C/EBPβ Is Involved in the Amplification of Early Precursors during Candidemia-Induced ''Emergency'' Granulopoiesis This information is current as of October 3, 2021. Sakiko Satake, Hideyo Hirai, Yoshihiro Hayashi, Nobuaki Shime, Akihiro Tamura, Hisayuki Yao, Satoshi Yoshioka, Yasuo Miura, Tohru Inaba, Naohisa Fujita, Eishi Ashihara, Jiro Imanishi, Teiji Sawa and Taira Maekawa

J Immunol 2012; 189:4546-4555; Prepublished online 28 Downloaded from September 2012; doi: 10.4049/jimmunol.1103007 http://www.jimmunol.org/content/189/9/4546 http://www.jimmunol.org/ Supplementary http://www.jimmunol.org/content/suppl/2012/09/28/jimmunol.110300 Material 7.DC1 References This article cites 37 articles, 15 of which you can access for free at: http://www.jimmunol.org/content/189/9/4546.full#ref-list-1

Why The JI? Submit online. by guest on October 3, 2021

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2012 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

C/EBPb Is Involved in the Amplification of Early Granulocyte Precursors during Candidemia-Induced “Emergency” Granulopoiesis

Sakiko Satake,*,†,1 Hideyo Hirai,*,1 Yoshihiro Hayashi,*,1 Nobuaki Shime,†,‡ Akihiro Tamura,* Hisayuki Yao,* Satoshi Yoshioka,* Yasuo Miura,* Tohru Inaba,x Naohisa Fujita,x Eishi Ashihara,*,{ Jiro Imanishi,‖ Teiji Sawa,† and Taira Maekawa*

Granulopoiesis is tightly regulated to meet host demands during both “steady-state” and “emergency” situations, such as infec- tions. The transcription factor CCAAT/enhancer binding protein b (C/EBPb) plays critical roles in emergency granulopoiesis, but the precise developmental stages in which C/EBPb is required are unknown. In this study, a novel flow cytometric method was developed that successfully dissected mouse cells undergoing granulopoiesis into five distinct subpopulations (#1–5) Downloaded from according to their levels of c-Kit and Ly-6G expression. After the induction of candidemia, rapid mobilization of mature and an increase in early granulocyte precursors accompanied by cell cycle acceleration was followed by a gradual increase in granulocytes originating from the immature populations. Upon infection, C/EBPb was upregulated at the protein level in all the granulopoietic subpopulations. The rapid increase in immature subpopulations #1 and #2 observed in C/EBPb knockout mice at 1 d postinfection was attenuated. Candidemia-induced cell cycle acceleration and proliferation of hematopoietic stem/ progenitors were also impaired. Taken together, these data suggest that C/EBPb is involved in the efficient amplification of early http://www.jimmunol.org/ granulocyte precursors during candidemia-induced emergency granulopoiesis. The Journal of Immunology, 2012, 189: 4546–4555.

