- Email: [email protected]
Bacteroides fragilis induce necrosis on mice peritoneal macrophages: In vitro and in vivo assays
Biochemical and Biophysical Research Communications 387 (2009) 627–632
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Bacteroides fragilis induce necrosis on mice peritoneal macrophages: In vitro and in vivo assays J.M.B.D. Vieira a,b,*, S.H. Seabra a, D.C. Vallim c, M.A. Américo d, S.E.L. Fracallanza d, R.C. Vommaro e, R.M.C.P. Domingues b a
Laboratório de Tecnologia em Cultura de Células, UEZO, Rio de Janeiro, Brazil Laboratório de Biologia de Anaeróbios, IMPPG, UFRJ, Rio de Janeiro, Brazil c Instituto Oswaldo Cruz, Rio de Janeiro, Brazil d Laboratório de Bacteriologia Médica, IMPPG, UFRJ, Rio de Janeiro, Brazil e Laboratório de Ultra-estrutura Celular Hertha Meyer, IBCCF, UFRJ, Brazil b
a r t i c l e
i n f o
Article history: Received 20 May 2009 Available online 2 June 2009
Keywords: Bacteroides fragilis Murine macrophages Bacteria/macrophage interactions NO TUNEL assay PI staining In vivo model Apoptosis Necrosis Scanning electron microscopy
a b s t r a c t Bacteroides fragilis is an anaerobic bacteria component of human intestinal microbiota and agent of infections. In the host B. fragilis interacts with macrophages, which produces toxic radicals like NO. The interaction of activated mice peritoneal macrophages with four strains of B. fragilis was evaluated on this study. Previously was shown that such strains could cause metabolic and morphologic alterations related to macrophage death. In this work propidium iodide staining showed the strains inducing macrophage necrosis in that the labeling was evident. Besides nitroblue tetrazolium test showed that B. fragilis stimulates macrophage to produce oxygen radicals. In vivo assays performed in BalbC mice have results similar to those for in vitro tests as well as scanning electron microscopy, which showed the same surface pore-like structures observed in vitro before. The results revealed that B. fragilis strains studied lead to macrophage death by a process similar to necrosis. Ó 2009 Elsevier Inc. All rights reserved.
Introduction Bacteroides fragilis is an anaerobic bacterium that colonizes the human intestinal tract [1,2]. Although B. fragilis do not represent the major anaerobic component in the gut, it is the most frequently recovered from human peritoneal infection specimens [3,4]. When it escapes the gut, as a result of a disease or intestinal surgery, it can generate a significant pathology, including abscess formation in multiple body sites like abdominal cavity and brain [5]. Such species present some virulence factors, like the capsular polysaccharide complex, which justify its pathogenic behavior [5,6–8]. Macrophages have functions like antigen presentation and secretion of cytokines such as tumor necrosis factor alpha (TNFa) and interleukin-1 (IL-1) [9,10]. The activated macrophages develop an enhanced microbicidal function, producing, for example,
* Corresponding author. Address: Laboratório de Tecnologia em Cultura de Células, UEZO, 23070-200 Rio de Janeiro, Brazil. Fax: +55 21 3332 5591. E-mail addresses: [email protected], [email protected] (J.M.B.D. Vieira). 0006-291X/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2009.05.124
nitric oxide (NO) [11,12]. Nevertheless, not only this production can be inhibited or modified by some pathogens but also the initial contact between cells, which seems to be a survival advantage for them [13–15]. The first step for macrophages eliminate pathogens is phagocytosis followed by intracellular digestion of them, which can result in macrophages apoptosis. The latter has been more frequently associated to the uptake and killing of intracellular pathogenic bacteria such as Mycobacteria, Legionella, Shigella and Salmonella [10,16–18]. Our group, studying the interaction between B. fragilis and mice peritoneal macrophages (M/), has described that B. fragilis interferes with iNOS activity and leads to pore formation in macrophage surface [19]. Besides, the macrophages ultrastructure observation after interaction assays with B. fragilis suggested that the phagocyte was in apoptosis. The aim of this study is to demonstrate if B. fragilis can modulate the respiratory burst in macrophages during interaction and answer a question, B. fragilis induces or not apoptosis on macrophages? So, we decided to realize B. fragilis/ M/ interactions in vitro and in vivo beyond verify what kind of cellular death occurs in resident macrophages after interaction.
628
J.M.B.D. Vieira et al. / Biochemical and Biophysical Research Communications 387 (2009) 627–632
Materials and methods Bacterial strains and culture conditions Four strains defined in a previous work [20] were chosen for this study: ES31-3 (isolated from Human Intestinal Microbiote), Eli1-1 (isolated from Human Intestinal Microbiote), 0789320 (isolated from Human Diarrheic Process) and 25285 (isolated from Abscess – American Type Culture Collection – ATCC). All the strains were maintained frozen at 20 °C as part of Anaerobe Biology Laboratory Culture Collection. The strains were reactivated in Brain Heart Infusion-pre-reduced anaerobically sterilized (BHI-PRAS/OxoidÒ – Basingstoke Hants, England) [21]. The bacterial suspensions used in experiments were obtained from culture in protease–peptone–yeast broth [22] at 37 °C. Peritoneal macrophages The macrophages were obtained as described previously [13]. The cells were stimulated with intraperitoneal inoculation of 30 mg/ml of tioglicolate in 21 days female BalbC mice. After 72 h, the macrophages were aseptically collected through peritoneal injection of cold Dulbeco’s Modified Eagles Medium (DMEM – SigmaTM – St. Louis, MO/USA). The collected cells were incubated in ice and adjusted to 2 106 cells/ml in cold DMEM. After 1 h of adhesion at 37 °C in 5% CO2 atmosphere, the wells were washed with DMEM without serum to remove non-adherent cells and the 24 wells plate was incubated with DMEM supplemented with 5% bovine fetal serum (BFS – LaborclinTM – Pinhais, Paraná, Brazil) inactivated by heating. Analysis of oxygen reactive intermediaries production in M/
with serum-free DMEM again and were mounted in the same medium to immediate observation in a LSM (LSM 310 – ZeissTM). Noninfected macrophages were the negative control. In vivo assays Chambers preparation. chambers were prepared using sterilized insulin needles parts, in which extremities were sealed with nonsterilizing 0.8 and 5 lm pore membranes. In each chamber a minimal hole was made in the needle part to inoculate the strains diluted according to McFarland Scale 2 (6 108 UFC/ml), afterwards the hole was heating closed. Chambers with serum-free BHI-PRAS (OxoidTM – Basingstoke Hants, England) broth were used as controls. Chambers insertion in mice. 21 days BalbC female mice were operated in anesthesia state using CO2 and chambers were inserted in their peritoneal cavity in the region near the intestinal loops, performing the proper aseptic procedures. Chambers extraction. occurred after 24 or 48 h by opening the peritoneal cavity, following the bacterial viability analysis through inoculation in Supplemented Blood Agar (SBA) [24]. Such inoculation was made after serial dilution of chambers contents (10 1 to 10 7 in Blanks-PRAS), in which each dilution was inoculated in SBA. After incubation for 48 h in anaerobiosis at 37 °C, the bacterial growth was analyzed for each dilution basing in experiment control (chambers with sterilized medium also inoculated in SBA). The biggest dilution title having UFCs were considered for calculus. The dilution factor multiplied by the used volume in chamber gives rise to UFCs/ml. Scanning electron microscopy – SEM
The oxidative burst in M/ infected with B. fragilis was detected as described previously [23]. The cells were washed with serumfree DMEM and nitroblue tetrazolium (NBT – SigmaTM) at 0.5 mg/ ml together with B. fragilis in an equivalent concentration of McFarland Scale 2. Yeasts were used as positive control. After 1 h of incubation at 37 °C in 5% CO2 atmosphere, the cells were washed with serum-free DMEM, then fixed in Bouin (7.5 ml of picric acid, 2.5 ml of formaldeide 37–40% and 5 ml of glacial acetic acid) and observed with Axioplan optic microscope (ZeissTM).
