Am J Physiol Gastrointest Liver Physiol 309: G181–G192, 2015. First published June 11, 2015; doi:10.1152/ajpgi.00053.2015.

Selective enrichment of commensal gut bacteria protects against Citrobacter rodentium-induced colitis Linda Vong,1 Lee J. Pinnell,1 Pekka Määttänen,1 C. William Yeung,1 Eberhard Lurz,1 and Philip M. Sherman1,2 1

Cell Biology Program, Research Institute, Division of Gastroenterology, Hepatology and Nutrition, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada; and 2Faculty of Dentistry, University of Toronto, Toronto, Ontario, Canada Submitted 20 February 2015; accepted in final form 31 May 2015

Vong L, Pinnell LJ, Määttänen P, Yeung CW, Lurz E, Sherman PM. Selective enrichment of commensal gut bacteria protects against Citrobacter rodentium-induced colitis. Am J Physiol Gastrointest Liver Physiol 309: G181–G192, 2015. First published June 11, 2015; doi:10.1152/ajpgi.00053.2015.—The intestinal microbiota plays a key role in shaping the host immune system. Perturbation of gut microbial composition, termed dysbiosis, is associated with an increased susceptibility to intestinal pathogens and is a hallmark of a number of inflammatory, metabolic, and infectious diseases. The prospect of mining the commensal gut microbiota for bacterial strains that can impact immune function represents an attractive strategy to counteract dysbiosis and resulting disease. In this study, we show that selective enrichment of commensal gut lactobacilli protects against the murine pathogen Citrobacter rodentium, a well-characterized model of enteropathogenic and enterohemorrhagic Escherichia coli infection. The lactobacilli-enriched bacterial culture prevented the expansion of Gammaproteobacteria and Actinobacteria and was associated with improved indexes of epithelial barrier function (dextran flux), transmissible crypt hyperplasia, and tissue inflammatory cytokine levels. Moreover, cultivation of gut bacteria from Citrobacter rodentium-infected mice reveals the differential capacity of bacterial subsets to mobilize neutrophil oxidative burst and initiate the formation of weblike neutrophil extracellular traps. Our findings highlight the beneficial effects of a lactobacilli-enriched commensal gut microenvironment and, in the context of an intestinal barrier breach, the ability of neutrophils to immobilize both commensal and pathogenic bacteria. Citrobacter rodentium; colitis; gut microbiota; lactobacilli; neutrophil extracellular traps

as a reservoir for tens of trillions of microorganisms (14). Separated from mucosal immune cells by a single layer of polarized epithelial cells, these symbiotic inhabitants (collectively termed the gut microbiota) play a key role in shaping the development of the mucosal immune system, as well as providing essential nutrients and restricting the colonization of pathogenic organisms (31, 55). To maintain intestinal homeostasis, the host immune system has evolved to confer immune tolerance to commensal bacteria, while retaining the ability to elicit an aggressive inflammatory response to invasive pathogenic bacteria. However, it has become increasingly clear that the composition of the gut microbiota, alongside host genetic, infectious, and environmental factors, also contributes to disease pathophysiology (66). In-

THE INTESTINAL TRACT SERVES

Address for reprint requests and other correspondence: P. M. Sherman, Cell Biology Program, Research Institute, Hospital for Sick Children, 555 Univ. Ave., Toronto, Ontario, Canada M5G 1X8 (e-mail: philip.sherman @sickkids.ca). http://www.ajpgi.org

deed, changes in gut microbial composition (referred to as dysbiosis) are evident in a number of chronic inflammatory and metabolic diseases, such as inflammatory bowel disease (IBD; including Crohn’s disease and ulcerative colitis) (38), obesity (74, 75), and atopy (58). While a definitive cause-and-consequence relationship between dysbiosis and disease has yet to be established in humans, decreased bacterial diversity and an altered ratio of beneficial vs. aggressive species is observed in one-third of patients with active IBD, including changes in Bacteroidetes and concomitant increases in Proteobacteria and Actinobacteria (21). Thus, where the global response to a healthy gut microbiota is immune tolerance, with dysbiosis, the host response is skewed toward aberrant inflammation. In Crohn’s disease, an increased prevalence of adherent-invasive Escherichia coli is reported in ileal biopsies (13), whereas levels of butyrate-producing Faecalibacterium prausnitzii are reduced, and is predictive of disease relapse (70, 71). Supplementation with Faecalibacterium prausnitzii in rodent models of experimental colitis ameliorates inflammation (39, 47). Other studies demonstrate the beneficial effects of probiotics (defined as live microorganisms that confer health benefits to the host) in both clinical IBD and in relevant experimental models (9, 24). Although these findings highlight the protective effects of beneficial microbes in the setting of dysbiosis and enterocolitis, the efficacy of probiotics is often strain and dose dependent (24, 53) and can be complicated by variations in host responses (36, 37, 53) and underlying disease severity (10). The prospect of mining the host commensal gut microbiota for beneficial bacterial strains that can impact immune function, and thus promote intestinal homeostasis, represents an attractive strategy to counteract dysbiosis and resulting disease. Indeed, rather than engage an absent or muted immune response, intestinal commensal bacteria can actively antagonize proinflammatory signaling (34, 48). In this study, we selectively cultivated subsets of commensal bacteria from the murine host and investigated the protective effects of these enriched commensal populations in the setting of C. rodentium-induced colitis, a well characterized murine model of both IBD (30) and infectious colitis (29). The gramnegative, murine-specific pathogen induces attaching-effacing lesions and injects bacterial effector proteins into the host cell via a locus of enterocyte effacement pathogenicity islandencoded, syringe-like type III secretion system (17). C. rodentium infection causes pronounced intestinal dysbiosis, involving overgrowth of the murine pathogen and a reduction in abundance and overall diversity of the commensal gut microbiota (43). Mice develop self-limited inflammation of the cecum and

0193-1857/15 Copyright © 2015 the American Physiological Society

G181

G182

COMMENSAL GUT MICROBES PROTECT AGAINST COLITIS

colon, transmissible crypt epithelial cell hyperplasia, and immune responses similar to those seen in human IBD. While C. rodentium infection initiates a vigorous Th1-driven proinflammatory response, neutrophil-dependent killing of the pathogen is also required to prevent bacterial dissemination and promote clearance of the infection (40). Neutrophils release novel DNA- and protease-rich extracellular structures, termed neutrophil extracellular traps (NETs), which function to capture, contain, and kill bacteria (5, 11, 49, 83). Our laboratory previously demonstrated the differential ability of beneficial and enteropathogenic bacteria to elicit NETs (77). In this study, we further delineate this paradigm and report on the capacity of subsets of intestinal commensal bacteria from C. rodentium-infected mice to mobilize reactive oxygen species (ROS) and activate NETs. MATERIALS AND METHODS

Ethics statement. All animal work was approved by the Hospital for Sick Children’s Animal Research Ethics Board (protocol approval no. 22577) and adhered to the Canadian Council on Animal Care guidelines for humane animal use. Bacterial strains and selective enrichment of intestinal microbes. To obtain a culture of lactobacilli-enriched gut bacteria (L-MRS), fecal pellets were harvested daily from treatment-naive mice (C57BL/6, aged 6 –7 wk, Charles River Laboratories, Wilmington, MA) and grown in static, nonaerated deMan Rogosa Sharp (MRS) broth (Difco Laboratories, Detroit, MI) for 16 h at 37°C. Cell density measurements (NanoDrop 2000c, Thermo Fischer Scientific, Wilmington, MA) were performed, and the bacterial culture adjusted to a density of 600-nm optical density (OD600): 2.03 ⫾ 0.05 nm in fresh MRS broth. For comparative studies, Lactobacillus rhamnosus strain GG (LGG; ATCC 53103) was cultured in static nonaerated MRS broth for 16 h at 37°C, as previously described (77). For infection studies, the attaching-effacing enteric pathogen Citrobacter rodentium, strain DBS 100 (kindly provided by the late David Schauer, Massachusetts Institute of Technology, Cambridge, MA) was cultured in static nonaerated Luria-Bertani (LB) broth (Difco Laboratories) for 16 h at 37°C, to yield a final concentration of 1 ⫻ 109 colony-forming units (CFU)/ml. To investigate the capacity of the C. rodentium-infected gut microbial community to stimulate production of ROS and activate NETs, fecal pellets from sham- or C. rodentium-infected mice were harvested and cultivated in Tryptic Soy (TS) broth (Difco Laboratories), which promotes the growth of aerobic and facultative anaerobic bacteria (OD600: 0.25 ⫾ 0.01 nm), LB broth (Difco Laboratories) to promote the growth of Escherichia coli (OD600: 0.19 ⫾ 0.01 nm), or MRS broth (Difco Laboratories), promoting the growth of lactobacilli (OD600: 0.58 ⫾ 0.03 nm), for 16 h at 37°C. Animals. Male C57BL/6 mice (Charles River Laboratories, Wilmington, MA) aged 6 –7 wk, were given access to water and standard chow ad libitum. All mice were housed for a minimum of 7 days before initiation of experimental protocols (day 0). Thereafter, mice were orogastrically gavaged with 100 ␮l of phosphate-buffered saline (PBS, Invitrogen, Burlington, ON), MRS broth (MRS; autoclaved for sterility), or the lactobacilli-enriched commensal gut bacteria (LMRS) once daily for 7 days. In some experiments, mice were administered 5 ⫻ 108 CFU/day of LGG for comparison. To investigate the impact of the enriched bacterial cultures on intestinal homeostasis, mice were either challenged with the murine pathogen C. rodentium (1 ⫻ 108 CFU/mouse, orogastric gavage), or sham infected with an equal volume of sterile LB broth. Mice were killed by cervical dislocation at 10 days postinfection (day 17), and colonic and fecal samples were processed for analysis of intestinal injury, microbial composition, and capacity to activate neutrophils, respectively. Fecal samples were stored at