eutrophilic granulocytes are the major cell type at the demands. Granulopoiesis, the process of granulocyte production front line of host defense (1, 2). Granulocytes are continu- in the BM, is under the control of both intrinsic and extrinsic cel- N ously produced in the bone marrow (BM) and supplied to lular factors. In mice deficient in the transcription factor CCAAT/ the peripheral or tissues, where they fight microorganisms enhancer binding protein (C/EBPa), granulocytes are completely via their bactericidal activities (1, 2). A shortage of granulocytes absent, suggesting a central role for C/EBPa in “steady-state” gran- rapidly causes fatal infections, whereas an excess of granulocytes ulopoiesis (4, 5). During “emergencies” such as infection, the sup- triggers conditions such as acute respiratory distress syndrome and ply of granulocytes from the marginal and BM pool increases, and by guest on October 3, 2021 sepsis-related other acute organ dysfunctions (1, 3). Therefore, the the cells are mobilized to the site of infection. In addition, BM number of granulocytes must be tightly regulated to meet host granulopoiesis is increased in response to various in- cluding IL-3, GM-CSF, and G-CSF (6–9). We and others have *Department of Transfusion Medicine and Cell Therapy, Kyoto University Hospital, previously shown that, in contrast to steady-state granulopoiesis, † Kyoto 606-8507, Japan; Department of Anesthesiology and Intensive Care, Kyoto which is dependent on C/EBPa, emergency granulopoiesis is de- Prefectural University of Medicine, Kyoto 602-8566, Japan; ‡Division of Intensive Care, Kyoto Prefectural University of Medicine, Kyoto 602-8566, Japan; xDepart- pendent on the transcription factor C/EBPb (8, 10). Although ment of Infection Control and Laboratory Medicine, Kyoto Prefectural University of C/EBPa plays a critical role in the transition from common myeloid Medicine, Kyoto 602-8566, Japan; {Department of Clinical and Translational Phys- iology, Kyoto Pharmaceutical University, Kyoto 607-8414, Japan; and ‖Meiji Uni- progenitors (CMPs) to granulocyte- progenitors (GMPs) versity of Integrative Medicine, Kyoto 629-0392, Japan (5), the precise developmental stage at which C/EBPb plays im- 1S.S., H.H. and Y.H. contributed equally to this work. portant roles during emergency granulopoiesis is unknown. Received for publication October 18, 2011. Accepted for publication August 31, All hematopoietic cells, including granulocytes, originate from 2012. a small number of hematopoietic stem cells (HSCs) residing This work was supported partly by grants from the Yasuda Medical Foundation, the within the BM (11, 12). During the process of differentiation, HSCs Fujiwara Foundation, and the Takeda Science Foundation (to H.H.) and a Grant-in- lose their multipotency and their ability for self-renewal and gain Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to H.H., N.S., Y.M., T.I., E.A., and T.M.), the Global lineage-specific functions. Recent advances in flow cytometry have Center of Excellence Program “Center for Frontier Medicine” and Integrative Re- enabled us to prospectively identify the intermediate stages be- search on Cancer Microenvironment Network from the Ministry of Education, Cul- ture, Sports, Science and Technology of Japan, the Kobayashi Foundation for Cancer tween HSCs and mature cells. Long-term HSCs gradually lose the Research, and the Senshin Medical Research Foundation (to T.M.). self-renewability as they differentiate and mature into granulocytes, Address correspondence and reprint requests to Dr. Hideyo Hirai, Department of first becoming short-term HSCs (13) and then multipotent pro- Transfusion Medicine and Cell Therapy, Kyoto University Hospital, 54 Kawahara-cho, genitors (14). Multipotent progenitors then begin to lose their Shogoin, Sakyo-ku, Kyoto 606-8507, Japan. E-mail address: [email protected]. ac.jp multipotency and give rise to specialized progenitors, including The online version of this article contains supplemental material. CMPs and GMPs (15–17). After passing through the GMP stage, Abbreviations used in this article: BC, ; BM, bone marrow; CMP, common granulocytic lineage cells gradually acquire specific features. This myeloid progenitor; GMP, granulocyte-macrophage progenitor; HSC, hematopoietic maturation process is nicely characterized by the development of stem cell; KO, knockout; MEP, –erythroid progenitor; MM, metamye- distinct morphological features and the expression of locyte; MMP9, matrix metallopeptidase 9; WT, wild-type. proteins. Giemsa staining clearly shows distinct differentiation steps Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00 in the progression from , , , www.jimmunol.org/cgi/doi/10.4049/jimmunol.1103007 The Journal of Immunology 4547 (MM), and band cells (BC) to segmented gran- Western blotting ulocytes. Three types of granule proteins are expressed during Sorted cells were diluted in Laemmli sample buffer and boiled for 10 min. specific developmental stages: azurophil granule proteins (primary Samples were then separated by 10% SDS-PAGE and transferred to poly- granules), specific granule proteins (secondary granules), and vinylidene fluoride membranes. A “Can Get” Signal immunoreaction en- gelatinase granule proteins (tertiary granules) (18, 19). However, hancer kit (Toyobo, Osaka, Japan) was used to dilute the primary and 235 188 despite the relatively well-known characteristics found among cells secondary Abs. Abs specific for phospho-C/EBPb (Thr or Thr in mouse) (3084; Cell Signaling Technology), C/EBPb (sc-150; Santa Cruz intermediate between HSCs and specialized progenitors, flow Biotechnology), and GAPDH (sc-25778; Santa Cruz Technology) were cytometric techniques that can dissect the process of granulocyte used as primary Abs. Immunoreactive proteins were detected using HRP- maturation have not been fully established. In this study, we de- conjugated anti-rabbit IgG (NA934V; GE Healthcare) and visualized using veloped a novel flow cytometric method for analyzing the process ECL (GE Healthcare). of mouse granulopoiesis and applied it to the investigation of can- Statistical analysis didemia-induced emergency granulopoiesis to determine the de- Statistical differences were determined using the Student t test. A p velopmental stages that require C/EBPb. value , 0.05 was considered statistically significant. Materials and Methods Mice Results Flow cytometric dissection of mouse granulopoiesis C57BL/6 mice (8–12 wk old) were purchased from Shimizu Laboratory Supplies (Kyoto, Japan) or CLEA Japan (Tokyo, Japan). Mice deficient in To elucidate the molecular mechanisms involved in granulopoi- C/EBPb (20) were bred and maintained under specific pathogen-free esis, we devised a novel flow cytometry method. The strategy is Downloaded from conditions in Kyoto University. Littermates were always used as controls shown in Fig. 1A. First, cells that have lost the potential to give rise for the experiments involving C/EBPb knockout (KO) mice. Approval for to granulocytes are removed from the target population (Fig. 1B– the animal protocols was obtained from the Committee on Animal Re- search of the Kyoto University Faculty of Medicine. D). Cells expressing lineage markers for T (CD4 and CD8), B lymphocytes (CD19 or B220), and erythroid cells (Ter119) Flow cytometric analysis and cell sorting are excluded, together with propidium iodide-stained dead cells high int Peripheral blood samples were obtained from the tail vein, and blood cell (R1 in Fig. 1B). Because a discrete population of SSC FSC count and differential were analyzed with a SE-9000 instrument (Sysmex, cells (R2 in Fig. 1C) was enriched in CD11b+Ly-6G27/4intSSChigh http://www.jimmunol.org/ Kobe, Japan). For flow cytometric analysis of peripheral blood, RBCs were eosinophilic granulocytes ( (9); Supplemental Fig. 1A–C), lysed by treatment with Pharm Lyse reagent (BD Biosciences). BM cells these cells were also removed. When the cells within R3 (Fig. 1C) were stained with fluorescence-conjugated Abs and analyzed either using high a FACSCalibur or a FACSAria instrument (BD Biosciences). The Abs used were analyzed for the expression of CD34 and c-Kit, the c-Kit were FITC-conjugated anti-CD34 (RAM34); PE-conjugated anti–Ly-6G CD34low cells (R4 in Fig. 1D) were megakaryocyte–erythroid (1A8); allophycocyanin-conjugated anti–c-Kit (2B8); and PerCP-Cy5.5– progenitors (MEP) that have been previously defined as c-Kithigh conjugated lineage markers, including CD4 (RM4-5), CD8 (53-6.7), CD19 CD34lowFcgRIII/IIlowlin2 cells (see Supplemental Fig. 1D) (15). (1D3), B220 (RA3-6B2), and Ter119. Allophycocyanin-conjugated anti– c-Kit PE-Cy7–conjugated anti–Sca-1, PE-conjugated anti-CD16/32, and Thus, R4 cells were also eliminated from the analysis. The re-

