∼40-fold increase in the number of CDllbloGr-1lo pro- C57BL/6 mice (Table I). However, due to the decrease in bone myelocytic cells, relative to uninfected mice; ∼15% of the cells in the marrow cellularity during infection (Ref. 23 and data not shown), spleens of IFN-gR–deficient mice were CDllbloGr-1lo cells. Thus, the change in number of total granulocytes was not significantly ehrlichial infection causes an expansion of immature and mature different between uninfected and infected mice. We observed granulocyte populations, and these changes are greatly exacer- a significant decrease in the number of CDllbloGr-1lo cells in bated in the absence of IFN-g signaling. These findings are con- C57BL/6 mice on day 8 postinfection, although the numbers sistent with related studies of Mycobacterium-infected mice (16). returned to normal by day 15 postinfection (Fig. 4B). In contrast, Cytological examination of neutrophils in spleen samples from the frequency of granulocytes increased from 37% to nearly 80% the infected mice revealed that the splenic neutrophils in the IFN- in the bone marrow of IFN-gR–deficient mice (Table I). In in- gR–deficient mice, but not C57BL/6 mice, harbored inclusions fected IFN-gR–deficient mice, we observed a significant increase in containing intracellular ehrlichiae (Fig. 4E). This observation was the numbers of CDllbloGr-1lo promyelocytes in the bone marrow by unexpected, as the ehrlichiae are largely monocytotropic and have day 15 postinfection (Fig. 4B). In the infected IFN-gR–deficient never been reported to infect neutrophils (22). Given that many mice, the CDllbloGr-1lo promyelocytic cell population constituted more phenotypically immature neutrophils were detected in the The Journal of Immunology 1037

FIGURE 4. Infection-induced granulopoiesis is altered in mice incapable of IFN-g signaling. A, Bone marrow harvested from C57BL/6 or IFN-gR– deficient mice was analyzed for expression of CDllb and Gr-1. B, The numbers of CDllbloGr-1lo, CDllbloGr-1hi, and CDllbhiGr-1hi cells identified in A are indicated. C and D, Splenocytes from the same mice were analyzed as in A, and the numbers of each population are indicated. E, Spleen cell preparations from mock and day 15 E. muris-infected mice were generated by touch impression and were analyzed by Giemsa staining. Neutrophils in the IFN-gR– deficient (bottom panels), but not in C57BL/6 mice (top panels), contained ehrlichia morulae (arrows). None of the neutrophils in the C57BL/6 mice were infected, but 24.7% of the IFN-gR–deficient neutrophils contained morulae (5 of 19, 3 of 17, and 10 of 33 cells enumerated contained morulae in the IFN- g–deficient mice; three individual fields were counted from three separate mice; original magnification 3100). spleens of IFN-gR–deficient mice, the altered ehrlichial cell tro- of monocytes, making them permissible to infection with E. muris; pism in the IFN-gR–deficient mice may also explain the higher however, these possibilities remain to be addressed. degree of bacterial infection. We interpret the altered bacterial tropism in the IFN-gR–deficient mice to suggest that neutrophils A direct role for IFN-g signaling in infection-induced that develop during infection, in the absence of IFN-g signaling, hematopoiesis can phagocytose but not kill the intracellular bacteria. An alter- It was possible that the persistence of myeloid progenitors, as well native explanation is that the cells identified as neutrophils in the as the enhanced granulopoietic response observed in the absence of spleens of infected IFN-gR–deficient mice have some characteristics IFN-g signaling, was mediated indirectly, perhaps by signaling on 1038 IFN-g–DEPENDENT MYELOPOIESIS

Table I. Frequencies of granulocyte subsets in the bone marrow and spleens of E. muris-infected mice

% Cell Population

Mouse Strain Tissue dpi CDllbloGr-1lo CDllbloGr-1hi CDllbhiGr-1hi Total C57BL/6 Bone marrow 0 12.15 6 1.78 27.13 6 3.11 0.66 6 0.49 39.94 Bone marrow 8 6.26 6 3.08 33.90 62.96 0.49 6 0.11 40.66 Bone marrow 15 18.30 6 3.70 28.27 61.67 0.96 6 0.21 47.50 IFN-gR2/2 Bone marrow 0 11.86 6 2.86 25.03 6 3.04 0.59 6 0.24 37.48 Bone marrow 8 17.73 6 1.19 27.63 6 5.48 0.49 6 0.25 45.86 Bone marrow 15 56.60 6 2.8 21.73 6 2.83 0.37 6 0.14 78.70 C57BL/6 Spleen 0 0.30 6 0.09 0.46 6 0.03 0.02 6 0.003 0.77 Spleen 15 0.07 6 0.07 0.82 6 0.072 0.06 6 0.023 1.53 IFN-gR2/2 Spleen 0 0.44 6 0.11 0.63 6 0.014 0.02 6 0.001 1.084 Spleen 15 12.40 6 1.27 8.07 65.48 0.62 6 0.17 21.09 Data indicate average frequency 6 SD. dpi, day postinfection. nonhematopoietic cells within the bone marrow. To address deficient cells. These data suggest that hematopoietic cells unable whether our observations were due to direct IFN-g signaling on to signal via the IFN-g pathway are retained as immature