Based on literature [25], after M//B. fragilis strains interactions, the cells were washed with DMEM without serum, fixed in Karnovsky (EMSTM) and 2% tanic acid (MerckTM) for 2 h, and washed with PBS. The cells were post fixed in 1% of osmium solution (OsO4 v/ v – SigmaTM) with tanic acid 2% in phosphate buffer, pH 7.2 for 1 h at room temperature (RT) and washed in PBS. Then the cells were dehydrated in ethanol gradual concentration (50–100%/ 10 min each solution – MerckTM), dried through CO2 critic point, mounted in metal support and covered with gold (20–30 nm) for observation in a Scanning Electron Microscope (JSM5310 – JeolTM, Japan).
TUNEL (terminal dUTP nick end labeling) assays: apoptosis analysis
Results
This evaluation was made using APO-BrdU TUNEL Assay kit (A – 23210, Molecular ProbesTM, São Paulo, Brazil). After interactions between B. fragilis and macrophages in 24 well plate, the cells were incubated with alcohol 70% at 4 °C, washed twice with PBS and incubated with 50 ll of DNA linking solution (provided by the kit) for 30 min at 37 °C, light protected. The cover slips were washed with washing buffer (provided by the kit) and incubated with antibody solution (quantity established by the kit) for 30 min at RT. The washing was made twice with washing buffer and the coverslips were mounted in N-propil galato (SigmaTM), following visualization in a Laser Scanning Microscope (LSM 310 – Zeiss, Switzerland). Non-infected macrophages were the negative control.
The four strains studied: ATCC25285, ELI1-1, S31-3, and 078320, were representative of the pulsotypes 4, 5, 16, and 24 [20]. Fig. 1 shows that despite apparent regulation of iNOS activity by B. fragilis based on previous results [19], none bacterial strain could regulate respiratory burst. For apoptosis analysis, interactions of B. fragilis ELI1-1 strain with macrophages were observed by fluorescence and compared with phase contrast. After 2 h of interaction, the macrophages did not suffer apoptosis (Inset Fig. 2F). Necrosis assays have similar results after B. fragilis ATCC25285/M/ interaction in same conditions (Fig. 2C and D). However, using the same interaction period to B. fragilis ELI1-1 strain (Fig. 2E and F) as well as for ES31-3 and 078320 strains (not shown) macrophages necrosis was observed. The same was seen in 4 h interaction experiments (fig. 2G–L) for all strains excepting ATCC25285 that did not discharge necrosis death in macrophages (Fig. 2I and J). To investigate the consequences of in vivo B. fragilis strains/M/ interactions were utilized propylene chambers with 0.8 or 5 lm membranes containing bacterial suspensions into mice peritoneal cavity (Fig. 3). After 24 h of interaction with chamber 0.8 lm, in-
Propidium iodide: necrosis evaluation After bacteria macrophage interaction in 24 well plates, a twice washing with DMEM without serum was performed and each coverslip was incubated with Propidium iodide (100 lg/ml – Molecular ProbesTM) in serum-free media. The coverslips were washed
J.M.B.D. Vieira et al. / Biochemical and Biophysical Research Communications 387 (2009) 627–632
629
fected macrophages did not present apoptosis (Insets Fig. 3D and F). In 24 h interaction experiments with 5 lm chamber some apoptotic macrophages were observed (insets Fig. 3J and L). Similar results were obtained after 48 h of interactions for both chambers (data not shown). However, macrophages necrosis after 24 h of ‘‘in vivo” interaction were observed (Fig. 3) with 0.8 lm (Fig. 3F) and 5 lm (Fig. 3J and L) chambers for 25285 and Eli1-1 strains. To evaluate macrophage membrane rupture in vivo, some experiments of SEM were made after 24 h and 48 h M//B. fragilis strains interaction using 0.8 lm chambers (Fig. 4A–H). The results showed pore-like structures on macrophages membranes (arrows in Fig. 4D, E, G and H). The same results were obtained with 5 lm chambers after 24 h interaction experiment (Fig. 4F and I). Besides, on Fig. 4F (star) is possible to see the modification on macrophage surface in comparison to control (Fig. 4C).
Discussion
Fig. 1. Oxidative burst: evidentiation of oxygen radicals production by macrophages after in vitro interaction with B. fragilis. (A) Negative control, non-infected Mf; (B) Positive control, Mf yeasts; (C) Mf 25285; (D) Mf 078320; (E) Mf Es31-3; (F) Mf Eli 1-1. Note formazane cristals presence (arrows). (A–F): interferential contrast. Barr 20 mm.
Bacteroides fragilis, a commensal anaerobe from intestinal microbiota, can cause infection when extrudes this site through virulence factors production [6–8]. To stay in host, it has to interact and escape from macrophages. The latter are able to destroy microorganisms by phagocyting and digesting them through the production of oxygen radicals [11,12]. In a previous work it was demonstrated that B. fragilis can induce alterations on NO synthesis by macrophages and can cause pore formation on the cell surface of these cells, a suggested way for actin extruding and iNOS release [19]. The intriguing fact that B. fragilis could modify the macrophage response diminishing the NO synthesis raised the question
Fig. 2. Necrosis Mf Evidenciation with PI after 2 h (A–F) and 4 h (G–L) of in vitro interaction with B. fragilis. (A and B)/(G and H) non-infected Mf; (C and D)/(I and J) Mf 25285; (E and F)/(K and L) Mf Eli1-1. PI staining is apparent to Eli1-1 strain (arrows). (A, C, E, G, I and K): phase contrast microscopy; (B, D, F, H, J and L): fluorescence microscopy. Insets shown results from in vitro apoptosis assays. Barr: 20 mm.
630
J.M.B.D. Vieira et al. / Biochemical and Biophysical Research Communications 387 (2009) 627–632
Fig. 3. PI staining of peritoneal M/ from mice infected with B. fragilis in 0.8 lm (A–F) and 5 lm (G–L) chambers for 24 h. (A and B)/(G and H) control, non-infected M/; (C and D)/(I and J) M/ 25285; (E and F)/(K and L) M/ Eli-1. Note the presence of necrotic M/ (arrows). (A, C, E, G, I and K) interpherencial contrast; (B, D, F, H, J and L) fluorescence. Insets: results for apoptosis. Bars: 20 lm.
Fig. 4. Scanning electron micrography of mice peritoneal M/ infected in vivo with B. fragilis strains in 0.8 lm chamber for 24 h (A, D and G) and 48 h (B, E and H), and 5 lm chamber for 24 h (C, F and I). (A, B and C) control, non-infected M/; (D, E and F) M/ 25285; (G, H and I) M/ Eli-1; (H) 48 h M/ Eli1-1. Note the presence of structures similar to pores on the surface of infected M/ (arrows).
if it could interfere with the oxidative burst as well. For all B. fragilis strains tested it was possible to see the presence of formasana
crystals in M/, indicating that even after interaction with B. fragilis, macrophages produce oxidative burst (Fig. 1).
