Portal de Periódicos da CAPES

Tumor Necrosis Factor and Its Receptors Are Crucial to Control Mycobacterium bovis Bacillus Calmette-Guerin Pleural Infection in a Murine Model

2016; Elsevier BV; Volume: 186; Issue: 9 Linguagem: Inglês

10.1016/j.ajpath.2016.05.015

1525-2191

Leslie Chávez‐Galán, Dominique Vesin, Noria Segueni, Pritha Prasad, Raphaële Buser-Llinares, Guillaume Blaser, Jean‐Claude Pache, Bernhard Ryffel, Valérie Quesniaux, Irene García,

Pneumonia and Respiratory Infections

Tumor necrosis factor (TNF) is crucial to control Mycobacterium tuberculosis infection, which remains a leading cause of morbidity and mortality worldwide. TNF blockade compromises host immunity and may cause reactivation of latent infection, resulting in overt pulmonary, pleural, and extrapulmonary tuberculosis. Herein, we investigate the roles of TNF and TNF receptors in the control of Mycobacterium bovis bacillus Calmette-Guerin (BCG) pleural infection in a murine model. As controls, wild-type mice and those with a defective CCR5, a receptor that is crucial for control of viral infection but not for tuberculosis, were used. BCG-induced pleural infection was uncontrolled and progressive in absence of TNF or TNF receptor 1 (TNFR1)/TNFR2 (TNFR1R2) with increased inflammatory cell recruitment and bacterial load in the pleural cavity, and heightened levels of pleural and serum proinflammatory cytokines and chemokines, compared to wild-type control mice. The visceral pleura was thickened with chronic inflammation, which was prominent in TNF−/− and TNFR1R2−/− mice. The parietal pleural of TNF−/− and TNFR1R2−/− mice exhibited abundant inflammatory nodules containing mycobacteria, and these mice developed nonresolving inflammation and succumbed from disseminated BCG infection. By contrast, CCR5−/− mice survived and controlled pleural BCG infection as wild-type control mice. In conclusion, BCG-induced pleurisy was uncontrolled in the absence of TNF or TNF receptors with exacerbated inflammatory response, impaired bacterial clearance, and defective mesothelium repair, suggesting a critical role of TNF to control mycobacterial pleurisy. Tumor necrosis factor (TNF) is crucial to control Mycobacterium tuberculosis infection, which remains a leading cause of morbidity and mortality worldwide. TNF blockade compromises host immunity and may cause reactivation of latent infection, resulting in overt pulmonary, pleural, and extrapulmonary tuberculosis. Herein, we investigate the roles of TNF and TNF receptors in the control of Mycobacterium bovis bacillus Calmette-Guerin (BCG) pleural infection in a murine model. As controls, wild-type mice and those with a defective CCR5, a receptor that is crucial for control of viral infection but not for tuberculosis, were used. BCG-induced pleural infection was uncontrolled and progressive in absence of TNF or TNF receptor 1 (TNFR1)/TNFR2 (TNFR1R2) with increased inflammatory cell recruitment and bacterial load in the pleural cavity, and heightened levels of pleural and serum proinflammatory cytokines and chemokines, compared to wild-type control mice. The visceral pleura was thickened with chronic inflammation, which was prominent in TNF−/− and TNFR1R2−/− mice. The parietal pleural of TNF−/− and TNFR1R2−/− mice exhibited abundant inflammatory nodules containing mycobacteria, and these mice developed nonresolving inflammation and succumbed from disseminated BCG infection. By contrast, CCR5−/− mice survived and controlled pleural BCG infection as wild-type control mice. In conclusion, BCG-induced pleurisy was uncontrolled in the absence of TNF or TNF receptors with exacerbated inflammatory response, impaired bacterial clearance, and defective mesothelium repair, suggesting a critical role of TNF to control mycobacterial pleurisy. Tuberculosis (TB) remains a major health problem causing high morbidity and mortality worldwide. Although the pulmonary form is the most common form of TB infection, extrapulmonary TB, including pleural TB, accounts for almost one third of reported TB cases.1Dye C. Raviglione M. Perspective: weigh all TB risks.Nature. 