cells hematopoietic cells, we first generated bone marrow chimeric within the bone marrow, perhaps as a consequence of their inability mice by irradiating IFN-gR–deficient and C57BL/6 mice and to undergo proper differentiation. In contrast, wild-type cells, ca- reconstituting them with congenic wild-type bone marrow. Eight pable of receiving IFN-g–induced signals, were not retained in the weeks post-reconstitution, the chimeric mice were infected with bone marrow, possibly because they differentiate and/or egress E. muris and bone marrow cells were analyzed on day 8 postin- from this site. Thus, we propose that in the absence of IFN-gR fection for progenitor cell phenotype. We found that the increased signaling, myeloid cells are prevented from completing their dif- frequency of Lin2c-Kit+Sca-1+ cells that emerged in infected ferentiation in response to infection-induced stress. wild-type mice was observed in both C57BL/6 and IFN-gR– We next examined the bone marrow and spleens of mock- deficient recipients (Fig. 5), supporting the idea that IFN-g acts infected and E. muris-infected chimeric mice to determine the directly on hematopoietic cells, and not indirectly, via interaction proportions of wild-type and IFN-gR–deficient cells within im- with nonhematopoietic stromal bone marrow cells. mature and mature granulocytic populations. In mock-infected It was possiblethat the enhancedgranulopoietic response observed chimeric mice, IFN-gR–deficient cells predominated within the in the absence of IFN-g signaling could also be a consequence of CDllbloGr-1lo bone marrow population, but postinfection wild- higher bacterial infection and inflammation in the IFN-gR–deficient type cells increased in proportion (Fig. 6F). There was little hosts. To address this concern we next generated radiation-induced change in the composition of the CDllbloGr-1hi population post- mixed bone marrow chimeric mice. Irradiated congenic mice were infection, and the IFN-gR–deficient cells predominated. However, reconstituted with a 1:1 ratio of GFP-expressing C57BL/6 (bACT- within the mature granulocyte population (CDllbhiGr-1hi) in the EGFP; CD45.2) and IFN-gR–deficient (CD45.2) bone marrow. bone marrow, we found that E. muris infection resulted in a sig- Donor cells and host cells were identified by CD45 allelism; nificant increase in the frequency of wild-type cells. Moreover, the wild-type cells (CD45.2+GFP+) were distinguished from IFN-gR– higher proportion of wild-type CDllbhiGr-1hi cells was also ob- deficient cells (CD45.2+GFP2) on the basis of GFP expression served in the periphery: CDllbhiGr-1hi cells were almost entirely (Fig. 6A). When mixed with C57BL/6 cells, the bone marrow- of wild-type origin in the chimeric mice, both prior to and after derived IFN-gR–deficient cells were fully capable of reconstitut- infection (Fig. 6H). There was little effect of IFN-g signaling ing recipient mice, and at 6 wk postreconstitution, IFN-gR–deficient cellsdominated in the peripheral blood (40.85 6 6.38% wild-type versus 55.91 6 6.43% IFN-gR–deficient). Moreover, IFN-gR–de- ficient T lymphocytes were found at higher frequency than wild- type T cells and, notably, double-positive CDllb and Gr-1 myeloid cells were predominantly of wild-type origin (data not shown). The chimeric mice were infected with E. muris, and the fre- quency and phenotype of wild-type and IFN-gR–deficient cells were analyzed within the bone marrow. Bacterial infection in the chimeric mice was similar to those in C57BL/6 mice, indicating that the chimeric mice controlled infection (Fig. 6B). We next addressed whether the donor cells were of wild-type or IFN-gR– deficient origin in the myeloid progenitor cell populations. In mock-infected mice, we observed an increased frequency of IFN- gR–deficient donor cells, relative to wild-type donor cells, within the Lin2IL-7Ra2c-KithiSca-12 (i.e., CMP) population (Fig. 6C, 2 2 FIGURE 5. IFN-g acts on hematopoietic progenitor cells during in- 6D). However, a greater proportion of the Lin IL-7Ra c-Kithi + fection. C57BL/6 mice (CD45.2) and IFN-gR–deficient mice (CD45.2) Sca-1 cells was of wild-type origin. Postinfection, we found that were lethally irradiated and were reconstituted with wild-type congenic the frequency of wild-type cells was greatly diminished within (CD45.1) bone marrow. Eight weeks postreconstitution, mice were either hi 2 hi + 2 both the c-Kit Sca-1 and c-Kit Sca-1 fractions of Lin IL- mock-infected or were infected with E. muris. Bone marrow was analyzed 2 2 2 7Ra cells; these two populations of Lin IL-7Ra myeloid pro- on day 8 postinfection for expression of c-Kit and Sca-1 on Lin2 donor genitors were instead composed almost entirely of IFN-gR– cells. The Journal of Immunology 1039