J.M.B.D. Vieira et al. / Biochemical and Biophysical Research Communications 387 (2009) 627–632
It has been described that several kind of microorganisms are able to hide themselves from the immune system by different mechanisms [26–28] and can even induce the host cell death, including those cells that act in the host immunity like macrophages [29–31]. In this context, a hypothesis took place: B. fragilis could also induce macrophage death like other extracellular bacteria? [10,31]. Some assays were performed in vitro to evaluate the kind of death, apoptosis or necrosis, induced by B. fragilis on M/. To achieve this purpose TUNEL technique was performed in vitro after 2 h of B.fragilis/M/ interaction (insets on Fig. 2). The results showed that this bacterial species do not induce macrophage apoptosis, since the specific labeling was not seen for the strains analyzed. The results obtained with the TUNEL in vitro raised the possibility that macrophages could be suffering necrosis instead of apoptosis after contact with B. fragilis. A study developed with Bordetella bronchiseptica showed that its secretion system type III (SSTIII) promotes a death process with similar characteristics from those of necrosis [29]. To evaluate the hypothesis for B. fragilis, IP labeling assays were realized after 2 (Fig. 2A–F) and 4 h (Fig. 2G–L) of interaction with M/. All the strains presented apparent labeling with IP, excepting the strain ATTCC 25285. This finding reinforced the idea that B. fragilis induce macrophages death by a process similar to necrosis. Some experiments were performed in vivo using chambers with different porosity (0.8 and 5 lm) inserted into mice peritoneal cavity for 24 and 48 h (Fig. 3) containing B. fragilis strains suspensions. The aim was compare the results obtained in vitro with data from in vivo assays. First, it was realized the experiments with TUNEL, since apoptosis induction in infection models for many kinds of microorganisms has been widely discussed [32]. Such assays realized after interaction in vivo for 24 (insets on Fig. 3) and 48 h (data not shown) presented the same results for the two kinds of chambers when compared to in vitro findings, which means that specific TUNEL labeling was not observed for any of the strains tested. The labeling observed on Fig. 3J and 3L insets represents the background and it is not an evidence of apoptosis. Similar experiments were performed to test the presence of necrotic macrophages on peritoneal cavity after infection for 24 (Fig. 3A–F) and 48 h (Fig. 3G–L) with B. fragilis. In this case was possible to observe labeling with IP for all the strains analyzed (Fig. 3), which corroborates the data obtained from in vitro analysis. Previously our group found, based on electron microscopy, that B. fragilis infection induced the rupture of macrophage membranes in vitro [19]. In order to make a comparison, we decided to do some tests in vivo. Thus, some SEM experiments were made after 24 h (Fig. 4A, D and G) and 48 h (Fig. 4B, E and H) of B. fragilis strains/ M/ interactions with 0.8 lm chamber and for 24 h in 5 lm chamber (Fig. 4C, F and I). It was not surprising that these findings were similar, since it was possible to observe pore-like structures on macrophages membranes for both chambers (arrows Fig. 4). Such results showed that the alterations on macrophages surface are induced by B. fragilis in vitro as described before [19] and in vivo, possibly leading to leak of cytoplasm contents, including iNOS. In the bacterial genus Yersinia, Shigella and Salmonella the secretion system type III is involved in causing apoptosis on macrophages [33,34] but this system was also described in B. bronchiseptica inducing the macrophage death by a process similar to necrosis [29]. On this context, it is possible that in B. fragilis case systems like this one are acting, being responsible to induce necrosis on infected macrophages. The fact that strain ATCC25285 revealed a different behavior in vitro in comparison to the others strains could be related to the lack of expression or basal expression of its secretion system. Anyway, this strain was able to cause alterations on macrophages metabolism, which was demonstrated by previous results [19] with immunocytochemistry for iNOS and consequent modification in NO production.
631
The utilization of animal models in this work intended to evaluate, based on the results obtained in vitro, particular aspects of M/ infected with B. fragilis, aiming to see the consequences of such interaction. Different models have been proposed to study anaerobe bacteria [35] but the system used in this work has never been used before, which can be seen as an experimental method of alternative tests in animals to study anaerobes. The B. fragilis species shows a wide genetic diversity and has been target of many analysis and discussions [36–39]. However, in our study this diversity did not influence the strains behavior even belonging to different pulsotypes described before [20] and having less than 30% of similarity. All four strains presented similar results in all experiments, excepting ATCC25285 for necrosis evaluation in vitro. In conclusion, it is possible to affirm that different Bacteroides fragilis clonal types induce mice activated peritoneal macrophages death by a process similar to necrosis. Nevertheless, future experiments are certainly needed to clarify this probable mechanism of immune system evasion. Acknowledgments We thank Joaquim Santos Filho for technical assistance. This work was supported by PRONEX-MCT, CNPq, FUJB, and FAPERJ. References [1] N.E.C. Moore, E.P. Cato, L.V. Holdeman, Anaerobic bacteria of gastrointestinal flora and their occurrence in clinical infections, J. Infect. Dis. 119 (1978) 641– 649. [2] I.R. Poxton, R. Brown, A. Sawyerr, A. Ferguson, The mucosal anaerobic Gramnegative bacteria of the human colon, Clin. Infect. Dis. 25 (Suppl. 2) (1997) S111–S113. [3] G.L. Simon, S.L. Gorbach, Intestinal health and disease, Gastroenterology 86 (1984) 174–193. [4] A. Willis, Abdominal sepsis, in: B.I. Duerden, B.S. Drasar (Eds.), Anaerobes in Human Diseases, Edward Arnold, London, 1991, pp. 197–223. [5] L. Pumbwe, C.A. Skilbeck, H.M. Wexler, The Bacteroides fragilis cell envelope: quarterback, linebacker, coach—or all three?, Anaerobe 12 (2006) 211–220 [6] A.B. Onderdonk, N.E. Moon, D.L. Kasper, J.G. Bartlett, Adherence of Bacteroides fragilis in vivo, Infect. Immun. 19 (1977) 1083. [7] S. Patrick, J. Parkhill, L.J. McCoy, N. Lennard, M.J. Larkin, M. Collins, M. Sczaniecka, G. Blakely, Multiple inverted DNA repeats Bacteroide fragilis that control polysaccharide antigenic variation are similar to the hin region inverted repeats of Salmonella typhimurium, Microbiology 149 (2003) 915–924. [8] A.O. Tzianabos, D.L. Kasper, A.B. Onderdonk, Structure and functions of Bacteroides fragilis capsular polysaccharides: relationship to induction and prevention of abscesses, Clin. Infect. Dis. 20 (Suppl. 2) (1995) S132. [9] F.C. Gibson III, A.B. Onderdonk, D.L. Kasper, A.O. Tzianabos, Cellular mechanism of intraabdominal abscess formation by Bacteroides fragilis, J. Immunol. 160 (1998) 5000–5006. [10] T. Frankenberg, S. Kirschnek, H. Häcker, G. Häcker, Phagocytosis-induced apoptosis of macrophages is linked to uptake, killing and degradation of bacteria, Eur. J. Immunol. 38 (2008) 204–215. [11] R.A. DaMatta, S.H. Seabra, L. Manhães, W. de Souza, Nitric oxide is not involved in the killing of Trypanosoma cruzi by chicken macrophages, Parasitol. Res. 86 (2000) 239–243. [12] A.J. Talati, H.J. Kim, Y.-K. Kim, A.I. Yi, B.K. English, Role of bacterial DNA in macrophage activation by group B. streptococci, Microbes Infect. 10 (2008) 1106–1113. [13] S.H. Seabra, W. de Souza, R.A. DaMatta, Toxoplasma gondii partially inhibits nitric oxide production of activated murine macrophages, Exp. Parasitol. 100 (2002) 62–70. [14] T.M. Stevanin, J.R. Laver, R.K. Poole, J.W.B. Moir, R.C. Read, Metabolism of nitric oxide by Neisseria meningitidis modifies release of NO-regulated cytokines and chemokines by human macrophages, Microbes Infect. 9 (2007) 981–987. [15] A. Lahiri, P. Das, D. Chakravortty, Arginase modulates Salmonella induced nitric oxide production in RAW264.7 macrophages and is required for Salmonella pathogenesis in mice model of infection, Microbes Infect. 10 (2008) 1166–1174. [16] Y. Weinrauch, A. Zychlinsky, The induction of apoptosis by bacterial pathogens, Annu. Rev. Microbiol. 53 (1999) 155–187. [17] W.W. Navarre, A. Zychlinsky, Pathogen-induced apoptosis of macrophages: a common end for different pathogenic strategies, Cell. Microbiol. 2 (2000) 265– 273. [18] M.L. Arcila, M.D. Sánchez, B. Ortiz, L.F. Barrera, L.F. García, M. Rojas, Activation of apoptosis, but not necrosis, during Mycobacterium tuberculosis infection correlated with decreased bacterial growth: role of TNF-a, IL-10, caspases and phospholipase A2, Cell. Immunol. 249 (2007) 80–93.