2013; 502: S13Crossref PubMed Scopus (8) Google Scholar, 2Ferrer J. Pleural tuberculosis.Eur Respir J. 1997; 10: 942-947PubMed Google Scholar, 3Kulchavenya E. Extrapulmonary tuberculosis: are statistical reports accurate?.Ther Adv Infect Dis. 2014; 2: 61-70Google Scholar Pleural TB is a frequent clinical problem with fluid and inflammatory cell accumulation in the pleural cavity after Mycobacterium tuberculosis infection.4Antony V.B. Immunological mechanisms in pleural disease.Eur Respir J. 2003; 21: 539-544Crossref PubMed Scopus (86) Google Scholar TB infection is controlled by a T helper 1-type immune response with a critical contribution, but a minority of infected individuals (5% to 10%) will develop active disease. Mycobacteria may remain latent but may be reactivated when host immunity declines.5Russell D.G. Who puts the tubercle in tuberculosis?.Nat Rev Microbiol. 2007; 5: 39-47Crossref PubMed Scopus (473) Google Scholar Pleural TB has been reported as either a primary TB infection or a result of reactivation of latent TB infection, which may be because of different factors, including tumor necrosis factor (TNF) blockade.5Russell D.G. Who puts the tubercle in tuberculosis?.Nat Rev Microbiol. 2007; 5: 39-47Crossref PubMed Scopus (473) Google Scholar, 6Keane J. Gershon S. Wise R.P. Mirabile-Levens E. Kasznica J. Schwieterman W.D. Siegel J.N. Braun M.M. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent.N Engl J Med. 2001; 345: 1098-1104Crossref PubMed Scopus (3265) Google Scholar TNF is an important cytokine involved in the pathogenesis of human inflammatory diseases, such as rheumatoid arthritis and Crohn's disease. Many studies using genetic models of TNF inactivation and treatment with TNF neutralizing molecules have provided evidence implicating TNF as a key factor in host protective immune responses against mycobacterial infections. Mouse models of experimental tuberculosis have shown impaired control of bactericidal mechanisms, defective granuloma formation, and an alteration of T helper 1-type immune responses in mice lacking TNF activity.7Flynn J.L. Chan J. Immunology of tuberculosis.Annu Rev Immunol. 2001; 19: 93-129Crossref PubMed Scopus (1730) Google Scholar, 8Garcia I. Olleros M.L. Quesniaux V.F. Jacobs M. Allie N. Nedospasov S.A. Szymkowski D.E. Ryffel B. Roles of soluble and membrane TNF and related ligands in mycobacterial infections: effects of selective and non-selective TNF inhibitors during infection.Adv Exp Med Biol. 2011; 691: 187-201Crossref PubMed Scopus (25) Google Scholar Transmembrane TNF and soluble TNF mediate protective immunity through TNFR1, whereas TNFR2 down-regulates immune functions and both TNF receptors interact with lymphotoxin α (LTα) produced by activated lymphocytes.8Garcia I. Olleros M.L. Quesniaux V.F. Jacobs M. Allie N. Nedospasov S.A. Szymkowski D.E. Ryffel B. Roles of soluble and membrane TNF and related ligands in mycobacterial infections: effects of selective and non-selective TNF inhibitors during infection.Adv Exp Med Biol. 2011; 691: 187-201Crossref PubMed Scopus (25) Google Scholar, 9Allie N. Grivennikov S.I. Keeton R. Hsu N.J. Bourigault M.L. Court N. Fremond C. Yeremeev V. Shebzukhov Y. Ryffel B. Nedospasov S.A. Quesniaux V.F. Jacobs M. Prominent role for T cell-derived tumour necrosis factor for sustained control of Mycobacterium tuberculosis infection.Sci Rep. 2013; 3: 1809Crossref PubMed Scopus (88) Google Scholar, 10Segueni N. Benmerzoug S. Rose S. Gauthier A. Bourigault M.