⫺80°C before processing, or immediately recultured in TS, MRS, and LB broth cultures, as described above. Morphology and histological analysis. Colons from sham- and C. rodentium-infected mice (administered PBS, MRS, or L-MRS) were excised, and their lengths measured following death (day 17). Histological analysis was performed to assess the extent of colonic injury. Briefly, distal colonic segments were harvested, fixed in 10% neutralbuffered formalin, and embedded in paraffin. Colonic samples were then sectioned, stained with hematoxylin and eosin, and visualized using a Leica DMI 6000B microscope equipped with a Leica DFC420 camera and Leica Application Suite Advanced Fluorescence software (Leica Microsystems, Concord, Ontario, Canada). Histopathology scoring (81) was performed on coded sections, and a numerical score assigned for severity of epithelial injury (graded 0 –3, from absent to mild, including superficial epithelial injury; moderate, including focal erosions; and severe, including multifocal erosion), the extent of inflammatory cell infiltrate (graded 0 –3, from absent to transmural), and goblet cell depletion (0 –1). Changes in crypt length, an indication of hyperplasia, were also measured and expressed as the average of 10 crypt length measurements per section of tissue, from 12 mice per group. Intestinal permeability to macromolecules. To evaluate alterations in intestinal epithelial barrier function in vivo, serum recovery of a 4-kDa fluorescein isothiocyanate (FITC)-conjugated dextran probe (Sigma-Aldrich, Oakville, Ontario, Canada) was measured (63). Briefly, mice were fasted overnight and then orogastrically gavaged with 100 ␮l of FITC-conjugated dextran (88 mg/ml; in sterile PBS). After 4 h, mice were killed, and whole blood collected by cardiac puncture. Serum was obtained by centrifugation at 1,000 g (4°C, 20 min), and levels of FITC-conjugated dextran determined by fluorometry (Victor X3, Perkin Elmer, Woodbridge, Ontario, Canada). Lipid peroxidation assay. Lipid peroxidation, employed as a marker of oxidative stress (16), was measured in the distal colonic segments of sham- and C. rodentium-infected mice (administered PBS, MRS, or L-MRS). Briefly, colonic segments were collected at time of death (10 days postinfection; day 17) and stored at ⫺80°C until used for measurement. The TBARS Assay Kit (Cayman Chemical, Ann Arbor, MI) was used to quantify malondialdehyde formation (a product of lipid peroxidation), following manufacturer’s instructions. All tissue samples were normalized by weight, and the malondialdehyde adduct quantified colorimetrically at 530 –540 nM on a microplate reader (Victor X3, Perkin Elmer). Microbial composition analyses. The microbial composition of murine fecal pellets was analyzed using quantitative RT-PCR (qPCR) and denaturing gradient gel electrophoresis (DGGE) (63). Briefly, prokaryotic DNA was extracted using the QIAamp stool mini kit (QIAGEN, Mississauga, ON) and reconstituted to 10 ng/␮l. qPCR was conducted using SsoFast Eva-Green Supermix (Bio-Rad Laboratories, Mississauga, ON) and a CFX96 C1000 Thermal Cycler (Bio-Rad Laboratories, Mississauga, ON). Gammaproteobacteria, Bacteroidetes, Firmicutes, Actinobacteria, and Lactobacillus were quantified using the 2⫺⌬⌬CT method (41) and primers and expressed as fold change relative to sham-infected mice, as previously described (1, 62). Relative numbers of C. rodentium, as an indicator of pathogen colonization (32, 50), were measured in fecal pellets using primers against espB, a gene encoding a secreted virulence factor (50). qPCR analysis for espB has previously been used to confirm the identity of C. rodentium bacterial colonies (44, 45), and transcriptional levels were shown to remain constant (17, 81). Bacterial 16S rRNA gene profiles of fecal bacteria, as well as selectively enriched (TS, LB, or MRS broth) bacterial populations from sham and C. rodentium-infected mice, were generated by DGGE. Briefly, gels were prepared using the primers 357f-GC and 518r, 30 –70% denaturing gradients, and 8% polyacrylamide gels. Gels were run at 80 V for 16 h (DGGE-2001, C.B.S. Scientific, San Diego, CA), stained with SYBR Gold (Invitrogen, Burlington, Ontario, Canada), and visualized with a Typhoon FLA 9500 Molecular Imager (GE Healthcare, Mississauga,

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00053.2015 • www.ajpgi.org

COMMENSAL GUT MICROBES PROTECT AGAINST COLITIS

Ontario, Canada). Community profiles were subjected to cluster analysis using Gelcompar II (Applied Maths, Sint-Marten-Latem, Belgium) with Dice’s band-matching coefficient and unweighted pair group method with arithmetic averages. Individual bands were excised, reamplified, and sequenced, as described previously (82). Sequencing was performed at the Centre for Applied Genomics (The Hospital for Sick Children, Toronto, ON, Canada) with their respective primers. The sequences were assembled and aligned by BioEdit [http://www.mbio.ncsu.edu/BioEdit/ bioedit.html (28)] with closely-related representatives (phylotypes) from NCBI BLASTN (http://ncbi.nlm.nih.gov/blast/), as well as novel complete and partial sequences obtained from GenBank. Phylogenetic relationships were constructed with evolutionary distances (Jukes-Cantor distances) and the neighbor-joining method using the MEGA 5.05 software package (72). The bootstrap analyses for the phylogenetic trees were calculated by running 1,000 replicates of the neighbor-joining data. Quantification of tissue cytokine levels. RNA was extracted from distal colonic segments using TRIZOL extraction reagent (Invitrogen, Burlington, Ontario, Canada), and cDNA synthesized using iScript (Bio-Rad Laboratories, Mississauga, Ontario, Canada) with 750 ng of template RNA. qPCR was performed, as described above, using primers against ␤-actin, interleukin (IL)-10, interferon-␥, and tumor necrosis factor-␣, and normalized against the reference genes GAPDH and proteasome, as previously described (1, 62). Bone marrow-derived neutrophil isolation. Murine bone marrowderived neutrophils (BMDN) were isolated from male C57BL/6 mice (77). Briefly, the tibia and femur were harvested and flushed with MEM-␣ (Invitrogen), and the resultant cell suspension centrifuged at 400 g for 10 min. The cell pellet was resuspended in PBS (⫺Ca2⫹/ Mg2⫹, pH 7.4; Invitrogen), layered onto a Percoll density gradient (80%/65%/55%; Sigma-Aldrich, Oakville, Ontario, Canada), and centrifuged at 1,000 g for 30 min at 4°C. BMDN at the 80%/65% interface were collected and washed in PBS, and contaminating red blood cells were removed with a brief hypotonic lysis in ice-cold dH2O before NaCl (3.6%) was added. The BMDN were subsequently resuspended in HBSS (⫹Ca2⫹/Mg2⫹; Invitrogen) and normalized to 1 ⫻ 106 cells/ml. For all experiments, 1 ⫻ 105 BMDN were employed. ROS production. ROS production by BMDN, in the absence or presence of fecal bacteria cultures, was quantified by luminol-enhanced chemiluminescence (77). Briefly, BMDN (1 ⫻ 105 cells/ reaction) were coincubated with a 10-␮l aliquot of TS, LB, or MRS broth-cultivated bacterial cultures (or sterile broth alone) from shamor C. rodentium-infected mice and equilibrated in HBSS for 30 min at 37°C in 5% CO2. Luminol (50 ␮M; Sigma-Aldrich) and horseradish peroxidase (1.2 U/ml; Sigma-Aldrich) were added to the reaction, and the resulting chemiluminescence was measured on a microplate reader (Victor X3, Perkin Elmer, Woodbridge, Ontario, Canada). Quantification and visualization of NETs. BMDN were incubated with TSB, LB, or MRS broth-cultivated fecal bacteria (10 ␮l of 1:10 diluted bacterial cultures, or sterile broth alone) as above, to a final volume of 300 ␮l HBSS in a black 96-well plate (3 h at 37°C in 5% CO2). NET activation by hydrogen peroxide (H2O2) (Sigma-Aldrich) and NET degradation with the addition of DNase (5 units) served as positive and negative controls, respectively. Extracellular DNA released from BMDN were stained with Sytox Green (5 ␮M, Invitrogen), a fluorescent membrane-impermeable DNA dye, and fluorescence was quantified using a microplate reader equipped with filters to detect excitation/emission maxima: 485/520 nm (Victor X3, Perkin Elmer), as previously described (77). Wells containing both bacteria and neutrophils were normalized against fluorescence emission from wells containing only bacteria. To visualize NETs, BMDN were plated onto 12-mm poly-L-lysine coated glass coverslips (BD Biosciences, Mississauga, Ontario, Canada) and coincubated with fecal bacterial cultures (3 h, 37°C, 5% CO2). Cells were subsequently fixed with 10% neutral-buffered formalin (45 min at room temperature), washed with PBS, and stored at 4°C. For confocal fluorescence analysis, cells were permeabilized with ice-cold methanol (10 min, ⫺20°C), and then stained with