632
J.M.B.D. Vieira et al. / Biochemical and Biophysical Research Communications 387 (2009) 627–632
[19] J.M.B.D. Vieira, D.C. Vallim, E.O. Ferreira, S.H. Seabra, R.C. Vommaro, K.E.S. Avelar, W. De Souza, M.C.S. Ferreira, R.M.C.P. Domingues, Bacteroides fragilis interferes with iNOS activity and leads to pore formation in macrophage surface, Biochem. Biophys. Res. Comm. 326 (2005) 607–613. [20] D.C. Vallim, I.C.M. Oliveira, E.N.F. Antunes, E.O. Ferreira, S.R. Moraes, G.R. Paula, M.C. Silva-Carvalho, A.M.S. Figueiredo, M.C.S. Ferreira, R.M.C.P. Domingues, Evaluation of genetic relatedness of Bacteroides fragilis strains isolated from different sources by AP-PCR and pulsed-field gel electrophoresis assays, Anaerobe 8 (2002) 192–199. [21] L.V. Holdeman, R.W. Kelly, W.E.C. Moore, Genus I. Bacteroides castellani and chalmers 1919, 959AL, in: N.R. Krieg, J.G. Holt (Eds.), Bergey’s Manual of Systematic Bacteriology, vol. 1, The Williams and Wilkins Co., Baltimore, 1984, pp. 604–631. [22] S. Patrick, A. Coffey, A.M. Emmerson, M.J. Larkin, The relationship between cell surface structure expression and haemagglutination in Bacteroides fragilis, FEMS Microbiol. Lett. 50 (1988) 67–71. [23] S. Rozental, C.S. Alviano, W. De Souza, The in vitro susceptibility of Fonsecaea pedrosoi to activated macrophages, Mycopathologia 126 (1994) 85–91. [24] H.R. Jousimies-Somer, P. Summanen, D.M. Citron, E.J. Baron, H.M. Wexler, S.M. Finegold, Anaerobic Bacteriology Manual, sixth ed., Belmont, California, 2002. [25] S. Favre-Bonte, A. Darfeuille-Michaud, C. Forestier, Aggregative adherence of Klebsiella pneumoniae to human intestine-407 cells, Infect. Immun. 63 (1995) 1318–1328. [26] S.H. Seabra, W. De Souza, R.A. DaMatta, Toxoplasma gondii exposes phosphatidylserine inducing a TGF-b1 autocrine effect orchestrating macrophage evasion, Biochem. Biophys. Res. Commun. 324 (2004) 744–752. [27] Y.-K. Shin, K.-Y. Kim, Y-P. Paik, Alterations of protein expression in macrophages in response to Candida albicans infection, Mol. Cells 20 (2) (2005) 271–279. [28] G. Greub, B. Desnues, D. Raoult, J.-L. Mege, Lack of microbicidal response in human macrophages infected with Parachlamydia acanthamoebae, Microbes Infect. (2005).
[29] K.E. Stockbauer, A. Foreman-Wykert, J.F. Miller, Bordetella type III secretion induces caspase 1-independent necrosis, Cell. Microbiol. 5 (2) (2003) 123–132. [30] L.D. Hernandez, M. Pypaert, R.A. Flavell, J.E. Galán, A Salmonella protein causes macrophage cell death by inducing autophagy, J. Cell Biol. 163 (5) (2003) 1123–1131. [31] F.R. Deleo, Modulation of phagocyte apoptosis by bacterial pathogens, Apoptosis 9 (2004) 399–413. [32] S.L. Fink, B. Cookson, Apoptosis, Pyroptosis and Necrosis: mechanistic description of dead and dying eukaryotic cells, Infect. Immun. 73 (4) (2005) 1907–1916. [33] D.M. Monack, J. Mecsas, N. Ghori, S. Falkow, Yersinia signals macrophages to undergo apoptosis and YopJ is necessary for this cell death, Proc. Natl. Acad. Sci. USA 94 (1997) 10385–10390. [34] K. Orth, L.E. Palmer, Z.Q. Bao, S. Stewart, A.E. Rudolph, J.B. Bliska, J.E. Dixon, Inhibition of the mitogen-activated protein kinase superfamily by a Yersinia effector, Science 285 (1999) 1920–1923. [35] A.B. ONDERDONK, Animals models simulating anaerobic infections, Anaerobe 11 (2005) 189–195. [36] S.R. Moraes, R.B. Gonçalves, C. Mouton, L. Seldin, M.C.S. Ferreira, R.M.C.P. Domingues, Bacteroides fragilis isolates compared by AP-PCR, Res. Microbiol. 150 (1999) 257–263. [37] S.R. Moraes, R.B. Gonçalves, C. Mouton, L. Seldin, M.C.S. Ferreira, R.M.C.P. Domingues, RepPCR to define genetic relatedness among Bacteroides fragilis strains, J. Med. Microbiol. 49 (2000) 279–284. [38] P.N. Sarma, Y.J. Tang, T.P. Prindiville, P.D. Osborne, S. Jang Jr., J. Silva, S.H. Cohen, Genotyping of Bacteroides fragilis isolates from stool specimens by arbitrarily primed-PCR, Diag. Microbiol. Infect. Dis. 37 (2000) 225–229. [39] E.R. Eribe, I. Olsen, Strain differentiation in Bacteroides fragilis by RAPD and dendron computer-assisted gel analysis, APMIS 108 (2000) 676–684.