FIGURE 6. Infection-induced myelopoiesis requires IFN-g signaling. A, A schematic of the experimental strategy and the gating and analysis approach used in the study are shown. Donor bone marrow-derived cells (all CD45.2) were identified, and Lin2IL-7Ra2 cells were further analyzed for GFP expression to identify the wild-type and IFN-gR–deficient donor cells. B, Bacterial infection was evaluated by quantitative PCR in spleen tissue from E. muris-infected C57BL/6, IFN-gR–deficient, and bone marrow chimeric mice. C, Lin2IL-7Ra2 cells were analyzed for expression of c-Kit and Sca-1 (left panels), and GFP expression was analyzed in the c-KithiSca-12 and c-KithiSca-1+ cells (right panels). D, The frequency of C57BL/6 (filled bars) or IFN- gR–deficient (open bars) donor cells within each population is shown. E, Bone marrow was harvested from mock and infected chimeric mice, and the immature and mature granulocyte populations were identified on the basis of expression of CDllb and Gr-1. F, The frequencies of C57BL/6 and IFN-gR– deficient donor-derived cells in bone marrow were determined for each of the indicated granulocyte populations. G and H, Spleens were harvested from mock and infected chimeric mice, and the granulocyte populations and the frequencies of C57BL/6 and IFN-gR–deficient donor-derived cells were de- termined as described for the bone marrow. The data are from one experiment that is representative of two experiments of similar design. on lymphoid lineage chimerism in mock-infected and E. muris- IFN-g–mediated transcriptional control of infection-induced infected mice (data not shown), suggesting that IFN-g acts on myelopoiesis myeloid, but not lymphoid, differentiation during infection. Our Although our data suggested that IFN-g acts directly on hemato- data indicate that IFN-g signaling is not required for the genera- poietic progenitor cells, we wanted to address whether the IFN-g tion of CDllbloGr-1lo cells, but is important for mediating the signaling pathway was activated in the bone marrow promyelocytic terminal differentiation of neutrophils. CD11bloGr-1lo cells that accumulated during infection. Therefore, 1040 IFN-g–DEPENDENT MYELOPOIESIS

CD11bloGr-1lo cells were purified by flow cytometry from mock- and day 15-infected wild-type and IFN-gR–deficient mice (.89% purity). Next, these cells were evaluated for the expression of genes known to be regulated by IFN-g, including the IFN regulatory factors (IRFs), a family of transcription factors that are critical for various aspects of hematopoiesis and immune cell function. The expression of several irf genes was induced during infection, in- cluding irf-1, irf-3, irf-4, irf-5, irf-7,andirf-8.Comparedtotheir expression in wild-type CD11bloGr-1lo cells, transcription of several of the irf genes was not induced in the IFN-gR–deficient cells harvested from infected mice (Fig. 7A), in particular irf-1 and irf-8. We also examined expression of a number of additional IFN-g– regulated genes, including adenosine deaminase acting on RNA (Adar)1,Irgm1, Ifng, Ifngr1, and others (Fig. 7B). Most of these genes, known to be regulated by IFN-g, were induced in wild-type CD11bloGr-1lo promyelocytes during infection, but not in IFN- gR–deficient cells. Thus, infection-induced IFN-g directly enhances myelopoiesis, likely via its ability to induce the expression, in de- veloping promyelocytes, of transcription factors known to be in- volved in hematopoiesis and granulopoiesis. These data also illus- trate that IFN-g is solely responsible for inducing transcription of irf-1, irf-8,Adar,Irgm1,Cxcl10, Mx1, Csf3,andCsf2rb during in- fection, as these genes were not induced in cells that could not FIGURE 7. IFN-g induces the transcription of myeloid-specific genes signal via this pathway. during infection. RNA was isolated from bone marrow CD11bloGr-1lo cells purified from mock-infected or E. muris-infected C57BL/6 or IFN-gR– deficient mice on day 15 postinfection, and mRNA expression was ana- Discussion lyzed. The data indicate gene expression in infected cells, relative to cells In this study, we demonstrate that an intracellular bacterial infection from mock-infected C57BL/6 mice (shaded histograms) or IFN-gR– causes major changes in the differentiation of hematopoietic pro- deficient mice (open histogram). The relative expression of irf1–irf8 (A) and other genes known to be regulated by IFN-g (B) are shown. *p , genitor cells via direct IFN-g signaling. We conclude that while 0.05; **p , 0.001 (for C57BL/6 mice); #p , 0.05; ##p , 0.001 (for IFN- homeostatic production of monocytes and granulocytes is not de- gR–deficient mice). pendent upon IFN-g signaling, infection elicits the production of both populations of myeloid cells, and that this infection-induced process is critically dependent on IFN-g signaling. These