G183

antibodies against histone H3 (D1H2 XP rabbit monoclonal antibody; Cell Signaling, Denvers, MA) and myeloperoxidase (MPO; mouse monoclonal antibody, R&D Systems, Minneapolis, MN) for 16 h at 4°C. Alexa Fluor 488 goat anti-rabbit secondary antibody (Invitrogen) and Alexa Fluor 532 goat anti-mouse secondary antibody (Invitrogen) were used to visualize histone H3 and MPO, respectively. 4=,6Diamidino-2-phenylindole (Invitrogen) was used to counterstain for DNA. Cells were mounted with Prolong Gold Anti-fade reagent (Invitrogen), and confocal fluorescence imaging was performed on a Quorum WaveFX Spinning Disk Confocal System mounted onto an Olympus IX81 microscope (Center Valley, PA). Images were acquired using a ⫻10 and ⫻20/0.75 objective, Hamamatsu C9100-13 EM-CCD, and 405-, 491-, and 561-nm laser lines (Hamamatsu City, Japan). Image acquisition was performed using Perkin Elmer Velocity 6.2.1 software. For analysis, three random fields from three separate experiments of each treatment were captured with a ⫻20 objective, and the total area occupied by DNA- and histone H3-stained cells was quantified using ImageJ (77). Results are expressed as percentage cell area (per field of view) normalized per 100 cells. Statistical analyses. Data are presented as means ⫾ SE. Comparisons among multiple groups were made using two-way analysis of variance, followed by Tukey’s post hoc analysis. Student’s t-test was used where indicated. An associated probability (P value of ⬍0.05) was considered significant. RESULTS

Selective enrichment of commensal gut bacteria protects against colitis. An acetate-rich culture medium (MRS broth) was used to enrich lactobacilli present in the murine healthy commensal gut microbiota. DGGE and sequence analysis of the bacterial culture (L-MRS) confirmed the predominance of Lactobacillus murinus (Fig. 1A, band c), which, in C57BL/6 mice, constitutes up to 72% of the culturable intestinal content (57). In fecal pellets of healthy mice, multiple bands corresponding to members of the Bacteroides genus were identified (Fig. 1A, band 1, Bacteroidales bacterium; band 2, Parabacteroides sp). To investigate the capacity of the lactobacilli-enriched bacterial culture (L-MRS) to protect against infectious colitis, mice were administered PBS, MRS broth (MRS), or L-MRS once daily, for 7 days. Mice were then infected with C. rodentium (or sham infected) and killed after 10 days. While there were no differences in colonic and crypt lengths of sham-infected mice, C. rodentium-infected mice given PBS had significantly shorter colons (P ⬍ 0.001 vs. PBS shaminfected; n ⫽ 12/group) than those administered L-MRS or MRS (P ⬎ 0.05 vs. L-MRS and MRS sham-infected mice; n ⫽ 12/group) (Fig. 1B). Moreover, C. rodentium-induced crypt hyperplasia in PBS-gavaged mice was significantly greater compared with mice administered L-MRS (P ⬍ 0.05; n ⫽ 8/group) (Fig. 1C). No differences in crypt length were found between sham- or C. rodentium-infected mice treated with L-MRS or MRS (P ⬎ 0.05; n ⫽ 8/group). Representative low (⫻10) and higher magnification (⫻40) images of hematoxylin and eosin-stained colonic sections from sham- and C. rodentiuminfected mice are shown in Fig. 1D. Epithelial disruption and adherent bacteria (42) (orange arrowheads) are visible in the colonic sections of PBS-treated C. rodentium-infected mice, and to a lesser extent the MRS-treated, but not L-MRS-treated mice, and are corroborated by the histopathology scores shown in Fig. 1E (n ⫽ 8/group). In contrast, mice administered commercially available LGG for 7 days before C. rodentium

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00053.2015 • www.ajpgi.org

G184

COMMENSAL GUT MICROBES PROTECT AGAINST COLITIS

Fig. 1. Lactobacilli-enriched commensal gut bacteria protects against infectious colitis. A: microbial community fingerprint [denaturing gradient gel electrophoresis (DGGE)] of murine fecal bacteria (pellet) and the lactobacilli-enriched bacterial culture (L-MRS). Sequence analysis identified the following bands: band c, Lactobacillus murinus; band 1, Bacteroidales bacterium; band 2, Parabacteroides sp. B: once-daily administration of phosphate-buffered saline (PBS), vehicle deMan Rogosa Sharp (MRS) broth, or LMRS for 7 days, followed by infection with C. rodentium or sham infection resulted in significant shortening of the colon in PBSbut not MRS- or L-MRS-treated mice infected with the murine pathogen (day 17, 10 days postinfection; ***P ⬍ 0.001 vs. PBS sham-infected; n ⫽ 12/group). C: whereas colonic crypt length increased in PBStreated, C. rodentium-infected mice (**P ⬍ 0.01 vs. PBS sham-infected; n ⫽ 12/group), crypt lengths of mice administered MRS or L-MRS were not different from sham-infected mice (P ⬎ 0.05; #P ⬍ 0.05, L-MRS C. rodentium-infected vs. PBS C. rodentium-infected, n ⫽ 12/group). D: representative hematoxylin and eosin-stained sections (⫻10 and ⫻40 original magnifications), bar ⫽ 75 ␮m. E: histopathology damage scores of colonic sections from sham- or C. rodentium-infected mice (day 17; ****P ⬍ 0.0001 vs. PBS sham-infected, #P ⬍ 0.01 vs. PBS C. rodentium-infected; P ⬎ 0.05, L-MRS sham vs. L-MRS C. rodentium-infected; n ⫽ 8). F: mice administered L. rhamnosus strain GG (LGG; 5 ⫻ 108 colony-forming units/ day) before pathogen infection were not protected from C. rodentium-induced crypt hyperplasia (P ⬎ 0.05; n ⫽ 5/group; **P ⬍ 0.01 vs. PBS sham-infected; n ⫽ 5/group). Values are means ⫾ SE, with significance calculated using two-way ANOVA with Tukey’s post hoc analysis.

infection (Fig. 1F) were not protected from crypt epithelial cell hyperplasia (P ⬎ 0.05; n ⫽ 5/group). Commensal lactobacilli bacteria ameliorate epithelial barrier dysfunction and intestinal inflammation. Increased intestinal epithelial cell permeability with C. rodentium infection contributes to enhanced bacterial translocation and activation of the host proinflammatory response (69). As expected, dextran flux was significantly elevated in PBS-treated, C. rodentium-infected mice (Fig. 2A, P ⬍ 0.01 vs. PBS sham-infected; n ⫽ 4 – 6/group). By contrast, mice administered either L-MRS or MRS for 7 days before C. rodentium infection had serum dextran levels comparable to those of sham-infected mice (P ⬎ 0.05; n ⫽ 4/group). In line with the protective effects of L-MRS and MRS administration, distal colonic expression of

the proinflammatory cytokines interferon-␥ and tumor necrosis factor-␣ were significantly reduced in these groups following C. rodentium infection, compared with PBS-treated, C. rodentium-infected mice (P ⬍ 0.0001; n ⫽ 4/group) (Fig. 2B). IL-10 expression levels remained unchanged between study groups. Protection against C. rodentium-associated intestinal dysbiosis and pathogen colonization. To assess alterations in gut microbial composition, fecal pellets were collected from mice before PBS, MRS, or L-MRS administration (day 0) and 10 days after sham or C. rodentium infection (day 17). There were no significant compositional differences (Gammaproteobacteria, Bacteroidetes, Firmicutes, Actinobacteria, and Lactobacillus) between each of the treatment groups before initiation of the treatment protocol (P ⬎ 0.05; n ⫽ 3– 4/group, Fig. 3A, top).

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00053.2015 • www.ajpgi.org

COMMENSAL GUT MICROBES PROTECT AGAINST COLITIS

G185

Fig. 2. Lactobacilli-enriched commensal gut bacteria ameliorate C. rodentium-induced epithelial barrier dysfunction and inflammation. A: increased levels of 4-kDa FITC-dextran was recovered in PBS-treated, C. rodentium-infected mice (**P ⬍ 0.01 vs. PBS, sham-infected; n ⫽ 4 – 6/group). By contrast, FITC-dextran levels from mice administered vehicle MRS broth (MRS) or lactobacilli-enriched bacterial culture (LMRS) and infected with C. rodentium were not significantly different from their sham-infected counterparts (P ⬎ 0.05 vs. sham infection; #P ⬍ 0.05, L-MRS C. rodentium-infected vs. PBS C. rodentium-infected; n ⫽ 4/group). B: quantitative PCR analysis of tissue inflammatory cytokines show elevated levels of IFN-␥ and TNF-␣ in PBS-treated, C. rodentium-infected mice (****P ⬍ 0.0001 vs. PBS, sham-infected, #P ⬍ 0.05 vs. MRS or L-MRS, C. rodentium-infected, n ⫽ 4/group). Values are means ⫾ SE, with significance calculated using two-way ANOVA with Tukey’s post hoc analysis.

