Advertisement

Perinatal antibiotic-induced shifts in gut microbiota have differential effects on inflammatory lung diseases

  • Author Footnotes
    ∗ These authors contributed equally to this work.
    Shannon L. Russell
    Footnotes
    ∗ These authors contributed equally to this work.
    Affiliations
    Department of Microbiology and Immunology, University of British Columbia, Vancouver, British Columbia, Canada

    Michael Smith Laboratories, University of British Columbia, Vancouver, British Columbia, Canada
    Search for articles by this author
  • Author Footnotes
    ∗ These authors contributed equally to this work.
    Matthew J. Gold
    Footnotes
    ∗ These authors contributed equally to this work.
    Affiliations
    Biomedical Research Center, University of British Columbia, Vancouver, British Columbia, Canada
    Search for articles by this author
  • Lisa A. Reynolds
    Affiliations
    Michael Smith Laboratories, University of British Columbia, Vancouver, British Columbia, Canada
    Search for articles by this author
  • Benjamin P. Willing
    Affiliations
    Michael Smith Laboratories, University of British Columbia, Vancouver, British Columbia, Canada

    Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Alberta, Canada
    Search for articles by this author
  • Pedro Dimitriu
    Affiliations
    Department of Microbiology and Immunology, University of British Columbia, Vancouver, British Columbia, Canada

    Life Sciences Institute, University of British Columbia, Vancouver, British Columbia, Canada
    Search for articles by this author
  • Lisa Thorson
    Affiliations
    Michael Smith Laboratories, University of British Columbia, Vancouver, British Columbia, Canada
    Search for articles by this author
  • Stephen A. Redpath
    Affiliations
    Department of Microbiology and Immunology, University of British Columbia, Vancouver, British Columbia, Canada

    Life Sciences Institute, University of British Columbia, Vancouver, British Columbia, Canada
    Search for articles by this author
  • Georgia Perona-Wright
    Affiliations
    Department of Microbiology and Immunology, University of British Columbia, Vancouver, British Columbia, Canada

    Life Sciences Institute, University of British Columbia, Vancouver, British Columbia, Canada
    Search for articles by this author
  • Marie-Renée Blanchet
    Affiliations
    Centre de Recherche, Institut Universitaire de Cardiologie et de Pneumologie de Québec, Quebec City, Quebec, Canada
    Search for articles by this author
  • William W. Mohn
    Affiliations
    Department of Microbiology and Immunology, University of British Columbia, Vancouver, British Columbia, Canada

    Life Sciences Institute, University of British Columbia, Vancouver, British Columbia, Canada
    Search for articles by this author
  • Author Footnotes
    ‡ These authors contributed equally to this work.
    B. Brett Finlay
    Correspondence
    B. Brett Finlay, PhD, Michael Smith Laboratories, 301 - 2185 East Mall, Vancouver, British Columbia, Canada V6T 1Z4.
    Footnotes
    ‡ These authors contributed equally to this work.
    Affiliations
    Department of Microbiology and Immunology, University of British Columbia, Vancouver, British Columbia, Canada

    Michael Smith Laboratories, University of British Columbia, Vancouver, British Columbia, Canada
    Search for articles by this author
  • Author Footnotes
    ‡ These authors contributed equally to this work.
    Kelly M. McNagny
    Correspondence
    Corresponding authors: Kelly M. McNagny, PhD, Biomedical Research Centre, University of British Columbia, 2222 Health Sciences Mall, Vancouver, British Columbia, Canada V6T 1Z3.
    Footnotes
    ‡ These authors contributed equally to this work.
    Affiliations
    Biomedical Research Center, University of British Columbia, Vancouver, British Columbia, Canada
    Search for articles by this author
  • Author Footnotes
    ∗ These authors contributed equally to this work.
    ‡ These authors contributed equally to this work.
Published:August 18, 2014DOI:https://doi.org/10.1016/j.jaci.2014.06.027

      Background

      Resident gut microbiota are now recognized as potent modifiers of host immune responses in various scenarios. Recently, we demonstrated that perinatal exposure to vancomycin, but not streptomycin, profoundly alters gut microbiota and enhances susceptibility to a TH2 model of allergic asthma.

      Objective

      Here we sought to further clarify the etiology of these changes by determining whether perinatal antibiotic treatment has a similar effect on the TH1/TH17-mediated lung disease, hypersensitivity pneumonitis.

      Methods

      Hypersensitivity pneumonitis was induced in C57BL/6 wild-type or recombination-activating gene 1–deficient mice treated perinatally with vancomycin or streptomycin by repeated intranasal administration of Saccharopolyspora rectivirgula antigen. Disease severity was assessed by measuring lung inflammation, pathology, cytokine responses, and serum antibodies. Microbial community analyses were performed on stool samples via 16S ribosomal RNA pyrosequencing and correlations between disease severity and specific bacterial taxa were identified.

      Results

      Surprisingly, in contrast to our findings in an allergic asthma model, we found that the severity of hypersensitivity pneumonitis was unaffected by vancomycin, but increased dramatically after streptomycin treatment. This likely reflects an effect on the adaptive, rather than innate, immune response because the effects of streptomycin were not observed during the early phases of disease and were abrogated in recombination-activating gene 1–deficient mice. Interestingly, Bacteroidetes dominated the intestinal microbiota of streptomycin-treated animals, while vancomycin promoted the expansion of the Firmicutes.

      Conclusions

      Perinatal antibiotics exert highly selective effects on resident gut flora, which, in turn, lead to very specific alterations in susceptibility to TH2- or TH1/TH17-driven lung inflammatory disease.

      Key words

      Abbreviations used:

      BAL (Bronchoalveolar lavage), DC (Dendritic cell), HP (Hypersensitivity pneumonitis), OTU (Operational taxonomic unit), RAG 1 (Recombination-activating gene 1), SR (Saccharopolyspora rectivirgula), Treg (Regulatory T)
      Hypersensitivity pneumonitis (HP), also known as extrinsic allergic alveolitis, is an exaggerated inflammatory response that develops after repeated exposure to a wide variety of inhaled environmental antigens. It is largely driven by inhalation of organic dust found in a range of occupational and residential settings. A number of environmental antigens are causative agents, including those derived from bacteria, fungi, parasites, animal and plant proteins, and inorganic chemical compounds (reviewed in Girard et al
      • Girard M.
      • Israel-Assayag E.
      • Cormier Y.
      Pathogenesis of hypersensitivity pneumonitis.
      ). Because of the diversity of environmental triggers and the lack of a consistent definition of HP, current epidemiological data are thought to severely underestimate the prevalence of the disease.
      • Girard M.
      • Lacasse Y.
      • Cormier Y.
      Hypersensitivity pneumonitis.
      Unlike asthma, which is classically defined as a TH2-mediated peribronchial airway inflammatory disease, HP is characterized by TH1/TH17-mediated inflammation affecting the parenchyma and small airways (alveoli) of the lungs (reviewed in Schuyler et al
      • Schuyler M.
      • Gott K.
      • Cherne A.
      Mediators of hypersensitivity pneumonitis.
      and McSharry et al
      • McSharry C.
      • Anderson K.
      • Bourke S.J.
      • Boyd G.
      Takes your breath away–the immunology of allergic alveolitis.
      ). In HP, initial antigen exposure induces an influx of neutrophils that, after repeated exposures, is replaced by the infiltration of lymphocytes secreting IFN-γ
      • Yamasaki H.
      • Ando M.
      • Brazer W.
      • Center D.M.
      • Cruikshank W.W.
      Polarized type 1 cytokine profile in bronchoalveolar lavage T cells of patients with hypersensitivity pneumonitis.
      and IL-17A.
      • Simonian P.L.
      • Roark C.L.
      • Born W.K.
      • O'Brien R.L.
      • Fontenot A.P.
      Gammadelta T cells and Th17 cytokines in hypersensitivity pneumonitis and lung fibrosis.
      • Simonian P.L.
      • Roark C.L.
      • Wehrmann F.
      • Lanham A.K.
      • Diaz del Valle F.
      • Born W.K.
      • et al.
      Th17-polarized immune response in a murine model of hypersensitivity pneumonitis and lung fibrosis.
      • Joshi A.D.
      • Fong D.J.
      • Oak S.R.
      • Trujillo G.
      • Flaherty K.R.
      • Martinez F.J.
      • et al.
      Interleukin-17-mediated immunopathogenesis in experimental hypersensitivity pneumonitis.
      Acute lymphocytosis involves mainly CD4+ T cells, whereas chronic forms require CD8+ T cells that can cause irreversible lung damage in the form of fibrosis or emphysema, resulting in significant morbidity and mortality.
      • McSharry C.
      • Anderson K.
      • Bourke S.J.
      • Boyd G.
      Takes your breath away–the immunology of allergic alveolitis.
      Treatment options are limited to corticosteroids and contact avoidance.
      • Girard M.
      • Lacasse Y.
      • Cormier Y.
      Hypersensitivity pneumonitis.
      The most widely used murine model of HP involves repeated intranasal exposure to antigens isolated from the gram-positive thermophile, Saccharopolyspora rectivirgula (SR), a bacterium isolated from moldy hay, which, in humans, causes the disease “farmer's lung.” Repeated inhalation of SR induces a TH1/TH17-type inflammation in the lung characterized by high levels of IFN-γ
      • Gudmundsson G.
      • Hunninghake G.W.
      Interferon-gamma is necessary for the expression of hypersensitivity pneumonitis.
      and IL-17A,
      • Simonian P.L.
      • Roark C.L.
      • Wehrmann F.
      • Lanham A.K.
      • Diaz del Valle F.
      • Born W.K.
      • et al.
      Th17-polarized immune response in a murine model of hypersensitivity pneumonitis and lung fibrosis.
      • Joshi A.D.
      • Fong D.J.
      • Oak S.R.
      • Trujillo G.
      • Flaherty K.R.
      • Martinez F.J.
      • et al.
      Interleukin-17-mediated immunopathogenesis in experimental hypersensitivity pneumonitis.
      antigen-specific IgG2a antibodies,
      • Girard M.
      • Israel-Assayag E.
      • Cormier Y.
      Pathogenesis of hypersensitivity pneumonitis.
      and formation of parenchymal granulomas.
      • Blanchet M.R.
      • Gold M.J.
      • McNagny K.M.
      Mouse models to evaluate the function of genes associated with allergic airway disease.
      In animal models, T-cell activation is involved in the development of HP
      • Israel-Assayag E.
      • Fournier M.
      • Cormier Y.
      Blockade of T cell costimulation by CTLA4-Ig inhibits lung inflammation in murine hypersensitivity pneumonitis.
      and IL-10 is an important modulator of disease severity.
      • Gudmundsson G.
      • Bosch A.
      • Davidson B.L.
      • Berg D.J.
      • Hunninghake G.W.
      Interleukin-10 modulates the severity of hypersensitivity pneumonitis in mice.
      Dendritic cells (DCs) are also critical for the initiation of antigen-specific T-cell responses; high DC maturation status
      • Girard M.
      • Israel-Assayag E.
      • Cormier Y.
      Mature CD11c(+) cells are enhanced in hypersensitivity pneumonitis.
      and CD34-dependent DC trafficking
      • Blanchet M.R.
      • Bennett J.L.
      • Gold M.J.
      • Levantini E.
      • Tenen D.G.
      • Girard M.
      • et al.
      CD34 is required for dendritic cell trafficking and pathology in murine hypersensitivity pneumonitis.
      are required for robust inflammatory responses to HP antigens. A number of innate immune cells have also been identified as early sources of IL-17A in experimental HP, including neutrophils, monocytes/macrophages, and γδ T cells.
      • Simonian P.L.
      • Roark C.L.
      • Born W.K.
      • O'Brien R.L.
      • Fontenot A.P.
      Gammadelta T cells and Th17 cytokines in hypersensitivity pneumonitis and lung fibrosis.
      • Hasan S.A.
      • Eksteen B.
      • Reid D.
      • Paine H.V.
      • Alansary A.
      • Johannson K.
      • et al.
      Role of IL-17A and neutrophils in fibrosis in experimental hypersensitivity pneumonitis.
      Mounting evidence in human and animal models suggests that gut microbes have a profound effect on the development of allergic asthma.
      • Leavy O.
      Mucosal immunology: the good the gut bugs do.
      • Russell S.L.
      • Finlay B.B.
      The impact of gut microbes in allergic diseases.
      • Brown E.M.
      • Arrieta M.C.
      • Finlay B.B.
      A fresh look at the hygiene hypothesis: how intestinal microbial exposure drives immune effector responses in atopic disease.
      By using germ-free mice
      • Herbst T.
      • Sichelstiel A.
      • Schar C.
      • Yadava K.
      • Burki K.
      • Cahenzli J.
      • et al.
      Dysregulation of allergic airway inflammation in the absence of microbial colonization.
      • Olszak T.
      • An D.
      • Zeissig S.
      • Vera M.P.
      • Richter J.
      • Franke A.
      • et al.
      Microbial exposure during early life has persistent effects on natural killer T cell function.
      and antibiotics,
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      • Russell S.L.
      • Gold M.J.
      • Willing B.P.
      • Thorson L.
      • McNagny K.M.
      • Finlay B.B.
      Perinatal antibiotic treatment affects murine microbiota, immune responses and allergic asthma.
      • Hill D.A.
      • Siracusa M.C.
      • Abt M.C.
      • Kim B.S.
      • Kobuley D.
      • Kubo M.
      • et al.
      Commensal bacteria-derived signals regulate basophil hematopoiesis and allergic inflammation.
      our laboratory and others have shown that the intestinal microbiota profoundly influence disease susceptibility in murine models of allergic asthma. In our studies, we found that mice receiving clinically relevant doses of vancomycin early in life develop more severe asthma than do untreated controls,
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      • Russell S.L.
      • Gold M.J.
      • Willing B.P.
      • Thorson L.
      • McNagny K.M.
      • Finlay B.B.
      Perinatal antibiotic treatment affects murine microbiota, immune responses and allergic asthma.
      which is likely mediated by a mechanism involving elevated IgE levels and reduced regulatory T (Treg) cells.
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      • Russell S.L.
      • Gold M.J.
      • Willing B.P.
      • Thorson L.
      • McNagny K.M.
      • Finlay B.B.
      Perinatal antibiotic treatment affects murine microbiota, immune responses and allergic asthma.
      Interestingly, this effect was not observed in mice given streptomycin, nor when either antibiotic was administered to adult mice.
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      Thus, our data point to a perinatal “window” when alterations in specific members of gut microbiota can influence asthma susceptibility. Whether this enhanced sensitivity to TH2-driven disease leads to protection from TH1/TH17-driven disease is unknown.
      Unlike allergic asthma,
      • Marra F.
      • Marra C.A.
      • Richardson K.
      • Lynd L.D.
      • Kozyrskyj A.
      • Patrick D.M.
      • et al.
      Antibiotic use in children is associated with increased risk of asthma.
      • Murk W.
      • Risnes K.R.
      • Bracken M.B.
      Prenatal or early-life exposure to antibiotics and risk of childhood asthma: a systematic review.
      there is currently no epidemiological evidence linking antibiotic use to an increased risk of HP, perhaps because of the diversity in causative agents or the complex nature of diagnosis and reporting strategies. Likewise, no studies have addressed the role of the microbiota in murine models of HP. Therefore, in this study, we have used a murine model to test the hypothesis that antibiotic perturbation of the intestinal microbiota alters susceptibility to experimental HP.

      Methods

       Mice

      C57BL/6J and B6.129S7-Rag1tm1Mom/J (Rag1KO) mice (Jackson Laboratories, Bar Harbor, Me) were bred and maintained in a specific pathogen-free facility at The Biomedical Research Centre. All experiments were conducted in accordance with the UBC Animal Care Committee guidelines.

       Antibiotic treatment

      C57BL/6J and RAG1KO breeding pairs were given vancomycin or streptomycin (Sigma-Aldrich, St Louis, Mo) at 200 mg/L in drinking water. Pups born from respective breeding pairs were reared on antibiotic-treated water for the duration of the experiment. These antibiotics were chosen because they are poorly absorbed when given orally, minimizing the risk of systemic effects on the host. The rationale behind the clinically relevant concentrations used has been described in detail previously.
      • Sekirov I.
      • Tam N.M.
      • Jogova M.
      • Robertson M.L.
      • Li Y.
      • Lupp C.
      • et al.
      Antibiotic-induced perturbations of the intestinal microbiota alter host susceptibility to enteric infection.

