Microbiota-Derived Compounds Drive Steady-State Granulopoiesis via MyD88/TICAM Signaling

This information is current as Maria L. Balmer, Christian M. Schürch, Yasuyuki Saito, of October 1, 2021. Markus B. Geuking, Hai Li, Miguelangel Cuenca, Larisa V. Kovtonyuk, Kathy D. McCoy, Siegfried Hapfelmeier, Adrian F. Ochsenbein, Markus G. Manz, Emma Slack and Andrew J. Macpherson J Immunol 2014; 193:5273-5283; Prepublished online 10 October 2014; Downloaded from doi: 10.4049/jimmunol.1400762 http://www.jimmunol.org/content/193/10/5273 http://www.jimmunol.org/ References This article cites 43 articles, 16 of which you can access for free at: http://www.jimmunol.org/content/193/10/5273.full#ref-list-1

Why The JI? Submit online.

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

by guest on October 1, 2021 • 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 © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Microbiota-Derived Compounds Drive Steady-State Granulopoiesis via MyD88/TICAM Signaling

Maria L. Balmer,* Christian M. Schurch,€ † Yasuyuki Saito,‡ Markus B. Geuking,* Hai Li,* Miguelangel Cuenca,x Larisa V. Kovtonyuk,‡ Kathy D. McCoy,* Siegfried Hapfelmeier,x Adrian F. Ochsenbein,† Markus G. Manz,‡ Emma Slack,*,{ and Andrew J. Macpherson*

Neutropenia is probably the strongest known predisposition to infection with otherwise harmless environmental or microbiota- derived species. Because initial swarming of at the site of infection occurs within minutes, rather than the hours required to induce “emergency granulopoiesis,” the relevance of having high numbers of these cells available at any one time is obvious. We observed that germ-free (GF) animals show delayed clearance of an apathogenic bacterium after systemic challenge. In this article, we show that the size of the myeloid cell pool correlates strongly with the complexity of the intestinal Downloaded from microbiota. The effect of colonization can be recapitulated by transferring sterile heat-treated serum from colonized mice into GF wild-type mice. TLR signaling was essential for microbiota-driven , as microbiota colonization or transferring serum from colonized animals had no effect in GF MyD882/2TICAM12/2 mice. Amplification of myelopoiesis occurred in the absence of microbiota-specific IgG production. Thus, very low concentrations of microbial Ags and TLR ligands, well below the threshold required for induction of adaptive immunity, sets the bone marrow myeloid cell pool size. Coevolution of mammals with their microbiota has probably led to a reliance on microbiota-derived signals to provide tonic stimulation to the systemic innate http://www.jimmunol.org/ immune system and to maintain vigilance to infection. This suggests that microbiota changes observed in dysbiosis, obesity, or antibiotic therapy may affect the cross talk between hematopoiesis and the microbiota, potentially exacerbating inflammatory or infectious states in the host. The Journal of Immunology, 2014, 193: 5273–5283.

n detection of wounding or infection, neutrophils are strates that this rapid response is essential for survival (4). How- recruited and swarm within minutes (1–3). The extreme ever, production of sufficient neutrophils is a costly procedure. An O sensitivity of neutropenic patients to infection demon- adult human produces thousands of neutrophils per second, which during health predominantly senesce (5). The possibility to “tune” by guest on October 1, 2021 the levels of production to the level of threat faced by *Division of Gastroenterology, Department of Clinical Research, University Clinic the organism would therefore appear to be beneficial. for Visceral Surgery and Medicine, University of Bern, 3010 Bern, Switzerland; †Department of Clinical Research, Tumor Immunology, University of Bern, 3010 It is well documented that during severe bacterial infection, Bern, Switzerland; ‡Division of Hematology, University Hospital Zurich, 8091 a process referred to as “emergency myelopoiesis” is induced (5, Zurich, Switzerland; xInstitute for Infectious Diseases, University of Bern, 3010 { 6). In this situation, a much greater proportion of multipotent Bern, Switzerland; and Institute of Microbiology, Swiss Federal Institute of Technology Zurich, 8093 Zurich, Switzerland progenitor cells are diverted into the myeloid lineage at the ex- Received for publication March 26, 2014. Accepted for publication September 9, pense of . This process appears to require sensing 2014. of pathogen-associated molecular patterns in a TLR-dependent A.J.M. was supported by the Swiss National Science Foundation (Grants 310030- manner (7). It is further known that massive production of IFN-g 124732 and 313600-123736), the Canadian Institutes of Health Research, and the can enhance the production of over and Genaxen Foundation. M.L.B. was supported by Oncosuisse and the Swiss National Science Foundation (Grant 313600-123736/1). S.H. and K.D.M. received funding can drive to start at sites outside the bone marrow from the European Research Council under the European Union’s Seventh Frame- (BM) (8). However, increased production of mature granulocytes work Programme (FP/2007-2013)/European Research Council Grants 281785 and requires hours to days in the human system, and although this 281904, respectively. A.F.O. was supported by the Swiss National Science Founda- tion (Grant 133132), Oncosuisse, and the Bernische Krebsliga. C.M.S. was supported increased production is central to the later control and eradication by the Gertrud-Hagmann-Stiftung fur€ Malignomforschung and the SwissLife of infection, the initial encounter with pathogenic or apathogenic Jubila¨umsstiftung. M.G.M. was supported by the Swiss National Science Foun- dation (Grant 310030_146528/1). E.S. (PZ00P3_136742) and M.B.G. were sup- bacteria requires the immediate action of the homeostatically ported by an Ambizione fellowship from the Swiss National Science Foundation. maintained circulating and marginated pools (1, 6). Address correspondence and reprint requests to Prof. Andrew J. Macpherson or We and others have previously noticed that germ-free (GF) mice Dr. Emma Slack, Division of Gastroenterology, Department of Clinical Research, show delayed kinetics of clearance of bacteria given i.v. (9, 10). GF University Clinic for Visceral Surgery and Medicine, University of Bern, Murtenstrasse 35, CH-3010 Bern, Switzerland (A.J.M.) or Institute of Microbiology, HCI G 413, mice are known to have an immature mucosal immune system, Vladimir-Prelog-Weg 1-5/10, CH-8093 Zurich, Switzerland (E.S.). E-mail addresses: including reduced secondary lymphoid tissues, lower levels of [email protected] (A.J.M.) or [email protected] (E.S.) secretory IgA, and fewer intestinal plasma cells (11). In addition, Abbreviations used in this article: BM, bone marrow; CLP, common lymphoid pro- GF mice have lower levels of serum Abs and increased suscepti- genitor; GF, germ-free; GMP, granulocyte- progenitor; HSPC, hematopoi- etic stem/progenitor cell; IVC, individually ventilated cage; LB, Luria–Bertani bility to infection with a number of bacterial pathogens, both in the (media or agar); LCM, low-complexity microbiota; LSK, lineage2, Sca1+, c-kit intestinal tract and systemically (11). It has been observed previ- “high” cells; MEP, -erythrocyte progenitor; RT, room temperature; ously that treatment of specific pathogen-free (SPF), but not GF SPF, specific pathogen-free. mice, with the lipid A–binding antibiotic polymyxin B reduces the Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 number of granulocyte-monocyte colonies formed during in vitro www.jimmunol.org/cgi/doi/10.4049/jimmunol.1400762 5274 THE MICROBIOTA DRIVES STEADY-STATE GRANULOPOIESIS culture of BM (12). Decreased numbers of granulocytes have also tubes, with the outside surface kept sterile, and imported into flexible film 10 been observed in the BM of kanamycin-treated mice (13). More isolators, where 500 ml (10 CFU) was gavaged into the stomachs of GF recent studies have identified increased bone mass and decreased C57BL/6 mice. Inoculum samples were then re-exported from the isolators for bacterial quantification by plating on supplemented agar plates. Fecal numbers in GF mice (14). Thus, we hypothesized that samples exported from the isolator were bacteriologically analyzed to cross talk between the microbiota and the BM is important to monitor HA107 shedding and bacteriological status of the inoculated mice. regulate the size of the steady-state myeloid cell pool. This regu- BrdU proliferation assay latory network may be implicated in the increased susceptibility to infections observed in GF mice and patients with dysbiosis. GF or separately housed SPF C57BL/6 mice were i.p. injected with 1 mg BrdU (BD Biosciences) 36 and 24 h before analysis and retro-orbitally bled 8, 24, or 72 h after the second injection into sterile EDTA tubes. Femurs and Materials and Methods tibias were collected and BM recovered by flushing in 2 ml RPMI 1640. Animal experiments Cells were surface-stained using Pacific blue–anti-Ly6G (Biolegend),