findings Thus, one possible explanation is that the Sca-12 cells have reveal a proposed, but largely unrecognized, role for IFN-g in the a greater ability to form colonies in vitro, whereas the infection- direct modulation of hematopoiesis during bacterial infection. Be- induced Lin2IL-7Ra2c-Kithi Sca-1+ population requires addi- cause many pathogens induce IFN-g production, these findings tional signals not provided in the standard in vitro culture con- have implications for understanding host defense against a wide ditions. Furthermore, in our studies of hematopoiesis in mixed range of microbial infections. This conclusion is supported by a bone marrow chimeric mice, we observed that infection caused report that IFN-g induces the expansion and proliferation potential a loss of wild-type cells within the Lin2IL-7Ra2c-Kithi Sca-12 of Lin2Sca-1+c-Kit+ cells (30). Moreover, studies of hematopoiesis and Sca-1+ populations; in these mice the progenitor cells were during malaria infection revealed the emergence of a unique pop- mostly derived from IFN-gR–deficient donor cells. We think that ulation of Lin2IL-7Ra+c-Kithi progenitor cells that was found to this finding is consistent with our interpretation that the wild-type require IFN-g signaling (20). We propose that these observations cells differentiated in response to infection-induced inflammation. together provide a paradigm whereby inflammatory cytokines, such An alternative explanation is that these cells were mobilized and as IFN-g, act to modulate the early differentiation of hematopoietic emigrated from the bone marrow. We also observed high fre- progenitor cells, thereby enhancing the production of the mature quencies of wild-type–derived granulocytes within the bone mar- blood cells required to combat infection. row and spleen of infected mice, suggesting that the wild-type Although our findings have demonstrated an important role for cells that were capable of IFN-g signaling proliferated and differ- IFN-g in vivo, they contrast with earlier studies that described a nega- entiated. Under these same conditions, differentiating promyelo- tive effect of IFN-g on hematopoiesis. In those studies, IFN-g in- cytes that could not signal via IFN-g accumulated in the bone hibited in vitro colony formation of human granulocyte-macrophage marrow and in the spleen. Thus, we propose that IFN-g promotes, progenitor cells (14, 15, 32), and even low levels of IFN-g inhibited not suppresses, hematopoietic differentiation in vivo. hematopoiesis in long-term bone marrow cultures (13). We observed Our studies reveal an important role for IFN-g in directing the that the frequency of in vitro bone marrow colony-forming cells de- development of both granulocytes and monocytes during infection. creased during infection in wild-type mice. Although this finding Although our previous studies of the bone marrow granulocyte would appear to be consistent with the earlier studies that demon- population prompted our investigation of these cells, we noted strated hematopoietic suppression, we propose instead that in vitro alterations to the monocyte population as well. The expansion of colony formation potential decreases because IFN-g promotes he- blood monocytes observed in C57BL/6 mice was reduced in IFN- matopoietic progenitor cell differentiation, thus depleting the bone gR–deficient mice, suggesting that IFN-g is required for the pro- marrow of progenitor cells. In our studies, the classically defined duction of infection-induced monocytes. IFN-gR–deficient mice myeloid progenitor cells were diminished postinfection in wild-type also exhibited pronounced neutrophilia postinfection, as compared mice but not in IFN-gR–deficient mice, and in the wild-type mice, with C57BL/6 mice, suggesting that granulopoiesis does not the frequency of Lin2IL-7Ra2c-KithiSca-1+ progenitors increased. require IFN-g signaling. However, more careful analysis of the The Journal of Immunology 1041 infection-induced granulocytes in the IFN-gR–deficient mice re- an important driving force in responding to infection, this factor may vealed that this population of cells was defective, as these cells also deplete stores of stem and progenitor cells within the hemato- were infected with E. muris. Additionally, wild-type cells that ex- poietic compartment. pressed the IFN-gR dominated the population of mature granulo- Although it is not clear how Sca-1 functions in progenitor cells, it cytes in the chimeric mice. Thus, our data support a requirement is thought that Sca-1 can regulate different signaling pathways by for IFN-g–dependent signaling in the terminal differentiation of modulating lipid raft composition, and it has been shown that both granulocytes and monocytes during ehrlichial infection. overexpression of Sca-1 