-L. Reverchon F. Philippeau A. Erard F. Le Bert M. Bouscayrol H. Wachter T. Garcia I. Kollias G. Jacobs M. Ryffel B. Quesniaux V.F.J. Innate myeloid cell TNFR1 mediates first line defence against primary Mycobacterium tuberculosis infection.Sci Rep. 2016; 6: 22454Crossref PubMed Scopus (31) Google Scholar, 11Garcia I, Olleros ML: The roles of tumor necrosis factor and other macrophage-derived cytokines in host defense mechanisms during the course of mycobacterium tuberculosis infection. Current Topics on The Profiles of Host Immunological Response To Mycobacterial Infections, Chapter 1. Edited by Tomioka H. Trivandrum, India: Research Signpost, 2009. pp. 1–46Google Scholar In humans, it has been shown that TNF also plays a critical role in host defense and the clinical use of TNF inhibitors has revealed an increased risk for opportunistic infections, including TB reactivation and new TB infections.6Keane J. Gershon S. Wise R.P. Mirabile-Levens E. Kasznica J. Schwieterman W.D. Siegel J.N. Braun M.M. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent.N Engl J Med. 2001; 345: 1098-1104Crossref PubMed Scopus (3265) Google Scholar Most reactivated TB, in patients treated with TNF inhibitors, develop extrapulmonary TB, which often delays the diagnosis and adequate treatment.6Keane J. Gershon S. Wise R.P. Mirabile-Levens E. Kasznica J. Schwieterman W.D. Siegel J.N. Braun M.M. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent.N Engl J Med. 2001; 345: 1098-1104Crossref PubMed Scopus (3265) Google Scholar, 12Dixon W.G. Hyrich K.L. Watson K.D. Lunt M. Galloway J. Ustianowski A. Symmons D.P. B S R B R Control Centre ConsortiumBSR Biologics RegisterDrug-specific risk of tuberculosis in patients with rheumatoid arthritis treated with anti-TNF therapy: results from the British Society for Rheumatology Biologics Register (BSRBR).Ann Rheum Dis. 2010; 69: 522-528Crossref PubMed Scopus (541) Google Scholar Several studies have noted that the proportion of extrapulmonary TB has increased from 16.4% in 2002 to 22.4% in 2011, but the causes so far have not been identified.3Kulchavenya E. Extrapulmonary tuberculosis: are statistical reports accurate?.Ther Adv Infect Dis. 2014; 2: 61-70Google Scholar, 13Peto H.M. Pratt R.H. Harrington T.A. LoBue P.A. Armstrong L.R. Epidemiology of extrapulmonary tuberculosis in the United States, 1993-2006.Clin Infect Dis. 2009; 49: 1350-1357Crossref PubMed Scopus (479) Google Scholar, 14Kruijshaar M.E. Abubakar I. Increase in extrapulmonary tuberculosis in England and Wales 1999-2006.Thorax. 2009; 64: 1090-1095Crossref PubMed Scopus (133) Google Scholar, 15Forssbohm M. Zwahlen M. Loddenkemper R. Rieder H.L. Demographic characteristics of patients with extrapulmonary tuberculosis in Germany.Eur Respir J. 2008; 31: 99-105Crossref PubMed Scopus (123) Google Scholar CCR5, the receptor for regulated on activation normal T cell expressed and secreted (RANTES), macrophage inflammatory protein (MIP)-1α, and MIP-1β, is an important chemokine receptor controlling cell recruitment in different pathological situations, including inflammation and viral infections.16Choe H. Farzan M. Sun Y. Sullivan N. Rollins B. Ponath P.D. Wu L. Mackay C.R. LaRosa G. Newman W. Gerard N. Gerard C. Sodroski J. The beta-chemokine receptors CCR3 and CCR5 facilitate infection by primary HIV-1 isolates.Cell. 1996; 85: 1135-1148Abstract Full Text Full Text PDF PubMed Scopus (2090) Google Scholar However, in contrast to TNF requirement for TB infection control, CCR5 has been shown to be dispensable, despite increased local lymphocytic recruitment, during pulmonary TB infection.17Algood H.M. Flynn J.L. CCR5-deficient mice control Mycobacterium tuberculosis infection despite increased pulmonary lymphocytic infiltration.J Immunol. 