       Induction of experimental HP

      HP was induced in 7-week-old mice as previously described.
      • Israel-Assayag E.
      • Fournier M.
      • Cormier Y.
      Blockade of T cell costimulation by CTLA4-Ig inhibits lung inflammation in murine hypersensitivity pneumonitis.
      • Blanchet M.R.
      • Bennett J.L.
      • Gold M.J.
      • Levantini E.
      • Tenen D.G.
      • Girard M.
      • et al.
      CD34 is required for dendritic cell trafficking and pathology in murine hypersensitivity pneumonitis.
      Mice were sensitized intranasally with 40 μL of 4 mg/mL SR antigen in PBS 3 d/wk for 3 weeks. Mice were sacrificed at the time points indicated. Mice were anesthetized with 500 mg/kg Avertin and blood was collected by cardiac puncture. After sacrifice, bronchoalveolar lavage (BAL) was performed by 3 × 1 mL washes with PBS. Total BAL cells were counted by a hemocytometer and immune cells were differentially quantified from cytospin preparations (Thermo Shandon, Pittsburg, Pa) stained with HemaStain (Fisher Scientific, Ottawa, Canada) on the basis of standard morphological criteria. Cell types were confirmed by flow cytometry (see below).

       Histology

      Lungs were collected and fixed in 10% formalin, embedded in paraffin, cut longitudinally into 5-μm sections, and stained with hematoxylin and eosin. Inflammation was blindly assessed from 5 fields per section, each graded on a scale of 0 to 5 (0 = no signs of disease, 5 = severe disease) for each of the following parameters: (1) peribronchial infiltration, (2) perivascular infiltration, (3) parenchymal infiltration, and (4) granulomas (number and size) for a maximum score of 20.

       Determination of SR-specific antibody responses

      SR-specific IgG1 and IgG2a in serum was measured by ELISA as previously described.
      • Blanchet M.R.
      • Bennett J.L.
      • Gold M.J.
      • Levantini E.
      • Tenen D.G.
      • Girard M.
      • et al.
      CD34 is required for dendritic cell trafficking and pathology in murine hypersensitivity pneumonitis.
      Plates were coated with 10 μg/mL SR antigen in 0.1 M sodium carbonate buffer overnight at 4°C. Serum was serially diluted from 1:2000 to 1:32000. SR-specific IgGs were detected using biotin-conjugated rat anti-mouse IgG1 and IgG2a followed by streptavidin-horseradish peroxidase and 3,3′,5,5′-Tetramethylbenzidine (TMB) substrate (both BD Biosciences, Franklin Lakes, NJ). Absorbance was measured at 450 nm using a Tecan Infinite 200 plate reader (Männedorf, Switzerland).

       RNA isolation and quantitative PCR

      RNA samples were prepared using an RNeasy Mini Kit, and cDNA was synthesized using the QuantiTect RT Kit (both Qiagen, Valencia, Calif). Real-time PCR was performed using a QuantiTect SYBR Green Master Mix (Qiagen) and a 7500 Fast Real-Time System (Applied Biosystems, Carlsbad, Calif). Values were normalized to the expression of glyceraldehyde 3-phosphate dehydrogenase for each sample. Primers for quantitative PCR: Ifng: 5′-TCAAGTGGCATAGATGTGGAAGAA-3′ (F) and 5′-TGGCTCTGCAGGATTTTCATG-3′ (R); Il17a: 5′-CTCTCCACCGCAATGAAGAC-3′ (F) and 5′-AGCTTTCCCTCCGCATTGA-3′ (R); Gapdh: 5′-ATTGTCAGCAATGCATCCTG-3′ (F) and 5′-ATGGACTGTGGTCATGAGCC-3′ (R).

       Cytokine analysis of lung homogenates and cell supernatants

      Lung lobes were collected and homogenized in PBS plus protease inhibitors (Roche, Branford, Conn). Supernatants were analyzed for cytokines using cytokine bead array flex sets for IL-17A and IFN-γ (BD Biosciences). Analysis was performed using an LSR II as described above. Cytokine concentrations were normalized to total protein, assessed by Bradford assay reagent (Pierce, Rockford, Ill).

       Isolation of immune cells and flow cytometry

      BAL, lung, and lung-draining lymph node (mediastinal lymph node) cells were isolated and stained with fluorochrome-conjugated antibodies against 7/4 (Abcam, Cambridge, Mass), B220, CD25, CD19, CD3, CD4, CD8, Gr-1, CD11b, CD11c, NK1.1, and CD45 (eBioscience, San Diego, Calif) and Siglec-F (BD Biosciences). Innate lymphoid cells were defined as CD25+, lineage (CD19, CD3, Gr-1, CD11c, CD11b, NK1.1), alveolar macrophages as CD11cHi SiglecF+, lymphocytes as CD3+B220+, and neutrophils as CD11cSiglecFCD11b+Gr-1+. Intracellular cytokine production was assessed by incubating 2 × 106 lung or lymph node cells with 0.5 μg/mL phorbol 12-myristate 13-acetate and 1 μg/mL ionomycin for 4 hours with Brefeldin A (10 μg/mL). Stimulated cells were stained intracellularly (BD Cytofix/Cytoperm) with fluorochrome-conjugated antibodies to IFN-γ or IL-17A (eBioscience). Flow cytometry was performed using an LSR II (BD Biosciences), and data were analyzed with FlowJo 8.7 software (TreeStar, Ashland, Ore). Lung cells were isolated and analyzed by using flow cytometry as previously described.
      • Blanchet M.R.
      • Bennett J.L.
      • Gold M.J.
      • Levantini E.
      • Tenen D.G.
      • Girard M.
      • et al.
      CD34 is required for dendritic cell trafficking and pathology in murine hypersensitivity pneumonitis.

       Microbial analysis

      For composition analyses, stool pellets from untreated or antibiotic-treated SR-challenged mice were homogenized using a 60s bead-beating method (FastPrep instrument, MP Biomedicals, Solon, Ohio) and total DNA was extracted (QIAamp DNA Stool Mini Kit, Qiagen). 16S ribosomal RNA gene fragments were amplified using 33 nucleotide-bar-coded primer pairs (27F; 5′-AGAGTTTGATCMTGGCTCAG-3′), (519R; 5′-GWATTACCGCGGCKGCTG-3′). PCR products were gel-purified (QIAquick gel extraction kit, Qiagen). Each amplicon (100 ng) was pooled and pyrosequenced using a 454 Titanium platform (Roche). Further information is listed in the Methods section in this article's Online Repository at www.jacionline.org.

       Bacterial taxa associated with disease severity

      We used Mann-Whitney U tests to compare the relative abundances of all OTUs identified in streptomycin-treated mice relative to untreated control mice after 3 weeks of SR exposure. Significantly different OTUs after false-discovery rate correction using the Benjamini-Hochberg method (P < .05) were used to generate a heat map, displaying log-transformed relative abundances (relative abundances were multiplied by 100 before log-transforming, so that a scale of 2 [most abundant, RED] to zero [below the limit of detection, BLUE] is created). This transformation produced the best visual separation among samples. Correlations between OTUs that differed significantly between control and streptomycin treatment were generated by comparing BAL counts from individual control and streptomycin-treated animals with relative abundances of these OTUs. OTUs that positively correlated with disease severity are shown in green, whereas OTUs that negatively correlated with disease severity are shown in brown. (R values shown in boxes). A Spearman correlation coefficient (R) was generated for each OTU in the relative abundance heat map. Coefficients are shown to the right of the heat map in green/brown boxes. Stronger correlations are indicated by the color and value of R.

       Statistics

      Differences between control and experimental groups were compared using 1-way ANOVA to calculate statistical significance (GraphPad Prism software, version 6.0, San Diego, Calif).