can inhibit myeloid differentiation (45). These studies also highlight a critical role for IFN-g in driving This observation is inconsistent with our hypothesis that IFN-g the transcriptional regulation of hematopoiesis during infection. drives myelopoiesis during infection. We propose that other genes, IFN-g mediates a range of signaling events within target cells. expressed during infection, modify the effects of Sca-1 signaling As our knowledge of IFN-g signaling has come primarily from invivo,asitiswellknownthatIFN-g signaling is context-dependent. studies of macrophages and other differentiated cell types in vitro, Indeed, the observation that Lin2Sca-1+c-Kit+ cells can be in- how such signals are regulated in hematopoietic progenitor cells duced to proliferate in response to IFN-g (30) suggests that the is unknown. CDllbloGr-1lo cells from infected IFN-gR–deficient outcome of IFN-g signaling may differ in less differentiated pro- mice expressed reduced amounts of irf-1 and irf-8 mRNA, relative genitor cells. In support of this idea, it has been shown that Sca-1 to cells from wild-type mice. IRF-1 and IRF-8 (also known as IFN is required for IFN-a–driven HSC proliferation (19). consensus sequence binding protein) are both known to play Several reports have also revealed that TLR signaling is critical critical roles in hematopoiesis (33). IRF-1 deficiency results in for HSC and progenitor cell differentiation (1, 3, 46–48), indicating increased numbers of immature granulocytic precursors (33), an that some pathogens interact directly with stem and progenitor observation that is similar to what we observed in infected IFN- cells. For example, monopoiesis was impaired in Listeria mono- gR–deficient mice. IRF-8 is required for normal monocyte and cytogenes-infected MyD88-deficient mice (48). We observed that dendritic cell development (34, 35), which is consistent with our MyD88-deficient mice exhibited a phenotype similar to that of observation that fewer blood monocytes, but increased neutrophils, E. muris-infected IFN-gR–deficient mice with respect to the re- were detected postinfection in the IFN-gR–deficient mice. IRF-8 duced Sca-1 upregulation on bone marrow progenitor cells (K. C. deficiency also results in myeloid hyperplasia (36) and in increased MacNamara and G. M. Winslow, unpublished observations). How- numbers of myeloid-derived suppressor cells (37); additionally, ever, E. muris is not known to encode canonical TLR ligands human myeloid malignancies show much lower than normal IRF- (i.e., LPS and peptidoglycan), and we failed to detect any differ- 8 expression (38). IRF-8 also regulates the proliferation-inducing ences in hematopoiesis in infected mice deficient in TLR2, TLR4, effects of GM-CSF by inducing the expression of neurofibromin 1, or TLR9 (K. C. MacNamara and G. M. Winslow, unpublished a tumor suppressor (39, 40). IRF-8 is also required for host de- observations). The TLR-associated signaling molecule MyD88 fense: IRF-8–deficient mice are highly sensitive to Mycobacterium forms a physical association with the IFN-gR, and MyD88 can tuberculosis and Plasmodium chabaudi infections (41, 42). IRF-8 stabilize IFN-g–induced gene transcripts (49). Thus, we propose acts to promote phagocyte development and maturation by in- that an alternative TLR-independent role for MyD88 in infection- ducing the expression of the phagocyte oxidase proteins gp91phox induced hematopoiesis is via the IFN-g pathway or in IL-1b or and gp67phox (43). Our observation that a high frequency of gra- IL-18 signaling. nulocytes contained morulae in the spleens of IFN-gR–deficient The role for IFN-g specifically during periods of stress and mice is consistent with the role of IRF-8 in promoting phagocyte infection is highlighted by the fact that under steady-state con- development during infection. Thus, the induction of irf-8 ex- ditions the bone marrow, blood, and spleen have nearly identical pression during ehrlichia infection likely plays an important role frequencies of leukocytes in both IFN-gR–deficient mice and in regulating myelopoiesis by controlling proliferation, differenti- C57BL/6 mice. However, infection of IFN-gR–deficient mice re- ation, and oxidative activity of phagocytes, and it is likely to be di- sulted in enhanced frequencies and numbers of myeloid progeni- rectly responsible for the defects in myelopoiesis in our infection tor cells, as compared with C57BL/6 mice. In one model of why model. Additionally, our studies show that IFN-g–mediated sig- myelopoiesis is enhanced in the absence of IFN-g, it was proposed naling is critical for regulating irf-8 gene expression during in- that IFN-g acts to suppress the innate immune response by acting fection. on myeloid progenitor cells (16). This model is in conflict with