2004; 173: 3287-3296Crossref PubMed Scopus (109) Google Scholar, 18Vesosky B. Rottinghaus E.K. Stromberg P. Turner J. Beamer G. CCL5 participates in early protection against Mycobacterium tuberculosis.J Leukoc Biol. 2010; 87: 1153-1165Crossref PubMed Scopus (67) Google Scholar Animal models of pleurisy using Mycobacterium bovis bacillus Calmette-Guerin (BCG) and heat-killed M. tuberculosis infections have shown recruitment of neutrophils, monocytes, eosinophils, and lymphocytes secreting cytokines and chemokines in the pleural spaces and participating to granuloma formation.19Antony V.B. Sahn S.A. Antony A.C. Repine J.E. Bacillus Calmette-Guerin-stimulated neutrophils release chemotaxins for monocytes in rabbit pleural spaces and in vitro.J Clin Invest. 1985; 76: 1514-1521Crossref PubMed Scopus (83) Google Scholar, 20Allen S.S. McMurray D.N. Coordinate cytokine gene expression in vivo following induction of tuberculous pleurisy in guinea pigs.Infect Immun. 2003; 71: 4271-4277Crossref PubMed Scopus (64) Google Scholar, 21Menezes-de-Lima-Junior O. Werneck-Barroso E. Cordeiro R.S. Henriques M.G. Effects of inhibitors of inflammatory mediators and cytokines on eosinophil and neutrophil accumulation induced by Mycobacterium bovis bacillus Calmette-Guerin in mouse pleurisy.J Leukoc Biol. 1997; 62: 778-785PubMed Google Scholar, 22D'Avila H. Roque N.R. Cardoso R.M. Castro-Faria-Neto H.C. Melo R.C. Bozza P.T. Neutrophils recruited to the site of Mycobacterium bovis BCG infection undergo apoptosis and modulate lipid body biogenesis and prostaglandin E production by macrophages.Cell Microbiol. 2008; 10: 2589-2604Crossref PubMed Scopus (69) Google Scholar, 23Souza M.C. Penido C. Costa M.F.S. Henriques M.G. Mechanisms of T-lymphocyte accumulation during experimental pleural infection induced by Mycobacterium bovis BCG.Infect Immun. 2008; 76: 5686-5693Crossref PubMed Scopus (16) Google Scholar In the present study, a mouse model of pleural M. bovis BCG infection has been used to analyze the immunopathology, cell recruitment, and accumulation of inflammatory mediators. Using mice deficient for TNF or for TNFR1 plus TNFR2, we studied the implication and outcomes of pleural mycobacterial disease in immunodeficient mice and compared them to the responses generated in normal mice and those lacking CCR5. We document the kinetics of BCG-induced pleurisy evolution, cell recruitment, pleural cytokine/chemokine production, and immunopathology of visceral and parietal pleural mesothelium. This study shows that TNF or TNF receptor deficiency is associated with exacerbation of proinflammatory cytokine and chemokine production, defective Bacillus killing, hyperplasia, and erosion of the visceral pleura, as well as nodularity of the parietal pleura. C57BL/6 mice and TNF−/−,24Marino M.W. Dunn A. Grail D. Inglese M. Noguchi Y. Richards E. Jungbluth A. Wada H. Moore M. Williamson B. Basu S. Old L.J. Characterization of tumor necrosis factor-deficient mice.Proc Natl Acad Sci U S A. 1997; 94: 8093-8098Crossref PubMed Scopus (693) Google Scholar TNFR1xTNFR2 (TNFR1R2−/−), and CCR5−/− (The Jackson Laboratory, Bar Harbor, ME) were housed in the Transgenose Institute animal facility (Transgenose Institute animal facility, CNRS UPS44, Orleans, France). For infection with M. bovis BCG, adult mice (8 to 12 weeks old) were kept in a biohazard animal unit (Orleans, France). Experiments were approved by the Ethics Committee for Animal Experimentation of CNRS Campus Orleans (number CLE CNRS Campus Orleans 2012-1001). Mycobacterium bovis BCG Pasteur strain 1173 P2 was grown in Middlebrook 7H9 broth containing ADC (Difco, France), and middle-log phase bacilli were washed and frozen aliquots kept at −80°C and then used for pleural