2/2 2/2 PerCP–Cy5.5–anti-Ly6C (eBioscience), allophycocyanin–Cy7–anti-CD11b C57BL/6, MyD88 TICAM1 , and NMRI mice were rederived GF as (Biolegend), and FITC–anti-B220 (Biolegend) Abs in 50 mlPBS/2%BSA, previously described (9, 11), and maintained GF in flexible film isolators at fixed, permeabilized using Cytofix/Cytoperm (BD Biosciences) and Cytofix/ the Clean Animal Facility, University of Bern, Switzerland. C57BL/6 Cytoperm Plus (BD Biosciences) solutions, and intracellularly stained with mouse colonies associated with a low-complexity microbiota (LCM) PE-conjugated anti-BrdU Ab or PE-isotype control (BD Biosciences) fol- consisting of 10–100 different bacterial species (15) were originally gen- lowing the manufacturer’s instructions. Samples were then acquired on a erated by selective colonization of GF mice. Conventionally raised labo- BD LSR II flow cytometer, and data were analyzed using FlowJo Software ratory SPF mice were housed in individually ventilated cages (IVCs) at the (Tree Star). Central Animal Facilities, University of Bern. All animals were housed in Bern, Switzerland, except for LCM mice shown in Figs. 2 and 4, which FACS analysis were maintained in a full-barrier facility in IVC cages at the Swiss Federal Downloaded from Institute of Technology (ETH) Zurich. In all experiments, mice were fed Fifty microliters whole or BM were incubated with 50 ml PBS/ the standard mouse chow Kilba Nafag 3437.P.M.L15 M/R autoclaved at BSA2% containing Pacific blue–anti-Ly6G (Biolegend), PerCP–Cy5.5– either 121˚C (SPF and LCM mice in IVC housing) or 132˚C (GF and anti-Ly6C (eBioscience), allophycocyanin–Cy7–anti-CD11b (Biolegend), gnotobiotic mice). All animal experiments were approved by the local FITC–anti-B220 (Biolegend), and PE–anti-CD54 (Biolegend) Abs, incu- Animal Care Committee. bated 20 min on ice, and RBC lysis was performed using FACSLysis so- lution (BD Biosciences). Cells were washed in PBS/BSA 2%, and all samples were acquired on a BD LSR II Flow Cytometer. Data analysis was Monocolonizations, triple colonizations, antibiotic treatment, http://www.jimmunol.org/ and i.v. injections performed using FlowJo Software (Tree Star), and frequencies were nor- malized to the total number of leukocytes as determined using a VetABC Stable Escherichia coli K-12 JM83 monocolonizations, Enterococcus animal blood counter (Medical Solution) or a hemocytometer (BM). My- faecalis (mouse isolate) monocolonizations, and triple colonizations with eloid cells were identified by expression of CD11b and absence of B220, E. coli K-12 JM83, S. xylosus (mouse isolate), and Enterococcus faecalis and then further subdivided into Ly6G+ granulocytes and Ly6C+ monocytes. were done by intragastric gavage of GF animals with 109 CFUs of pure- cultured bacteria in 500 ml. Inocula were aseptically prepared and imported Bacterial FACS analysis into flexible film isolators. This mixture of bacteria was chosen based on Bacterial FACS analysis was performed as described previously (11, 17). In our previously published work demonstrating absence of pathology in 2/2 2/2 brief, single colonies of plated bacteria were inoculated into 5 ml LB MyD88 TICAM1 animals (9), ease of culture and quantification, and medium and cultured overnight at 37˚C. Bacteria were then gently pelleted to examine two relevant isolates commonly found in SPF mouse colonies for 3 min at 7000 rpm in an Eppendorf minifuge and washed three times by guest on October 1, 2021 alongside an apathogenic Gammaproteobacteria strain. Antibiotic-treated with sterile-filtered PBS/2% BSA/azide before determining the OD600 and mice were administered 1 mg/ml metronidazole, 1 mg/ml ampicillin, 0.5 resuspending at ∼107 bacteria/ml. Mouse serum was diluted 1:10 in PBS/ mg/ml neomycin sulfate, and 1 mg/ml vancomycin in the drinking water 2% BSA/azide and heat-inactivated at 60˚C for 30 min. The serum solution (refreshed every 4 d) for 4 wk before analysis. GF, E. faecalis mono- 7 was then spun at 13,000 rpm in an Eppendorf minifuge for 10 min to colonized, and triple-colonized mice received 10 live E. coli K-12 i.v. into remove any bacteria-sized contaminants, and the supernatant was used to the tail vein. Blood was collected retro-orbitally using sterile glass Pasteur perform serial dilutions. Serum solution and bacterial suspension were pipettes under isoflurane anesthesia (Halocarbon Laboratories) 30 min, 3 h, then mixed at a ratio of 1:1 and incubated at 4˚C for 1 h. Bacteria were or 6 h after injection of live bacteria. Animals were euthanized and spleen, washed twice before resuspending in monoclonal PE anti-mouse IgG1 and liver, lungs, and BM were removed aseptically. Fecal pellets were col- allophycocyanin anti-mouse IgM (BD Pharmingen). After a further hour lected, the cecum was opened, and an aliquot of cecal content was taken. of incubation, the bacteria were washed and then resuspended in 2% Organs were homogenized in 0.5% Tergitol/PBS using a Tissuelyser paraformaldehyde/PBS for acquisition by FACSArray using forward (Qiagen) and sterile stainless-steel ball bearings. Cecal contents, blood, and scatter and side scatter parameters in logarithmic mode. organ suspensions were then plated on Luria–Bertani (LB) containing appropriate antibiotics for overnight culture at 37˚C and CFU counting. Progenitor cell analysis i.v. injection of heat-killed bacteria and serum Femurs and tibias were flushed into 20 ml RPMI 1640, and whole BM cells were counted in a hemocytometer using Trypan blue staining to exclude Eight-week-old GF C57BL/6 mice were exported into sterile IVCs and dead cells. BM lineage depletion was performed using biotinylated Abs 4 injected into the tail vein with heat-killed (20 min at 121˚C) E. coli 10 or against red cell precursors (anti-Ter119), B cells (anti-CD19, clone 6D5), 6 10 diluted in PBS to a final volume of 500 ml. Pooled serum from SPF or T cells (anti-CD3ε, clone 145-2C11), and myeloid cells (anti-Gr1, clone 2/2 2/2 GF JH or RAG mice was sterile filtered and 400 ml injected per RB6-8C5; all Biolegend), MACS anti-biotin beads, and LS columns mouse. PBS was autoclaved, sterile filtered, and 500 ml injected per (Miltenyi Biotec). Cells were then stained with anti-CD127–FITC (SB/ mouse. All mice were analyzed after 24 h, and GF status was confirmed by 199), anti-CD135–PE (A2F10), anti-CD16/32–PE–Cy7, anti-Thy1.1– plating of feces on blood-agar plates and anaerobic incubation for 7 d at allophycocyanin (OX-7), anti–c-kit–allophycocyanin–Alexa750 (all Bio- 37˚C, as well as liquid cultures in thioglycollate medium (#CM0173; legend); anti–Sca-1–PerCP–Cy5.5 (D7) and anti-CD34–eFluor 450 (RAM34; Oxoid) for 2 wk. eBioscience). The hematopoietic stem/progenitor cell (HSPC)–enriched fraction was identified by the lack of lineage markers (lin2) and by the Reversible HA107 colonization and microbiology expression of Sca-1 and c-kit (lineage2, Sca1+, c-kit “high” cells [LSK]). Granulocyte-monocyte progenitors (GMPs) were identified from the my- Reversible colonization was performed as described previously (16). In + 2 2 brief, D-Ala (200 mg/ml)/m-diaminopimelic acid (50 mg/ml)–supple- eloid lineage-restricted progenitors (c-kit , Sca-1 , CD127 ) according mented LB cultures were aseptically inoculated from single colonies of to FcgR and CD34 expression. E. coli HA107 and incubated with shaking at 160 rpm at 37˚C for 18 h. Total BM CFU assay Bacteria were harvested by centrifugation (15 min, 3500 3 g, 4˚C) in a sterile aerosol-proof assembly, washed in sterile PBS, and concentrated For Figs. 2 and 3, total BM cells (3.0 3 103 cells/well) were cultured for to a density of 2 3 1010 CFU/ml in PBS, all performed aseptically under 12–14 d in methylcellulose media (MethoCult M3234; Stemcell Technol- a sterile laminar flow hood. The bacterial suspensions were sealed in sterile ogies) supplemented with mouse IL-3 (10 ng/ml), human IL-6 (10 ng/ml), The Journal of Immunology 5275 mouse (10 ng/ml), mouse GM-CSF (10 ng/ml), mouse E. coli K-12 had been induced in the colonized animals (data not thrombopoietin (10 ng/ml), and human erythropoietin (10 U/ml; all shown), this isotype at mucosal surfaces is not thought to augment Peprotech). The colonies were counted on the basis of their morphological bacterial clearance from the peripheral blood (19), and coloniza- characteristics (18). 