      Results

       Antibiotic treatment increases the severity of HP

      HP was induced in mice exposed pre- and perinatally to vancomycin or streptomycin by intranasal instillation of SR antigen 3 times a week for 3 weeks. In contrast to what we observed in a TH2 model of allergic asthma,
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      • Russell S.L.
      • Gold M.J.
      • Willing B.P.
      • Thorson L.
      • McNagny K.M.
      • Finlay B.B.
      Perinatal antibiotic treatment affects murine microbiota, immune responses and allergic asthma.
      mice treated with streptomycin, but not vancomycin, exhibited exacerbated disease after repeated exposure to SR as assessed by total leukocytes in the BAL (Fig 1, A). Differential analyses of infiltrating airway leukocytes from SR-challenged animals revealed that streptomycin-treated mice had increased numbers of lymphocytes and macrophages (Fig 1, B), consistent with these cell types playing important roles in HP pathogenesis. Interestingly, relative frequencies of these cell types in the BAL of streptomycin-treated mice were not different from those of control SR-challenged mice (see Fig E1, A, in this article's Online Repository at www.jacionline.org), arguing for increased severity but not a major alteration in disease course. As expected, no differences in total cell numbers or cell types were observed in the BAL of PBS-challenged control or antibiotic-treated animals, with alveolar macrophages making up the vast majority of leukocytes in the airways (Fig 1, A; see Fig E1, B).
      Figure thumbnail gr1
      Fig 1Early life streptomycin treatment exacerbates HP. A and B, BAL and differential leukocyte counts of control or antibiotic-treated mice challenged for 3 weeks with SR antigen or PBS and sacrificed on day 20 after initial SR exposure. C, Pathological scores and representative H&E-stained lung sections. Scale bar, 300 μm. Error bars are means ± SEM, representative of at least 3 independent experiments (n = 3-6). Statistics shown are based on comparisons to SR-challenged controls. H&E, Hematoxylin and eosin; ND, none detected; Strep, streptomycin; Vanco, vancomycin. *P < .05 and **P < .01.
      The effects of vancomycin and streptomycin on the severity of HP were also examined histologically. Unlike allergic asthma in which lung inflammation is predominantly localized to the bronchi and neighboring arteries, HP is characterized by the formation of granulomatous cell aggregates deep in the alveolar spaces.
      • Sharma O.P.
      • Fujimura N.
      Hypersensitivity pneumonitis: a noninfectious granulomatosis.
      Pathology scores were consistently higher in SR-challenged, streptomycin-treated animals (Fig 1, C) than in control mice across all parameters evaluated: perivascular infiltration, peribronchial infiltration, parenchymal infiltration, and granuloma formation. Interestingly, vancomycin-treated, SR-challenged mice scored higher than SR-challenged control animals in a few categories (perivascular infiltration and granuloma formation). No pathology was observed in PBS-treated control mice and PBS-treated vancomycin and streptomycin-treated animals.
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      In summary, on the basis of an evaluation of BAL infiltrates and histopathology, we find that streptomycin treatment enhances the severity of experimental HP.

       Streptomycin-treated SR-challenged mice have elevated TH1 and TH17 responses

      Next, we characterized the immune response generated in control and antibiotic-treated mice after SR exposure. We detected equivalent levels of SR antigen-specific IgG2a and IgG1 in the serum of untreated and antibiotic-treated mice (Fig 2, A), indicating that antibiotics did not dramatically alter the humoral immune responses in HP. This is consistent with human data that suggest that antigen-specific antibody titers are not good indicators of disease severity and that individuals with high serum antibody titers to HP-inducing antigens can remain asymptomatic.
      • Cormier Y.
      • Belanger J.
      • Durand P.
      Factors influencing the development of serum precipitins to farmer's lung antigen in Quebec dairy farmers.
      Figure thumbnail gr2
      Fig 2IFN-γ and IL-17A are upregulated in streptomycin-exacerbated HP. A, Serum SR-specific IgG1 and IgG2a responses after 3 weeks of SR exposure. B and C, Ifng and Il17a mRNA expression and IFN-γ and IL-17A cytokine production in lung homogenates from control and antibiotic-treated mice (means of 3-6 mice per group ± SEM, 1 of at least 2 independent experiments). Data in Fig 2 (C) are pooled from 2 experiments. Statistics shown are based on comparisons to SR-challenged controls. Eosinos, Eosinophils; Lymphos, lymphocytes; Macs, macrophages; ND, none detected; Neutros, neutrophils; Strep, streptomycin; Vanco, vancomycin. *P < .05 and **P < .01.
      It is well established that TH1 and TH17 immune responses are important contributors to disease pathogenesis in HP.
      • Simonian P.L.
      • Roark C.L.
      • Wehrmann F.
      • Lanham A.K.
      • Diaz del Valle F.
      • Born W.K.
      • et al.
      Th17-polarized immune response in a murine model of hypersensitivity pneumonitis and lung fibrosis.
      • Abdelsamed H.A.
      • Desai M.
      • Nance S.C.
      • Fitzpatrick E.A.
      T-bet controls severity of hypersensitivity pneumonitis.
      • Schuyler M.
      • Gott K.
      • Cherne A.
      • Edwards B.
      Th1 CD4+ cells adoptively transfer experimental hypersensitivity pneumonitis.
      Correspondingly, we found increased mRNA transcripts for Ifng and Il17a in lungs of streptomycin-treated mice relative to vancomycin-treated and control mice after SR challenge (Fig 2, B), and elevated IFN-γ and IL-17A cytokines in lung homogenates from the same mice (Fig 2, C). Lung cells from SR-challenged mice restimulated with SR antigen trended toward increased IL-17A production in the streptomycin-treated mice; however, this did not reach significance (data not shown). There were no differences in the frequency of lung or mediastinal lymph node CD4+ T cells producing IL-17A or IFN-γ in streptomycin-treated mice compared with control animals after SR exposure.
      Several members of the intestinal microbiota have been implicated in the development of TH1,
      • Mazmanian S.K.
      • Liu C.H.
      • Tzianabos A.O.
      • Kasper D.L.
      An immunomodulatory molecule of symbiotic bacteria directs maturation of the host immune system.
      TH17,
      • Ivanov II,
      • Atarashi K.
      • Manel N.
      • Brodie E.L.
      • Shima T.
      • Karaoz U.
      • et al.
      Induction of intestinal Th17 cells by segmented filamentous bacteria.
      and Treg-cell responses.
      • Atarashi K.
      • Tanoue T.
      • Oshima K.
      • Suda W.
      • Nagano Y.
      • Nishikawa H.
      • et al.
      Treg induction by a rationally selected mixture of Clostridia strains from the human microbiota.
      • Atarashi K.
      • Tanoue T.
      • Shima T.
      • Suda W.
      • Nagano Y.
      • Nishikawa H.
      • et al.
      Induction of colonic regulatory T cells by indigenous Clostridium species.
      To determine whether there was an intrinsic alteration in a TH1 or TH17 pathway, we investigated whether the DCs or T cells from antibiotic-treated mice were skewed toward TH1 or TH17 responses in the steady state. There were no differences in cytokine secretion by OT-II T cells conditioned by ovalbumin-pulsed DCs from control, vancomycin-treated, or streptomycin-treated mice (see Fig E2, A, in this article's Online Repository at www.jacionline.org). Similarly, splenic T cells isolated from control or antibiotic-treated mice, cultured in neutral, TH1-inducing, or TH17-inducing conditions, did not differ in their intracellular IFN-γ or IL-17A production, suggesting that there is no systemic skewing of T-cell responsiveness after streptomycin treatment (see Fig E2, B).