Following C. rodentium infection, a 10.1 ⫾ 2.4-fold increase in Gammaproteobacteria and 6.5 ⫾ 1.6-fold increase in Actinobacteria was observed in mice treated with PBS (P ⬍ 0.05 vs. PBS sham-infected mice; n ⫽ 3– 4/group, Fig. 3A, bottom). In contrast, levels of Gammaproteobacteria and Actinobacteria were not significantly different in mice administered MRS or the lactobacilli-enriched bacterial culture, whether sham or C. rodentium infected (P ⬎ 0.05; n ⫽ 3– 4/group). Interestingly, Lactobacillus levels in MRS-treated, C. rodentium-infected mice were elevated 4.0 ⫾ 1.0-fold (P ⬍ 0.05; n ⫽ 3– 4/group) compared with the sham-infected counterparts. Hierarchical cluster analysis of the 16S bacterial community fingerprint is presented in Fig. 3B, where, at day 17, bacterial communities from the C. rodentium-infected mice administered L-MRS clustered alongside sham-infected mice (66.7% similarity), and separately from C. rodentium-infected mice administered PBS or MRS (37.4% similarity). Intensity of band A, identified as dcrB, a major facilitator superfamily protein unique to C. rodentium (59), was reduced in L-MRS-treated, C. rodentiuminfected mice compared with PBS- and MRS-treated counterparts. Results from sequence analysis of bands 1– 4 are presented in a phylogenetic tree (Fig. 3C). Analysis of fecal espB mRNA, employed as an indication of C. rodentium colonization, revealed a significant reduction in espB in mice administered the lactobacilli-enriched culture (0.23 ⫾ 0.1fold change, P ⬍ 0.05; n ⫽ 7– 8/group) compared with mice administered PBS (Fig. 3D). These observations are in line with

observed relative changes in levels of Gammaproteobacteria. espB was undetectable in fecal pellets obtained from shaminfected mice. Together, these data indicate that C. rodentium had a reduced capacity to colonize the murine gut in the presence of L-MRS, and that the lactobacilli-enriched culture conferred protection against C. rodentium-induced dysbiosis. Bacterial communities cultivated from C. rodentium-infected mice stimulate neutrophil ROS production. The primary function of intestinal neutrophils is to kill invading microbes that have traversed the epithelial barrier; however, aberrant neutrophil recruitment and activation also play a role in intestinal pathologies (20, 25). To delineate the effects of dysbiotic microbial communities on neutrophil function, subsets of intestinal bacteria from either sham- or C. rodentium-infected mice were cultivated (in TS, MRS, or LB broth), and their capacity to activate neutrophils investigated. Hierarchical cluster analysis of the bacterial communities cultivated from sham- and C. rodentium-infected mice is shown in Fig. 4A. Three distinct cluster profiles emerged in the dendrogram. All bacterial cultures that were cultivated in MRS broth (whether from sham- or C. rodentium-infected mice) clustered closely together (ⱖ87.9% similarity); TS and LB broth-cultivated bacteria from sham-infected mice, as well as lactobacilli-treated, pathogen-infected mice clustered together (ⱖ47.4% similarity), whereas TS and LB broth cultivated bacteria from C. rodentium-infected mice administered PBS or MRS clustered separately (22.9% similarity with the other

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00053.2015 • www.ajpgi.org

G186

COMMENSAL GUT MICROBES PROTECT AGAINST COLITIS

Fig. 3. Gut microbial dysbiosis is reversed by lactobacilli-enriched commensal bacterial culture. A: quantitative PCR analysis of fecal microbial composition before once-daily administration of PBS, MRS broth (MRS), or lactobacilli-enriched gut bacterial culture (L-MRS) (day 0; top), and 10 days postsham or C. rodentium infection (day 17; bottom). Whereas mice administered PBS and infected with C. rodentium had increased levels of Gammaproteobacteria and Actinobacteria (*P ⬍ 0.05 vs. PBS, sham-infected, n ⫽ 4/group), there were no changes in mice administered L-MRS (P ⬎ 0.05 vs. L-MRS, sham-infected, n ⫽ 4/group). Increased levels of Lactobacillus were present in fecal pellets of MRS-treated, C. rodentium-infected mice. Values are means ⫾ SE, with significance calculated using two-way ANOVA with Tukey’s post hoc analysis. B: DGGE and cluster analysis of the gut microbial community were assessed at day 17 (10 days postinfection). C. rodentium-infected mice administered L-MRS clustered separately from mice administered PBS or MRS broth. Band A, a gene encoding dcrB, a C. rodentium-specific protein, is reduced in LMRS-treated, C. rodentium-infected mice. C: sequence analysis of bands 1– 4 are presented in the phylogenetic tree. D: mRNA levels of espB, a surrogate marker for C. rodentium colonization (32, 50), were reduced in C. rodentium-infected mice previously administered L-MRS (*P ⬍ 0.05 vs. PBS, C. rodentium-infected; n ⫽ 7– 8/group). Values are means ⫾ SE, with significance calculated using the unpaired Student’s t-test.

treatment groups). Sequence analysis of band A, identified as dcrB (as above), confirms the presence of the murine pathogen in TS and LB broth-cultivated bacterial populations from infected and PBS- and MRS-treated mice, but not in mice administered L-MRS. There was no evidence of C. rodentium in MRS broth-cultivated bacterial populations of pathogeninfected mice. A phylogenetic tree of sequenced bands (labeled a– c) is shown in Fig. 4B. In sham-infected mice, L. murinus

(band c) was absent or below the threshold of detection from mice administered PBS (TS, LB, MRS broth cultivation), whereas in groups that were administered either MRS or L-MRS, the relative abundance (band intensity) was increased. These observations indicate that the lactobacilli-enriched bacterial culture successfully colonized the recipient mice. Moreover, detection of L. murinus in sham-infected mice administered MRS broth highlights the capacity of the nutrient-rich

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00053.2015 • www.ajpgi.org

COMMENSAL GUT MICROBES PROTECT AGAINST COLITIS

G187

Fig. 4. Composition of fecal microbial communities cultured from sham- and C. rodentium-infected mice. A: fecal pellets from either sham- or C. rodentium-infected mice [administered PBS, vehicle MRS broth (MRS) or lactobacilli-enriched bacterial cultures (L-MRS)], were cultivated in Tryptic Soy (TS), MRS, or Luria-Bertani (LB) broth (16 h at 37°C) and evaluated by DGGE and cluster analysis. Fecal communities clustered in 3 distinct groups: MRS broth cultivated bacteria (ⱖ87.9% similarity), TS and LB broth cultivated bacteria from sham or C. rodentium-infected mice treated with the lactobacilli-enriched culture (ⱖ47.4% similarity), and TS and LB broth cultivated bacteria from C. rodentiuminfected mice administered PBS or MRS (ⱖ70.5% similarity). Band A, dcrB, which encodes a C. rodentium-specific protein, was present in TS- and LB-cultivated bacterial cultures, but not the MRScultivated bacterial cultures of pathogen-infected mice. SAB, similarity coefficient. B: sequencing results of bands a– c are presented in the phylogenetic tree. IBD, inflammatory bowel disease.

media to support the growth of lactobacilli in vivo. Interestingly, C. rodentium infection further increased L. murinus in all treatment groups. While expansion of L. murinus has previously been reported during dysbiosis (4), our results suggest that this likely confers protective effects. The capacity of the bacterial populations to mobilize neutrophil ROS production was investigated. LB broth-cultivated bacteria (containing predominately E. coli) from sham-infected mice administered PBS or MRS activated BMDN ROS production (P ⬍ 0.05 vs. BMDN alone; n ⫽ 8/group) (Fig. 5A). In contrast, MRS broth-cultivated bacteria had no effect on ROS production (P ⬎ 0.05 vs. BMDN alone; n ⫽ 8/group). In mice infected with C. rodentium, both TS- (containing E. coli and lactobacilli) and LB-cultivated bacteria from mice treated with PBS (P ⬍ 0.05 vs. BMDN alone, n ⫽ 8/group) elicited more ROS than mice administered L-MRS (P ⬍ 0.05; n ⫽ 8/group). Together, these results demonstrate that intestinal dysbiosis in response to C. rodentium infection, particularly TS and LB broth-cultivated communities, has the capacity to activate neutrophil ROS production; however, this can be ameliorated by treatment with the lactobacilli-enriched bacterial culture. While the capacity of neutrophils to mobilize ROS was increased in the presence of the dysbiotic bacterial communi-

ties, this did not translate to an imbalance in redox status in vivo. Analysis of distal colonic lipid peroxidation levels, a measure of oxidative stress, revealed no differences between sham- and C. rodentium-infected mice (P ⬎ 0.05; n ⫽ 4/group) (Fig. 5B). Intestinal bacteria activate NETs. Given the key role of ROS in the activation cascade of NETs, we next investigated the capacity of NETs to be elicited by the TS, LB, and MRS broth-cultivated bacterial communities from C. rodentiuminfected mice. Fluorescent detection of extracellular DNA (using a cell-impermeable, DNA-intercalating dye, SYTOX green) was used to quantify NET release. As shown in Fig. 6A, TS broth-cultivated bacteria from pathogen-infected mice (administered PBS, MRS broth, or L-MRS) elicited robust weblike NET formation (P ⬍ 0.05 vs. BMDN alone; n ⫽ 8/group). In contrast, MRS broth-cultivated bacteria did not induce NETs (P ⬎ 0.05 vs. BMDN alone; n ⫽ 8/group). Data are presented as %total DNA, and the addition of H2O2 and DNase served as positive and negative controls, respectively. Representative images of NETs activated by TS, MRS, and LB-broth cultivated bacterial populations are shown in Fig. 6B (left), stained with antibodies against histone H3 (green), MPO (red),