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Bacteroides fragilis induce necrosis on mice peritoneal macrophages: In vitro and in vivo assays J.M.B.D. Vieira a,b,*, S.H. Seabra a, D.C. Vallim c, M.A. Américo d, S.E.L. Fracallanza d, R.C. Vommaro e, R.M.C.P. Domingues b a
Laboratório de Tecnologia em Cultura de Células, UEZO, Rio de Janeiro, Brazil Laboratório de Biologia de Anaeróbios, IMPPG, UFRJ, Rio de Janeiro, Brazil c Instituto Oswaldo Cruz, Rio de Janeiro, Brazil d Laboratório de Bacteriologia Médica, IMPPG, UFRJ, Rio de Janeiro, Brazil e Laboratório de Ultra-estrutura Celular Hertha Meyer, IBCCF, UFRJ, Brazil b
a r t i c l e
i n f o
Article history: Received 20 May 2009 Available online 2 June 2009
Keywords: Bacteroides fragilis Murine macrophages Bacteria/macrophage interactions NO TUNEL assay PI staining In vivo model Apoptosis Necrosis Scanning electron microscopy
a b s t r a c t Bacteroides fragilis is an anaerobic bacteria component of human intestinal microbiota and agent of infections. In the host B. fragilis interacts with macrophages, which produces toxic radicals like NO. The interaction of activated mice peritoneal macrophages with four strains of B. fragilis was evaluated on this study. Previously was shown that such strains could cause metabolic and morphologic alterations related to macrophage death. In this work propidium iodide staining showed the strains inducing macrophage necrosis in that the labeling was evident. Besides nitroblue tetrazolium test showed that B. fragilis stimulates macrophage to produce oxygen radicals. In vivo assays performed in BalbC mice have results similar to those for in vitro tests as well as scanning electron microscopy, which showed the same surface pore-like structures observed in vitro before. The results revealed that B. fragilis strains studied lead to macrophage death by a process similar to necrosis. Ó 2009 Elsevier Inc. All rights reserved.
Introduction Bacteroides fragilis is an anaerobic bacterium that colonizes the human intestinal tract [1,2]. Although B. fragilis do not represent the major anaerobic component in the gut, it is the most frequently recovered from human peritoneal infection specimens [3,4]. When it escapes the gut, as a result of a disease or intestinal surgery, it can generate a significant pathology, including abscess formation in multiple body sites like abdominal cavity and brain [5]. Such species present some virulence factors, like the capsular polysaccharide complex, which justify its pathogenic behavior [5,6–8]. Macrophages have functions like antigen presentation and secretion of cytokines such as tumor necrosis factor alpha (TNFa) and interleukin-1 (IL-1) [9,10]. The activated macrophages develop an enhanced microbicidal function, producing, for example,
* Corresponding author. Address: Laboratório de Tecnologia em Cultura de Células, UEZO, 23070-200 Rio de Janeiro, Brazil. Fax: +55 21 3332 5591. E-mail addresses: [email protected], [email protected] (J.M.B.D. Vieira). 0006-291X/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2009.05.124
nitric oxide (NO) [11,12]. Nevertheless, not only this production can be inhibited or modified by some pathogens but also the initial contact between cells, which seems to be a survival advantage for them [13–15]. The first step for macrophages eliminate pathogens is phagocytosis followed by intracellular digestion of them, which can result in macrophages apoptosis. The latter has been more frequently associated to the uptake and killing of intracellular pathogenic bacteria such as Mycobacteria, Legionella, Shigella and Salmonella [10,16–18]. Our group, studying the interaction between B. fragilis and mice peritoneal macrophages (M/), has described that B. fragilis interferes with iNOS activity and leads to pore formation in macrophage surface [19]. Besides, the macrophages ultrastructure observation after interaction assays with B. fragilis suggested that the phagocyte was in apoptosis. The aim of this study is to demonstrate if B. fragilis can modulate the respiratory burst in macrophages during interaction and answer a question, B. fragilis induces or not apoptosis on macrophages? So, we decided to realize B. fragilis/ M/ interactions in vitro and in vivo beyond verify what kind of cellular death occurs in resident macrophages after interaction.
628
J.M.B.D. Vieira et al. / Biochemical and Biophysical Research Communications 387 (2009) 627–632
Materials and methods Bacterial strains and culture conditions Four strains defined in a previous work [20] were chosen for this study: ES31-3 (isolated from Human Intestinal Microbiote), Eli1-1 (isolated from Human Intestinal Microbiote), 0789320 (isolated from Human Diarrheic Process) and 25285 (isolated from Abscess – American Type Culture Collection – ATCC). All the strains were maintained frozen at 20 °C as part of Anaerobe Biology Laboratory Culture Collection. The strains were reactivated in Brain Heart Infusion-pre-reduced anaerobically sterilized (BHI-PRAS/OxoidÒ – Basingstoke Hants, England) [21]. The bacterial suspensions used in experiments were obtained from culture in protease–peptone–yeast broth [22] at 37 °C. Peritoneal macrophages The macrophages were obtained as described previously [13]. The cells were stimulated with intraperitoneal inoculation of 30 mg/ml of tioglicolate in 21 days female BalbC mice. After 72 h, the macrophages were aseptically collected through peritoneal injection of cold Dulbeco’s Modified Eagles Medium (DMEM – SigmaTM – St. Louis, MO/USA). The collected cells were incubated in ice and adjusted to 2 106 cells/ml in cold DMEM. After 1 h of adhesion at 37 °C in 5% CO2 atmosphere, the wells were washed with DMEM without serum to remove non-adherent cells and the 24 wells plate was incubated with DMEM supplemented with 5% bovine fetal serum (BFS – LaborclinTM – Pinhais, Paraná, Brazil) inactivated by heating. Analysis of oxygen reactive intermediaries production in M/
with serum-free DMEM again and were mounted in the same medium to immediate observation in a LSM (LSM 310 – ZeissTM). Noninfected macrophages were the negative control. In vivo assays Chambers preparation. chambers were prepared using sterilized insulin needles parts, in which extremities were sealed with nonsterilizing 0.8 and 5 lm pore membranes. In each chamber a minimal hole was made in the needle part to inoculate the strains diluted according to McFarland Scale 2 (6 108 UFC/ml), afterwards the hole was heating closed. Chambers with serum-free BHI-PRAS (OxoidTM – Basingstoke Hants, England) broth were used as controls. Chambers insertion in mice. 21 days BalbC female mice were operated in anesthesia state using CO2 and chambers were inserted in their peritoneal cavity in the region near the intestinal loops, performing the proper aseptic procedures. Chambers extraction. occurred after 24 or 48 h by opening the peritoneal cavity, following the bacterial viability analysis through inoculation in Supplemented Blood Agar (SBA) [24]. Such inoculation was made after serial dilution of chambers contents (10 1 to 10 7 in Blanks-PRAS), in which each dilution was inoculated in SBA. After incubation for 48 h in anaerobiosis at 37 °C, the bacterial growth was analyzed for each dilution basing in experiment control (chambers with sterilized medium also inoculated in SBA). The biggest dilution title having UFCs were considered for calculus. The dilution factor multiplied by the used volume in chamber gives rise to UFCs/ml. Scanning electron microscopy – SEM
The oxidative burst in M/ infected with B. fragilis was detected as described previously [23]. The cells were washed with serumfree DMEM and nitroblue tetrazolium (NBT – SigmaTM) at 0.5 mg/ ml together with B. fragilis in an equivalent concentration of McFarland Scale 2. Yeasts were used as positive control. After 1 h of incubation at 37 °C in 5% CO2 atmosphere, the cells were washed with serum-free DMEM, then fixed in Bouin (7.5 ml of picric acid, 2.5 ml of formaldeide 37–40% and 5 ml of glacial acetic acid) and observed with Axioplan optic microscope (ZeissTM).