       Adaptive immunity likely contributes to streptomycin-exacerbated HP

      To determine whether the enhanced severity of HP in mice treated with streptomycin was due to altered innate or adaptive immune responses, we challenged mice with SR antigen for 1 or 3 consecutive days and measured lung inflammation 24 hours later to highlight contributions from innate cells before robust adaptive responses could be generated. After 1 or 3 SR exposures, there were no significant differences in total inflammatory cells in the BAL between SR-challenged, streptomycin-treated animals and untreated controls (Fig 3, A). Early SR exposure is characterized by an influx of neutrophils, which we observed on differential analysis of the BAL. However, there were no significant differences in the frequency of neutrophils, macrophages, or lymphocytes in the BAL of mice treated with streptomycin relative to controls (Fig 3, B). Similarly, there were no differences in Ifng or Il17a expression (Fig 3, C) or IL-17A+/IFN-γ+–producing innate lymphoid cells (Fig 3, D) in the lung after short-term SR exposure.
      Figure thumbnail gr3
      Fig 3Streptomycin does not alter early innate immune responses in HP. A, Total BAL counts of control or streptomycin-treated mice challenged with SR for 1 (1 × SR) or 3 (3 × SR) consecutive days. B, Frequency of leukocyte subsets in BAL of SR-challenged mice. C, Ifng and Il17a mRNA expression in lung tissue from SR-challenged control and streptomycin-treated mice. D, Intracellular IFN-γ or IL-17A production by innate immune cells (CD45+CD25+lineage) isolated from lung tissue of SR-challenged mice. Means of 3 to 6 mice per group ± SEM, representative of 2 independent experiments. NS, Not significant; Strep, streptomycin.
      Because IL-17A–producing γδ T cells have been implicated in HP pathogenesis,
      • Simonian P.L.
      • Roark C.L.
      • Born W.K.
      • O'Brien R.L.
      • Fontenot A.P.
      Gammadelta T cells and Th17 cytokines in hypersensitivity pneumonitis and lung fibrosis.
      we investigated the prevalence of IL-17A–producing, RORγt+ γδ T cells in the lung after early SR exposure in antibiotic-treated mice. We observed no differences in RORγt+ γδ T cells in streptomycin- or vancomycin-treated mice compared with controls (data not shown).
      In a further attempt to determine whether adaptive immunity plays a role in streptomycin-exacerbated HP, we treated recombination-activating gene 1 (RAG 1)-deficient mice (lacking mature T and B cells) with vancomycin or streptomycin perinatally, and then exposed them to SR for 3 weeks. Interestingly, although RAG 1-deficient mice showed increased BAL cell counts after SR exposure, no differences were detected in BAL total cells (Fig 4, A) or differential cell counts (Fig 4, B and C) between streptomycin-treated and control SR-challenged RAG 1–deficient mice, suggesting that an adaptive immune cell lineage is likely mediating this streptomycin-induced HP phenotype in wild-type mice.
      Figure thumbnail gr4
      Fig 4Adaptive immunity plays a role in streptomycin-exacerbated HP. A, BAL counts of control or antibiotic-treated, SR-challenged (3 weeks) RAG 1–deficient or wild-type (WT) mice. Total number (B) or frequency (C) of leukocyte subsets in BAL of SR-challenged mice. Data shown are means of 3 to 6 mice per group ± SEM. Statistics shown are based on comparisons to WT controls. Eosinos, Eosinophils; Lymphos, lymphocytes; Macs, macrophages; Neutros, neutrophils; NK, natural killer; Strep, streptomycin; Vanco, vancomycin. *P < .05 and ** P < .01.

       Streptomycin treatment alters the composition of the intestinal microbiota

      Previously, we found that perinatal vancomycin treatment profoundly altered the gut microbiota of mice pre- and postinduction of experimental asthma, whereas streptomycin treatment resulted in more subtle shifts in microbial composition.
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      We therefore sought to determine whether these same trends existed after SR-induced HP.
      Changes in gut composition after antibiotic treatment and HP induction were assessed by pyrosequencing the V1 to V2 regions of bacterial 16S ribosomal RNA genes (64,659 quality reads obtained from 33 samples). As expected, there were distinct differences among antibiotic-treated groups and untreated controls after HP induction (Fig 5, A). As previously observed,
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      vancomycin induced profound shifts in microbial community composition at the family level relative to control and streptomycin-treated animals (Fig 5, A). Also, at the operational taxonomic unit (OTU) level, the effect of vancomycin was much greater than that of streptomycin with respect to community similarity, demonstrated by principle coordinate analyses (Fig 5, B), Bray-Curtis similarity index (control vs control, 64.7; control vs streptomycin, 29.1; control vs vancomycin, 1.81) and diversity (Inverse Simpson's diversity index: control, 12.3 ± 1.6; streptomycin, 14.0 ± 1.0; vancomycin, 2.7 ± 0.3). Despite their similar diversity, there were notable changes at the family level between control and streptomycin-treated mice. Some of the bacterial families were similarly depleted in both vancomycin- and streptomycin-treated mice. For example, Clostridiaceae 1 had a relative abundance of 1.8% ± 0.7% in control mice but were undetectable in both antibiotic-treated groups (Fig 5, A). Conversely, many bacterial families were differently affected by the 2 antibiotic treatments. Vancomycin-treated mice exhibited a loss/reduction in a number of bacterial families (predominantly Bacteroidetes), which were largely replaced by Firmicutes (Lactobacillus sp), Paenibacillaceae (Paenibacillus), Verrucomicrobia (specifically Akkermansia), and Enterobacteriaceae. In contrast, streptomycin-treated mice exhibited an expansion of Bacteroidetes from 52.2% ± 6.4% to 76.5% ± 18.0% relative abundance, involving only 2 families: Porphyromonadaceae and Bacteroidaceae.
      Figure thumbnail gr5
      Fig 5Bacteroidetes are enriched in mice treated with streptomycin. Pyrosequencing analysis of 16S rRNA gene amplicons was performed on fecal DNA samples from the indicated mice. A, Relative abundance (%) of OTUs grouped at the family level from the indicated SR-challenged mice. Classification scheme: k, kingdom; p, phylum; c, class; o, order; f, family. Each bar represents an average of 8 mice per group, from 2 independent experiments. Rare taxa were removed from the legend, but still included in the graph. B, Principal coordinate analysis (PCO) of bacterial communities. Each colored circle represents 1 mouse. rRNA, Ribosomal RNA; Strep, streptomycin; Vanco, vancomycin.

       Specific members of the microbiota correlate with disease severity in HP

      There were 26 OTUs significantly affected by streptomycin treatment post-HP induction (Fig 6, B). All these OTUs were undetected in vancomycin-treated mice. All OTUs identified as Clostridiales were depleted in mice treated with streptomycin, whereas 12 of 14 OTUs classified within the Bacteroidetes phylum that differed between streptomycin and control were increased after streptomycin treatment. These data coincide with the increase in Bacteroidetes at the family and phylum levels after streptomycin treatment (Fig 5, A). Microbiota profiles of individual mice from SR-challenged control and streptomycin-treated groups were highly variable while following the above trends (see Fig E3 in this article's Online Repository at www.jacionline.org).
      Figure thumbnail gr6
      Fig 6Specific bacterial taxa correlate with disease severity in HP. A, Relative abundance (%) of all OTUs classified as Bacteroidetes or Firmicutes was plotted against BAL infiltrates for individual streptomycin-treated mice. B, Heatmap displaying OTUs enriched in either control (C) or streptomycin (S)-treated mice. Samples were hierarchically clustered using the complete linkage method. Colored “cells” represent log-transformed % relative abundances and vary between blue and red. The numbered vertical bar represents Spearman correlations between differentially abundant OTUs and BAL counts in the lungs of control and streptomycin-treated, SR-challenged mice. Correlation coefficient values are represented by colors ranging from brown (negative correlations) to green (positive correlation). Each column represents 1 mouse.
      There was also substantial variation within the streptomycin-treated mice with respect to HP severity. Thus, to determine whether this was due to the presence or absence of particular bacterial groups, the abundance of select taxa was compared with disease severity (measured by total cells in BAL from streptomycin-treated, SR-challenged mice compared with control-challenged mice; Fig 6, A and B; see Fig E4 in this article's Online Repository at www.jacionline.org). First, given the apparent shift toward OTUs belonging to Bacteroidetes after streptomycin treatment, and Firmicutes after vancomycin treatment, we sought to determine whether there was a correlation between disease severity and all OTUs belonging to Bacteroidetes and Firmicutes. Intriguingly, members of the microbiota belonging to Bacteroidetes (all OTUs) correlated positively with disease severity, while those belonging to Firmicutes correlated negatively with disease severity (Fig 6, A).
      Individual OTUs (Fig 6, B, green/brown boxes) and select taxa (Fig E4) also correlated with disease severity. Interestingly, the Bacteroidetes (all OTUs, R = 0.87, and specific OTUs that corresponded to unclassified members of Bacteroidetes phylum, R = 0.89, and Bacteroidales order, R = 0.78) positively correlated with disease severity (P < .05), while the Firmicutes (all OTUs, R = −0.75, and Lachnospiraceae, R = −0.80) negatively correlated with disease severity. The abundance of other related taxa, such as Bacteroidaceae, Clostridiaceae, and Ruminococcaceae, showed no correlation with disease severity (Fig E4).

