- Email: [email protected]
Bacillus subtilis spores reduce susceptibility to Citrobacter rodentium-mediated enteropathy in a mouse model
Research in Microbiology 157 (2006) 891–897 www.elsevier.com/locate/resmic
Bacillus subtilis spores reduce susceptibility to Citrobacter rodentium-mediated enteropathy in a mouse model Rossana D’Arienzo a , Francesco Maurano a , Giuseppe Mazzarella a , Diomira Luongo a , Rosita Stefanile a , Ezio Ricca b , Mauro Rossi a,∗ a Istituto di Scienze dell’Alimentazione, CNR, via Roma 52, 83100 Avellino, Italy b Dipartimento di Biologia Strutturale e Funzionale, Università Federico II, via Cinthia complesso Monte Sant’Angelo, 80126 Naples, Italy
Received 6 March 2006; accepted 5 June 2006 Available online 26 September 2006
Abstract The present work was aimed at investigating whether Bacillus subtilis spores, widely used in probiotic as well as pharmaceutical preparations for mild gastrointestinal disorders, can suppress enteric infections. To address this issue, we developed a mouse model of infection using the mouse enteropathogen Citrobacter rodentium, a member of a family of human and animal pathogens which includes the clinically significant enteropathogenic (EPEC) and enterohemorrhagic (EHEC) Escherichia coli strains. This group of pathogens causes transmissible colonic hyperplasia by using attaching and effacing (A/E) lesions to colonize the host colon. Because of its similarities to human enteropathogens, C. rodentium is now widely used as an in vivo model for gastrointestinal infections. Swiss NIH mice were orally administered B. subtilis spores one day before infection with C. rodentium. Mice were sacrificed on day 15 after infection, and distal colon, liver and mesenteric lymph nodes were removed for bacteria counts, morphology, immunohistology and IFNγ mRNA analysis. We observed that spore predosing was effective in significantly decreasing infection and enteropathy in suckling mice infected with a dose of C. rodentium sufficient to cause colon colonization, crypt hyperplasia and high mortality rates. Moreover, in mice predosed with spores, the number of CD4+ cells and IFNγ transcript levels remained high. These results thus indicate that our newly established model of C. rodentium infection is a suitable system for analyzing the effects of probiotic bacteria on enteroinfections and that B. subtilis spores are efficient at reducing C. rodentium infection in mice, leaving unaltered the immune response against the pathogen. © 2006 Elsevier Masson SAS. All rights reserved. Keywords: Citrobacter rodentium; Bacillus subtilis; Enteropathy; Mucosal immunity; Mesenteric lymph nodes; Spores
1. Introduction The human gastrointestinal tract (GIT) is normally colonized by a large variety of microorganisms, forming the so-called endogenous microflora and including at least 400 microbial species. In this symbiotic interaction, microbes form a first line of defense against invasion by pathogenic organisms and provide the host with essential products, such as vitamins, amino acids and organic acids. These contributions are reciprocated by stable conditions of temperature, pH, osmolarity and food supply for the microorganisms [14,19]. * Corresponding author.