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00053.2015 • www.ajpgi.org

G188

COMMENSAL GUT MICROBES PROTECT AGAINST COLITIS

Fig. 5. Fecal microbial communities from C. rodentiuminfected mice vary in production of reactive oxygen species (ROS) by neutrophils. A: TS, LB, and MRS broth-cultivated microbial communities from sham- or C. rodentium-infected mice [administered PBS, MRS broth (MRS), or the lactobacilli-enriched gut bacterial community (L-MRS) once daily for 7 days] were coincubated with murine bone marrow-derived neutrophils (BMDN) and assayed for ROS production. LB-cultivated bacteria from PBS- or MRS-treated, sham-infected mice mobilized ROS (*P ⬍ 0.05 vs. BMDN; n ⫽ 8/group). By contrast, both TS and LB broth-cultivated bacteria from PBS- and MRS-treated, C. rodentiuminfected mice induced ROS production (*P ⬍ 0.05 vs. BMDN; n ⫽ 8/group). Significantly reduced levels of ROS were mobilized by bacteria from L-MRS-treated, C. rodentium-infected mice (#P ⬍ 0.05 vs. TS- and LB-cultivated, PBS C. rodentium-infected mice; n ⫽ 8/group). B: distal colonic lipid peroxidation levels, measured as an marker of oxidative stress, were not different between sham- and C. rodentium-infected mice administered PBS, MRS, or L-MRS (P ⬎ 0.05; n ⫽ 4/group). Values are means ⫾ SE, with significance calculated using two-way ANOVA with Tukey’s post hoc analysis. MDA, malondialdehyde.

and the DNA counterstain 4=,6-diamidino-2-phenylindole (blue), and quantified in Fig. 6B (right). DISCUSSION

The intestinal microbiota plays a key role in shaping the host immune system. Perturbations in microbial composition (for example, following antibiotic exposure or as occurs during acute mucosal infections) render the host more susceptible to intestinal pathogens (67, 79) and disease in later life (58, 65). Significantly, intestinal dysbiosis is a hallmark of a number of inflammatory, metabolic, and infectious diseases (38, 58, 74, 75) and provides, therefore, a compelling therapeutic target to restore health and homeostasis. In this study, we utilized commensal gut microbes resident in the healthy murine host to selectively cultivate a population of lactobacilli-enriched bacteria. We hypothesized that, by initially expanding the population of commensal beneficial bacteria in recipient mice, the capacity of C. rodentium to colonize and induce intestinal dysbiosis would be inhibited. The amino acid- and nitrogen-rich selective media (15) employed supports the growth of gut lactobacilli and contains acetate, magnesium, and manganese, which provide essential growth factors, while inhibiting the replication of other bacteria (64, 68). Our results demonstrate that the lactobacillienriched bacterial cultures dramatically protect against pathogen infection and dysbiosis. Inhibition of C. rodentium colonization was confirmed using a combination of molecular and

histological protocols, including quantification of Gammaproteobacteria, and detection of espB and dcrB levels. Moreover, these effects were replicated, albeit to a lesser extent, with administration of the selective culture media alone, indicating that the nutrient-supplemented environment of the intestine was skewed to support the growth of beneficial lactobacilli in vivo. Indeed, the Lactobacillus genus, along with Bifidobacteria, represents a source of lactic acid-producing probiotic bacteria (22). Reported benefits of lactobacilli include their ability to activate the host immune system, prevent the duration and intensity of diarrheal episodes, enhance colonization resistance, and produce bacteriocins (pathogen inhibitory compounds) (76, 80). Importantly, the time point of administration is crucial in determining the outcome of infection: beneficial lactobacilli have previously been shown to protect against infectious colitis when administered up to 3 days after pathogen challenge (63). Sequence analysis of the beneficial bacterial cultures confirmed the predominance of L. murinus, which, alongside L. reuteri, constitutes up to 72% of culturable lactobacilli present in the intestinal content of C57BL/6 mice (57). Given their ability to survive low pH and bile salts, lactobacilli stably colonize the intestinal tract, with the number of resident Lactobacillus species reported to correlate negatively with susceptibility to colitis (57). L. murinus is one of eight bacterial species (collectively termed altered Schaedler’s flora) used to restore cecal morphology of germ-free mice to a structure more comparable to specific pathogen-free mice (18). L. murinus

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00053.2015 • www.ajpgi.org

G189

COMMENSAL GUT MICROBES PROTECT AGAINST COLITIS

Fig. 6. Gut microbial communities from C. rodentium-infected mice vary in their capacity to induce neutrophil extracellular traps (NETs). A: murine BMDNs were incubated with TS, LB, or MRS-cultivated bacteria from sham- or C. rodentium-infected mice [administered PBS, vehicle MRS broth (MRS), or L-MRS once-daily, 7 days] and assayed for their ability to elicit NETs. Release of extracellular DNA was quantified using the cell-impermeable, DNA intercalating dye Sytox Green (SYTOX). NET activation by hydrogen peroxide (H2O2) and inhibition by DNase served as positive and negative controls, respectively. TS brothcultivated bacteria from C. rodentium-infected mice activated NET formation (*P ⬍ 0.05 vs. BMDN; n ⫽ 8/group). ***P ⬍ 0.001 vs. BMDN; n ⫽ 8/group. B, left: bacterial cultures were plated onto poly-Llysine coated coverslips with BMDN to visualize NETs. Cells were labeled with myeloperoxidase (red) and histone H3 (green) and counterstained with 4=,6-diamidino-2phenylindole (blue). Bar ⫽ 90 ␮m. Right: quantification of microscopy images are shown and are representative of 3 independent experiments. Values are means ⫾ SE, with significance calculated using two-way ANOVA with Tukey’s post hoc analysis.

produces a number of low-molecular-weight anti-microbial compounds (54). It has been reported to protect against Salmonella Typhimurium invasion of epithelial cells in vitro, and, when used in combination with other probiotic strains, L. murinus alleviates symptoms of S. Typhimurium infection and reduces pathogen shedding in infected pigs in vivo (7, 8). Whereas C. rodentium infection typically induces marked intestinal dysbiosis and overgrowth of the murine pathogen (43), our analyses of the gut microbial composition revealed no changes in Gammaproteobacteria levels in mice administered L-MRS. Enhanced nitrate levels (78) and Paneth cell death (61) are thought to contribute to Gammaproteobacteria expansion, with increased levels of E. coli, but not commensals (such as bacteria from the Bifidobacteria and Lactobacillus/Lacto-

coccus genera), found in proximity to the intestinal epithelium of pathogen-infected mice (51). Aside from promoting colonization resistance to prevent C. rodentium colonization of the colon, lactobacilli-enriched cultures may also confer protection through modification of the metabolic landscape (76). L. paracasei and L. rhamnosus modify a number of metabolic parameters, including production of short-chain fatty acids (46). Recent evidence highlights the role of succinate, produced by the commensal symbiont Bacteroides thetaiotaomicron, in potentiating C. rodentium virulence (12). Induction of the succinate-to-butyrate pathway has also been reported to promote proliferation and colonization of the enteric pathogen Clostridium difficile (19). While lactobacilli and E. coli contribute to the accumulation of succinate in antibiotic-associated