       Discussion

      On the basis of our previous findings that perinatal vancomycin, but not streptomycin, treatment had profound effects on disease severity in a TH2 model of allergic asthma,
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      • Russell S.L.
      • Gold M.J.
      • Willing B.P.
      • Thorson L.
      • McNagny K.M.
      • Finlay B.B.
      Perinatal antibiotic treatment affects murine microbiota, immune responses and allergic asthma.
      we set out to investigate how these animals responded in a TH1/TH17 lung disease model of HP in order to gain further insights into how 2 different antibiotics promote distinct immunological outcomes during disease development. Vancomycin and streptomycin shift the composition of the intestinal microbiota in very distinct/contrasting ways, suggesting that alterations in resident gut microbiota may be responsible for their differential effects on asthma severity. Several potential immune mechanisms could account for the increased susceptibility observed in vancomycin-treated mice: (1) a general, microbiota-driven immune skewing away from TH1/17 responses and toward TH2 responses, as has been suggested by others,
      • Hill D.A.
      • Siracusa M.C.
      • Abt M.C.
      • Kim B.S.
      • Kobuley D.
      • Kubo M.
      • et al.
      Commensal bacteria-derived signals regulate basophil hematopoiesis and allergic inflammation.
      (2) a general, microbiota-driven loss of a key immunosuppressive population (eg, Treg cells), or (3) a much more selective set of microbiota-driven changes that lead to distinct alterations in a specific immune response pathway. To explore these possibilities, we assessed the influence of vancomycin and streptomycin on susceptibility to the TH1/TH17-mediated disease, HP. Surprisingly, the severity of HP was relatively unaffected by vancomycin, but was exacerbated by streptomycin as evidenced by increased airway inflammation, enhanced recruitment of lymphocytes and macrophages, increased IFN-γ and IL-17A production, and more severe lung pathology. Analyses of the intestinal microbiota in these mice revealed a subtle shift in community membership by streptomycin, similar to what we described previously.
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      Thus, the data suggest that these antibiotics do not induce global immune skewing toward a specific subset of TH cells, but instead select for microbial communities that directly or indirectly alter specific immunological pathways involved in pulmonary responses to environmental antigens.
      A number of innate and adaptive immune cells play important roles in HP, so we attempted to investigate mechanistically whether streptomycin treatment enhanced innate or adaptive arms of the immune response. We observed no striking differences in total BAL counts, neutrophils, γδ T cells, innate lymphoid cells, or IFN-γ and IL-17A cytokines in mice treated with streptomycin compared with untreated controls after exposing mice to SR for just 1 or 3 consecutive days. Because this time frame would be too early to elicit a specific adaptive immune response, the data suggest that early innate immune responses are not responsible for the exacerbated HP phenotype we observed in streptomycin-treated mice. In addition, we observed no differences in disease severity between streptomycin-treated and control SR-challenged RAG 1–deficient mice. Although this experiment needs to be interpreted with caution because inactivation of the RAG 1 gene could itself shift the microbiota in ways that obscure the effect of streptomycin, at face value, it further supports a role for the adaptive immune response in the streptomycin-induced exacerbation of disease. Although we were unable to identify the specific cell type involved in mediating streptomycin-exacerbated HP, we were able to rule out a number of possibilities: polarized DCs and T cells, systemic SR-specific IgG antibodies, and a number of innate immune cell lineages. Thus, our data would argue for heightened adaptive, T-cell–mediated responses in streptomycin-treated mice exposed repeatedly to SR.
      Previously, we provided evidence that vancomycin treatment increases asthma severity by mechanisms involving elevated production of IgE and a decreased frequency of Treg cells.
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      • Russell S.L.
      • Gold M.J.
      • Willing B.P.
      • Thorson L.
      • McNagny K.M.
      • Finlay B.B.
      Perinatal antibiotic treatment affects murine microbiota, immune responses and allergic asthma.
      The lack of an overt exacerbation of HP pathology or cytokine production in vancomycin-treated mice would argue against a systemwide defect in the immunosuppressive capacity of Treg cells. In addition, it was previously shown that overexpression of the TH2-polarizing transcription factor GATA3 provides protection against HP
      • Matsuno Y.
      • Ishii Y.
      • Yoh K.
      • Morishima Y.
      • Haraguchi N.
      • Nikuchi N.
      • et al.
      Overexpression of GATA-3 protects against the development of hypersensitivity pneumonitis.
      rather than exacerbation. Together, these observations suggest that global TH2 skewing does not mediate vancomycin-exacerbated asthma because these animals were not protected from experimental HP. Thus, different antibiotics appear to exert more subtle effects on responses in specific disease states.
      Examination of the gut bacterial communities in mice exposed to vancomycin or streptomycin perinatally, compared with untreated controls, reveals intriguing changes that may account for the differences in susceptibility to asthma
      • Russell S.L.
      • Gold M.J.
      • Hartmann M.
      • Willing B.P.
      • Thorson L.
      • Wlodarska M.
      • et al.
      Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
      and HP. Remarkably, despite the drastic shift in microbiota composition induced by vancomycin (with diversity measures dropping more than 50% compared with controls), and the minimal shifts in composition induced by streptomycin, only streptomycin had profound effects on HP. Regardless of the degree of perturbation, it is clear that distinct phyla dominate microbiota treated with vancomycin (Firmicutes) versus streptomycin (Bacteroidetes). Bacteroidetes demonstrated the strongest positive association with HP disease severity in streptomycin-treated animals, while loss of select members of the Firmicutes (specifically those belonging to Lachnospiraceae and Clostridiaceae 1) also correlated with disease severity, suggesting that the proportions of 2 dominant bacterial phyla in the gut may play a defining role in HP pathogenesis. In addition, community composition, rather than diversity, appears to be the major driver of dysregulated immune responses in the HP disease model.
      Bacteroidetes (the most over-represented bacterial group after streptomycin treatment) is a diverse phylum, with limited characterization of specific clades. The most well-studied organism, Bacteroides fragilis, has been shown to play a role in the maturation of TH1-driven immune responses early in life by increasing IFN-γ production by CD4+ T cells,
      • Mazmanian S.K.
      • Liu C.H.
      • Tzianabos A.O.
      • Kasper D.L.
      An immunomodulatory molecule of symbiotic bacteria directs maturation of the host immune system.
      a result that could be important in the context of HP. Intriguingly, there is an enterotoxigenic strain of B fragilis that promotes colon tumorigenesis in a Stat3 and TH17-dependent manner,
      • Wu S.
      • Rhee K.J.
      • Albesiano E.
      • Rabizadeh S.
      • Wu X.
      • Yen H.R.
      • et al.
      A human colonic commensal promotes colon tumorigenesis via activation of T helper type 17 T cell responses.
      which further supports a role for this group in influencing TH17 disease.
      Although the precise immunological mechanisms underlying streptomycin-exacerbated HP remain undefined, there is compelling evidence to suggest that members of the microbiota elicit distinct immune cell responses (eg, Treg cell,
      • Atarashi K.
      • Tanoue T.
      • Oshima K.
      • Suda W.
      • Nagano Y.
      • Nishikawa H.
      • et al.
      Treg induction by a rationally selected mixture of Clostridia strains from the human microbiota.
      • Atarashi K.
      • Tanoue T.
      • Shima T.
      • Suda W.
      • Nagano Y.
      • Nishikawa H.
      • et al.
      Induction of colonic regulatory T cells by indigenous Clostridium species.
      • Round J.L.
      • Mazmanian S.K.
      Inducible Foxp3+ regulatory T-cell development by a commensal bacterium of the intestinal microbiota.
      TH17
      • Ivanov II,
      • Atarashi K.
      • Manel N.
      • Brodie E.L.
      • Shima T.
      • Karaoz U.
      • et al.
      Induction of intestinal Th17 cells by segmented filamentous bacteria.
      ) and alter pathogenesis in several disease models including colitis, allergy, and arthritis. Microbiome studies reveal great interindividual variation in the bacterial species that reside in the intestinal microbiota of humans
      • Gill S.R.
      • Pop M.
      • Deboy R.T.
      • Eckburg P.B.
      • Turnbaugh P.J.
      • Samuel B.S.
      • et al.
      Metagenomic analysis of the human distal gut microbiome.
      and mice.
      • Hildebrand F.
      • Nguyen T.L.
      • Brinkman B.
      • Yunta R.G.
      • Cauwe B.
      • Vandenabeele P.
      • et al.
      Inflammation-associated enterotypes, host genotype, cage and inter-individual effects drive gut microbiota variation in common laboratory mice.
      This has fueled the notion that different commensal species provide similar functional roles in different microbial environments. Thus, instead of focusing on the presence or absence of individual species in different disease settings, the microbiome can be viewed as a complex network of interactions that impinge on key functions of the host by the community as a whole. In the context of HP, perhaps streptomycin promotes the overgrowth of bacteria that promote an “inflammatory” milieu that, directly or indirectly, exaggerates TH1/TH17 adaptive immune responses. Further studies will be needed to delineate these mechanisms, but it will be essential to compare and contrast the microbes, their nutrient environments, and host immune cell homeostasis in streptomycin- versus vancomycin-treated backgrounds.
      In summary, we have described a new role for the commensal microbiota in a TH1/TH17-driven inflammatory lung disease model. Our data suggest that the previously documented shifts in gut microbiota that exacerbate allergic asthma have no effect on HP and that instead, HP is exacerbated by a more subtle shift in microbiota induced by streptomycin. Together, these studies argue for a more nuanced role for the microbiota in altering disease development and demonstrate how divergent shifts in microbiota have very different consequences in 2 immunologically distinct disease models. These novel findings have important implications for clinical interventions with both antibiotics and probiotics.
      Key messages
      • Perinatal antibiotic-induced shifts in gut microbiota enhance susceptibility to a murine model of hypersensitivity pneumonitis.
      • Microbiotas dominated by Firmicutes or Bacteroidetes have differential effects on TH2 and TH1/TH17-driven disease models.
      We thank Michael Hughes and Frann Antignano for their helpful advice, Justin Wong and the UBC Flow Cytometry Facility, and members of the BRC Animal Facility for expert animal care.