Based on literature [25], after M//B. fragilis strains interactions, the cells were washed with DMEM without serum, fixed in Karnovsky (EMSTM) and 2% tanic acid (MerckTM) for 2 h, and washed with PBS. The cells were post fixed in 1% of osmium solution (OsO4 v/ v – SigmaTM) with tanic acid 2% in phosphate buffer, pH 7.2 for 1 h at room temperature (RT) and washed in PBS. Then the cells were dehydrated in ethanol gradual concentration (50–100%/ 10 min each solution – MerckTM), dried through CO2 critic point, mounted in metal support and covered with gold (20–30 nm) for observation in a Scanning Electron Microscope (JSM5310 – JeolTM, Japan).
TUNEL (terminal dUTP nick end labeling) assays: apoptosis analysis
Results
This evaluation was made using APO-BrdU TUNEL Assay kit (A – 23210, Molecular ProbesTM, São Paulo, Brazil). After interactions between B. fragilis and macrophages in 24 well plate, the cells were incubated with alcohol 70% at 4 °C, washed twice with PBS and incubated with 50 ll of DNA linking solution (provided by the kit) for 30 min at 37 °C, light protected. The cover slips were washed with washing buffer (provided by the kit) and incubated with antibody solution (quantity established by the kit) for 30 min at RT. The washing was made twice with washing buffer and the coverslips were mounted in N-propil galato (SigmaTM), following visualization in a Laser Scanning Microscope (LSM 310 – Zeiss, Switzerland). Non-infected macrophages were the negative control.
The four strains studied: ATCC25285, ELI1-1, S31-3, and 078320, were representative of the pulsotypes 4, 5, 16, and 24 [20]. Fig. 1 shows that despite apparent regulation of iNOS activity by B. fragilis based on previous results [19], none bacterial strain could regulate respiratory burst. For apoptosis analysis, interactions of B. fragilis ELI1-1 strain with macrophages were observed by fluorescence and compared with phase contrast. After 2 h of interaction, the macrophages did not suffer apoptosis (Inset Fig. 2F). Necrosis assays have similar results after B. fragilis ATCC25285/M/ interaction in same conditions (Fig. 2C and D). However, using the same interaction period to B. fragilis ELI1-1 strain (Fig. 2E and F) as well as for ES31-3 and 078320 strains (not shown) macrophages necrosis was observed. The same was seen in 4 h interaction experiments (fig. 2G–L) for all strains excepting ATCC25285 that did not discharge necrosis death in macrophages (Fig. 2I and J). To investigate the consequences of in vivo B. fragilis strains/M/ interactions were utilized propylene chambers with 0.8 or 5 lm membranes containing bacterial suspensions into mice peritoneal cavity (Fig. 3). After 24 h of interaction with chamber 0.8 lm, in-
Propidium iodide: necrosis evaluation After bacteria macrophage interaction in 24 well plates, a twice washing with DMEM without serum was performed and each coverslip was incubated with Propidium iodide (100 lg/ml – Molecular ProbesTM) in serum-free media. The coverslips were washed
J.M.B.D. Vieira et al. / Biochemical and Biophysical Research Communications 387 (2009) 627–632
629
fected macrophages did not present apoptosis (Insets Fig. 3D and F). In 24 h interaction experiments with 5 lm chamber some apoptotic macrophages were observed (insets Fig. 3J and L). Similar results were obtained after 48 h of interactions for both chambers (data not shown). However, macrophages necrosis after 24 h of ‘‘in vivo” interaction were observed (Fig. 3) with 0.8 lm (Fig. 3F) and 5 lm (Fig. 3J and L) chambers for 25285 and Eli1-1 strains. To evaluate macrophage membrane rupture in vivo, some experiments of SEM were made after 24 h and 48 h M//B. fragilis strains interaction using 0.8 lm chambers (Fig. 4A–H). The results showed pore-like structures on macrophages membranes (arrows in Fig. 4D, E, G and H). The same results were obtained with 5 lm chambers after 24 h interaction experiment (Fig. 4F and I). Besides, on Fig. 4F (star) is possible to see the modification on macrophage surface in comparison to control (Fig. 4C).
Discussion
Fig. 1. Oxidative burst: evidentiation of oxygen radicals production by macrophages after in vitro interaction with B. fragilis. (A) Negative control, non-infected Mf; (B) Positive control, Mf yeasts; (C) Mf 25285; (D) Mf 078320; (E) Mf Es31-3; (F) Mf Eli 1-1. Note formazane cristals presence (arrows). (A–F): interferential contrast. Barr 20 mm.
Bacteroides fragilis, a commensal anaerobe from intestinal microbiota, can cause infection when extrudes this site through virulence factors production [6–8]. To stay in host, it has to interact and escape from macrophages. The latter are able to destroy microorganisms by phagocyting and digesting them through the production of oxygen radicals [11,12]. In a previous work it was demonstrated that B. fragilis can induce alterations on NO synthesis by macrophages and can cause pore formation on the cell surface of these cells, a suggested way for actin extruding and iNOS release [19]. The intriguing fact that B. fragilis could modify the macrophage response diminishing the NO synthesis raised the question
Fig. 2. Necrosis Mf Evidenciation with PI after 2 h (A–F) and 4 h (G–L) of in vitro interaction with B. fragilis. (A and B)/(G and H) non-infected Mf; (C and D)/(I and J) Mf 25285; (E and F)/(K and L) Mf Eli1-1. PI staining is apparent to Eli1-1 strain (arrows). (A, C, E, G, I and K): phase contrast microscopy; (B, D, F, H, J and L): fluorescence microscopy. Insets shown results from in vitro apoptosis assays. Barr: 20 mm.
630
J.M.B.D. Vieira et al. / Biochemical and Biophysical Research Communications 387 (2009) 627–632
Fig. 3. PI staining of peritoneal M/ from mice infected with B. fragilis in 0.8 lm (A–F) and 5 lm (G–L) chambers for 24 h. (A and B)/(G and H) control, non-infected M/; (C and D)/(I and J) M/ 25285; (E and F)/(K and L) M/ Eli-1. Note the presence of necrotic M/ (arrows). (A, C, E, G, I and K) interpherencial contrast; (B, D, F, H, J and L) fluorescence. Insets: results for apoptosis. Bars: 20 lm.
Fig. 4. Scanning electron micrography of mice peritoneal M/ infected in vivo with B. fragilis strains in 0.8 lm chamber for 24 h (A, D and G) and 48 h (B, E and H), and 5 lm chamber for 24 h (C, F and I). (A, B and C) control, non-infected M/; (D, E and F) M/ 25285; (G, H and I) M/ Eli-1; (H) 48 h M/ Eli1-1. Note the presence of structures similar to pores on the surface of infected M/ (arrows).
if it could interfere with the oxidative burst as well. For all B. fragilis strains tested it was possible to see the presence of formasana
crystals in M/, indicating that even after interaction with B. fragilis, macrophages produce oxidative burst (Fig. 1).