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00053.2015 • www.ajpgi.org

G190

COMMENSAL GUT MICROBES PROTECT AGAINST COLITIS

diarrhea (73), our results demonstrate that selective nutritionsupported growth of beneficial bacteria effectively restricted the proliferation and colonization of the murine pathogen C. rodentium. Moreover, indexes of epithelial barrier dysfunction, mucosal inflammation, and intestinal dysbiosis were preserved in the setting of pathogen challenge and comparable to shaminfected mice. Interestingly, Lactobacillus levels were increased in the MRS-treated, C. rodentium-infected mice, but not sham-infected mice, 10 days postinfection. Although lactobacilli have been reported to be increased during intestinal dysbiosis (4), the underlying mechanism remains unclear. In support of the superior protective effects of the commensal gut flora, it is noteworthy that the lactobacilli-enriched bacterial community afforded greater protection than L. rhamnosus strain GG alone. Neutrophils play a major role in microbial clearance. They are recruited to the intestine by IL-17A (52) and are required for the immobilization and killing of expanding Gammaproteobacteria populations during intestinal infections (51). Absence of MyD88-dependent neutrophil recruitment and mobilization during C. rodentium infection result in exacerbated disease and are characterized by a failure to control pathogen colonization and dissemination beyond the gut (40). Given that C. rodentium induces a self-resolving colitis (the murine pathogen is cleared from the intestinal tract by 6 wk postinfection in immune-sufficient animals) (2), it serves as a model to investigate interactions between a transient dysbiotic bacterial community and neutrophils. Neutrophils are well-equipped to combat infection and readily liberate NETs to capture, contain, and kill bacteria (5, 11, 49, 83). Composed predominantly of decondensed chromatin, these weblike extracellular structures are highly decorated with antimicrobial histones and granular proteins, including elastase and MPO (56). The release of NETs results in auto cell death (lytic NETosis) (5, 23, 26); although neutrophils can also retain host defense functions (vital NETosis), such as recruitment, chemotaxis, and phagocytosis (11, 60, 83). Importantly, while the inability to mobilize NETs is associated with persistent and recurrent infections (3, 23), excessive release of cell-free DNA and proteases, on the other hand, can damage surrounding cells and contribute to disease pathology (27, 35). In the context of the intestine, the differential capacity of commensal and pathogenic bacteria to activate NETs likely serves to minimize bystander injury (77). Our laboratory previously reported that enterohemorrhagic E. coli strain CL56 and adherent-invasive E. coli strain LF82 robustly activate NETs in vitro, whereas commensal E. coli strains had no effect (77). In the present study, we used the C. rodentium model of colitis to investigate the capacity of dysbiotic bacterial communities to induce NETs. To delineate which populations of gut bacteria activate neutrophils and elicit NET formation, we cultivated subsets of bacteria using selective culture media: TS broth to cultivate both beneficial and potentially pathogenic gut bacteria (lactobacilli and E. coli); LB broth to cultivate E. coli; and MRS broth to promote the growth of lactobacilli. We hypothesized that the overall balance in immunosuppressive vs. immunostimulatory effects of the gut microbial communities would determine the extent of neutrophil activation. Indeed, in support of this premise, we demonstrate the capacity of the LB broth-cultivated bacteria (from PBS- and MRS-treated, sham-infected

mice), but not the TS broth-cultivated community, to induce oxidative burst from BMDN. MRS broth-cultivated bacterial communities failed to mobilize ROS. The ability of subsets of the commensal gut microbiome to activate neutrophils highlights the ease with which potentially aggressive bacteria, in the presence of select environmental pressures, can bloom to become pathobionts (33). Infection with C. rodentium dramatically increased Gammaproteobacteria and Actinobacteria levels, which correlated with an increased capacity of all subsets of bacteria from PBS-treated, C. rodentium-infected mice to enhance ROS production in BMDN. By contrast, bacterial populations derived from pathogen-infected mice treated with the lactobacilli-enriched culture had no effect on BMDN, providing further experimental support for the beneficial effects of this bacterial population. Indeed, NETs can be activated by a variety of stimuli, including cell-to-cell interactions (11), phorbol esters, fungi, parasites, microbial components such as LPS, and ROS (6). In this study, we have shown that TS broth-cultivated bacteria from each of the three C. rodentium-infected treatment groups robustly form NETs, despite lack of ROS mobilization in mice administered a lactobacilli-enriched culture. Evidence of ROSindependent activation of NETs has been reported previously (60) in response to Staphylococcus aureus. However, it is noteworthy that MRS-cultivated lactobacilli from each of the three treatment groups failed to induce NETs. Our laboratory previously reported that L. rhamnosus, strain GG, effectively inhibits microbial and pharmacological activation of NETs (77), mediated in part by antioxidative activity. In summary, in this study, we demonstrate that expansion of commensal gut lactobacilli communities, by either selective reconstitution or nutritional supplementation, protects against C. rodentium infection and colitis. During dysbiosis, neutrophils encountering intestinal bacteria can be activated to form NETs. In the context of an inflamed intestine, the induction of NETs likely serves to bolster host innate immune responses to potentially invasive bacteria, whether pathogens or commensals. GRANTS This work was supported by operating grants from the Canadian Institutes of Health Research (MOP-89894 and IOP-92890). P. M. Sherman is the recipient of a Canada Research Chair in Gastrointestinal Disease. DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: L.V. and P.M.S. conception and design of research; L.V., L.J.P., and P.M. performed experiments; L.V., L.J.P., P.M., C.W.Y., and E.L. analyzed data; L.V., L.J.P., C.W.Y., and E.L. interpreted results of experiments; L.V., L.J.P., and C.W.Y. prepared figures; L.V. drafted manuscript; L.V., L.J.P., P.M., E.L., and P.M.S. edited and revised manuscript; L.V., L.J.P., P.M., C.W.Y., E.L., and P.M.S. approved final version of manuscript. REFERENCES 1. Assa A, Vong L, Pinnell LJ, Avitzur N, Johnson-Henry KC, Sherman PM. Vitamin D deficiency promotes epithelial barrier dysfunction and intestinal inflammation. J Infect Dis 210: 1296 –1305, 2014. 2. Barthold SW. The microbiology of transmissible murine colonic hyperplasia. Lab Anim Sci 30: 167–173, 1980. 3. Bianchi M, Hakkim A, Brinkmann V, Siler U, Seger RA, Zychlinsky A, Reichenbach J. Restoration of NET formation by gene therapy in CGD controls aspergillosis. Blood 114: 2619 –2622, 2009.

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00053.2015 • www.ajpgi.org

COMMENSAL GUT MICROBES PROTECT AGAINST COLITIS 4. Bindels LB, Beck R, Schakman O, Martin JC, De Backer F, Sohet FM, Dewulf EM, Pachikian BD, Neyrinck AM, Thissen JP, Verrax J, Calderon PB, Pot B, Grangette C, Cani PD, Scott KP, Delzenne NM. Restoring specific lactobacilli levels decreases inflammation and muscle atrophy markers in an acute leukemia mouse model. PLoS One 7: e37971, 2012. 5. Brinkmann V, Reichard U, Goosmann C, Fauler B, Uhlemann Y, Weiss DS, Weinrauch Y, Zychlinsky A. Neutrophil extracellular traps kill bacteria. Science 303: 1532–1535, 2004. 6. Brinkmann V, Zychlinsky A. Neutrophil extracellular traps: is immunity the second function of chromatin? J Cell Biol 198: 773–783, 2012. 7. Casey PG, Casey GD, Gardiner GE, Tangney M, Stanton C, Ross RP, Hill C, Fitzgerald GF. Isolation and characterization of anti-Salmonella lactic acid bacteria from the porcine gastrointestinal tract. Lett Appl Microbiol 39: 431–438, 2004. 8. Casey PG, Gardiner GE, Casey G, Bradshaw B, Lawlor PG, Lynch PB, Leonard FC, Stanton C, Ross RP, Fitzgerald GF, Hill C. A five-strain probiotic combination reduces pathogen shedding and alleviates disease signs in pigs challenged with Salmonella enterica Serovar Typhimurium. Appl Environ Microbiol 73: 1858 –1863, 2007. 9. Claes IJ, De Keersmaecker SC, Vanderleyden J, Lebeer S. Lessons from probiotic-host interaction studies in murine models of experimental colitis. Mol Nutr Food Res 55: 1441–1453, 2011. 10. Claes IJ, Lebeer S, Shen C, Verhoeven TL, Dilissen E, De Hertogh G, Bullens DM, Ceuppens JL, Van Assche G, Vermeire S, Rutgeerts P, Vanderleyden J, De Keersmaecker SC. Impact of lipoteichoic acid modification on the performance of the probiotic Lactobacillus rhamnosus GG in experimental colitis. Clin Exp Immunol 162: 306 –314, 2010. 11. Clark SR, Ma AC, Tavener SA, McDonald B, Goodarzi Z, Kelly MM, Patel KD, Chakrabarti S, McAvoy E, Sinclair GD, Keys EM, AllenVercoe E, Devinney R, Doig CJ, Green FH, Kubes P. Platelet TLR4 activates neutrophil extracellular traps to ensnare bacteria in septic blood. Nat Med 13: 463–469, 2007. 12. Curtis MM, Hu Z, Klimko C, Narayanan S, Deberardinis R, Sperandio V. The gut commensal Bacteroides thetaiotaomicron exacerbates enteric infection through modification of the metabolic landscape. Cell Host Microbe 16: 759 –769, 2014. 13. Darfeuille-Michaud A, Boudeau J, Bulois P, Neut C, Glasser AL, Barnich N, Bringer MA, Swidsinski A, Beaugerie L, Colombel JF. High prevalence of adherent-invasive Escherichia coli associated with ileal mucosa in Crohn’s disease. Gastroenterology 127: 412–421, 2004. 14. Dave M, Higgins PD, Middha S, Rioux KP. The human gut microbiome: current knowledge, challenges, and future directions. Transl Res 160: 246 –257, 2012. 15. De Man JC, Rogosa M, Sharp ME. A medium for the cultivation of Lactobacilli. J Appl Bacteriol 23: 130 –135, 1960. 16. Del Rio D, Stewart AJ, Pellegrini N. A review of recent studies on malondialdehyde as toxic molecule and biological marker of oxidative stress. Nutr Metab Cardiovasc Dis 15: 316 –328, 2005. 17. Deng W, Li Y, Hardwidge PR, Frey EA, Pfuetzner RA, Lee S, Gruenheid S, Strynakda NC, Puente JL, Finlay BB. Regulation of type III secretion hierarchy of translocators and effectors in attaching and effacing bacterial pathogens. Infect Immun 73: 2135–2146, 2005. 18. Dewhirst FE, Chien CC, Paster BJ, Ericson RL, Orcutt RP, Schauer DB, Fox JG. Phylogeny of the defined murine microbiota: altered Schaedler flora. Appl Environ Microbiol 65: 3287–3292, 1999. 19. Ferreyra JA, Wu KJ, Hryckowian AJ, Bouley DM, Weimer BC, Sonnenburg JL. Gut microbiota-produced succinate promotes C. difficile infection after antibiotic treatment or motility disturbance. Cell Host Microbe 16: 770 –777, 2014. 20. Fournier BM, Parkos CA. The role of neutrophils during intestinal inflammation. Mucosal Immunol 5: 354 –366, 2012. 21. Frank DN, St Amand AL, Feldman RA, Boedeker EC, Harpaz N, Pace NR. Molecular-phylogenetic characterization of microbial community imbalances in human inflammatory bowel diseases. Proc Natl Acad Sci U S A 104: 13780 –13785, 2007. 22. Frei R, Akdis M, O’Mahony L. Prebiotics, probiotics, synbiotics, and the immune system: experimental data and clinical evidence. Curr Opin Gastroenterol 31: 153–158, 2015. 23. Fuchs TA, Abed U, Goosmann C, Hurwitz R, Schulze I, Wahn V, Weinrauch Y, Brinkmann V, Zychlinsky A. Novel cell death program leads to neutrophil extracellular traps. J Cell Biol 176: 231–241, 2007. 24. Gareau MG, Sherman PM, Walker WA. Probiotics and the gut microbiota in intestinal health and disease. Nat Rev Gastroenterol Hepatol 7: 503–514, 2010.