E-mail address: [email protected] (M. Rossi). 0923-2508/$ – see front matter © 2006 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.resmic.2006.06.001
A normally balanced microbiota is important in preventing intestinal disorders as well as microbial infections and, for these reasons, there has been much interest in developing probiotic products containing live bacteria able to survive inside the GIT and display positive effects on human health [26]. Although the use of probiotics is not new, with some products licensed for human use and marketed in the late ’50s [35], it is only recently that this field has begun to receive an increasing level of scientific interest. Lactic acid bacteria (e.g., Lactobacillus spp. and Bifidobacterium spp.) are the microorganisms most commonly used as probiotics. These bacteria are normal colonizers of the human and animal GIT and their use in probiotic products is thought to contribute to restoring the natural microflora of the gut. Mi-
892
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
croorganisms not normally present in the GIT are also widely used as probiotics. The Gram-positive spore-forming bacteria, and normally members of the genus Bacillus, belong to this group [18]. Although these bacteria have long been considered soil microorganisms, a number of recent reports have pointed out that they are commonly found associated with the GIT of humans and animals [1,13]. In addition, molecular as well as immunological data have convincingly shown that within the animal GIT Bacillus spores are able to carry on a complete life cycle, germinating in response to water availability and nutritional stimuli, proliferating and resporulating [4,11,12,17,38]. Starting with the pioneering work of Nurmi and Rantala [29], the concept of using probiotics to suppress or reduce pathogenic infections has been elaborated as a potentially protective approach [15]. A series of studies has demonstrated that a mixture of Streptococcus thermophilus and Lactobacillus acidophilus protects epithelial cells from enteroinvasive Escherichia coli strains [30], that Saccharomyces boulardii is effective in preventing the occurrence of Clostridium difficile [9] and Escherichia coli [8] infections, that Bacillus subtilis spores may be successfully used as competitive exclusion agents in poultry [25]. To expand those studies and to determine whether sporebased probiotic products can be successfully used to prevent infections by enteric pathogens, we began developing a new animal model. As the model pathogen, we used Citrobacter rodentium, a member of a family of human and animal pathogens that use attaching and effacing (A/E) lesions to colonize the host colon. The epithelial lesions caused by C. rodentium are indistinguishable from those of enteropathogenic (EPEC) [31] and enterohemorrhagic (EHEC) [10] E. coli strains. Following oral infection of mice, bacteria colonize the distal colon, causing crypt hyperplasia, mucosal thickening and uneven epithelial surface. A feature common to all these bacteria is the production of A/E lesions [3,33], characterized by localized destruction (effacement) of brush border microvilli and attachment of the bacterium to the host cell membrane, followed by the formation of an underlying pedestal-like structure in the host cell [5,40]. 2. Materials and methods 2.1. Mice NIH Swiss mice were obtained from Harlan-Nossan (Milan, Italy). A colony of pathogen-free mice was then developed at our animal facility. Mice were kept in sterilized cages and fed autoclaved food and water. The protocols employed were in direct accordance with guidelines drafted by the Veterinary Department of the Italian Health Ministry. 2.2. Bacterial strains and growth C. rodentium (formerly Citrobacter freundii strain biotype 4280) strain DBS100 [32] was grown in rich (LB) medium at 37 ◦ C in aerobic conditions and the number of cells (CFUs) determined prior to mouse inoculation. Early stationary phase
cells were always used to inoculate mice. A spontaneous nalidixic-acid-resistant (nalr ) strain was obtained by plating 5 × 109 cells of DBS100 on LB plates supplemented with 100 µg/ml of nalidixic acid. Resistant colonies grown on the selective plate were purified and used to inoculate mice. B. subtilis strain PY79 [41] was used to obtain spores by the exhaustion method [7]. In brief, cells were grown at 37 ◦ C in aerobic conditions in Difco sporulation medium (DSM; Difco Laboratories) for 36 h and spores were collected by centrifugation, washed several times and purified by lysozyme treatment to eliminate vegetative cells, as previously described [28]. The number of purified spores obtained was measured by direct counting with a Burker chamber under an optical microscope (Olympus BH-2 with 40× lenses). 