J.M.B.D. Vieira et al. / Biochemical and Biophysical Research Communications 387 (2009) 627–632
It has been described that several kind of microorganisms are able to hide themselves from the immune system by different mechanisms [26–28] and can even induce the host cell death, including those cells that act in the host immunity like macrophages [29–31]. In this context, a hypothesis took place: B. fragilis could also induce macrophage death like other extracellular bacteria? [10,31]. Some assays were performed in vitro to evaluate the kind of death, apoptosis or necrosis, induced by B. fragilis on M/. To achieve this purpose TUNEL technique was performed in vitro after 2 h of B.fragilis/M/ interaction (insets on Fig. 2). The results showed that this bacterial species do not induce macrophage apoptosis, since the specific labeling was not seen for the strains analyzed. The results obtained with the TUNEL in vitro raised the possibility that macrophages could be suffering necrosis instead of apoptosis after contact with B. fragilis. A study developed with Bordetella bronchiseptica showed that its secretion system type III (SSTIII) promotes a death process with similar characteristics from those of necrosis [29]. To evaluate the hypothesis for B. fragilis, IP labeling assays were realized after 2 (Fig. 2A–F) and 4 h (Fig. 2G–L) of interaction with M/. All the strains presented apparent labeling with IP, excepting the strain ATTCC 25285. This finding reinforced the idea that B. fragilis induce macrophages death by a process similar to necrosis. Some experiments were performed in vivo using chambers with different porosity (0.8 and 5 lm) inserted into mice peritoneal cavity for 24 and 48 h (Fig. 3) containing B. fragilis strains suspensions. The aim was compare the results obtained in vitro with data from in vivo assays. First, it was realized the experiments with TUNEL, since apoptosis induction in infection models for many kinds of microorganisms has been widely discussed [32]. Such assays realized after interaction in vivo for 24 (insets on Fig. 3) and 48 h (data not shown) presented the same results for the two kinds of chambers when compared to in vitro findings, which means that specific TUNEL labeling was not observed for any of the strains tested. The labeling observed on Fig. 3J and 3L insets represents the background and it is not an evidence of apoptosis. Similar experiments were performed to test the presence of necrotic macrophages on peritoneal cavity after infection for 24 (Fig. 3A–F) and 48 h (Fig. 3G–L) with B. fragilis. In this case was possible to observe labeling with IP for all the strains analyzed (Fig. 3), which corroborates the data obtained from in vitro analysis. Previously our group found, based on electron microscopy, that B. fragilis infection induced the rupture of macrophage membranes in vitro [19]. In order to make a comparison, we decided to do some tests in vivo. Thus, some SEM experiments were made after 24 h (Fig. 4A, D and G) and 48 h (Fig. 4B, E and H) of B. fragilis strains/ M/ interactions with 0.8 lm chamber and for 24 h in 5 lm chamber (Fig. 4C, F and I). It was not surprising that these findings were similar, since it was possible to observe pore-like structures on macrophages membranes for both chambers (arrows Fig. 4). Such results showed that the alterations on macrophages surface are induced by B. fragilis in vitro as described before [19] and in vivo, possibly leading to leak of cytoplasm contents, including iNOS. In the bacterial genus Yersinia, Shigella and Salmonella the secretion system type III is involved in causing apoptosis on macrophages [33,34] but this system was also described in B. bronchiseptica inducing the macrophage death by a process similar to necrosis [29]. On this context, it is possible that in B. fragilis case systems like this one are acting, being responsible to induce necrosis on infected macrophages. The fact that strain ATCC25285 revealed a different behavior in vitro in comparison to the others strains could be related to the lack of expression or basal expression of its secretion system. Anyway, this strain was able to cause alterations on macrophages metabolism, which was demonstrated by previous results [19] with immunocytochemistry for iNOS and consequent modification in NO production.
631
The utilization of animal models in this work intended to evaluate, based on the results obtained in vitro, particular aspects of M/ infected with B. fragilis, aiming to see the consequences of such interaction. Different models have been proposed to study anaerobe bacteria [35] but the system used in this work has never been used before, which can be seen as an experimental method of alternative tests in animals to study anaerobes. The B. fragilis species shows a wide genetic diversity and has been target of many analysis and discussions [36–39]. However, in our study this diversity did not influence the strains behavior even belonging to different pulsotypes described before [20] and having less than 30% of similarity. All four strains presented similar results in all experiments, excepting ATCC25285 for necrosis evaluation in vitro. In conclusion, it is possible to affirm that different Bacteroides fragilis clonal types induce mice activated peritoneal macrophages death by a process similar to necrosis. Nevertheless, future experiments are certainly needed to clarify this probable mechanism of immune system evasion. Acknowledgments We thank Joaquim Santos Filho for technical assistance. This work was supported by PRONEX-MCT, CNPq, FUJB, and FAPERJ. References [1] N.E.C. Moore, E.P. Cato, L.V. Holdeman, Anaerobic bacteria of gastrointestinal flora and their occurrence in clinical infections, J. Infect. Dis. 119 (1978) 641– 649. [2] I.R. Poxton, R. Brown, A. Sawyerr, A. Ferguson, The mucosal anaerobic Gramnegative bacteria of the human colon, Clin. Infect. Dis. 25 (Suppl. 2) (1997) S111–S113. [3] G.L. Simon, S.L. Gorbach, Intestinal health and disease, Gastroenterology 86 (1984) 174–193. [4] A. Willis, Abdominal sepsis, in: B.I. Duerden, B.S. Drasar (Eds.), Anaerobes in Human Diseases, Edward Arnold, London, 1991, pp. 197–223. [5] L. Pumbwe, C.A. Skilbeck, H.M. Wexler, The Bacteroides fragilis cell envelope: quarterback, linebacker, coach—or all three?, Anaerobe 12 (2006) 211–220 [6] A.B. Onderdonk, N.E. Moon, D.L. Kasper, J.G. Bartlett, Adherence of Bacteroides fragilis in vivo, Infect. Immun. 19 (1977) 1083. [7] S. Patrick, J. Parkhill, L.J. McCoy, N. Lennard, M.J. Larkin, M. Collins, M. Sczaniecka, G. Blakely, Multiple inverted DNA repeats Bacteroide fragilis that control polysaccharide antigenic variation are similar to the hin region inverted repeats of Salmonella typhimurium, Microbiology 149 (2003) 915–924. [8] A.O. Tzianabos, D.L. Kasper, A.B. Onderdonk, Structure and functions of Bacteroides fragilis capsular polysaccharides: relationship to induction and prevention of abscesses, Clin. Infect. Dis. 20 (Suppl. 2) (1995) S132. [9] F.C. Gibson III, A.B. Onderdonk, D.L. Kasper, A.O. Tzianabos, Cellular mechanism of intraabdominal abscess formation by Bacteroides fragilis, J. Immunol. 160 (1998) 5000–5006. [10] T. Frankenberg, S. Kirschnek, H. Häcker, G. Häcker, Phagocytosis-induced apoptosis of macrophages is linked to uptake, killing and degradation of bacteria, Eur. J. Immunol. 38 (2008) 204–215. [11] R.A. DaMatta, S.H. Seabra, L. Manhães, W. de Souza, Nitric oxide is not involved in the killing of Trypanosoma cruzi by chicken macrophages, Parasitol. Res. 86 (2000) 239–243. [12] A.J. Talati, H.J. Kim, Y.-K. Kim, A.I. Yi, B.K. English, Role of bacterial DNA in macrophage activation by group B. streptococci, Microbes Infect. 10 (2008) 1106–1113. [13] S.H. Seabra, W. de Souza, R.A. DaMatta, Toxoplasma gondii partially inhibits nitric oxide production of activated murine macrophages, Exp. Parasitol. 100 (2002) 62–70. [14] T.M. Stevanin, J.R. Laver, R.K. Poole, J.W.B. Moir, R.C. Read, Metabolism of nitric oxide by Neisseria meningitidis modifies release of NO-regulated cytokines and chemokines by human macrophages, Microbes Infect. 9 (2007) 981–987. [15] A. Lahiri, P. Das, D. Chakravortty, Arginase modulates Salmonella induced nitric oxide production in RAW264.7 macrophages and is required for Salmonella pathogenesis in mice model of infection, Microbes Infect. 10 (2008) 1166–1174. [16] Y. Weinrauch, A. Zychlinsky, The induction of apoptosis by bacterial pathogens, Annu. Rev. Microbiol. 53 (1999) 155–187. [17] W.W. Navarre, A. Zychlinsky, Pathogen-induced apoptosis of macrophages: a common end for different pathogenic strategies, Cell. Microbiol. 2 (2000) 265– 273. [18] M.L. Arcila, M.D. Sánchez, B. Ortiz, L.F. Barrera, L.F. García, M. Rojas, Activation of apoptosis, but not necrosis, during Mycobacterium tuberculosis infection correlated with decreased bacterial growth: role of TNF-a, IL-10, caspases and phospholipase A2, Cell. Immunol. 249 (2007) 80–93.