FITC-conjugated anti-CD34 Abs were used for staining HSCs and myeloid maining cells (R5) were then analyzed for the expression of c-Kit by guest on October 3, 2021 progenitor cells. All Abs were purchased from BD Biosciences, eBio- and Ag 6 complex, locus G (Ly-6G; a member of the science, BioLegend, or Caltag Laboratories. To exclude dead cells, BM Ly-6 family of GPI-anchored proteins), which is specifically ex- cells were stained with propidium iodide. Data were analyzed using the FlowJo software (Tree Star). pressed by granulocytes (Fig. 1E) (22). During the process of granulocytic differentiation, cells are expected to acquire Ly-6G Infection with Candida albicans expression and gradually lose the expression of c-Kit. The cells Experiments involving infection with C. albicans were carried out as within R5 were then divided into five cell subpopulations (#1–#5), 2 previously described, with minor modifications (21). Briefly, C. albicans as shown in Fig. 1E. Subpopulation #1 was c-KithighLy-6G ,#2 (18804; American Type Culture Collection) was plated onto Luria–Bertani was c-KitintLy-6G2, and #5 was c-KitlowLy-6Ghigh discrete pop- broth agar plates, and the plates were stored at 4˚C for a maximum of 4 wk. ulation. The remaining cells between #2 and #5 were divided Before each experiment, several colonies were picked from the plate and grown in 3 ml Sabouraud dextrose broth (Sigma-Aldrich, St. Louis, MO) at equally into subpopulations #3 and #4, according to the expression low/neg 2 37˚C for 24 h. The fungi were washed twice with pyrogen-free PBS and levels of Ly-6G (Fig. 1C). The c-Kit Ly-6G population resuspended in PBS, and 4 3 106 CFU/20 g body weight/mouse were i.v. comprised cells that stained negative for all other markers and injected via the tail vein to induce disseminated candidiasis. was enriched in expressing M-CSF receptor (Supple- Quantitative RT-PCR mental Fig. 1E, 1F). During granulopoiesis, the cells are hypoth- esized to progress from stage 1 to 5 (in that order). Total RNA from sorted cells was extracted using an RNeasy Micro Kit (Qiagen), according to the manufacturer’s instructions. RNA was reverse Five subpopulations dissected by flow cytometry precisely transcribed and subsequently amplified using a LightCycler System reflect the stepwise progression of granulopoiesis (Roche Diagnostics) programmed with the following parameters: 95˚C for 10 min, followed by 45 cycles at 95˚C for 10 s and 60˚C for 30 s. The PCR To validate the flow cytometric dissection, the morphological mixture contained TaqMan Master Mix (Roche Diagnostics), cDNA, the characteristics of these five subpopulations were analyzed. After relevant primer pairs, and a TaqMan probe (Universal Probe Library, sorting, the cells were stained with Wright–Giemsa solution (Fig. Roche Diagnostics). The sequences of the primers and probes are listed in Table I. The expression level of each gene was normalized against that of 2A). Subpopulation #1 comprised mainly myeloblasts with large, GAPDH. round nuclei and a narrow, basophilic cytoplasm (71.5 6 3.5%). Subpopulation #2 contained an abundance of PMs with oval- In vivo BrdU incorporation analysis shaped nuclei and a wider, less basophilic cytoplasm (78.5 6 2.1%). Mice were injected i.p. with 1 mg BrdU 1 h before euthanasia. BM cells Segmentation of the nuclei became gradually more evident be- were harvested and stained with fluorescent-conjugated Abs. Labeled BM tween stages 3 and 4 (myelocytes, 71.5 6 5.0% in subpopulation cells were fixed, permeabilized, and treated with DNase to expose the 6 incorporated BrdU using a BrdU Flow Kit (BD Biosciences). Cells were #3; MM, 64.0 7.1% in subpopulation #4); subpopulation #5 then stained with an FITC-labeled anti-BrdU and analyzed using a mainly comprised cells with donut-shaped nuclei (BC and seg- FACSCalibur instrument. mented cells; 60.5 6 2.1 and 29.5 6 3.5%, respectively). Cyto- 4548 INVOLVEMENT OF C/EBPb IN EMERGENCY GRANULOPOIESIS

Table I. Primers and probes used for quantitative RT-PCR

Gene Forward Primer Reverse Primer Universal Probe (No.) Cathepsin G 59-CAACGGTTCTGGAAAGATGC-39 59-CTTCTCGGCCTCCAATGAT-39 15 Myeloperoxidase 59-GGAAGGAGACCTAGAGGTTGG-39 59-TAGCACAGGAAGGCCAATG-39 7 Elastase 2 59-ACTCTGGCTGCCATGCTACT-39 59-GCCACCAACAATCTCTGAGG-39 107 Proteinase 3 59-AGCTACCCATCCCCCAAG-39 59-TCGTGCCCACCTACAATCTT-39 4 Lactoferrin 59-GGGCAAGTGCGGTTTAGTT-39 59-CCATTGCTTTTGGAGGATTT-39 53 MMP9 59-ACGACATAGACGGCATCCA-39 59-TGTCGGCTGTGGTTCAGTT-39 77 C/EBPa 59-CCTTCAACGACGAGTTCCTG-39 59-TGGCCTTCTCCTGCTGTC-39 11 C/EBPb 59-ATCGACTTCAGCCCCTACCT-39 59-TAGTCGTCGGCGAAGAGG-39 55 GAPDH 59-TGTCCGTCGTGGATCTGAC-39 59-CCTGCTTCACCACCTTCTTG-39 80 plasmic staining became weaker as the cells moved toward stage expressed after the stage, was highly expressed within 5. These morphological changes within the defined subpopulations subpopulation #4 (Fig. 2C). The gelatinase matrix metallopeptidase correlate well with granulocytic maturation. 9, a granule protein expressed at the MM and BC stages (18, 23), Next, the expression of granule proteins by each population was was expressed in subpopulation #4 and, at much higher levels, in examined (Fig. 2B–D, Table I), because granule proteins are #5 (Fig. 2D). The expression patterns of the different granule known to be expressed only at specific stages in differentiation proteins within the defined subpopulations, together with the ob-