IFN-g–driven hematopoiesis can be viewed as a favorable re- our finding that in mice containing both wild-type and IFN-gR– sponse to infection, due to the production of pathogen-killing leu- deficient cells, most mature neutrophils were found to be of wild- kocytes. However, several studies have suggested that IFN sig- type origin. Our data support the notion that IFN-g induced during naling in HSCs may also have a negative effect. In this study, we infection is required for the production of mature neutrophils. have reported direct action of IFN-g on myeloid progenitor cells, However, an alternative possibility is that IFN-g acts to increase but it is likely that IFN-g also acts on HSCs during ehrlichial the lifespan of mature neutrophils, thus favoring the accumulation infection, as several IFN-g–inducible genes are vital to HSC main- of wild-type cells in the spleen. We propose a model whereby tenance (17, 18). The immunity-related GTPase Irgm1 (also known IFN-g acts directly on progenitor cells and immature cells, thereby as Lrg-47) was induced in HSCs during infection and was required accelerating their movement out of the pool of earlier progenitor for the maintenance of the HSC pool (17). The IFN-g–inducible cells. This would explain why the pool of Lin2IL-7Ra2cKithi gene Adar1 was also shown to be critical for HSC maintenance: cells is composed of IFN-gR–deficient cells in the infected chi- loss of ADAR1 in hematopoietic cells led to rapid apoptosis and meric mice: in the absence of IFN-g signaling they fail to dif- the induction of IFN-responsive genes (18). These studies provide ferentiate and, thus, dominate this pool. strong evidence that IFN-g acts directly on HSCs, and it was re- Our data suggest that the hematological alterations observed cently shown that IFN-g signaling leads to HSC exhaustion during during ehrlichiosis are due to the direct effects of inflammation on chronic mycobacterial infection (31). This conclusion is consistent hematopoietic stem and progenitor cells. We propose that anemia with the observation that HSCs lacking IRF-2, a transcriptional re- and thrombocytopenia, which are hallmarks of ehrlichiosis, occur pressor of type I IFN signaling, undergo exhaustion when chronically at the expense of promoting the production of granulocytes. The stimulated with IFN-a (44). Thus, although IFN-g signaling may be observation that the onset of these hematological abnormalities is 1042 IFN-g–DEPENDENT MYELOPOIESIS delayed in IFN-gR–deficient mice, where we also see a reduced 16.Murray,P.J.,R.A.Young,andG.Q.Daley. 1998. Hematopoietic remodeling in interferon-g-deficient mice infected with mycobacteria. Blood 91: 2914– loss of classically defined myeloid progenitor cells, supports our 2924. conclusion that anemia and thrombocytopenia are a direct result of 17. Feng, C. G., D. C. Weksberg, G. A. Taylor, A. Sher, and M. A. Goodell. 2008. altered functional capacity of myeloid progenitor cells. The cost The p47 GTPase Lrg-47 (Irgm1) links host defense and hematopoietic stem cell proliferation. Cell Stem Cell 2: 83–89. of producing more granulocytes may therefore be the transient 18. Hartner, J. C., C. R. Walkley, J. Lu, and S. H. Orkin. 2009. ADAR1 is essential loss of hemoglobin and platelet production; this outcome may be for the maintenance of hematopoiesis and suppression of interferon signaling. favorable to the host, as wild-type mice survive infection and IFN- Nat. Immunol. 10: 109–115. 19. Essers, M. A., S. Offner, W. E. Blanco-Bose, Z. Waibler, U. Kalinke, gR–deficient mice are unable to control the infection. M. A. Duchosal, and A. Trumpp. 2009. IFNa activates dormant haematopoietic Most infectious diseases cause inflammation and induce the stem cells in vivo. Nature 458: 904–908. expression of IFN-g, so similar alterations in hematopoiesis likely 20. Belyaev, N. N., D. E. Brown, A. I. Diaz, A. Rae, W. Jarra, J. Thompson, J. Langhorne, and A. J. Potocnik. 2010. Induction of an IL7-R+c-Kithi myelo- occur in many infections. A better understanding of how infec- lymphoid progenitor critically dependent on IFN-g signaling during acute tions alter hematopoiesis and progenitor cell function and capacity malaria. Nat. Immunol. 11: 477–485. 21. Olano, J. P., G. Wen, H. M. Feng, J. W. McBride, and D. H. Walker. 2004. is critical not only for our overall understanding of immunity, but Histologic, serologic, and molecular analysis of persistent ehrlichiosis in a mu- for our general knowledge of how hematopoiesis is modulated in rine model. Am. J. Pathol. 165: 997–1006. the face of inflammation and immunologic stress. 22. Paddock, C. D., and J. E. Childs. 2003. Ehrlichia chaffeensis: a prototypical emerging pathogen. Clin. Microbiol. Rev. 16: 37–64. 