2 tion of clean mice induces mucosal IgA independently of serum Lin CFU assay IgG induction (20). In the serum, there was no measurable specific For Fig. 4, a total of 3.3 3 103 MACS-purified lin2 BM cells were IgG or IgM (Fig. 1B, 1C) directed against E. coli K-12 in either plated into MethoCult M3134 medium (Stemcell Technologies) sup- GF or colonized animals. As a positive control, we compared the plemented with 15% FCS, 20% BIT (50 mg/ml BSA in IMDM, 1.44 U/ titer of E. coli–specific IgG present in the serum of E. coli K-12 ml human insulin [Actrapid; Novo Nordisk], and 250 ng/ml human holo monocolonized MyD882/2TICAM12/2 mice, which are known transferrin [Prospec]), 100 mM 2-ME, 100 U/ml penicillin, 100 mg/ml streptomycin, 2 mM L-glutamine, and 50 ng/ml mouse SCF, 10 ng/ml to mount strong systemic Ab responses to their intestinal bacteria mouse IL-3, 10 ng/ml human IL-6, and 50 ng/ml mouse Flt3-ligand (all because of failed mucosal containment (9) (Fig. 1B, inset). Sup- from Prospec). Colonies were enumerated after 7 d ($30 cells/colony) porting the interpretation that the differences in innate immunity on a DMIL inverted microscope (Leica) equipped with an Intensilight likely underlie the delay in systemic vascular bacterial clearance C-HGFI unit (Nikon) (17). in the GF state, our previous data demonstrated an identical Serum lipocalin 2 ELISA phenomenon when we compared i.v. E. coli K-12 challenge of GF Lipocalin 2 ELISAs were performed according to the manufacturer’s wild-type mice and mice colonized with the evolutionarily distant instructions with a few modifications (murine; R&D Systems). Nunc Firmicute E. faecalis (9). No Ab cross-reactivity could be ob- ELISA plates were coated with 50 ml capture Ab (1:200 in PBS) overnight served between these two bacterial strains (9). We also confirmed at 4˚C in a humidified chamber. After washing in PBS/Tween 0.05% experimentally that innate immune responses were enhanced m Downloaded from (Sigma-Aldrich) and blocking in 150 l PBS/BSA 2% for 15 min at room during peripheral blood bacteremia in mice carrying intestinal temperature, samples and standards were added in 3-fold dose titrations starting at 1:10 (serum and standard) and incubated overnight at 4˚C in microbes. After i.v. challenge there were enhanced responses of a humidified chamber. After washing in PBS/Tween 0.05%, 50 ml detec- serum lipocalin 2 (Fig. 1D) and the myeloid-associated tion Ab (1:200 in PBS/BSA 2%) was added and plates were incubated for and chemokines IL-6 (Fig. 1E), MCP1 (Fig. 1F), and TNF 1 h at room temperature. Plates were then washed in PBS/Tween 0.05% (Fig. 1G) in colonized compared with GF mice, despite equal and 100 ml HRP-streptavidin (1:1000 in PBS; Biolegend) was added for 1 h. Plates were then washed and developed with 100 ml of substrate so- steady-state levels of serum lipocalin 2 in unmanipulated animals, http://www.jimmunol.org/ lution (10 ml substrate buffer, 1 mg ABTS, 10 mlH2O2 [both from Sigma- regardless of their colonization status (Fig. 1H). Aldrich]). OD was measured at 415 nm and four-parameter curves were Initial clearance of bacteria from the blood involves both generated to compare EC50 values of samples and standards. and bactericidal serum factors such as the complement Serum measurements cascade. As murine complement activity is difficult to measure because of its labile nature (21) and complement factor production Serum cytokines were measured using cytometric bead arrays (BD Bio- sciences). Serum was 1:4 diluted and incubated for 1 h with 0.5 ml capture has previously been described to be normal in GF and Ag-free beads per sample (Mouse IL-6 flex set B4 #558301, Mouse TNF flex set mice (22, 23), we decided to determine whether the microbiota C8 #558299, Mouse MCP-1 flex set B7 #558342; BD Biosciences). After was linked to steady-state myelopoiesis. We first quantified blood washing in PBS, 0.5 ml PE-detection beads was added per sample and and BM HSPCs (which are found in the LSK fraction), gran- by guest on October 1, 2021 incubated 1 h at room temperature. Samples were then washed in PBS and ulocyte precursor populations, and mature granulocytes (Fig. 2A, acquired on a FACS Array (BD Biosciences). Four-parameter standard curves were fitted for each cytokine standard, and concentrations of 2B) from mice housed under four different hygiene conditions cytokines in the serum samples were calculated. with increasing microbiota complexity: 1) GF, that is, mice housed stringently in isolators that have never encountered live microbes; Serum LPS measurements 2) “triple-colonized mice,” which were generated by experimen- Serum LPS concentrations were determined using the Limulus Assay kit tally colonizing GF mice from pure cultures of E. coli K-12, according to the manufacturer’s protocol (50-647U, 50-648U; Lonza). E. faecalis,andS. xylosus; 3) LCM mice harboring between 20 and Statistical analysis 100 bacterial strains but lacking any Gammaproteobacteria (4, 15) SPF mice, which were bred and maintained with a full microbiota Differences were analyzed for statistical significance using Prism 4 for . Macintosh (GraphPad Software). The details of the test carried out are of 100 bacterial species, but screened to be free of known mouse indicated in the figure legends. Where data were approximately normally pathogens. In agreement with the literature on fully GF and distributed, values were compared using either a Student t test for single antibiotic-treated mice (7, 12, 13), the number of myeloid cells, variable or two-way ANOVA for two variables. Approximate p values mature granulocytes, mature monocytes, and GMPs in BM all were computed for two-way ANOVA. Where data were not normally distributed (e.g., bacterial CFU counts close to or equal to zero), non- increased consistently with increasing microbiota complexity parametric two-tailed Mann–Whitney U tests were applied. In all cases, (Fig. 2C–F), suggesting the existence of a microbiota-dependent p , 0.05 was considered significant. signal that increases myeloid cell production in healthy animals. The LSK cell population was only increased in the presence of Results a complete “SPF” microbiota, suggesting that amplification of this We and others have previously reported that GF mice show im- population requires either a stronger microbiota-associated signal paired clearance of systemically administered bacteria, even when orthepresenceofmicrobialspecies absent from gnotobiotic the bacterial strain given is fully avirulent (9, 10). To confirm these mice (Fig. 2G). The size of the common lymphoid progenitor observations, we compared clearance of an i.v. dose of ampicillin- (CLP) population did not differ significantly with hygiene status resistant E. coli K-12 from the blood of GF C57BL/6 mice, with (Fig. 2H). These changes were also observed in classical CFU clearance from littermates that had been colonized for 21 d with assays of total BM from GF and SPF mice, confirming a differ- E. faecalis, S. xylosus, and ampicillin-susceptible E. coli K-12. At ence in the HSPC frequencies on a functional basis (Fig. 2I). 6 h after injection, GF mice had higher counts of ampicillin-resistant In addition, SPF mice treated with broad-spectrum antibiotics E. coli K-12 in their blood (Fig. 1A). Because E. coli K-12 was showed a corresponding decrease in BM granulocyte and in the present in the intestines of the colonized mice, we first sought to LSK cell numbers (Fig. 3A, 3B). The antibiotic-induced decrease exclude a role for Ag-specific adaptive immunity in the improved in myelopoiesis could also be observed at the functional level in bacterial clearance. Although intestinal secretory IgA specific for total BM CFU assays (Fig. 3C). Thus, the presence of an intestinal 5276 THE MICROBIOTA DRIVES STEADY-STATE GRANULOPOIESIS Downloaded from