(19). First, the expression of azurophil granule proteins, including served morphological characteristics, confirmed that the process of Downloaded from cathepsin G, proteinase 3, myeloperoxidase, and elas- granulopoiesis progresses from subpopulations #1 to #5 in this tase 2, which are synthesized at the promyelocytic stage, was ex- order. These results validate the flow cytometric method for the amined (Fig. 2B). The level of cathepsin G transcripts was high in analysis of granulopoiesis. subpopulations #1 and #2, much lower in #3, and almost unde- Flow cytometric analysis of emergency granulopoiesis tectable in #4 and #5. This expression pattern was similar for all the azurophil granule proteins tested. Lactoferrin, a granule protein To assess changes in granulopoiesis during the early phase of an infection, the flow cytometry methodwasappliedtoamouse http://www.jimmunol.org/ candidemia model. C. albicans (4 3 106 CFU/20 g body weight) was i.v. injected into mice and peripheral WBC counts were taken on days 0–4 (Fig. 3A, 3B). The number of WBCs and gran- ulocytes, as well as the frequency of CD11b+Ly-6G+ granulocytes, increased during the observation period, suggesting that gran- ulocyte production and mobilization were significantly enhanced, as described previously (8, 21). The BM cells from infected mice

were then analyzed on days 0–4 postinfection (Fig. 4A). BM cells by guest on October 3, 2021 from mice in the steady state were always analyzed before flow cytometric analysis of cells from infected mice. The gates were then set as described in Fig. 1 and applied to the analysis of BM cells from infected mice. Downregulation of c-Kit was observed in subpopulations #2–#5 during the early phase of infection (days 1 and 2); therefore, the gates were adjusted on with the y-axis (corresponding to c-Kit expression). To verify whether the five subpopulations defined during the steady state were still present during this emergency situation, the expression levels of various granule proteins were measured within each subpopulation on days 0, 2, and 4 postinfection. As shown in Fig. 3C, the expression profile of each of the granule proteins was identical to that ob- served during the steady state (Fig. 2B–D), suggesting that the flow cytometric method was also valid for the analysis of emergency granulopoiesis. Candidemia-induced dynamic changes in granulopoiesis The flow cytometric profiles of BM cells during candidemia from days 0 to 4 are shown in Fig. 4A. The cell numbers per 5 3 105 BM cells within each subpopulation are shown in Fig. 4B. The number of total BM cells did not significantly change throughout the observation period (data not shown); the cell numbers shown FIGURE 1. Flow cytometric analysis of murine granulopoiesis. (A) in Fig. 4B reflect the absolute numbers in the BM. On day 1, the B E Strategy used for flow cytometric analysis of mouse granulopoiesis. ( – ) cell number (per 5 3 105 BM cells) in subpopulation #1 (which Staining and gating of mouse BM cells. First, cells that had lost the potential + B D included c-Kit hematopoietic stem/progenitor cells) significantly to give rise to granulocytes were removed from the target population ( – ). 6 6 6 , The remaining cells (R5) were then analyzed for expression of c-Kit and Ly- increased from 8,076 868 (mean SD) to 12,776 831 (p 6G (E). Subpopulation #1 was c-KithighLy-6G2, #2 was c-KitintLy-6G2,and 0.01; n = 3), and the cells in subpopulation #5 decreased from #5 was a discrete c-KitlowLy-6Ghigh population. The remaining cells between 125,675 6 13,707 to 71,996 6 15,920 (p , 0.01; n = 3). The cell #2 and #5 were divided into subpopulations #3 and #4 according to their number in subpopulation #5 started to increase on day 2, whereas expression levels of Ly-6G. (E). Ery, Erythrocyte; PI, propidium iodide. the increase in cell number for subpopulations #2, #3, and #4 The Journal of Immunology 4549 Downloaded from http://www.jimmunol.org/ by guest on October 3, 2021

FIGURE 2. The flow cytometric dissection precisely reflects the differentiation and maturation stages of the granulocytes. (A) Wright–Giemsa staining and differential counting of the sorted subpopulations (#1–#5) from Fig. 1E (original magnification 3400). (B)–(D) show mRNA levels for the azurophil granule proteins (primary granules), specific granule proteins (secondary granules), and gelatinase granule proteins (tertiary granules), respectively. Data are representative of three independent experiments. MB, ; MC, myelocyte; MPO, myeloperoxidase; MMP9, matrix metalloproteinase 9; PM, ; SC, segmented cell. began after day 3. Within the granulopoietic compartment (Fig. granulopoietic precursors (subpopulation #1) had already begun to 4C), the frequency of subpopulation #5 cells decreased, all the increase, and therefore, cell cycle changes at these early times other subpopulation cell levels increased on day 1, and the timing of were examined. BM cells were harvested 1 h after i.p. injection of the next peak level (arrowhead in Fig. 4C) gradually progressed from BrdU on day 1 postinfection (Fig. 5A). The cells in subpopulations #1 to #5. These data suggest that, upon infection, mature gran- #1–#3 had already incorporated BrdU. In contrast, very few ulocytes (#5) were immediately mobilized from the BM and that numbers of the cells in subpopulation #5 showed BrdU uptake the enhancement of granulopoiesis was initiated in subpopulation during the steady state (day 0 in Fig. 5B, 5C). By 24 h after the #1 on day 1, and this population differentiated to stage 5 over time. induction of candidemia, the frequency of BrdU-positive cells within subpopulations #1 and #2 significantly increased (from The cell cycle was accelerated in subpopulations #1 and #2 in 28.3 6 4.5 to 41.8 6 4.1%, p , 0.05; and 40.6 6 2.0 to 53.5 6 the early phase of emergency granulopoiesis 3.1%, p , 0.01, respectively), whereas BrdU incorporation by To further investigate the dynamic changes in the cell populations subpopulations #3–#5 was not affected, suggesting that the during the candidemia-induced emergency response, the cell cycle enhancement of granulopoiesis in response to infection was reg- status of the five subpopulations was analyzed using in vivo BrdU ulated at the level of the earliest granulocytic precursors, the cells labeling experiments. On day 1 postinfection, the numbers of early in subpopulations #1 and #2. 4550 INVOLVEMENT OF C/EBPb IN EMERGENCY GRANULOPOIESIS Downloaded from http://www.jimmunol.org/ by guest on October 3, 2021 FIGURE 3. The robustness of the flow cytometric methods in candidemia-induced emergency granulopoiesis. (A) Total WBC counts and granulocyte numbers in the peripheral blood of mice after the induction of candidemia. CD11b+Ly-6G+ cells in the peripheral blood are also shown in (B). (C) mRNA levels for azurophil granule proteins (primary granules), specific granule proteins (secondary granules), and gelatinase granule proteins (tertiary granules) on days 0, 2, and 4 after the induction of candidemia. Data are representative of two independent experiments. MMP9, Matrix metalloproteinase 9; MPO, myeloperoxidase.