23. MacNamara, K. C., R. Racine, M. Chatterjee, D. Borjesson, and G. M. Winslow. Acknowledgments 2009. Diminished hematopoietic activity associated with alterations in innate and adaptive immunity in a mouse model of human monocytic ehrlichiosis. We thank Susan Wittmer (Trudeau Institute) and Naomi Walker (University Infect. Immun. 77: 4061–4069. of California, Davis) for excellent technical assistance. We also acknowl- 24. Bitsaktsis, C., B. Nandi, R. Racine, K. C. MacNamara, and G. Winslow. 2007. edge Renjie Song, in the Wadsworth Center Immunology Core Facility, and T-cell-independent humoral immunity is sufficient for protection against fatal Chang-Uk Lim, in the Ordway Research Institute Flow Cytometry Facility, intracellular ehrlichia infection. Infect. Immun. 75: 4933–4941. 25. Stevenson, H. L., J. M. Jordan, Z. Peerwani, H. Q. Wang, D. H. Walker, and for excellent technical expertise. N. Ismail. 2006. An intradermal environment promotes a protective type-1 re- sponse against lethal systemic monocytotropic ehrlichial infection. Infect. Immun. 74: 4856–4864. Disclosures 26. Racine, R., M. Chatterjee, and G. M. Winslow. 2008. CD11c expression iden- The authors have no financial conflicts of interest. tifies a population of extrafollicular antigen-specific splenic plasmablasts re- sponsible for CD4 T-independent antibody responses during intracellular bacterial infection. J. Immunol. 181: 1375–1385. 27. Kondo, M., I. L. Weissman, and K. Akashi. 1997. Identification of clono- References genic common lymphoid progenitors in mouse bone marrow. Cell 91: 661– 1. Singh, P., Y. Yao, A. Weliver, H. E. Broxmeyer, S. C. Hong, and C. H. Chang. 672. 2008. Vaccinia virus infection modulates the hematopoietic cell compartments in 28. Akashi, K., D. Traver, T. Miyamoto, and I. L. Weissman. 2000. A clonogenic the bone marrow. Stem Cells 26: 1009–1016. common myeloid progenitor that gives rise to all myeloid lineages. Nature 404: 2. Zhang, P., S. Nelson, G. J. Bagby, R. Siggins, II, J. E. Shellito, and D. A. Welsh. 193–197. 2008. The lineage2c-Kit+Sca-1+ cell response to Escherichia coli bacteremia in 29. Sinclair, A., B. Daly, and E. Dzierzak. 1996. The Ly-6E.1 (Sca-1) gene requires Balb/c mice. Stem Cells 26: 1778–1786. a39 chromatin-dependent region for high-level g-interferon-induced hemato- 3. Ya´n˜ez, A., C. Murciano, J. E. O’Connor, D. Gozalbo, and M. L. Gil. 2009. poietic cell expression. Blood 87: 2750–2761. Candida albicans triggers proliferation and differentiation of hematopoietic stem 30. Zhao, X., G. Ren, L. Liang, P. Z. Ai, B. Zheng, J. A. Tischfield, Y. Shi, and 2 and progenitor cells by a MyD88-dependent signaling. Microbes Infect. 11: 531– C. Shao. 2010. Brief report: interferon-g induces expansion of Lin Sca-1+C- 535. Kit+ cells. Stem Cells 28: 122–126. 4. Basu, S., G. Hodgson, H. H. Zhang, M. Katz, C. Quilici, and A. R. Dunn. 2000. 31. Baldridge, M. T., K. Y. King, N. C. Boles, D. C. Weksberg, and M. A. Goodell. “Emergency” granulopoiesis in G-CSF-deficient mice in response to Candida 2010. Quiescent haematopoietic stem cells are activated by IFN-g in response to albicans infection. Blood 95: 3725–3733. chronic infection. Nature 465: 793–797. 5. Burg, N. D., and M. H. Pillinger. 2001. The neutrophil: function and regulation 32. Greenberg, P. L., and S. A. Mosny. 1977. Cytotoxic effects of interferon in vitro in innate and humoral immunity. Clin. Immunol. 99: 7–17. on granulocytic progenitor cells. Cancer Res. 37: 1794–1799. 6. Ueda, Y., D. W. Cain, M. Kuraoka, M. Kondo, and G. Kelsoe. 2009. IL-1R type 33. Testa, U., E. Stellacci, E. Pelosi, P. Sestili, M. Venditti, R. Orsatti, A. Fragale, I-dependent hemopoietic stem cell proliferation is necessary for inflammatory E. Petrucci, L. Pasquini, F. Belardelli, et al. 2004. Impaired myelopoiesis in mice granulopoiesis and reactive neutrophilia. J. Immunol. 182: 6477–6484. devoid of interferon regulatory factor 1. Leukemia 18: 1864–1871. 7. Ueda, Y., M. Kondo, and G. Kelsoe. 2005. Inflammation and the reciprocal 34. Tsujimura, H., T. Nagamura-Inoue, T. Tamura, and K. Ozato. 2002. IFN production of granulocytes and lymphocytes in bone marrow. J. Exp. Med. 201: consensus sequence binding protein/IFN regulatory factor-8 guides bone 1771–1780. marrow progenitor cells toward the macrophage lineage. J. Immunol. 169: 8. Ueda, Y., K. Yang, S. J. Foster, M. Kondo, and G. Kelsoe. 2004. Inflammation 1261–1269. controls B lymphopoiesis by regulating chemokine CXCL12 expression. J. Exp. 35. Tamura, T., P. Tailor, K. Yamaoka, H. J. Kong, H. Tsujimura, J. J. O’Shea, Med. 199: 47–58. H. Singh, and K. Ozato. 2005. IFN regulatory factor-4 and -8 govern dendritic cell 9. Cheers, C., A. M. Haigh, A. Kelso, D. Metcalf, E. R. Stanley, and A. M. Young. subset development and their functional diversity. J. Immunol. 174: 2573–2581. 