632
J.M.B.D. Vieira et al. / Biochemical and Biophysical Research Communications 387 (2009) 627–632
[19] J.M.B.D. Vieira, D.C. Vallim, E.O. Ferreira, S.H. Seabra, R.C. Vommaro, K.E.S. Avelar, W. De Souza, M.C.S. Ferreira, R.M.C.P. Domingues, Bacteroides fragilis interferes with iNOS activity and leads to pore formation in macrophage surface, Biochem. Biophys. Res. Comm. 326 (2005) 607–613. [20] D.C. Vallim, I.C.M. Oliveira, E.N.F. Antunes, E.O. Ferreira, S.R. Moraes, G.R. Paula, M.C. Silva-Carvalho, A.M.S. Figueiredo, M.C.S. Ferreira, R.M.C.P. Domingues, Evaluation of genetic relatedness of Bacteroides fragilis strains isolated from different sources by AP-PCR and pulsed-field gel electrophoresis assays, Anaerobe 8 (2002) 192–199. [21] L.V. Holdeman, R.W. Kelly, W.E.C. Moore, Genus I. Bacteroides castellani and chalmers 1919, 959AL, in: N.R. Krieg, J.G. Holt (Eds.), Bergey’s Manual of Systematic Bacteriology, vol. 1, The Williams and Wilkins Co., Baltimore, 1984, pp. 604–631. [22] S. Patrick, A. Coffey, A.M. Emmerson, M.J. Larkin, The relationship between cell surface structure expression and haemagglutination in Bacteroides fragilis, FEMS Microbiol. Lett. 50 (1988) 67–71. [23] S. Rozental, C.S. Alviano, W. De Souza, The in vitro susceptibility of Fonsecaea pedrosoi to activated macrophages, Mycopathologia 126 (1994) 85–91. [24] H.R. Jousimies-Somer, P. Summanen, D.M. Citron, E.J. Baron, H.M. Wexler, S.M. Finegold, Anaerobic Bacteriology Manual, sixth ed., Belmont, California, 2002. [25] S. Favre-Bonte, A. Darfeuille-Michaud, C. Forestier, Aggregative adherence of Klebsiella pneumoniae to human intestine-407 cells, Infect. Immun. 63 (1995) 1318–1328. [26] S.H. Seabra, W. De Souza, R.A. DaMatta, Toxoplasma gondii exposes phosphatidylserine inducing a TGF-b1 autocrine effect orchestrating macrophage evasion, Biochem. Biophys. Res. Commun. 324 (2004) 744–752. [27] Y.-K. Shin, K.-Y. Kim, Y-P. Paik, Alterations of protein expression in macrophages in response to Candida albicans infection, Mol. Cells 20 (2) (2005) 271–279. [28] G. Greub, B. Desnues, D. Raoult, J.-L. Mege, Lack of microbicidal response in human macrophages infected with Parachlamydia acanthamoebae, Microbes Infect. (2005).
[29] K.E. Stockbauer, A. Foreman-Wykert, J.F. Miller, Bordetella type III secretion induces caspase 1-independent necrosis, Cell. Microbiol. 5 (2) (2003) 123–132. [30] L.D. Hernandez, M. Pypaert, R.A. Flavell, J.E. Galán, A Salmonella protein causes macrophage cell death by inducing autophagy, J. Cell Biol. 163 (5) (2003) 1123–1131. [31] F.R. Deleo, Modulation of phagocyte apoptosis by bacterial pathogens, Apoptosis 9 (2004) 399–413. [32] S.L. Fink, B. Cookson, Apoptosis, Pyroptosis and Necrosis: mechanistic description of dead and dying eukaryotic cells, Infect. Immun. 73 (4) (2005) 1907–1916. [33] D.M. Monack, J. Mecsas, N. Ghori, S. Falkow, Yersinia signals macrophages to undergo apoptosis and YopJ is necessary for this cell death, Proc. Natl. Acad. Sci. USA 94 (1997) 10385–10390. [34] K. Orth, L.E. Palmer, Z.Q. Bao, S. Stewart, A.E. Rudolph, J.B. Bliska, J.E. Dixon, Inhibition of the mitogen-activated protein kinase superfamily by a Yersinia effector, Science 285 (1999) 1920–1923. [35] A.B. ONDERDONK, Animals models simulating anaerobic infections, Anaerobe 11 (2005) 189–195. [36] S.R. Moraes, R.B. Gonçalves, C. Mouton, L. Seldin, M.C.S. Ferreira, R.M.C.P. Domingues, Bacteroides fragilis isolates compared by AP-PCR, Res. Microbiol. 150 (1999) 257–263. [37] S.R. Moraes, R.B. Gonçalves, C. Mouton, L. Seldin, M.C.S. Ferreira, R.M.C.P. Domingues, RepPCR to define genetic relatedness among Bacteroides fragilis strains, J. Med. Microbiol. 49 (2000) 279–284. [38] P.N. Sarma, Y.J. Tang, T.P. Prindiville, P.D. Osborne, S. Jang Jr., J. Silva, S.H. Cohen, Genotyping of Bacteroides fragilis isolates from stool specimens by arbitrarily primed-PCR, Diag. Microbiol. Infect. Dis. 37 (2000) 225–229. [39] E.R. Eribe, I. Olsen, Strain differentiation in Bacteroides fragilis by RAPD and dendron computer-assisted gel analysis, APMIS 108 (2000) 676–684.