Candidemia upregulates C/EBPb protein in all granulopoietic candidemia in C/EBPb KO mice (Fig. 7). On day 1 postinfection, subpopulations an increase in the levels of subpopulations #1 and #2 cells (from 6 6 6 6 To determine the precise developmental stage at which C/EBPb 3.6 1.6 to 10.2 2.4% and 4.7 0.8to11.6 4.1%, re- 6 plays a critical role, the mRNA and protein levels of C/EBPb were spectively) and a decrease in subpopulation #5 cells (from 66.3 6 measured in each of the subpopulations on days 0 and 1 postin- 2.9 to 41.2 11.5%) were observed in wild-type (WT) mice (Fig. fection (Fig. 6A, 6B). During the steady state, the mRNA levels of 7A, left panels,Fig.7B).InC/EBPb KO mice, the distribution C/EBPb were upregulated in line with the progression of granu- of cells at steady state was identical to that of WT mice (Fig. 7A, locytic maturation (from subpopulations #1 to #5; Fig. 6A). The upper panels, 7B). In contrast, the increases in subpopulations #1 mRNA levels on day 1 postinfection were unchanged in all the and #2 cells and the decreases in subpopulation #5 cells were sig- 6 6 6 granulopoietic subpopulations. nificantlyimpairedinKOmice(from3.1 1.3 to 5.9 1.8, 4.5 6 6 6 In addition to transcriptional regulation, C/EBPb is also regu- 1.0 to 6.4 1.5, and 67.5 4.7 to 58.6 6.5%, respectively; Fig. lated at the translational and posttranslational levels, including the 7A, right panels, 7B). The levels of mRNA expression for granule phosphorylation of a threonine residue (Thr188 in mice) (24). proteins within each subpopulation from WT and KO mice were Western blotting analysis revealed that both phospho-C/EBPb and identical at both steady state and during infection (Supplemental total C/EBPb increased in all the granulopoietic subpopulations by Fig. 2). When the cell cycle status of these models was evaluated day 1 postinfection. These data suggested that C/EBPb is upreg- using in vivo BrdU labeling experiments, incorporation of BrdU in ulated at the translational level across all stages of granulopoiesis subpopulations #1 and #2 in C/EBPb KO mice was always slightly during the early phase of an infection. lower than in WT mice, but the differences were not statistically significant. These findings suggest that C/EBPb is required for b C/EBP is required for candidemia-induced increases in the efficient proliferation of early granulocytic precursors. early granulopoietic subpopulation b Our previous findings showed that emergency granulopoiesis in- C/EBP is involved in the amplification of early granulocyte duced by candidemia is impaired in C/EBPb KO mice (8). To precursors further characterize the defect in granulopoiesis of C/EBPb KO To elucidate the roles of C/EBPb in the proliferation of the early mice, the novel flow cytometric method was used to examine granulopoietic subpopulations, the HSC and myeloid progenitor The Journal of Immunology 4551

FIGURE 4. Flow cytometric analysis of candidemia-induced granulopoiesis. (A) Flow cytometric analysis of candi- demia-induced emergency granulopoi- esis. BM cells were harvested on days 0–4 after i.v. injection of C. albicans and analyzed by flow cytometry. (B) Chro- nological changes in the cell numbers of each population per 5 3 105 BM cells after the onset of infection. (C) Chronological changes in the frequency Downloaded from of the indicated granulopoietic subpopu- lations within the granulopoietic com- partment, including subpopulations #1– #5. Arrowheads indicate the peak levels after day 2 within the observation pe- riod. Data are representative of three in- , dependent experiments (n =3;*p http://www.jimmunol.org/ 0.05). by guest on October 3, 2021

compartments were analyzed in WT and KO mice (Figs. 8, 9). were identical between WT mice and C/EBPb KO mice during the The frequency and number of c-Kit+Sca-1+lineage markers2 HSC steady state and were not significantly affected on day 1 postin-

FIGURE 5. Cell cycle changes during emergency granulopoiesis. (A) In vivo BrdU incorporation assay. Flow cytometric analyses of the BrdU-positive cells in each population are shown in (B) and (C). Data are representative of two independent experiments (n = 3; *p , 0.05). 4552 INVOLVEMENT OF C/EBPb IN EMERGENCY GRANULOPOIESIS

FIGURE 6. C/EBPb expression in granulocytic subpopulations. C/EBPb mRNA levels were measured by quantitative RT-PCR (A), and protein levels were measured by Western blotting (B) before (day 0) and 1 d postinfection. Data are representative of at least two independent experiments (n =3;*p , 0.05). Downloaded from fection (Fig. 8A–C). Myeloid progenitors, including CMP, macrophage-granulocyte progenitors (GMP), and MEP, were detected at the same frequency in WT mice and C/EBPb KO mice before infection with C. albicans (Fig. 8D–F). Induction of can-

didemia significantly increased the frequency and number of GMP http://www.jimmunol.org/ in WT mice, and these increases were significantly attenuated in C/EBPb KO mice. At steady state, the cell cycle status of HSC and myeloid progenitors (as assessed by in vivo BrdU labeling ex- periments) was identical in both WT and C/EBPb KO mice (Fig. 9A, 9B). Upon induction of candidemia, the frequency of BrdU- positive cells in the HSC and CMP populations from WT mice increased significantly; however, an increase of BrdU-positive cells was observed only within the HSC compartment in C/EBPb KO mice and at a lower level than that in WT mice (Fig. 9A, 9B). These by guest on October 3, 2021 results suggest that C/EBPb is involved in the efficient prolifer- ation of HSCs and myeloid progenitors.