1988. Production of colony-stimulating factors (CSFs) during infection: separate 36. Holtschke, T., J. Lo¨hler, Y. Kanno, T. Fehr, N. Giese, F. Rosenbauer, J. Lou, determinations of macrophage-, granulocyte-, granulocyte-macrophage-, and K. P. Knobeloch, L. Gabriele, J. F. Waring, et al. 1996. Immunodeficiency and multi-CSFs. Infect. Immun. 56: 247–251. chronic myelogenous leukemia-like syndrome in mice with a targeted mutation 10. Zhang, P., A. Iwama, M. W. Datta, G. J. Darlington, D. C. Link, and D. G. Tenen. of the ICSBP gene. Cell 87: 307–317. 1998. Upregulation of interleukin 6 and granulocyte colony-stimulating factor 37. Stewart, T. J., D. J. Liewehr, S. M. Steinberg, K. M. Greeneltch, and S. I. Abrams. receptors by transcription factor CCAAT enhancer binding protein a (C/EBPa) 2009. Modulating the expression of IFN regulatory factor 8 alters the protumori- is critical for granulopoiesis. J. Exp. Med. 188: 1173–1184. genic behavior of CD11b+Gr-1+ myeloid cells. J. Immunol. 183: 117–128. 11. Metcalf, D., C. G. Begley, G. R. Johnson, N. A. Nicola, A. F. Lopez, and 38. Schmidt, M., S. Nagel, J. Proba, C. Thiede, M. Ritter, J. F. Waring, D. J. Williamson. 1986. Effects of purified bacterially synthesized murine multi- F. Rosenbauer, D. Huhn, B. Wittig, I. Horak, and A. Neubauer. 1998. Lack of CSF (IL-3) on hematopoiesis in normal adult mice. Blood 68: 46–57. interferon consensus sequence binding protein (ICSBP) transcripts in human 12. Hirai, H., P. Zhang, T. Dayaram, C. J. Hetherington, S. Mizuno, J. Imanishi, myeloid leukemias. Blood 91: 22–29. K. Akashi, and D. G. Tenen. 2006. C/EBPb is required for “emergency” gran- 39. Zhang, Y. Y., T. A. Vik, J. W. Ryder, E. F. Srour, T. Jacks, K. Shannon, and ulopoiesis. Nat. Immunol. 7: 732–739. D. W. Clapp. 1998. Nf1 regulates hematopoietic progenitor cell growth and Ras 13. Selleri, C., J. P. Maciejewski, T. Sato, and N. S. Young. 1996. Interferon-g signaling in response to multiple cytokines. J. Exp. Med. 187: 1893–1902. constitutively expressed in the stromal microenvironment of human mar- 40. Huang, W., E. Horvath, and E. A. Eklund. 2007. PU.1, interferon regulatory row cultures mediates potent hematopoietic inhibition. Blood 87: 4149– factor (IRF) 2, and the interferon consensus sequence-binding protein (ICSBP/ 4157. IRF8) cooperate to activate NF1 transcription in differentiating myeloid cells. 14. Toretsky, J. A., N. T. Shahidi, and J. L. Finlay. 1986. Effects of recombinant J. Biol. Chem. 282: 6629–6643. human interferon g on hematopoietic progenitor cell growth. Exp. Hematol. 14: 41. Marquis, J. F., R. LaCourse, L. Ryan, R. J. North, and P. Gros. 2009. Dissem- 182–186. inated and rapidly fatal tuberculosis in mice bearing a defective allele at IFN 15. Broxmeyer, H. E., L. Lu, E. Platzer, C. Feit, L. Juliano, and B. Y. Rubin. 1983. regulatory factor 8. J. Immunol. 182: 3008–3015. Comparative analysis of the influences of human g, a and b interferons on 42. Turcotte, K., S. Gauthier, D. Malo, M. Tam, M. M. Stevenson, and P. Gros. 2007. human multipotential (CFU-GEMM), erythroid (BFU-E) and granulocyte- Icsbp1/IRF-8 is required for innate and adaptive immune responses against in- macrophage (CFU-GM) progenitor cells. J. Immunol. 131: 1300–1305. tracellular pathogens. J. Immunol. 179: 2467–2476. The Journal of Immunology 1043

43. Eklund, E. A., and R. Kakar. 1999. Recruitment of CREB-binding protein by Andrian. 2007. Immunosurveillance by hematopoietic progenitor cells traffick- PU.1, IFN-regulatory factor-1, and the IFN consensus sequence-binding protein ing through blood, lymph, and peripheral tissues. Cell 131: 994–1008. is necessary for IFN-g-induced p67phox and gp91phox expression. J. Immunol. 47. Welner, R. S., R. Pelayo, Y. Nagai, K. P. Garrett, T. R. Wuest, D. J. Carr, 163: 6095–6105. L. A. Borghesi, M. A. Farrar, and P. W. Kincade. 2008. Lymphoid precursors are 44. Sato, T., N. Onai, H. Yoshihara, F. Arai, T. Suda, and T. Ohteki. 2009. Interferon directed to produce dendritic cells as a result of TLR9 ligation during herpes regulatory factor-2 protects quiescent hematopoietic stem cells from type I infection. Blood 112: 3753–3761. interferon-dependent exhaustion. Nat. Med. 15: 696–700. 48. Serbina, N. V., T. M. Hohl, M. Cherny, and E. G. Pamer. 2009. Selective ex- 45. Holmes, C., and W. L. Stanford. 2007. Concise review: stem cell antigen-1: pansion of the monocytic lineage directed by bacterial infection. J. Immunol. expression, function, and enigma. Stem Cells 25: 1339–1347. 183: 1900–1910. 46. Massberg, S., P. Schaerli, I. Knezevic-Maramica, M. Ko¨llnberger, N. Tubo, 49. Sun, D., and A. Ding. 2006. MyD88-mediated stabilization of interferon-g- E. A. Moseman, I. V. Huff, T. Junt, A. J. Wagers, I. B. Mazo, and U. H. von induced cytokine and chemokine mRNA. Nat. Immunol. 7: 375–381.