infection. Colony-forming units (CFUs) were evaluated in the pleural fluids at day 14 or 28 after infection and in spleen at day 28 after infection, as previously reported.25Olleros M.L. Vesin D. Bisig R. Santiago-Raber M.L. Schuepbach-Mallepell S. Kollias G. Gaide O. Garcia I. Membrane-bound TNF induces protective immune responses to M. bovis BCG infection: regulation of memTNF and TNF receptors comparing two memTNF molecules.PLoS One. 2012; 7: e31469Crossref PubMed Scopus (23) Google Scholar Mycobacterial pleurisy was generated by intrapleural cavity injection of 106 CFUs of BCG Pasteur in 100 μL of saline, as previously reported.22D'Avila H. Roque N.R. Cardoso R.M. Castro-Faria-Neto H.C. Melo R.C. Bozza P.T. Neutrophils recruited to the site of Mycobacterium bovis BCG infection undergo apoptosis and modulate lipid body biogenesis and prostaglandin E production by macrophages.Cell Microbiol. 2008; 10: 2589-2604Crossref PubMed Scopus (69) Google Scholar, 23Souza M.C. Penido C. Costa M.F.S. Henriques M.G. Mechanisms of T-lymphocyte accumulation during experimental pleural infection induced by Mycobacterium bovis BCG.Infect Immun. 2008; 76: 5686-5693Crossref PubMed Scopus (16) Google Scholar, 26Mubarak K.K. Montes-Worboys A. Regev D. Nasreen N. Mohammed K.A. Faruqi I. Hensel E. Baz M.A. Akindipe O.A. Fernandez-Bussy S. Nathan S.D. Antony V.B. Parenchymal trafficking of pleural mesothelial cells in idiopathic pulmonary fibrosis.Eur Respir J. 2012; 39: 133-140Crossref PubMed Scopus (63) Google Scholar Injection into the lung parenchyma results in bilateral pneumothorax and death as mice have only one pleural cavity.26Mubarak K.K. Montes-Worboys A. Regev D. Nasreen N. Mohammed K.A. Faruqi I. Hensel E. Baz M.A. Akindipe O.A. Fernandez-Bussy S. Nathan S.D. Antony V.B. Parenchymal trafficking of pleural mesothelial cells in idiopathic pulmonary fibrosis.Eur Respir J. 2012; 39: 133-140Crossref PubMed Scopus (63) Google Scholar Mice were monitored twice a week and sacrificed 14 or 28 days after infection or observed for survival analysis. Groups of naïve littermates referred as uninfected or day 0 were killed at the same time and analyzed similarly to infected mice. Thoracic cavities from naïve and infected mice were washed with 1 mL of 2 mmol/L EDTA–phosphate-buffered saline, samples were centrifuged, and supernatants containing pleural fluid were frozen at −80°C for cytokine, chemokine, and CFU evaluation. Pleural cells were suspended in phosphate-buffered saline–1% bovine serum albumin, counted, and used for cytospin, followed by May-Grünwald-Giemsa or Ziehl-Neelsen staining. For evaluation of monocytes/macrophages, lymphocytes, neutrophils, and infected cells, 400 to 500 cells were counted on cytospin and reported as percentage of total cell number in each pleural cell sample. Cytokines and chemokines that have been shown to be activated by an i.v. BCG infection were analyzed in the pleural fluid and in the serum of naïve littermates, and at day 14 and 28 after infection.25Olleros M.L. Vesin D. Bisig R. Santiago-Raber M.L. Schuepbach-Mallepell S. Kollias G. Gaide O. Garcia I. Membrane-bound TNF induces protective immune responses to M. bovis BCG infection: regulation of memTNF and TNF receptors comparing two memTNF molecules.PLoS One. 2012; 7: e31469Crossref PubMed Scopus (23) Google Scholar Interferon (IFN)-γ, IL-12p70, IL-6, TNF, IL-10, and monocyte chemotactic protein (MCP)-1 were quantified using a flow cytometry bead array assay (BD, San Jose, CA). Samples were acquired using a FACSCyan Cytometer and analyzed with FCAP Array software version 3.0. RANTES, MIP-1α, and keratinocyte-derived cytokine or CXCL1/KC were quantified by enzyme-linked immunosorbent assay (R&D Systems, Abingdon, UK). Histologic analyses were performed in naïve and infected mouse tissues. Tissues were fixed in 4% phosphate-buffered formalin and paraffin embedded. Sections (5 μm thick) were stained with hematoxylin and eosin or Ziehl-Neelsen. Histologic scores and free alveolar spaces were analyzed, as previously reported.27Segueni N. Vigne S. Palmer G. Bourigault M.