2.3. Bacterial administration in mice Mice were infected by oral gavage with 0.1 ml and 20 µl broth for adult (21-day-old) and suckling (7-day-old) mice, respectively, containing different CFUs of C. rodentium (day 0). In spore predosing experiments, mice were orally administered different amounts of B. subtilis spores in the same volumes used for C. rodentium one day before infection. Mice were sacrificed on day 15 after infection; following careful dissection the distal colon (4 cm proximal to the anal verge), liver and mesenteric lymph nodes (MLN) were aseptically removed for analysis, as described below. 2.4. Animal studies 2.4.1. Mortality rate assessment Following infection, mice were monitored every day and when any mouse became moribund, it was immediately sacrificed. Data were expressed as percentages of the initial number. 2.4.2. Bacterial counts Colonic tissue, liver and MLN were resuspended in 4 ml PBS and homogenized on ice. Serial dilutions of the homogenates were plated onto LB-agar plates containing 100 µg/ml nalidixic acid and incubated at 37 ◦ C. Bacterial colonies were enumerated the following day. 2.4.3. Mucosal hyperplasia score Full-thickness colonic tissues were fixed in 10% neutral buffered formalin. Sections (3 µm) were cut and stained with hematoxylin and eosin. Crypt height was measured by micrometry with 10 measurements for each mouse by counting only well-oriented crypts. 2.4.4. Immunohistochemistry Cryostat sections (5 µm) were air-dried at room temperature and fixed in acetone for 10 min. All sections were repeatedly washed at room temperature in Tris-buffered saline (TBS), incubated in normal rabbit serum and diluted 1:100 in TBS for 30 min. Sections were then incubated with the monoclonal antibody YTS 191 (anti-CD4; BD Biosciences, Heidelberg, Germany) for 1 h at room temperature. Next, sections were incu-
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
893
bated for 30 min with rabbit anti-mouse serum (Dako, Milan, Italy) and for a further 30 min with peroxidase/anti-peroxidase (PAP) complex (Dako). Slides were washed in TBS for 10 min and subsequently developed by a final incubation of five min with 2 amino-9 ethyl-carbazole (Sigma). As control, nonimmune mouse IgGs of the same isotype were used. Sections were finally stained with Mayer’s hematoxylin and mounted. CD4+ cell number in the colonic lamina propria (LP) compartment was determined by counting the number of stained cells within a total area of 1 mm2 of LP using a microscope with a calibrated ocular aligned parallel to the muscularis mucosae. Counts were independently analyzed in a blinded manner by two observers. 2.4.5. Real-time PCR analysis of IFNγ transcript levels Total RNA was extracted from colonic tissue using a single step guanidine isothiocianate-phenol-chloroform isolation method [6]. Reverse transcription of 1 µg RNA was primed using oligo-(dT)12–18 and different aliquots of cDNA were then used for PCR. Real-time quantitative PCR was performed and analyzed using the iCycler (BIO-RAD Laboratories Inc., Hercules, CA, USA). The PCR reaction buffer contained 1× SYBR Green supermix (BIO-RAD) and 500 nM of each primer. The temperature profile of the amplification consisted of 50 cycles of 30 s denaturation at 95 ◦ C, 30 s annealing at 54.6 ◦ C (L32, housekeeping gene) or 56.4 ◦ C (IFNγ) and 40 s extension at 72 ◦ C. Negative controls were performed by omitting RNA from the cDNA synthesis and specific PCR amplification. The oligonucleotides used for amplification of IFNγ were previously described [36]. The primers used for amplification of the housekeeping gene L32 were designed to avoid amplification from contaminating genomic DNA by using Beacon designer Software (BIO-RAD). Their sequences are: 5 -AGCAGAGCTGGAGTCGCTTT (sense); 5 -GGAGCTGCCATCCAAAAGATACTA (anti-sense). The amount of IFNγ transcript was normalized to L32 according IFNγ to the formula: 2-CT where CT = CT − CL32 T . Normalized IFNγ mRNA levels were expressed relative to the expression in control mice according to the formula: 2-CT where sample CT = CT − Ccontrol . Thus, 1 arbitrary unit (AU) T corresponds to the relative levels of IFNγ in control mice. 2.5. Statistical analysis Student’s t test was used to compare results from two different experimental groups. Results were expressed as mean ± SD. Statistical significance was reached when P values were less than 0.05. 3. Results 3.1. Spore predosing in adult NIH Swiss mice In order to test the effect of B. subtilis spores at preventing or reducing enteroinfections, we decided to establish an infection model using the mouse enteropathogen C. rodentium. In preliminary experiments, adult NIH Swiss mice (21-days-old)
Fig. 1. Predosing with B. subtilis spores of adult mice. 21-day-old mice were orally administered different amounts of spores one day before infection with C. rodentium (1 × 1010 CFUs). On day 15 after infection, mice were sacrificed and distal colon removed for bacteria counts (A), evaluation of crypt length (B) and infiltrating CD4+ cells in the LP (C). There were six mice per group and the experiment was repeated three times with analogous results. In the control group (ctr) mice were non-infected. * P < 0.05. ** P < 0.001.