      Methods

       DC-T-cell polarization assay

      DCs (CD11c+, CD64) were sorted using fluorescence-activated cell sorting from the mesenteric lymph nodes of untreated or antibiotic-treated mice, and CD4+ T cells were magnetically sorted (EasySep Mouse CD4+ T cell Isolation Kit; StemCell Technologies, Vancouver, British Columbia, Canada) from the peripheral lymph nodes of untreated OT-II transgenic mice. DCs and OT-II CD4+ T cells were cultured together for 3 days in the presence of ovalbumin peptide (Invivogen, San Diego, Calif). Culture supernatants were then taken and analyzed for the presence of cytokines using cytokine bead array assays (BD Biosciences).

       Splenic T-cell polarization assay

      CD4+ T cells were isolated from the spleens and lymph nodes of control, vancomycin-, and streptomycin-treated mice by negative selection with magnetic beads (StemCell Technologies). Purified CD4+ cells (2.5 × 105) were cultured for 6 days in culture media (Iscove modified Dulbecco medium supplemented with 10% FBS, 100 U/mL penicillin, 100 μg/mL streptomycin, and 150 μM monothioglycerol) with 1 μg/mL each of plate-bound anti-CD3 (145-2C11) and anti-CD28 (37.51) in the presence of neutral (10 ng/mL IL-2), TH1 (10 ng/mL IL-2, 10 ng/mL IL-12, and 10 μg/mL anti-IL-4 [1B11]), or TH17 (10 ng/mL IL-1b, 20 ng/mL IL-6, 10 ng/mL IL-23, 10 ng/mL TNF-α, 1 ng/mL TGF-β, 10 μg/mL anti-IL-4 [1B11], and 10 μg/mL anti-IFN-γ [XMG1.2]) conditions. For intracellular staining, polarized cells were stimulated for 4 hours with 50 ng/mL phorbol 12-myristate 13-acetate and 750 ng/mL ionomycin in the presence of brefeldin A (GolgiPlug, BD). Cells were stained with fixable viability dye (eBioscience), stained for CD4 (GK1.5, eBioscience), fixed and permeabilized with intracellular staining buffers (eBioscience), and then stained with antibodies against IL-17A (17B7) and IFN-γ (XMG1.2, eBioscience).

       Bioinformatics for microbial analyses

      High-quality sequence reads were determined using mothur according to standard operating procedures described by Schloss et al.
      • Schloss P.D.
      • Westcott S.L.
      • Ryabin T.
      • Hall J.R.
      • Hartmann M.
      • Hollister E.B.
      • et al.
      Introducing mother: open-source, platform-independent, community-supported software for describing and comparing microbial communities.
      Quality sequences were obtained by removing sequences with ambiguous bases, read lengths less than 200 bases, and chimeras (identified using chimera.uchime). Quality sequences were aligned to the Silva bacterial reference alignment and OTUs were generated using a dissimilarity cutoff of 0.03. Sequences were classified using the classify.seqs command within the Ribosomal Database Project as reference. For Fig 5, A, sequences were grouped on the basis of terminal classification, family level and above. If sequences could not be classified at the family level, they were grouped into the taxonomic group (ie, order, class, phylum, etc) in which they could be confidently classified.

       Multivariate analysis of community structures and diversity

      Figure thumbnail fx1
      Fig E1Abundance of immune cell subsets in BAL of SR-challenged antibiotic-treated mice. Frequency (%) of leukocyte subsets relative to total BAL cells from (A) 3-week SR-challenged or (B) 3-week PBS-challenged antibiotic-treated mice, quantified by cytospin or flow cytometry. Means of 3 to 6 mice per group ± SEM, representative of at least 2 independent experiments. Strep, Streptomycin; Vanco, vancomycin.
      Figure thumbnail fx2
      Fig E2Streptomycin treatment does not polarize systemic DC or T-cell responses toward TH1/TH17 effector phenotypes. A, IFN-γ and IL-17A secretion by OT-II cells cultured for 3 days with DCs isolated from mesenteric lymph nodes of control or antibiotic-treated naive animals with or without OVA in the presence of LPS or anti-CD3. Supernatant cytokines from cocultures were quantified by cytokine bead array. B, Intracellular IFN-γ and IL-17A cytokine production by splenic CD4+ T cells cultured for 7 days in neutral, TH1-, or TH17-inducing conditions. Means of 3 mice per group ± SEM and representative data from 1 (A) or 2 (B) independent experiments. OVA, Ovalbumin; Strep, streptomycin; Vanco, vancomycin.
      Figure thumbnail fx3
      Fig E3Interindividual differences exist between streptomycin-treated and control microbial communities. Family-level phylogenetic classification of 16S rRNA gene frequencies in fecal DNA from SR-challenged control (C) and streptomycin-treated (S) animals. Those indicated with a classification level other than family level (f) could be identified confidently only to the level indicated: k, kingdom; p, phylum; c, class; o, order; f, family. Each bar represents 1 mouse. Rare taxa were removed from the legend, but still included in the graph. rRNA, Ribosomal RNA.
      Figure thumbnail fx4
      Fig E4Specific bacterial taxa correlate with disease severity in streptomycin-treated mice. Relative abundance (%) of select bacterial taxa grouped at family level (unless otherwise specified) was plotted against total BAL infiltrates of individual mice treated with streptomycin and challenged for 3 weeks with SR antigen. Significant positive or negative correlation coefficients are indicated beside each corresponding graph. Those with a classification level other than family (f) could be identified confidently only to the level indicated: p, phylum; c, class; o, order; f, family. Each dot on the graph represents abundance/BAL from 1 mouse.