G191

25. Grainger JR, Wohlfert EA, Fuss IJ, Bouladoux N, Askenase MH, Legrand F, Koo LY, Brenchley JM, Fraser ID, Belkaid Y. Inflammatory monocytes regulate pathologic responses to commensals during acute gastrointestinal infection. Nat Med 19: 713–721, 2013. 26. Hakkim A, Fuchs TA, Martinez NE, Hess S, Prinz H, Zychlinsky A, Waldmann H. Activation of the Raf-MEK-ERK pathway is required for neutrophil extracellular trap formation. Nat Chem Biol 7: 75–77, 2011. 27. Hakkim A, Furnrohr BG, Amann K, Laube B, Abed UA, Brinkmann V, Herrmann M, Voll RE, Zychlinsky A. Impairment of neutrophil extracellular trap degradation is associated with lupus nephritis. Proc Natl Acad Sci U S A 107: 9813–9818, 2010. 28. Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp Ser 41: 95–98, 1999. 29. Higgins LM, Frankel G, Connerton I, Goncalves NS, Dougan G, MacDonald TT. Role of bacterial intimin in colonic hyperplasia and inflammation. Science 285: 588 –591, 1999. 30. Higgins LM, Frankel G, Douce G, Dougan G, MacDonald TT. Citrobacter rodentium infection in mice elicits a mucosal Th1 cytokine response and lesions similar to those in murine inflammatory bowel disease. Infect Immun 67: 3031–3039, 1999. 31. Honda K, Takeda K. Regulatory mechanisms of immune responses to intestinal bacteria. Mucosal Immunol 2: 187–196, 2009. 32. Johnson-Henry KC, Pinnell LJ, Waskow AM, Irrazabal T, Martin A, Hausner M, Sherman PM. Short-chain fructo-oligosaccharide and inulin modulate inflammatory responses and microbial communities in Caco2bbe cells and in a mouse model of intestinal injury. J Nutr 144: 1725– 1733, 2014. 33. Kamada N, Chen GY, Inohara N, Nunez G. Control of pathogens and pathobionts by the gut microbiota. Nat Immunol 14: 685–690, 2013. 34. Kelly D, Campbell JI, King TP, Grant G, Jansson EA, Coutts AG, Pettersson S, Conway S. Commensal anaerobic gut bacteria attenuate inflammation by regulating nuclear-cytoplasmic shuttling of PPARgamma and RelA. Nat Immunol 5: 104 –112, 2004. 35. Kessenbrock K, Krumbholz M, Schonermarck U, Back W, Gross WL, Werb Z, Grone HJ, Brinkmann V, Jenne DE. Netting neutrophils in autoimmune small-vessel vasculitis. Nat Med 15: 623–625, 2009. 36. Kim SC, Tonkonogy SL, Albright CA, Tsang J, Balish EJ, Braun J, Huycke MM, Sartor RB. Variable phenotypes of enterocolitis in interleukin 10-deficient mice monoassociated with two different commensal bacteria. Gastroenterology 128: 891–906, 2005. 37. Kim SC, Tonkonogy SL, Karrasch T, Jobin C, Sartor RB. Dualassociation of gnotobiotic IL-10-/- mice with 2 nonpathogenic commensal bacteria induces aggressive pancolitis. Inflamm Bowel Dis 13: 1457–1466, 2007. 38. Kostic AD, Xavier RJ, Gevers D. The microbiome in inflammatory bowel disease: current status and the future ahead. Gastroenterology 146: 1489 –1499, 2014. 39. Laval L, Martin R, Natividad J, Chain F, Miquel S, de Maredsous CD, Capronnier S, Sokol H, Verdu E, van Hylckama Vlieg JE, BermudezHumaran L, Smokvina T, Langella P. Lactobacillus rhamnosus CNCM I-3690 and the commensal bacterium Faecalibacterium prausnitzii A2165 exhibit similar protective effects to induced barrier hyper-permeability in mice. Gut Microbes 6: 1–9, 2015. 40. Lebeis SL, Bommarius B, Parkos CA, Sherman MA, Kalman D. TLR signaling mediated by MyD88 is required for a protective innate immune response by neutrophils to Citrobacter rodentium. J Immunol 179: 566 – 577, 2007. 41. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2[-delta delta C(T)] method. Methods 25: 402–408, 2001. 42. Luperchio SA, Schauer DB. Molecular pathogenesis of Citrobacter rodentium and transmissible murine colonic hyperplasia. Microbes Infect 3: 333–340, 2001. 43. Lupp C, Robertson ML, Wickham ME, Sekirov I, Champion OL, Gaynor EC, Finlay BB. Host-mediated inflammation disrupts the intestinal microbiota and promotes the overgrowth of Enterobacteriaceae. Cell Host Microbe 2: 204, 2007. 44. Maaser C, Housley MP, Iimura M, Smith JR, Vallance BA, Finlay BB, Schreiber JR, Varki NM, Kagnoff MF, Eckmann L. Clearance of Citrobacter rodentium requires B cells but not secretory immunoglobulin A (IgA) or IgM antibodies. Infect Immun 72: 3315–3324, 2004. 45. Manthey CF, Calabio CB, Wosinski A, Hanson EM, Vallance BA, Groisman A, Martin MG, Wang JY, Eckmann L. Indispensable func-

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00053.2015 • www.ajpgi.org

G192

46.

47.

48. 49. 50.

51.

52. 53.

54.

55. 56. 57.

58. 59.

60.

61.

62. 63. 64.