FIGURE 1. GF mice show delayed clearance of apathogenic bacteria and blunted inflammatory responses. (A) GF C57BL/6 mice 10 wk of age were triple-colonized by oral gavage of 107 CFU each of E. coli K-12, S. xylosus, and E. faecalis. After 21 d of colonization, triple-colonized mice and GF controls were then i.v. injected with 107 CFU E. coli K-12, and bacterial counts were determined 6 h later in the peripheral blood. Each point represents one mouse from one of two independent experiments, and horizontal lines show geometric means. (B and C) Serum IgG and IgM titers against E. coli K-12 of the mice shown in (A). Dose titrations of serum were incubated with E. coli K-12, and specific IgG- and IgM-binding were visualized by FACS. All curves http://www.jimmunol.org/ represent individual mice from one of two independent experiments. (B, inset)SerumIgGagainstE. coli K-12 from wild-type (open symbols) and Myd882/2 TICAM12/2 mice monocolonized with E. coli K-12. (D) Serum lipocalin 2 values 6 h after i.v. injection of 107 CFU E. coli K-12 in the same mice as in (A) and (B). Each point represents one mouse and horizontal lines show geometric means. (E–G) Serum cytokine levels 30 min, 3 h, and 6 h after i.v. injection of 107 CFU E. coli K-12 in GF (open circles and dashed lines) or triple-colonized (filled circles and continuous lines) mice. Each point represents one mouse from one of two independent experiments. (H) Serum lipocalin 2 values in unmanipulated GF and triple-colonized mice. Each point represents one mouse and horizontal lines show geometric means. Unpaired t test was used to compare GF and triple-colonized mice, *p # 0.05, **p # 0.01. microbiota appeared to specifically amplify myelopoiesis in the established early after colonization, but no steady-state kinetic BM, and with simple microbiota, this amplification occurred differences in myelopoiesis between GF and SPF mice. However, by guest on October 1, 2021 mainly between the production of multipotent progenitors and the fraction of BrdU+ cells within any FACS marker–defined cell their differentiation into GMPs. population is a complex function of influx from precursor pop- We next examined whether changes in BM populations induced ulations, the rate of cell division within the defined population, the by the microbiota were dynamic or represented an irreversible rate of death within the population, and the rate of differentiation developmental shift in response to colonization. Thus, GF mice into subsequent populations (24): any of these components may were treated with the strongly auxotrophic E. coli K-12 mutant differ in GF and colonized animals. Alternatively, because the HA107 (16). This bacterium colonizes the intestine for 12–24 h absolute numerical difference between GMPs in GF and SPF BM postgavage but cannot replicate in vivo, so GF status is regained was ∼2-fold, either transient differences at the time of initial between 24 and 48 h postgavage (16). At 12 h postgavage, an colonization or very small steady-state differences in the prolif- expansion of the BM myeloid cell pool was observed (Fig. 4A, eration, immigration, and emigration rates that are not detectable 4B). However, 14 d postgavage, that is, 13 d after regaining GF within the resolution of our BrdU analysis could also sufficiently status, HA107-treated mice were equivalent to GF mice in their explain the observed data. BM myeloid cell status (Fig. 4A, 4B). Although the changes in Blood granulocytes are terminally differentiated cells that are myeloid progenitor cells did not reach significance in these thought to be incapable of further cell division (postmitotic) and are experiments (Fig. 4C), we could observe an increase in CFU directly derived from the nonreplicating BM granulocyte pool. formation 12 h after HA107 treatment that returned to baseline Thus, the fraction of BrdU+ granulocytes in blood depends only on 14 d after regaining GF status (Fig. 4D). Together, these data the rate of cell influx from the BM pool and the rate of granulocyte imply that changes in myeloid cell production are regulated dy- removal by death or tissue emigration. Because the steady-state namically depending on the current level of microbiota exposure. number of granulocytes in blood is similar in GF and SPF mice In contrast with the BM, no significant differences were observed (Fig. 5B), these two rates must be approximately equal. Thus, the in peripheral blood granulocyte or monocyte numbers between significantly decreased fraction of BrdU+ blood granulocytes that any of the colonization statuses (Fig. 5A–C). We therefore used a is measured 48 h postlabeling in GF mice (Fig. 6C) represents 24-h pulse of BrdU labeling to address the kinetics of granulocyte a decreased rate of influx of granulocytes into the blood and production and release in GF and SPF mice. Although the total a correspondingly decreased rate of clearance of granulocytes number of HSPCs, GMPs, and granulocytes in BM was consis- from the blood in GF mice. tently higher in SPF mice, the fractions of these populations in- Extending this logic, the rate of influx into the blood granulocyte corporating label over time was identical between GF and SPF population must be equal to the rate of efflux from the BM animals (Fig. 6A, 6B). This is consistent with published obser- granulocyte population. However, no difference in the labeling vations (7) and may indicate that different population levels are kinetics between GF and SPF BM granulocytes was observed The Journal of Immunology 5277 Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 2. Size of myeloid pool corresponds to microbiota complexity. (A and B) Gating strategies used in (C)–(H). The HSPC-enriched fraction was defined as LSK. GMPs were identified from the myeloid lineage-restricted progenitors (c-kit+, Sca-12, CD1272) according to FcgR and CD34 expression. Myeloid cells were identified by expression of CD11b and absence of B220, and then further subdivided into Ly6G+ granulocytes and Ly6C+ monocytes. (C–H) Absolute numbers of myeloid cells, granulocytes, monocytes, GMP, LSK, and CLP in the BM of C57BL/6 mice 7–9 wk of age kept GF, colonized for 21 d with E. coli, S. xylosus, and E. faecalis by oral gavage (triple), harboring an LCM or kept under SPF conditions. Each point represents an individual mouse pooled from one to four independent experiments, and horizontal lines show means. One-way ANOVA and Tukey’s posttest were used to compare the groups. *p # 0.05, **p # 0.01, ***p # 0.001. (I) CFU numbers per hind leg (femur + tibia) of C57BL/6 mice kept under GF or SPF conditions. Student t test was performed to compare total or each type of colonies between the groups. *p , 0.05 (for total CFU and CFU-M). BFU-E, burst forming unit- erythrocyte; CFU-G, CFU-granulocyte; CFU-GEMM, CFU-granulocyte/erythrocyte//megakaryocyte; CFU-GM, CFU-granulocyte/macrophage; CFU-M, CFU-macrophage; CFU-Mk, CFU-megakaryocyte.