Discussion This paper describes a novel flow cytometric method for the analysis of mouse granulopoiesis and its use in elucidating the molecular mechanisms underlying emergency granulopoiesis. The prospective identification and isolation of distinct cellular populations is of fundamental importance, especially in the fields of hematology and immunology, and flow cytometry is a powerful tool for this purpose. For successful analysis of a specific hema- topoietic lineage, the choice of lineage-specific surface Ag is a central issue in identifying the cell population. In this study, Ly-6G was used as a marker for granulocytic differentiation. An anti- granulocyte receptor 1 Ab (RB6-8C5), which has been widely used for the detection of mouse granulocytes (21, 25, 26), rec- FIGURE 7. Requirement of C/EBPb in the proliferation and differenti- ognizes both Ly-6G and Ly-6C (22, 27). However, the expression ation of granulocytes in the early phase of candidemia. (A) Flow cytometric of Ly-6C is not restricted to granulocytes, and so, the Ab cross- analysis of WT and C/EBPb KO mice on days 0 and 1 postinfection. (B) reacts with nongranulocytic lineages such as dendritic cells (28), The frequencies of subpopulations #1–#5 within the granulopoietic com- C monocytes (29, 30), and a subpopulation of lymphocytes (31). In partment on days 0 and 1. ( ) In vivo BrdU incorporation. The frequency of the BrdU-positive cells in each population is shown. Data are representative contrast, Ly-6G, which was cloned as a member of the Ly-6 al- of at least two independent experiments (n =3;*p , 0.05). loantigen family (32), is expressed exclusively on granulocytes (22, 33), and treatment with an anti–Ly-6G Ab specifically depletes granulocytes in vivo (34, 35). Although the physiological tions identified by our flow cytometry method precisely reflect the ligand and biological function of Ly-6G are unclear, the gran- stepwise differentiation of granulocytes, confirming that this method ulocyte-specific expression of Ly-6G makes it an ideal marker is useful for analyzing stage-specific gene expression and func- of granulocytic differentiation. To our knowledge, this study shows tion during both steady state and emergency granulopoiesis. for the first time that Ly-6G expression correlates with the degree Granulopoiesis is a continuous process for producing sufficient of granulocytic differentiation and maturation. The five subpopula- numbers of granulocytes for the demands of the host during both The Journal of Immunology 4553 Downloaded from http://www.jimmunol.org/

FIGURE 8. C/EBPb is required for an efficient increase in myeloid progenitors in response to candidemia. (A) Flow cytometric analysis of lineage marker (lin)2 BM cells from WT and C/EBPb KO mice on days 0 and 1 postinfection. Numbers indicate the percentages of c-Kit+Sca-12 cells (left panels) and c-Kit+Sca-1+ cells within the lin2 cell population. The frequency (B) and number (per 1 3 106 BM cells) (C) of c-Kit+Sca-1+lin2 HSC are shown. (D) Flow cytometric analysis of c-Kit+Sca-12lin2 BM cells from WT and C/EBPb KO mice on days 0 and 1 postinfection. Numbers indicate the percentages of CMP, GMP, and MEP within the c-Kit+Sca-12lin2 BM cell population. The frequency (E) and cell number (per 1 3 106 BM cells) (F) of the myeloid , , progenitors are shown. K, KO; W, WT. Data are representative of at least two independent experiments (n =4;*p 0.05, Dp 0.10). by guest on October 3, 2021 steady state and emergency conditions. In this study, upon infec- The expression of C/EBPb is regulated at the transcriptional, tion, mice exhibited an immediate (day 1) decrease in subpopu- translational, and posttranslational levels; posttranslational mod- lation #5 and an increase in subpopulation #1. Subsequently, ifications include phosphorylation and sumoylation (36). We have increases in the intermediate subpopulations were observed (Fig. previously shown that C/EBPb transcripts were upregulated after 4A, 4B). These results suggested that the immediate mobilization stimulation or infection (8). However, in this study, of mature granulocytes from the BM was followed by a more C/EBPb mRNA levels were unchanged in all the subpopulations, gradual enhancement of granulopoiesis originating from changes although the protein levels were markedly upregulated on day 1 in the earliest granulopoietic precursors. The cell cycle status of postinfection. In fact, in the current series of experiments, C/EBPb each subpopulation was also evaluated. A cell cycle acceleration mRNA levels were substantially upregulated by day 4 postinfection was observed in subpopulation #1 on day 1 and was accompanied (data not shown), which suggests that C/EBPb is upregulated at the by an increase in the number of cells in subpopulation #1. A cell translational level as an immediate response and, subsequently, at cycle acceleration was also observed in subpopulation #2 on day 2 the transcription level as a prolonged response. but did not result in an increase in cell numbers within subpop- Our previous work identified distinct roles for C/EBPb and ulation #2 (Figs. 4B, 5C). However, it did lead to an increase in C/EBPa during granulopoiesis (8). C/EBPa is prerequisite during cell numbers within subpopulation #5 on day 2. Acceleration of steady-state granulopoiesis; naive mice deficient for C/EBPa lack the cell cycle within a subpopulation did not necessarily correlate granulocytes in healthy steady-state conditions (4, 5). In vivo well with an increase in the cell number of that subpopulation. production of granulocytes can be induced in C/EBPa KO mice When each subpopulation was sorted and subjected to propidium only in the presence of cytokine stimulation (8). In contrast, gran- iodide staining after fixation, sub-G1 cells were rarely observed, ulopoietic responses to cytokine stimulation or infection are im- either in the presence or absence of infection; this suggests that paired in C/EBPb KO mice, whereas steady-state granulopoiesis apoptosis was not involved in the process (Supplemental Fig. 3). is normal, suggesting a specific requirement for C/EBPb during Taken together, these results suggest that cell cycle acceleration emergency granulopoiesis (8, 10). In agreement with these previ- during granulopoiesis is tightly coupled to extremely rapid dif- ous findings, we showed in this study that C/EBPb is dispensable ferentiation and maturation (from subpopulations #2 to #5). Such in steady-state granulopoiesis but is clearly required for efficient am- fine-tuning of the proliferation and differentiation of granulocytic plification of early granulopoietic subpopulation #1 and #2 cells, on precursors is requisite for emergency granulopoiesis, and our flow day 1 postinfection. Subpopulation #1 comprised more immature cytometric method provides an excellent platform for investigat- hematopoietic cells, including HSCs and myeloid progenitors. ing the underlying regulatory mechanisms. Further analysis of HSC and myeloid progenitors clearly showed 4554 INVOLVEMENT OF C/EBPb IN EMERGENCY GRANULOPOIESIS Downloaded from