-L. Olleros M.L. Vesin D. Garcia I. Ryffel B. Quesniaux V.F.J. Gabay C. Limited contribution of IL-36 versus IL-1 and TNF pathways in host response to mycobacterial infection.PLoS One. 2015; 10: e0126058Crossref Scopus (32) Google Scholar Staining of paraffin-embedded, formaldehyde-fixed lung sections was evaluated for mesothelin expression. After deparaffinization, sections were incubated with anti-mouse mesothelin rabbit IgG affinity-purified antibody (C-ERC; Takara Bio Europe, France) for 90 minutes at room temperature. Second antibody, a goat anti-rabbit IgG horseradish peroxidase conjugate, was incubated for 30 minutes and detected with diaminobenzidine-peroxide substrate solution, and sections were counterstained with hematoxylin. Statistical analyses were performed with GraphPad Prism software version 6.01 (GraphPad Software, La Jolla, CA). P < 0.05 was considered statistically significant. Experiments with two groups were analyzed with an unpaired Student's t test, and experiments with more than two groups with one-way analysis of variance and nonparametric Kruskal-Wallis test. BCG-induced pleural tuberculosis was generated by injecting 106 CFUs BCG Pasteur strain into the pleural cavity. Previous studies have shown that BCG pleural infection induces an early neutrophil influx, followed by monocytes, and lymphocytes in the pleural cavity.22D'Avila H. Roque N.R. Cardoso R.M. Castro-Faria-Neto H.C. Melo R.C. Bozza P.T. Neutrophils recruited to the site of Mycobacterium bovis BCG infection undergo apoptosis and modulate lipid body biogenesis and prostaglandin E production by macrophages.Cell Microbiol. 2008; 10: 2589-2604Crossref PubMed Scopus (69) Google Scholar To examine the importance of TNF, TNF receptors, and CCR5, deficient and wild-type (WT) control mice were infected, and the inflammatory cell recruitment into the pleural cavity at different time points after infection was determined and compared to those observed in naïve littermates. No differences between animal groups were observed in pleural cells from naïve mice. At day 14 after infection, there was a significant increase in the total number of cells (threefold to fivefold) in WT mice, which was more prominent in TNF−/− and TNFR1R2−/− mice; these differences were maintained at day 28 after infection (Figure 1A). This increase concerned lymphocytes and monocytes/macrophages, which were higher in mice deficient for TNF or TNF receptors (Figure 1, B and C). Neutrophils were highly augmented in all infected mice without any difference between the groups (Figure 1D). At day 28 after infection, few cells were positively stained for mesothelin, a protein expressed by pleural mesothelial cells, indicating that at late infection, mesothelial cells can be found in the pleural cavity (Supplemental Figure S1A). Microscopic analysis of pleural cells showed that the most striking changes between groups were the presence of multinucleated giant macrophages (size, 30 to 50 μm), which were more frequently observed in the pleural cavity of TNF−/− and TNFR1R2−/− mice (Figure 1, E–G). These giant macrophages contained numerous acid fast bacilli, as visualized by Ziehl-Neelsen staining, and vacuoles, suggesting a weakened killing activity in TNF−/− and TNFR1R2−/− mice compared to WT or CCR5−/− mice, which could normally kill BCG and avoid intracellular accumulation of vacuoles and bacilli (Figure 1, F and G). We addressed the question if inducible nitric oxide synthase (iNOS) expression was impaired in TNF−/− and TNFR1R2−/− macrophages and observed that iNOS was weakly immunostained in few macrophages of TNF−/− and TNFR1R2−/− compared with WT pleural macrophages (Figure 1H). Thus, the absence of TNF pathway, but not of CCR5, induces an increased accumulation of infected giant macrophages in the pleural cavity, indicating that TNF may play a key role in allowing phagocytes to kill intracellular BCG by activating iNOS. In view of the drastic inflammatory immune cell response, the accumulation of cytokines and chemokines mediating host reaction against mycobacteria was expected, as previously reported, using a different model of BCG infection.25Olleros M.L. Vesin D. Bisig R. Santiago-Raber M.L. Schuepbach-Mallepell S. Kollias G. Gaide O. Garcia I. Membrane-bound TNF induces protective immune responses to M. bovis BCG infection: regulation of memTNF and TNF receptors comparing two memTNF molecules.PLoS One. 2012; 7: e31469Crossref PubMed Scopus (23) Google Scholar Indeed, IFN-γ levels increased in the pleural cavity of BCG-infected WT mice, and increased even >65-fold more in TNF−/− and TNFR1R2−/− mice on day 14 and 28 after infection (Figure 2A). IL-12p70 levels increased in all gene-deficient mice at late infection as compared to WT mice (Figure 2B). By contrast, IL-6 levels in pleural fluid only increased in TNF−/− and TNFR1R2−/− as compared to WT and CCR5−/−mice (Figure 2C). These data suggest exacerbated inflammation in the pleura of TNF−/− and TNFR1R2−/− mice associated with high activity of accumulated cells. TNF release was 25-fold increased in TNFR1R2−/− mice, reaching 4 to 5 ng/mL at day 14 to 28 after infection (Figure 2D). The IL-10 level in the pleural cavity was increased twofold by day 28 after infection in TNF−/− mice as compared to WT mice, but not in TNFR1R2−/− mice (Figure 2E). The differences between TNF−/− and TNFR1R2−/− mice can be explained by the fact that homotrimeric LTα, which plays a limited role in the presence of TNF during BCG infection, signals through TNF receptors.28Allie N. Keeton R. Court N. Abel B. Fick L. Vasseur V. Vacher R. Olleros M.L. Drutskaya M.S. Guler R. Nedospasov S.A. Garcia I. Ryffel B. Quesniaux V.F. Jacobs M. Limited role for lymphotoxin {alpha} in the host immune response to Mycobacterium tuberculosis.J Immunol. 2010; 185: 4292-4301Crossref PubMed Scopus (17) Google Scholar Both TNF−/− and TNFR1R2−/− mice had higher CXCL1/KC and MCP-1 chemokines than WT mice (Figure 2, F and G). Two chemokines signaling through CCR5 were evaluated in the pleura, RANTES and MIP-1α. High levels of RANTES were found in CCR5−/− mice and TNFR1R2−/− but not in TNF−/− mice and WT mice (Figure 2H). In contrast, MIP-1α levels were increased by the infection but similar in all groups of mice (data not shown). We further examined the levels of mediators in the serum of infected mice and observed that pleural BCG infection activated IFN-γ, but TNF−/− and TNFR1R2−/− mice had higher levels of IFN-γ than WT and CCR5−/− mice (Figure 3A). IL-12p70 was augmented in the serum of infected mice, but levels were higher in TNF−/− mice at day 14 after infection (Figure 3B). IL-6 was lightly released in the serum of all mice, but CCR5−/− mice showed low levels at day 28 after infection (Figure 3C). As previously observed in the pleural cavity, TNFR1R2−/− had high levels of TNF in serum (Figure 3D). TNF−/− mice presented high levels of IL-10 in the serum unlike WT, TNFR1R2−/−, or CCR5−/− mice (Figure 3E). MCP-1 was increased by the infection, but no difference between groups of mice was found (Figure 3F). Therefore, absence of TNF/TNFR1R2 pathway exacerbates the pleural proinflammatory environment, which is likely because of uncontrolled infection. The pleural cavity is covered by the visceral