(Fig. 6D). This may be because of matched decreased differenti- In summary, in GF mice, the BM granulocyte pool and myeloid ation and proliferation rates throughout the whole myeloid line- precursors are present at half to one quarter of the normal numbers age, or may simply be too small an effect to see above biological observed in SPF animals. This corresponds to slower kinetics of noise, as only 1–2% of the total granulocyte population is in the granulocyte turnover in the peripheral blood. Using our current circulation at any one time (25). methods, no change in the kinetics of BM myelopoiesis could be 5278 THE MICROBIOTA DRIVES STEADY-STATE GRANULOPOIESIS

FIGURE 3. Treatment with broad-spectrum antibiotics decreases myelopoiesis. (A and B) Absolute numbers of granulocytes (A) and LSK cells (B) in the BM of C57BL/6 mice maintained with antibiotics (metronidazole, neomycin, ampicillin, and vancomycin: Abx) or without antibiotics (control) in drinking water. Each point represents an individual mouse pooled from three independent experiments, and horizontal lines show means. Student t test was used to compare the groups. (C) CFU numbers per hind leg of C57BL/6 mice kept for 4 wk with antibiotics (Abx) or without antibiotics (control) in drinking water. Data show means + SEM and are pooled from three independent experiments with a total of three or five mice (control or Abx, respectively) per group. Student t test was performed to compare total or each type of colony between the groups. *p , 0.05, ***p , 0.001. BFU-E, burst-forming unit-erythrocyte; CFU-G, CFU-granulocyte; CFU-GEMM, CFU-granulocyte/erythrocyte/macrophage/megakaryocyte; CFU-GM, CFU-granulocyte/macrophage; CFU-M, CFU-macrophage; CFU-Mk, CFU-megakaryocyte. Downloaded from observed between GF and SPF animals. Taken together with the phenomenon could be observed already at 24 h after serum rapid dynamic responses seen during reversible colonization with transfer, indicating a very rapid effect on the myelopoietic lineage. HA107, this suggests that the increased output in steady-state is Correspondingly, these treatments were also associated with an rather related to expansion of the BM that has occurred soon after increase in the fraction of immature granulocytes in the blood, as colonization (shortly after birth) in SPF mice. revealed by a mild “left shift” of the granulocyte population (18) To define how the intestinal microbiota might be communicating (Fig. 7C, 7D). with the BM, we attempted to replicate the colonization phenotype The heat stability of the active serum compound driving mye- http://www.jimmunol.org/ by sterile transfer of serum from SPF mice or GF mice into oth- lopoiesis suggested that microbial products in the serum might be erwise unmanipulated GF mice. This transfer of SPF but not GF necessary to maintain a sufficient myeloid cell pool. This was serum was sufficient to greatly expand the BM myeloid cell pools a surprising suggestion given the known firewall effect of the (Fig. 7A, 7B) without affecting the peripheral blood granulocyte mesenteric lymph nodes and liver in preventing systemic spread of numbers (data not shown). Intriguingly, transfer of an equivalent live microbes (26, 27), and the absence of microbiota-specific volume of serum in which the majority of protein had been fully systemic adaptive immunity in healthy mice (9, 28). We there- denatured (data not shown) and precipitated by boiling at 90˚C for fore hypothesized that only concentrations of microbial products

30 min, followed by high-speed centrifugation and sterile filtra- well below the threshold required to activate systemic adaptive by guest on October 1, 2021 tion, showed an enhanced effect on BM myelopoiesis when immunity would be sufficient to recapitulate the myeloid cell compared with untreated serum, demonstrating that the active expansion observed on colonization. Wild-type GF mice received component is highly heat stable (Fig. 7A, 7B). Of interest, this injections of either 104 or 106 heat-killed E. coli i.v. to test this.

FIGURE 4. Microbiota-driven myelopoiesis is dynamically regulated. Absolute numbers of BM myeloid cells (A), granulocytes (B), GMPs (C), and lin2 CFU (D) in 6- to 8-wk-old C57BL/6 mice. GF mice were gavaged four times over 2 wk with 1010 E. coli HA107 and analyzed 12 h after the last gavage, when the intestine was still colonized (“HA107 12h transiently colonized,” filled squares), or 14 d later, when the mice were GF again (“HA107 d14 GF,” open squares) in com- parison with GF mice (open circles). Pooled data from two independent experiments are shown. Each point represents an individual mouse, and hori- zontal lines show means. One-way ANOVA and Tukey’s posttest were used to compare the groups. *p # 0.05. The Journal of Immunology 5279

FIGURE 5. Peripheral blood myeloid cell numbers are similar with different levels of microbiota complexity. (A–C) Absolute numbers of peripheral blood myeloid cells, granulocytes, and monocytes in C57BL/6 mice at 7–9 wk of age, kept GF, colonized for 21 d with E. coli, S. xylosus, and E. faecalis by oral gavage (triple), LCM, and SPF. Each point represents one individual mouse pooled from one to four independent experiments, and horizontal lines show means. Myeloid cells were identified by expression of CD11b and absence of B220, and then further subdivided into Ly6G+ granulocytes and Ly6C+ monocytes as shown in Fig. 2B.