COMMENSAL GUT MICROBES PROTECT AGAINST COLITIS

tions of ABL and PDGF receptor kinases in epithelial adherence of attaching/effacing pathogens under physiological conditions. Am J Physiol Cell Physiol 307: C180 –C189, 2014. Martin FP, Wang Y, Sprenger N, Yap IK, Lundstedt T, Lek P, Rezzi S, Ramadan Z, van Bladeren P, Fay LB, Kochhar S, Lindon JC, Holmes E, Nicholson JK. Probiotic modulation of symbiotic gut microbial-host metabolic interactions in a humanized microbiome mouse model. Mol Syst Biol 4: 157, 2008. Martin R, Chain F, Miquel S, Lu J, Gratadoux JJ, Sokol H, Verdu EF, Bercik P, Bermudez-Humaran LG, Langella P. The commensal bacterium Faecalibacterium prausnitzii is protective in DNBS-induced chronic moderate and severe colitis models. Inflamm Bowel Dis 20: 417–430, 2014. Mazmanian SK, Round JL, Kasper DL. A microbial symbiosis factor prevents intestinal inflammatory disease. Nature 453: 620 –625, 2008. McDonald B, Urrutia R, Yipp BG, Jenne CN, Kubes P. Intravascular neutrophil extracellular traps capture bacteria from the bloodstream during sepsis. Cell Host Microbe 12: 324 –333, 2012. McKeel R, Douris N, Foley PL, Feldman SH. Comparison of an espB gene fecal polymerase chain reaction assay with bacteriologic isolation for detection of Citrobacter rodentium infection in mice. Comp Med 52: 439 –444, 2002. Molloy MJ, Grainger JR, Bouladoux N, Hand TW, Koo LY, Naik S, Quinones M, Dzutsev AK, Gao JL, Trinchieri G, Murphy PM, Belkaid Y. Intraluminal containment of commensal outgrowth in the gut during infection-induced dysbiosis. Cell Host Microbe 14: 318 –328, 2013. Monteleone I, Pallone F, Monteleone G. Th17-related cytokines: new players in the control of chronic intestinal inflammation. BMC Med 9: 122, 2011. Moran JP, Walter J, Tannock GW, Tonkonogy SL, Sartor RB. Bifidobacterium animalis causes extensive duodenitis and mild colonic inflammation in monoassociated interleukin-10-deficient mice. Inflamm Bowel Dis 15: 1022–1031, 2009. Nardi RM, Santoro MM, Oliveira JS, Pimenta AM, Ferraz VP, Benchetrit LC, Nicoli JR. Purification and molecular characterization of antibacterial compounds produced by Lactobacillus murinus strain L1. J Appl Microbiol 99: 649 –656, 2005. O’Hara AM, Shanahan F. The gut flora as a forgotten organ. EMBO Rep 7: 688 –693, 2006. Papayannopoulos V, Metzler KD, Hakkim A, Zychlinsky A. Neutrophil elastase and myeloperoxidase regulate the formation of neutrophil extracellular traps. J Cell Biol 191: 677–691, 2010. Pena JA, Li SY, Wilson PH, Thibodeau SA, Szary AJ, Versalovic J. Genotypic and phenotypic studies of murine intestinal lactobacilli: species differences in mice with and without colitis. Appl Environ Microbiol 70: 558 –568, 2004. Penders J, Stobberingh EE, van den Brandt PA, Thijs C. The role of the intestinal microbiota in the development of atopic disorders. Allergy 62: 1223–1236, 2007. Petty NK, Bulgin R, Crepin VF, Cerdeno-Tarraga AM, Schroeder GN, Quail MA, Lennard N, Corton C, Barron A, Clark L, Toribio AL, Parkhill J, Dougan G, Frankel G, Thomson NR. The Citrobacter rodentium genome sequence reveals convergent evolution with human pathogenic Escherichia coli. J Bacteriol 192: 525–538, 2010. Pilsczek FH, Salina D, Poon KK, Fahey C, Yipp BG, Sibley CD, Robbins SM, Green FH, Surette MG, Sugai M, Bowden MG, Hussain M, Zhang K, Kubes P. A novel mechanism of rapid nuclear neutrophil extracellular trap formation in response to Staphylococcus aureus. J Immunol 185: 7413–7425, 2010. Raetz M, Hwang SH, Wilhelm CL, Kirkland D, Benson A, Sturge CR, Mirpuri J, Vaishnava S, Hou B, Defranco AL, Gilpin CJ, Hooper LV, Yarovinsky F. Parasite-induced TH1 cells and intestinal dysbiosis cooperate in IFN-gamma-dependent elimination of Paneth cells. Nat Immunol 14: 136 –142, 2013. Rautava J, Pinnell LJ, Vong L, Akseer N, Assa A, Sherman PM. Oral microbiome composition changes in mouse models of colitis. J Gastroenterol Hepatol 30: 521–527, 2014. Rodrigues DM, Sousa AJ, Johnson-Henry KC, Sherman PM, Gareau MG. Probiotics are effective for the prevention and treatment of Citrobacter rodentium-induced colitis in mice. J Infect Dis 206: 99 –109, 2012. Roe AJ, O’Byrne C, McLaggan D, Booth IR. Inhibition of Escherichia coli growth by acetic acid: a problem with methionine biosynthesis and homocysteine toxicity. Microbiology 148: 2215–2222, 2002.

65. Russell SL, Gold MJ, Hartmann M, Willing BP, Thorson L, Wlodarska M, Gill N, Blanchet MR, Mohn WW, McNagny KM, Finlay BB. Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma. EMBO Rep 13: 440 –447, 2012. 66. Sekirov I, Russell SL, Antunes LC, Finlay BB. Gut microbiota in health and disease. Physiol Rev 90: 859 –904, 2010. 67. Sekirov I, Tam NM, Jogova M, Robertson ML, Li Y, Lupp C, Finlay BB. Antibiotic-induced perturbations of the intestinal microbiota alter host susceptibility to enteric infection. Infect Immun 76: 4726 –4736, 2008. 68. Silver S, Johnseine P, Whitney E, Clark D. Manganese-resistant mutants of Escherichia coli: physiological and genetic studies. J Bacteriol 110: 186 –195, 1972. 69. Skinn AC, Vergnolle N, Zamuner SR, Wallace JL, Cellars L, MacNaughton WK, Sherman PM. Citrobacter rodentium infection causes iNOS-independent intestinal epithelial dysfunction in mice. Can J Physiol Pharmacol 84: 1301–1312, 2006. 70. Sokol H, Pigneur B, Watterlot L, Lakhdari O, Bermudez-Humaran LG, Gratadoux JJ, Blugeon S, Bridonneau C, Furet JP, Corthier G, Grangette C, Vasquez N, Pochart P, Trugnan G, Thomas G, Blottiere HM, Dore J, Marteau P, Seksik P, Langella P. Faecalibacterium prausnitzii is an anti-inflammatory commensal bacterium identified by gut microbiota analysis of Crohn disease patients. Proc Natl Acad Sci U S A 105: 16731–16736, 2008. 71. Sokol H, Seksik P, Furet JP, Firmesse O, Nion-Larmurier I, Beaugerie L, Cosnes J, Corthier G, Marteau P, Dore J. Low counts of Faecalibacterium prausnitzii in colitis microbiota. Inflamm Bowel Dis 15: 1183–1189, 2009. 72. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28: 2731–2739, 2011. 73. Tsukahara T, Ushida K. Succinate accumulation in pig large intestine during antibiotic-associated diarrhea and the constitution of succinateproducing flora. J Gen Appl Microbiol 48: 143–154, 2002. 74. Turnbaugh PJ, Hamady M, Yatsunenko T, Cantarel BL, Duncan A, Ley RE, Sogin ML, Jones WJ, Roe BA, Affourtit JP, Egholm M, Henrissat B, Heath AC, Knight R, Gordon JI. A core gut microbiome in obese and lean twins. Nature 457: 480 –484, 2009. 75. Turnbaugh PJ, Ley RE, Mahowald MA, Magrini V, Mardis ER, Gordon JI. An obesity-associated gut microbiome with increased capacity for energy harvest. Nature 444: 1027–1031, 2006. 76. Turpin W, Humblot C, Thomas M, Guyot JP. Lactobacilli as multifaceted probiotics with poorly disclosed molecular mechanisms. Int J Food Microbiol 143: 87–102, 2010. 77. Vong L, Lorentz RJ, Assa A, Glogauer M, Sherman PM. Probiotic Lactobacillus rhamnosus inhibits the formation of neutrophil extracellular traps. J Immunol 192: 1870 –1877, 2014. 78. Winter SE, Winter MG, Xavier MN, Thiennimitr P, Poon V, Keestra AM, Laughlin RC, Gomez G, Wu J, Lawhon SD, Popova IE, Parikh SJ, Adams LG, Tsolis RM, Stewart VJ, Baumler AJ. Host-derived nitrate boosts growth of E. coli in the inflamed gut. Science 339: 708–711, 2013. 79. Wlodarska M, Willing B, Keeney KM, Menendez A, Bergstrom KS, Gill N, Russell SL, Vallance BA, Finlay BB. Antibiotic treatment alters the colonic mucus layer and predisposes the host to exacerbated Citrobacter rodentium-induced colitis. Infect Immun 79: 1536 –1545, 2011. 80. Wolvers D, Antoine JM, Myllyluoma E, Schrezenmeir J, Szajewska H, Rijkers GT. Guidance for substantiating the evidence for beneficial effects of probiotics: prevention and management of infections by probiotics. J Nutr 140: 698S–712S, 2010. 81. Wu X, Vallance BA, Boyer L, Bergstrom KS, Walker J, Madsen K, O’Kusky JR, Buchan AM, Jacobson K. Saccharomyces boulardii ameliorates Citrobacter rodentium-induced colitis through actions on bacterial virulence factors. Am J Physiol Gastrointest Liver Physiol 294: G295–G306, 2008. 82. Yeung CW, Woo M, Lee K, Greer CW. Characterization of the bacterial community structure of Sydney Tar Ponds sediment. Can J Microbiol 57: 493–503, 2011. 83. Yipp BG, Petri B, Salina D, Jenne CN, Scott BN, Zbytnuik LD, Pittman K, Asaduzzaman M, Wu K, Meijndert HC, Malawista SE, de Boisfleury CA, Zhang K, Conly J, Kubes P. Infection-induced NETosis is a dynamic process involving neutrophil multitasking in vivo. Nat Med 18: 1386 –1393, 2012.

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00053.2015 • www.ajpgi.org