pleural mesothelium enveloping the lung parenchyma and the parietal pleura, which lines the inner chest and is composed of the mesothelium and a submesothelium fat layer. We examined both the visceral and the parietal pleura by microscopy to characterize the main pathology occurring in BCG-induced pleurisy. At day 14 after infection, all mice presented hyperplasia of the visceral pleura with increased thickness and no striking differences between the groups (Figure 4). At day 28 of infection, infected mutant mice displayed thick visceral pleura with enlargement of submesothelium at the site of inflammation. TNF−/− and TNFR1R2−/− mice and, to a lesser extent, CCR5−/− mice showed pulmonary lesions, including pleuritis, bronchiolitis, granuloma, pneumonia with increased pleural hyperplasia, decrease of free alveolar spaces, and higher pulmonary cell infiltration than WT mice (Figure 4, A and B). Immunostaining of mesothelin, a protein expressed by the mesothelium, evidenced alteration of the mesothelium barrier in mutant mice, and hyperplasia was observed in TNF−/− and TNFR1R2−/− mice, but was less in CCR5−/− mice at day 14 after infection (Figure 5). At day 28 after infection, TNF−/− and TNFR1R2−/− mice showed mesothelium hyperplasia and focal necrosis with discontinuity of the mesothelial barrier and occasional ulceration of the visceral pleural mesothelium (Figure 5 and Supplemental Figure S1, B and C). We then explored how the parietal pleural was modified by the BCG infection. At day 14 after infection, macroscopic examination evidenced the presence of several white foci resembling fat-like structures adherent to the dorsal part of the chest and diaphragm (Supplemental Figure S2A). White foci were more prominent in TNF−/− and TNFR1R2−/− compared to WT and CCR−/− mice (Figure 6A). Microscopic examination revealed that white nodules were formed by myeloid cells, lymphoid cells, and adipocytes (Figure 6B). The presence of BCG was identified by Ziehl-Neelsen staining, and they were more abundant in TNF−/− and TNFR1R2−/− mouse tissues than in WT and CCR5−/− mice (Figure 6C). At day 28, white nodules were less frequent or absent in WT and CCR5−/− mice but abundant in TNF−/− and TNFR1R2−/− mice. WT mice showed a predominance of adipocytes in the tissue, whereas TNF−/− and TNFR1R2−/− mice had fewer or absent adipocytes and enlarged regions of inflammatory cells (Figure 7A and Supplemental Figure S2B). Granulomas were observed in WT mouse white nodules that covered the parietal pleura, but only disorganized granulomatous structures and lymphocytic accumulation were observed in TNF−/− and TNFR1R2−/− mice by day 28 after infection (Figure 7, A and B). WT mouse nodules exhibited few mycobacteria in macrophages located at the central part of a granuloma (Figure 7C). In contrast, TNF−/− and TNFR1R2−/− mice showed no structured granulomas and mycobacteria were widespread in inflamed pleural submesothelium (Figure 7, B and C).Figure 5Mesothelin immunostaining in naïve littermates (uninfected) and infected mice. Visceral pleural mesothelium is altered by bacillus Calmette-Guerin (BCG) infection. Lung immunostaining of mesothelin detecting the mesothelial cells in uninfected or naïve littermates and after 14 and 28 days of BCG pleural infection. A: No differences are found in uninfected WT, TNF−/−, TNFR1R2−/−, and CCR5−/− mice. B: At day 14 after infection, mice show thickening of the visceral pleura. TNF−/− and TNFR1R2−/− mice show mesothelium hyperplasia and discontinuity of the barrier. C: At day 28 after infection, hyperplasia of the visceral pleural is exhibited in TNF−/− and TNFR1R2−/− mice, which also show eventual necrosis of the mesothelium (arrow). Images are representative of three mice per group. Scale bars = 50μm.View Large