The lowest dose of heat-killed bacteria accurately recapitulated required for signaling via IL-1 family cytokine receptors, but the effects of colonization 24 h postinjection, whereas higher because these cytokines should be destroyed by the vigorous heat- doses resulted in complex phenotypes, most likely associated with inactivation process used and were undetectable in the heat-treated Downloaded from high levels of cytokine production and rapid release and con- sera (data not shown), these data strongly suggested that heat- sumption of BM granulocytes, as would be expected in sepsis- stable microbial compounds present in SPF mouse serum are associated responses (1, 6, 29) (Fig. 7E, 7F). necessary to drive sufficient steady-state myelopoiesis. Corre- Strong candidates for recognition and signaling downstream of spondingly, MyD882/2TICAM12/2 mice monocolonized with systemic microbial compounds are TLRs (30). Indeed, lower rates E. faecalis did not display improved clearance of E. coli (Fig. 8F). of emergency myelopoiesis have been previously reported in http://www.jimmunol.org/ MyD882/2 animals (5, 18, 29), although continuously elevated Discussion systemic bacterial counts in SPF MyD882/2 mice (9) could be It is now well established that the is required associated with exhaustion of the myeloid system, complicating to maintain homeostasis with the microbiota (9, 31–33). However, interpretation of these results. To determine whether TLR sig- the details of cross talk between the intestinal microbiota and the naling is necessary for microbiota-induced increases in the mye- host are only starting to be realized. In this article, we reveal that loid cell pool, we carried out “triple” recolonizations of GF systemic recognition of microbiota-derived products by TLRs is MyD882/2 TICAM12/2 mice and examined BM myeloid cell necessary to maintain a sufficient pool of BM myeloid cells. This populations 21 d postgavage. In agreement with observations in suggests that coevolution of the host with its resident microbes has by guest on October 1, 2021 SPF mice, no amplification of the BM myeloid cell pool was led to a reliance on microbiota-derived signals to maintain vigi- observed in MyD882/2TICAM12/2 mice (Fig. 8A–C), despite lance to infection. At first, it seems an odd evolutionary trajectory elevated systemic bacterial exposure (9). In addition, transfer of to have been selected, as surely it would be safer to genetically sterile-filtered, boiled serum from wild-type SPF mice into GF encode this level of myeloid cell production within the host ge- MyD882/2TICAM12/2 mice failed to induce any expansion of nome. In contrast, changes in microbiota composition can be the BM myeloid compartment (Fig. 8D, 8E). MyD88 is also strong indicators of increased infectious challenge, for example,

FIGURE 6. BM BrdU labeling follows similar kinetics in GF and SPF mice, but labeled gran- ulocytes accumulate faster in the peripheral blood of SPF mice. (A–D) Time course of appearance of labeled LSK, GMP, and granulocytes in BM (A, B, and D) or peripheral blood (C) after two i.p. injections of 1 mg BrdU in 7-wk-old C57BL/6 GF (open circles and dashed lines) or SPF (filled dots and continuous lines) mice. Data show means 6 SD from two independent experiments with n =3 mice per group. Unpaired t test was used to com- pare the groups. *p # 0.05. Myeloid cells were identified by expression of CD11b and absence of B220, and then further subdivided into Ly6G+ granulocytes and Ly6C+ monocytes (as in Fig. 2B). The HSPC-enriched fraction was defined as lin2, Sca-1+, c-kithi (LSK). GMPs were identified from the myeloid lineage-restricted progenitors (c-kit+, Sca-12, CD1272) according to FcgR and CD34 expression (as in Fig. 2A). 5280 THE MICROBIOTA DRIVES STEADY-STATE GRANULOPOIESIS Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 7. Transfer of serum from SPF mice or low numbers of heat-killed E. coli can mimic the effects of colonization. (A and B) Absolute numbers of BM myeloid cells and granulocytes in 8-wk-old C57BL/6 GF mice before (GF) or 24 h after i.v. injection of 400 ml PBS or sterile-filtered serum (serum- transfer). Serum was obtained from Ab-deficient GF or SPF mice and transferred before (SPF) and after heat inactivation (HI-SPF) for 30 min at 90˚C. Each point represents one mouse, and horizontal lines show means. Data shown are pooled from two independent experiments. *p # 0.05. (C) Percentage of CD11b+, Ly6Glow immature granulocytes among all granulocytes in peripheral blood of the same mice as in (A) and (B). (D) Representative FACS plots of peripheral blood leukocytes, pregated on Ly6G+ and Ly6C2 cells as shown in Fig. 2, 24 h after serum transfer of GF (left panel) or heat-inactivated SPF- serum (right panel). (E and F) Absolute numbers of BM myeloid cells and granulocytes in 8-wk-old GF NMRI mice 24 h after i.v. injection of heat-killed E. coli K-12 in two different concentrations (104 and 106) or PBS control (PBS). Each point represents an individual mouse from one experiment, and horizontal lines show means. One-way ANOVA was used to compare the groups. *p # 0.05. overgrowth of Gammaproteobacteria can be associated with in- products are much stronger stimuli to the systemic innate immune creased susceptibility to intestinal colonization with pathogens of system than others, and this will be an important area for future the same phylum (15). It further seems likely that “microbiota systems-biochemistry research. complexity” alone is not sufficient to explain the stimulus for In this study, animals were not perfused before analysis, and we myeloid cell expansion. Microbial species display highly variable therefore do not directly distinguish between “marginated” and relationships to the host, both with respect to the niche that they “circulating” granulocytes in the BM. Nevertheless, because the colonize (adhesion to epithelial cells, mucus, luminal) and to the concentration of granulocytes in peripheral blood does not differ microbial products that they produce (34). It remains highly significantly between GF and colonized animals, the increase in possible that particular species and their constituent molecular mature BM granulocytes would be likely indicative of an increase The Journal of Immunology 5281 Downloaded from