FIGURE 9. C/EBPb is involved in cell cycle ac- celeration in HSC and progenitors during emergency granulopoiesis induced by candidemia. (A)Invivo BrdU incorporation assay. Flow cytometric analyses of the BrdU-positive cells in each population are shown http://www.jimmunol.org/ in (A) and (B). HSC, c-Kit+Sca-1+lin2 HSC. Data are representative of two independent experiments (n =4; *p , 0.05). by guest on October 3, 2021

that C/EBPb is necessary for cell cycle acceleration in both HSC C/EBPb in the proliferation of granulocytic precursors is neces- and CMP and for the increase in GMP numbers during the early sary to understand the molecular mechanism underlying emergency phase of candidemia-induced emergency granulopoiesis (Figs. 8, 9). granulopoiesis. Previous findings showed that C/EBPb and C/EBPa share the Granulopoiesis involves the loss of self-renewal by HSCs, pro- ability to induce granulocyte differentiation but that the strong in- liferation, and lineage specification, followed by the maturation hibitory effect on the cell cycle is specific to C/EBPa (8, 37), and acquisition of specific functions. As increased levels of C/EBPb which is downregulated during emergency granulopoiesis (data not protein were observed in all the granulopoietic subpopulations, shown) (8). The defects in emergency granulopoiesis observed in C/EBPb may play a role in many of these processes. In this study, C/EBPb KO mice may be attributable to an imbalance between the expression levels of granule proteins by each subpopulation proliferation and differentiation of granulocytic precursors caused were not affected in the absence of C/EBPb, suggesting that up- by the downregulation of C/EBPa in addition to the loss of regulation of C/EBPb is not directly involved in the differentiation C/EBPb. Further investigation of the target molecules driven by or maturation of granulocytes in response to infection. Therefore, The Journal of Immunology 4555 we have not yet fully elucidated the stage-specific significance 14. Adolfsson, J., O. J. Borge, D. Bryder, K. Theilgaard-Mo¨nch, I. Astrand- Grundstro¨m, E. Sitnicka, Y. Sasaki, and S. E. Jacobsen. 2001. Upregulation of of C/EBPb during emergency situations. We are currently inves- Flt3 expression within the bone marrow Lin‑Sca1+c-Kit+ stem cell compartment tigating the role of C/EBPb in HSC regulation and granulocyte is accompanied by loss of self-renewal capacity. Immunity 15: 659–669. function during emergencies using C/EBPb KO mice. 15. Akashi, K., D. Traver, T. Miyamoto, and I. L. Weissman. 2000. A clonogenic common myeloid progenitor that gives rise to all myeloid lineages. Nature 404: In summary, our novel flow cytometric analysis method revealed 193–197. that C/EBPb is involved in the efficient amplification of early 16. Manz, M. G., T. Miyamoto, K. Akashi, and I. L. Weissman. 2002. Prospective granulocytic precursors. The method developed in this study will isolation of human clonogenic common myeloid progenitors. Proc. Natl. Acad. Sci. USA 99: 11872–11877. provide an excellent platform for studies investigating granulopoi- 17. Adolfsson, J., R. Ma˚nsson, N. Buza-Vidas, A. Hultquist, K. Liuba, C. T. Jensen, esis, will further elucidate the molecular mechanisms involved in D. Bryder, L. Yang, O. J. Borge, L. A. Thoren, et al. 2005. Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised the shift from C/EBPa- to C/EBPb-dependency during emergency road map for adult blood lineage commitment. Cell 121: 295–306. granulopoiesis, and will facilitate a greater understanding of the 18. Borregaard, N., O. E. Sørensen, and K. Theilgaard-Mo¨nch. 2007. Neutrophil hematopoietic system as a part of host defense. granules: a library of innate immunity proteins. Trends Immunol. 28: 340–345. 19. Ha¨ger, M., J. B. Cowland, and N. Borregaard. 2010. Neutrophil granules in health and disease. J. Intern. Med. 268: 25–34. 20. Screpanti, I., L. Romani, P. Musiani, A. Modesti, E. Fattori, D. Lazzaro, Acknowledgments C. Sellitto, S. Scarpa, D. Bellavia, G. Lattanzio, et al. 1995. Lymphoproliferative We thank Y. Nakagawa, M. Katakami, and Y. Manabe for excellent tech- disorder and imbalanced T-helper response in C/EBPb-deficient mice. EMBO J. nical assistance and all the laboratory members in the Department of 14: 1932–1941. 21. Basu, S., G. Hodgson, H. H. Zhang, M. Katz, C. Quilici, and A. R. Dunn. 2000. Transfusion Medicine and Cell Therapy for valuable discussions. “Emergency” granulopoiesis in G-CSF‑deficient mice in response to Candida albicans infection. Blood 95: 3725–3733. 22. Fleming, T. J., M. L. Fleming, and T. R. Malek. 1993. Selective expression of Disclosures Ly-6G on myeloid lineage cells in mouse bone marrow: RB6-8C5 mAb to Downloaded from The authors have no financial conflicts of interest. granulocyte-differentiation antigen (Gr-1) detects members of the Ly-6 family. J. Immunol. 151: 2399–2408. 23. Chakrabarti, S., J. M. Zee, and K. D. Patel. 2006. Regulation of matrix metalloproteinase-9 (MMP-9) in TNF-stimulated : novel pathways References for tertiary granule release. J. Leukoc. Biol. 79: 214–222. 1. Christopher, M. J., and D. C. Link. 2007. Regulation of neutrophil homeostasis. 24. Nakajima, T., S. Kinoshita, T. Sasagawa, K. Sasaki, M. Naruto, T. Kishimoto, Curr. Opin. Hematol. 14: 3–8. and S. Akira. 1993. Phosphorylation at threonine-235 by a ras-dependent