      References

        • Girard M.
        • Israel-Assayag E.
        • Cormier Y.
        Pathogenesis of hypersensitivity pneumonitis.
        Curr Opin Allergy Clin Immunol. 2004; 4: 93-98
        • Girard M.
        • Lacasse Y.
        • Cormier Y.
        Hypersensitivity pneumonitis.
        Allergy. 2009; 64: 322-334
        • Schuyler M.
        • Gott K.
        • Cherne A.
        Mediators of hypersensitivity pneumonitis.
        J Lab Clin Med. 2000; 136: 29-38
        • McSharry C.
        • Anderson K.
        • Bourke S.J.
        • Boyd G.
        Takes your breath away–the immunology of allergic alveolitis.
        Clin Exp Immunol. 2002; 128: 3-9
        • Yamasaki H.
        • Ando M.
        • Brazer W.
        • Center D.M.
        • Cruikshank W.W.
        Polarized type 1 cytokine profile in bronchoalveolar lavage T cells of patients with hypersensitivity pneumonitis.
        J Immunol. 1999; 163: 3516-3523
        • Simonian P.L.
        • Roark C.L.
        • Born W.K.
        • O'Brien R.L.
        • Fontenot A.P.
        Gammadelta T cells and Th17 cytokines in hypersensitivity pneumonitis and lung fibrosis.
        Transl Res. 2009; 154: 222-227
        • Simonian P.L.
        • Roark C.L.
        • Wehrmann F.
        • Lanham A.K.
        • Diaz del Valle F.
        • Born W.K.
        • et al.
        Th17-polarized immune response in a murine model of hypersensitivity pneumonitis and lung fibrosis.
        J Immunol. 2009; 182: 657-665
        • Joshi A.D.
        • Fong D.J.
        • Oak S.R.
        • Trujillo G.
        • Flaherty K.R.
        • Martinez F.J.
        • et al.
        Interleukin-17-mediated immunopathogenesis in experimental hypersensitivity pneumonitis.
        Am J Respir Crit Care Med. 2009; 179: 705-716
        • Gudmundsson G.
        • Hunninghake G.W.
        Interferon-gamma is necessary for the expression of hypersensitivity pneumonitis.
        J Clin Invest. 1997; 99: 2386-2390
        • Blanchet M.R.
        • Gold M.J.
        • McNagny K.M.
        Mouse models to evaluate the function of genes associated with allergic airway disease.
        Curr Opin Allergy Clin Immunol. 2012; 12: 467-474
        • Israel-Assayag E.
        • Fournier M.
        • Cormier Y.
        Blockade of T cell costimulation by CTLA4-Ig inhibits lung inflammation in murine hypersensitivity pneumonitis.
        J Immunol. 1999; 163: 6794-6799
        • Gudmundsson G.
        • Bosch A.
        • Davidson B.L.
        • Berg D.J.
        • Hunninghake G.W.
        Interleukin-10 modulates the severity of hypersensitivity pneumonitis in mice.
        Am J Respir Cell Mol Biol. 1998; 19: 812-818
        • Girard M.
        • Israel-Assayag E.
        • Cormier Y.
        Mature CD11c(+) cells are enhanced in hypersensitivity pneumonitis.
        Eur Respir J. 2009; 34: 749-756
        • Blanchet M.R.
        • Bennett J.L.
        • Gold M.J.
        • Levantini E.
        • Tenen D.G.
        • Girard M.
        • et al.
        CD34 is required for dendritic cell trafficking and pathology in murine hypersensitivity pneumonitis.
        Am J Respir Crit Care Med. 2011; 184: 687-698
        • Hasan S.A.
        • Eksteen B.
        • Reid D.
        • Paine H.V.
        • Alansary A.
        • Johannson K.
        • et al.
        Role of IL-17A and neutrophils in fibrosis in experimental hypersensitivity pneumonitis.
        J Allergy Clin Immunol. 2013; 131: 1663-1673
        • Leavy O.
        Mucosal immunology: the good the gut bugs do.
        Nat Rev Immunol. 2012; 12: 319
        • Russell S.L.
        • Finlay B.B.
        The impact of gut microbes in allergic diseases.
        Curr Opin Gastroenterol. 2012; 28: 563-569
        • Brown E.M.
        • Arrieta M.C.
        • Finlay B.B.
        A fresh look at the hygiene hypothesis: how intestinal microbial exposure drives immune effector responses in atopic disease.
        Semin Immunol. 2013; 25: 378-387
        • Herbst T.
        • Sichelstiel A.
        • Schar C.
        • Yadava K.
        • Burki K.
        • Cahenzli J.
        • et al.
        Dysregulation of allergic airway inflammation in the absence of microbial colonization.
        Am J Respir Crit Care Med. 2011; 184: 198-205
        • Olszak T.
        • An D.
        • Zeissig S.
        • Vera M.P.
        • Richter J.
        • Franke A.
        • et al.
        Microbial exposure during early life has persistent effects on natural killer T cell function.
        Science. 2012; 336: 489-493
        • Russell S.L.
        • Gold M.J.
        • Hartmann M.
        • Willing B.P.
        • Thorson L.
        • Wlodarska M.
        • et al.
        Early life antibiotic-driven changes in microbiota enhance susceptibility to allergic asthma.
        EMBO Rep. 2012; 13: 440-447
        • Russell S.L.
        • Gold M.J.
        • Willing B.P.
        • Thorson L.
        • McNagny K.M.
        • Finlay B.B.
        Perinatal antibiotic treatment affects murine microbiota, immune responses and allergic asthma.
        Gut Microbes. 2013; 4: 158-164
        • Hill D.A.
        • Siracusa M.C.
        • Abt M.C.
        • Kim B.S.
        • Kobuley D.
        • Kubo M.
        • et al.
        Commensal bacteria-derived signals regulate basophil hematopoiesis and allergic inflammation.
        Nat Med. 2012; 18: 538-546
        • Marra F.
        • Marra C.A.
        • Richardson K.
        • Lynd L.D.
        • Kozyrskyj A.
        • Patrick D.M.
        • et al.
        Antibiotic use in children is associated with increased risk of asthma.
        Pediatrics. 2009; 123: 1003-1010
        • Murk W.
        • Risnes K.R.
        • Bracken M.B.
        Prenatal or early-life exposure to antibiotics and risk of childhood asthma: a systematic review.
        Pediatrics. 2011; 127: 1125-1138
        • Sharma O.P.
        • Fujimura N.
        Hypersensitivity pneumonitis: a noninfectious granulomatosis.
        Semin Respir Infect. 1995; 10: 96-106
        • Cormier Y.
        • Belanger J.
        • Durand P.
        Factors influencing the development of serum precipitins to farmer's lung antigen in Quebec dairy farmers.
        Thorax. 1985; 40: 138-142
        • Abdelsamed H.A.
        • Desai M.
        • Nance S.C.
        • Fitzpatrick E.A.
        T-bet controls severity of hypersensitivity pneumonitis.
        J Inflamm (Lond). 2011; 8: 15
        • Schuyler M.
        • Gott K.
        • Cherne A.
        • Edwards B.
        Th1 CD4+ cells adoptively transfer experimental hypersensitivity pneumonitis.
        Cell Immunol. 1997; 177: 169-175
        • Mazmanian S.K.
        • Liu C.H.
        • Tzianabos A.O.
        • Kasper D.L.
        An immunomodulatory molecule of symbiotic bacteria directs maturation of the host immune system.
        Cell. 2005; 122: 107-118
        • Ivanov II,
        • Atarashi K.
        • Manel N.
        • Brodie E.L.
        • Shima T.
        • Karaoz U.
        • et al.
        Induction of intestinal Th17 cells by segmented filamentous bacteria.
        Cell. 2009; 139: 485-498
        • Atarashi K.
        • Tanoue T.
        • Oshima K.
        • Suda W.
        • Nagano Y.
        • Nishikawa H.
        • et al.
        Treg induction by a rationally selected mixture of Clostridia strains from the human microbiota.
        Nature. 2013; 500: 232-236
        • Atarashi K.
        • Tanoue T.
        • Shima T.
        • Suda W.
        • Nagano Y.
        • Nishikawa H.
        • et al.
        Induction of colonic regulatory T cells by indigenous Clostridium species.
        Science. 2011; 331: 337-341
        • Matsuno Y.
        • Ishii Y.
        • Yoh K.
        • Morishima Y.
        • Haraguchi N.
        • Nikuchi N.
        • et al.
        Overexpression of GATA-3 protects against the development of hypersensitivity pneumonitis.
        Am J Respir Crit Care Med. 2007; 176: 1015-1025
        • Wu S.
        • Rhee K.J.
        • Albesiano E.
        • Rabizadeh S.
        • Wu X.
        • Yen H.R.
        • et al.
        A human colonic commensal promotes colon tumorigenesis via activation of T helper type 17 T cell responses.
        Nat Med. 2009; 15: 1016-1022
        • Round J.L.
        • Mazmanian S.K.
        Inducible Foxp3+ regulatory T-cell development by a commensal bacterium of the intestinal microbiota.
        Proc Natl Acad Sci U S A. 2010; 107: 12204-12209
        • Gill S.R.
        • Pop M.
        • Deboy R.T.
        • Eckburg P.B.
        • Turnbaugh P.J.
        • Samuel B.S.
        • et al.
        Metagenomic analysis of the human distal gut microbiome.
        Science. 2006; 312: 1355-1359
        • Hildebrand F.
        • Nguyen T.L.
        • Brinkman B.
        • Yunta R.G.
        • Cauwe B.
        • Vandenabeele P.
        • et al.
        Inflammation-associated enterotypes, host genotype, cage and inter-individual effects drive gut microbiota variation in common laboratory mice.
        Genome Biol. 2013; 14: R4
        • Sekirov I.
        • Tam N.M.
        • Jogova M.
        • Robertson M.L.
        • Li Y.
        • Lupp C.
        • et al.
        Antibiotic-induced perturbations of the intestinal microbiota alter host susceptibility to enteric infection.
        Infect Immun. 2008; 76: 4726-4736