FIGURE 8. Microbiota-driven myelopoiesis requires MyD88/TICAM1 signaling. (A–C) Percentage of BM myeloid cells, granulocytes, and GMPs in 10- http://www.jimmunol.org/ wk-old GF and triple-colonized wild-type (B6) and MyD882/2TICAM12/2 mice (MyD88/TICAM1). Triple colonizations were performed by oral gavage of 107 CFU each of E. coli K-12, S. xylosus, E. faecalis, and animals analyzed after 21 d. Each point represents an individual mouse from one experiment, and horizontal lines show means. One-way ANOVA and Tukey’s posttest were used, *p # 0.05. (D and E) Absolute numbers of BM myeloid cells and granulocytes in 8-wk-old GF MyD882/2TICAM12/2 mice 24 h after i.v. injection of 400 ml sterile-filtered serum from Ab-deficient SPF mice or 400 mlof the same serum after 30 min of heat-inactivation at 90˚C. Each point represents an individual mouse from one experiment, and horizontal lines show means. (F) Bacterial counts in the spleens 6 h after i.v. injection of 107 CFU E. coli K-12 JM83 into 6- to 8-wk-old MyD882/2TICAM12/2 mice maintained GF or monocolonized with E. faecalis for 21 d. Each point represents an individual mouse from one experiment, and horizontal lines show means. Unpaired t test was used to compare the groups. by guest on October 1, 2021 in the marginated pool of granulocytes, as well as an increased turnover in GF mice (Fig. 6C) is indicative that there is a lower rate rate of release and consumption, as revealed by BrdU labeling. of granulocyte consumption in GF mice. Numerous previous The importance of the marginated pool size is illustrated by hu- studies have indicated that the diurnal pattern of granulocyte re- man individuals with benign ethnic , in whom ex- lease and removal can strongly influence the rate of granulocyte pansion of this pool in the BM is thought to fully compensate the production (38). However, Ab-induced neutropenia was shown to low circulating granulocyte numbers in the peripheral blood (35). increase myelopoiesis identically in GF and SPF mice (7), sug- Interestingly, although we saw no differences in the blood neu- gesting that this mechanism is microbiota independent. Other trophil numbers between GF and colonized mice, as the com- studies have suggested that IL-17 (39, 40) and IFN-g (8) play plexity of the microbiota increased, there was a 2- to 3-fold important roles in steady-state granulopoiesis, and that the pro- expansion of the total BM granulocyte population, which may duction of these cytokines is known to be lower in the intestines of represent the marginated pool, although this compartmentalization GF mice than of SPF animals (41, 42). Additional molecules was not directly addressed. Steady-state granulopoiesis contrasts originating from the microbiota, such as bile acid metabolites or markedly with the situation in emergency myelopoiesis, where the short-chain fatty acids, are also known to shape the host immune marginated pool is rapidly mobilized into the circulation, reducing system, for example, in the differentiation of regulatory T cells the total BM granulocyte numbers, at the same time as massively (39) or the activation status of intestinal (43). Such increasing the rate of granulocyte production at the expense of components are likely to be present in SPF mouse serum and be other hematopoietic lineages (5, 6). In keeping with the homeo- absent from GF mouse serum, and may not be fully inactivated by static nature of colonization-induced myelopoiesis, CLPs are heat treatment; thus, we cannot exclude their role in modulation of unaffected by colonization. A further possible complication in granulocyte production. However, the complete absence of re- interpreting the data is the existence of reverse migration of sponse of MyD882/2TICAM12/2 mice to either recolonization or neutrophils to BM, either in steady-state (36) or as a clearance serum transfer suggests that TLR stimulation is both necessary mechanism for aged cells (37). Nevertheless, the reverse migrating and epistatic to any other regulatory mechanisms. population, where observed, makes up ∼0.25% of circulating Intriguingly, TLR signaling is also essential for induction of neutrophils in a healthy animal, so this phenomenon is unlikely to emergency hematopoiesis (6, 29), suggesting the existence of contribute dramatically to the doubling in the BM granulocyte a quantitative, rather than a “black and white,” switch from the pool observed upon colonization of GF mice. steady-state. It is currently unclear whether there are differences in Within the scope of this study we cannot exclude the role of other TLR sensitivity of different cell types, or differential signaling mechanisms in controlling the size of the myeloid cell pool in GF induced by low and high concentrations of TLR ligands within the and colonized animals. Indeed, the decreased blood granulocyte same cells. Because of technical and ethical limitations in making 5282 THE MICROBIOTA DRIVES STEADY-STATE GRANULOPOIESIS axenic BM chimeras and maintaining them in strict GF conditions, 11. Smith, K., K. D. McCoy, and A. J. Macpherson. 2007. Use of axenic animals in studying the adaptation of mammals to their commensal intestinal microbiota. it was not feasible to use this technique to determine whether the Semin. Immunol. 19: 59–69. stromal or hematopoietic compartment was required to be MyD88 12. Goris, H., F. de Boer, and D. van der Waaij. 1985. Myelopoiesis in experi- sufficient. Future generation of GF MyD88-Flox and cell-type– mentally contaminated specific-pathogen-free and germfree mice during oral administration of polymyxin. Infect. Immun. 50: 437–441. specific Cre-expressing lines (29) would permit accurate dis- 13. Tada, T., S. Yamamura, Y. Kuwano, and T. Abo. 1996. Level of myelopoiesis in crimination of these effects. the bone marrow is influenced by intestinal flora. Cell. Immunol. 173: 155–161. Despite the easily measurable differences in the population sizes 14. Sjo¨gren, K., C. Engdahl, P. Henning, U. H. Lerner, V. Tremaroli, M. K. Lagerquist, F. Ba¨ckhed, and C. Ohlsson. 2012. The gut microbiota reg- of BM myeloid cells, we did not find a difference in the kinetics of ulates bone mass in mice. J. Bone Miner. Res. 27: 1357–1367. replication and differentiation by BrdU labeling. This may be 15. Stecher, B., S. Chaffron, R. Ka¨ppeli, S. Hapfelmeier, S. Freedrich, T. C. Weber, J. Kirundi, M. Suar, K. D. McCoy, C. von Mering, et al. 2010. Like will to like: simply because of a larger BM niche in colonized mice, established abundances of closely related species can predict susceptibility to intestinal at the time of colonization, as suggested by the observed lower bone colonization by pathogenic and commensal bacteria. PLoS Pathog. 6: e1000711. density (14). It remains likely that a transient increase of precursor 16. Hapfelmeier, S., M. A. Lawson, E. Slack, J. K. Kirundi, M. Stoel, M. Heikenwalder, J. Cahenzli, Y. Velykoredko, M. L. Balmer, K. Endt, et al. proliferation and/or differentiation rates would be observed at the 2010. Reversible microbial colonization of germ-free mice reveals the dynamics moment of colonization, accompanied by generation of osteo- of IgA immune responses. Science 328: 1705–1709. clasts and expansion of the BM niche, that subsequently settles 17. Schurch,€ C. M., C. Riether, and A. F. Ochsenbein. 2014. Cytotoxic CD8+ T cells stimulate hematopoietic progenitors by promoting cytokine release from bone into an overall larger myeloid precursor pool that divides and marrow mesenchymal stromal cells. Cell Stem Cell 14: 460–472. replicates with indiscernibly different rates (24). 18. Boettcher, S., P. Ziegler, M. A. Schmid, H. Takizawa, N. van Rooijen, M. Kopf, M. Heikenwalder, and M. G. Manz. 2012. Cutting edge: LPS-induced emergency The presence of microbial products in serum that are capable of myelopoiesis depends on TLR4-expressing nonhematopoietic cells. J. Immunol. stimulating the production of a sufficient BM myeloid cell pool is 188: 5824–5828. an intriguing observation with potential clinical relevance in the 19. Michetti, P., M. J. Mahan, J. M. Slauch, J. J. Mekalanos, and M. R. Neutra. 1992. Downloaded from Monoclonal secretory immunoglobulin A protects mice against oral challenge with management of patients with microbial dysbiosis, or patients on the invasive pathogen Salmonella typhimurium. Infect. Immun. 60: 1786–1792. long-term antibiotic therapy. In combination with the observation 20. Macpherson, A. J., D. Gatto, E. Sainsbury, G. R. Harriman, H. Hengartner, and that steady-state levels of granulocyte production are dynamically R. M. Zinkernagel. 2000. A primitive T cell-independent mechanism of intes- tinal mucosal IgA responses to commensal bacteria. Science 288: 2222–2226. regulated based on the tonic stimulus received when using re- 21. Lachmann, P. J. 2010. Preparing serum for functional complement assays. versibly colonizing auxotrophic E. coli mutants, this suggests J. Immunol. Methods 352: 195–197.