2. Summers, C., S. M. Rankin, A. M. Condliffe, N. Singh, A. M. Peters, and mitogen-activated protein kinase cascade is essential for transcription factor http://www.jimmunol.org/ E. R. Chilvers. 2010. Neutrophil kinetics in health and disease. Trends Immunol. NF-IL6. Proc. Natl. Acad. Sci. USA 90: 2207–2211. 31: 318–324. 25. Lagasse, E., and I. L. Weissman. 1996. Flow cytometric identification of murine 3. Burg, N. D., and M. H. Pillinger. 2001. The neutrophil: function and regulation neutrophils and monocytes. J. Immunol. Methods 197: 139–150. in innate and humoral immunity. Clin. Immunol. 99: 7–17. 26. Guibal, F. C., M. Alberich-Jorda, H. Hirai, A. Ebralidze, E. Levantini, A. Di 4. Zhang, D. E., P. Zhang, N. D. Wang, C. J. Hetherington, G. J. Darlington, and Ruscio, P. Zhang, B. A. Santana-Lemos, D. Neuberg, A. J. Wagers, et al. 2009. D. G. Tenen. 1997. Absence of granulocyte colony-stimulating factor signaling Identification of a myeloid committed progenitor as the cancer-initiating cell in and neutrophil development in CCAAT enhancer binding protein a-deficient acute promyelocytic leukemia. Blood 114: 5415–5425. mice. Proc. Natl. Acad. Sci. USA 94: 569–574. 27. Nagendra, S., and A. J. Schlueter. 2004. Absence of cross-reactivity between 5. Zhang, P., J. Iwasaki-Arai, H. Iwasaki, M. L. Fenyus, T. Dayaram, B. M. Owens, murine Ly-6C and Ly-6G. Cytometry A 58: 195–200. H. Shigematsu, E. Levantini, C. S. Huettner, J. A. Lekstrom-Himes, et al. 2004. 28. Shortman, K., and S. H. Naik. 2007. Steady-state and inflammatory dendritic- Enhancement of repopulating capacity and self-renewal cell development. Nat. Rev. Immunol. 7: 19–30.

in the absence of the transcription factor C/EBP a. Immunity 21: 853–863. 29. Jutila, D. B., S. Kurk, and M. A. Jutila. 1994. Differences in the expression of by guest on October 3, 2021 6. Cheers, C., A. M. Haigh, A. Kelso, D. Metcalf, E. R. Stanley, and A. M. Young. Ly-6C on neutrophils and monocytes following PI-PLC hydrolysis and cellular 1988. Production of colony-stimulating factors (CSFs) during infection: separate activation. Immunol. Lett. 41: 49–57. determinations of macrophage-, granulocyte-, granulocyte-macrophage-, and 30. Jutila, M. A., F. G. Kroese, K. L. Jutila, A. M. Stall, S. Fiering, L. A. Herzenberg, multi-CSFs. Infect. Immun. 56: 247–251. E. L. Berg, and E. C. Butcher. 1988. Ly-6C is a /macrophage and 7. Watari, K., S. Asano, N. Shirafuji, H. Kodo, K. Ozawa, F. Takaku, and endothelial cell differentiation antigen regulated by interferon-g. Eur. J. Immu- S. Kamachi. 1989. Serum granulocyte colony-stimulating factor levels in healthy nol. 18: 1819–1826. volunteers and patients with various disorders as estimated by enzyme immu- 31. Kung, J. T., M. Castillo, P. Heard, K. Kerbacher, and C. A. Thomas, III. 1991. noassay. Blood 73: 117–122. Subpopulations of CD8+ cytotoxic T cell precursors collaborate in the absence of 8. Hirai, H., P. Zhang, T. Dayaram, C. J. Hetherington, S. Mizuno, J. Imanishi, conventional CD4+ helper T cells. J. Immunol. 146: 1783–1790. K. Akashi, and D. G. Tenen. 2006. C/EBPb is required for “emergency” gran- 32. Fleming, T. J., C. O’hUigin, and T. R. Malek. 1993. Characterization of two ulopoiesis. Nat. Immunol. 7: 732–739. novel Ly-6 genes: protein sequence and potential structural similarity to 9. Ueda, Y., D. W. Cain, M. Kuraoka, M. Kondo, and G. Kelsoe. 2009. IL-1R type a-bungarotoxin and other neurotoxins. J. Immunol. 150: 5379–5390. I-dependent hemopoietic stem cell proliferation is necessary for inflammatory 33. Zhang, X., L. Majlessi, E. Deriaud, C. Leclerc, and R. Lo-Man. 2009. Coac- granulopoiesis and reactive neutrophilia. J. Immunol. 182: 6477–6484. tivation of Syk kinase and MyD88 adaptor protein pathways by bacteria pro- 10. Akagi, T., T. Saitoh, J. O’Kelly, S. Akira, A. F. Gombart, and H. P. Koeffler. motes regulatory properties of neutrophils. Immunity 31: 761–771. 2008. Impaired response to GM-CSF and G-CSF, and enhanced apoptosis in 34. Daley, J. M., A. A. Thomay, M. D. Connolly, J. S. Reichner, and J. E. Albina. C/EBPb-deficient hematopoietic cells. Blood 111: 2999–3004. 2008. Use of Ly6G-specific monoclonal antibody to deplete neutrophils in mice. 11. Osawa, M., K. Hanada, H. Hamada, and H. Nakauchi. 1996. Long-term lym- J. Leukoc. Biol. 83: 64–70. phohematopoietic reconstitution by a single CD34-low/negative hematopoietic 35. Abbitt, K. B., M. J. Cotter, V. C. Ridger, D. C. Crossman, P. G. Hellewell, and stem cell. Science 273: 242–245. K. E. Norman. 2009. Antibody ligation of murine Ly-6G induces , 12. Matsuzaki, Y., K. Kinjo, R. C. Mulligan, and H. Okano. 2004. Unexpectedly blood flow cessation, and death via complement-dependent and independent efficient homing capacity of purified murine hematopoietic stem cells. Immunity mechanisms. J. Leukoc. Biol. 85: 55–63. 20: 87–93. 36. Nerlov, C. 2010. Transcriptional and translational control of C/EBPs: the case for 13. Yang, L., D. Bryder, J. Adolfsson, J. Nygren, R. Ma˚nsson, M. Sigvardsson, and “deep” genetics to understand physiological function. Bioessays 32: 680–686. ‑ ‑ S. E. Jacobsen. 2005. Identification of Lin Sca1+kit+CD34+Flt3 short-term he- 37. Duprez, E., K. Wagner, H. Koch, and D. G. Tenen. 2003. C/EBPb: a major PML- matopoietic stem cells capable of rapidly reconstituting and rescuing myeloab- RARA‑responsive gene in retinoic acid-induced differentiation of APL cells. lated transplant recipients. Blood 105: 2717–2723. EMBO J. 22: 5806–5816.