22. Kawakami, M., I. Ihara, A. Suzuki, and Y. Harada. 1982. Properties of a new http://www.jimmunol.org/ a system amenable to medical manipulation. A previous report complement-dependent bactericidal factor specific for Ra chemotype salmonella observed the NOD1 ligand in both serum and BM of colonized in sera of conventional and germ-free mice. J. Immunol. 129: 2198–2201. mice (44), and although a potential effect on myeloid cell numbers 23. Czuprynski, C. J., and E. Balish. 1981. Killing of Listeria monocytogens by conventional and germfree rat sera. Infect. Immun. 33: 348–354. was not addressed, a positive effect of the NOD1 ligand on 24. Bonhoeffer, S., H. Mohri, D. Ho, and A. S. Perelson. 2000. Quantification of cell granulocyte function was shown. Modulation of the intestinal turnover kinetics using 5-bromo-29-deoxyuridine. J. Immunol. 164: 5049–5054. pool of microbial molecules or their systemic derivatives may 25. Semerad, C. L., F. Liu, A. D. Gregory, K. Stumpf, and D. C. Link. 2002. G-CSF is an essential regulator of neutrophil trafficking from the bone marrow to the thus provide a novel way to adjuvant antibiotic therapy for im- blood. Immunity 17: 413–423. proved patient outcomes or to dampen inappropriate inflammatory 26. Macpherson, A. J., E. Slack, M. B. Geuking, and K. D. McCoy. 2009. The responses. mucosal firewalls against commensal intestinal microbes. Semin. Immunopathol. 31: 145–149. by guest on October 1, 2021 27. Balmer, M. L., E. Slack, A. de Gottardi, M. A. Lawson, S. Hapfelmeier, Acknowledgments L. Miele, A. Grieco, H. Van Vlierberghe, R. Fahrner, N. Patuto, et al. 2014. The We thank J. Kirundi, A. Huguenin, and B. Flogerzi for technical support liver may act as a firewall mediating mutualism between the host and its gut commensal microbiota. Sci. Transl. Med. 6:237ra266. and C. Benarafa for helpful comments. 28. Macpherson, A. J., and T. Uhr. 2004. Induction of protective IgA by intestinal dendritic cells carrying commensal bacteria. Science 303: 1662–1665. Disclosures 29. Boettcher, S., R. C. Gerosa, R. Radpour, J. Bauer, F. Ampenberger, M. Heikenwalder, M. Kopf, and M. G. Manz. 2014. Endothelial cells translate The authors have no financial conflicts of interest. pathogen signals into G-CSF-driven emergency granulopoiesis. Blood 124: 1393–1403. 30. Song, D. H., and J. O. Lee. 2012. Sensing of microbial molecular patterns by References Toll-like receptors. Immunol. Rev. 250: 216–229. 1. Kolaczkowska, E., and P. Kubes. 2013. Neutrophil recruitment and function in 31. Vaishnava, S., C. L. Behrendt, A. S. Ismail, L. Eckmann, and L. V. Hooper. 2008. health and inflammation. Nat. Rev. Immunol. 13: 159–175. Paneth cells directly sense gut commensals and maintain homeostasis at the 2. La¨mmermann, T., P. V. Afonso, B. R. Angermann, J. M. Wang, W. Kastenmuller,€ intestinal host-microbial interface. Proc. Natl. Acad. Sci. USA 105: 20858– C. A. Parent, and R. N. Germain. 2013. Neutrophil swarms require LTB4 and 20863. integrins at sites of cell death in vivo. Nature 498: 371–375. 32. Elinav, E., T. Strowig, A. L. Kau, J. Henao-Mejia, C. A. Thaiss, C. J. Booth, 3. Ribeiro-Gomes, F. L., and D. Sacks. 2012. The influence of early neutrophil- D. R. Peaper, J. Bertin, S. C. Eisenbarth, J. I. Gordon, and R. A. Flavell. 2011. Leishmania interactions on the host immune response to infection. Front. Cell. NLRP6 inflammasome regulates colonic microbial ecology and risk for colitis. Infect. Microbiol. 2: 59. Cell 145: 745–757. 4. Lekstrom-Himes, J. A., and J. I. Gallin. 2000. Immunodeficiency diseases caused 33. Cullender, T. C., B. Chassaing, A. Janzon, K. Kumar, C. E. Muller, J. J. Werner, by defects in phagocytes. N. Engl. J. Med. 343: 1703–1714. L. T. Angenent, M. E. Bell, A. G. Hay, D. A. Peterson, et al. 2013. Innate and 5. Takizawa, H., S. Boettcher, and M. G. Manz. 2012. Demand-adapted regulation adaptive immunity interact to quench microbiome flagellar motility in the gut. of early hematopoiesis in infection and inflammation. Blood 119: 2991–3002. Cell Host Microbe 14: 571–581. 6. Manz, M. G., and S. Boettcher. 2014. Emergency granulopoiesis. Nat. Rev. 34. Lozupone, C. A., J. I. Stombaugh, J. I. Gordon, J. K. Jansson, and R. Knight. Immunol. 14: 302–314. 2012. Diversity, stability and resilience of the human gut microbiota. Nature 489: 7. Bugl, S., S. Wirths, M. P. Radsak, H. Schild, P. Stein, M. C. Andre´,M.R.Muller,€ 220–230. E. Malenke, T. Wiesner, M. Ma¨rklin, et al. 2013. Steady-state neutrophil ho- 35. Haddy, T. B., S. R. Rana, and O. Castro. 1999. Benign ethnic neutropenia: what meostasis is dependent on TLR4/TRIF signaling. Blood 121: 723–733. is a normal absolute neutrophil count? J. Lab. Clin. Med. 133: 15–22. 8. de Bruin, A. M., S. F. Libregts, M. Valkhof, L. Boon, I. P. Touw, and 36. Buckley, C. D., E. A. Ross, H. M. McGettrick, C. E. Osborne, O. Haworth, M. A. Nolte. 2012. IFNg induces monopoiesis and inhibits neutrophil devel- C. Schmutz, P. C. Stone, M. Salmon, N. M. Matharu, R. K. Vohra, et al. 2006. opment during inflammation. Blood 119: 1543–1554. Identification of a phenotypically and functionally distinct population of long- 9.Slack,E.,S.Hapfelmeier,B.Stecher,Y.Velykoredko,M.Stoel,M.A.E.Lawson, lived neutrophils in a model of reverse endothelial migration. J. Leukoc. Biol. 79: M. B. Geuking, B. Beutler, T. F. Tedder, W.-D. Hardt, et al. 2009. Innate and 303–311. adaptive immunity cooperate flexibly to maintain host-microbiota mutualism. 37. Rankin, S. M. 2010. The bone marrow: a site of neutrophil clearance. J. Leukoc. Science 325: 617–620. Biol. 88: 241–251. 10. Deshmukh, H. S., Y. Liu, O. R. Menkiti, J. Mei, N. Dai, C. E. O’Leary, 38. Casanova-Acebes, M., C. Pitaval, L. A. Weiss, C. Nombela-Arrieta, R. Che`vre, P. M. Oliver, J. K. Kolls, J. N. Weiser, and G. S. Worthen. 2014. The microbiota N. A-Gonza´lez, Y. Kunisaki, D. Zhang, N. van Rooijen, L. E. Silberstein, et al. regulates neutrophil homeostasis and host resistance to Escherichia coli K1 2013. Rhythmic modulation of the hematopoietic niche through neutrophil sepsis in neonatal mice. Nat. Med. 20: 524–530. clearance. Cell 153: 1025–1035. The Journal of Immunology 5283

39. Arpaia, N., C. Campbell, X. Fan, S. Dikiy, J. van der Veeken, P. deRoos, H. Liu, 42. Ohnmacht, C., R. Marques, L. Presley, S. Sawa, M. Lochner, and G. Eberl. 2011. J. R. Cross, K. Pfeffer, P. J. Coffer, and A. Y. Rudensky. 2013. Metabolites Intestinal microbiota, evolution of the immune system and the bad reputation of produced by commensal bacteria promote peripheral regulatory T-cell genera- pro-inflammatory immunity. Cell. Microbiol. 13: 653–659. tion. Nature 504: 451–455. 43. Chang, P. V., L. Hao, S. Offermanns, and R. Medzhitov. 2014. The microbial 40. Stark, M. A., Y. Huo, T. L. Burcin, M. A. Morris, T. S. Olson, and K. Ley. 2005. metabolite butyrate regulates intestinal macrophage function via histone of apoptotic neutrophils regulates granulopoiesis via IL-23 and IL- deacetylase inhibition. Proc. Natl. Acad. Sci. USA 111: 2247–2252. 17. Immunity 22: 285–294. 44. Clarke, T. B., K. M. Davis, E. S. Lysenko, A. Y. Zhou, Y. Yu, and J. N. Weiser. 41. Jarchum, I., and E. G. Pamer. 2011. Regulation of innate and adaptive immunity 2010. Recognition of peptidoglycan from the microbiota by Nod1 enhances by the commensal microbiota. Curr. Opin. Immunol. 23: 353–360. systemic innate immunity. Nat. Med. 16: 228–231. Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021