were orally challenged with a nalr derivative of the DBS100 strain at different doses and it was found that the optimal dose for inducing colon colonization and enteropathy was 1 × 1010 CFUs (data not shown). However, no dead mice were observed after oral challenge. Based on these results, we decided to test three doses of B. subtilis spores, 1 × 109 , 1 × 1010 and 1 × 1011 spores/mouse, to evaluate their effectiveness in reducing infection caused by C. rodentium. Purified spores were administered one day before infection and mice sacrificed 15 days after infection to assess colon colonization and mucosal hyperplasia. As shown in Fig. 1A, a significant (P < 0.05) reduction in colonic colonization was observed only in mice predosed with the highest amounts of spores. However, no differences in crypt length (Fig. 1B) nor in the number of CD4+ cells infiltrating the
894
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
Fig. 2. Predosing with B. subtilis spores (1 × 109 spores) of suckling mice infected with C. rodentium (1 × 103 CFUs): effects on enteropathy. (A) diagram of crypt length; (B) hematoxylin- and eosin-stained colonic section of mock-infected (sterile broth) mouse showing normal colonic architecture; (C) colonic section of C. rodentium-infected mouse showing crypt hyperplasia, loss of surface epithelial cells and increased vascularization of LP; (D) colonic section of infected mouse pretreated with spores showing absence of enteropathy. There were six mice per group and the experiment was repeated three times with analogous results. In the control group (ctr) mice were non-infected. Panels B–D are at an identical level of magnification (100×). * P < 0.05. ** P < 0.001.
mucosa (Fig. 1C) were observed between predosed and nonpredosed infected mice at any dose. A possible interpretation of these results is that spore administration was potentially effective at reducing enteropathy, as suggested by the reduction in colon colonization, but that the high dose of pathogen used to induce the disease in adult mice masked this effect. 3.2. Suckling NIH Swiss mice are a sensitive infection model for C. rodentium To test whether our interpretation of data was correct, we decided to develop a more sensitive infection model that would enable us to use a lower dose of pathogen for detecting clear symptoms of infection. With this aim, we tested suckling (7 days old) NIH Swiss mice with decreasing doses of C. rodentium, from 1 × 107 to 1 × 103 CFUs. We registered high mortality rates, ranging between 58 and 78%, at all examined doses (data not shown). Mice infected with 1 × 103 CFUs of the nalr strain of C. rodentium that survived the challenge were sacrificed 15 days after infection and were found to have increased crypt length (Fig. 2A), high levels of colon colonization (Fig. 3A) and increased numbers of CD4+ cells (Fig. 4A). Colonic sections of infected and control non-infected mice were hematoxylin- and eosin-stained and microscopically analyzed. As shown in panels B and C of Fig. 2, infected mice showed clear signs of
enteropathy with crypt hyperplasia, loss of surface epithelial cells and increased vascularization of LP. 3.3. Spore predosing in suckling NIH Swiss mice On the basis of these findings, we decided to test whether B. subtilis spores were able to reduce enteropathy in suckling mice infected with 1 × 103 CFUs of C. rodentium. We predosed mice with a single dose of 1 × 109 /mouse, since the small volume of the inoculum used with suckling mice (20 µl) did not allow us to use higher amounts of spores. Purified spores were administered one day before infection with 1 × 103 CFUs of C. rodentium, and mice were sacrificed 15 days after infection to assess colonization of internal organs and mucosal hyperplasia. In a parallel experiment, groups of mice were followed for 28 days after infection to assess the mortality rate. We observed that spore predosing was successful at preventing enteropathy, as indicated by the analysis of various morphological parameters characterizing C. rodentium infection, such as loss of epithelial cell surface and increased vascularization of LP. In comparison with infected mice (Fig. 2C), absence of epithelial damage, normal vascularization (Fig. 2D) and reduced crypt length (Fig. 2A) were observed in spore-predosed mice. In addition, spore predosing significantly reduced colon colonization (Fig. 3A); however, the low levels of colonization of internal organs detected in infected mice resulted unchanged by spore predosing (Fig. 3B–3C). Most important, in this model, spore predosing significantly decreased the mortality rate: in-
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
895
Fig. 4. Modulation of the immune response following B. subtilis spore predosing. Mice treated as described in the legend of Fig. 3 were analyzed for CD4+ cell number (A) and IFNγ transcript expression in the colon (B). For panel A, there were six mice per group and the experiment was repeated two times with analogous results. For panel B, reported results are representative of three independent experiments. In the control group (ctr) mice were non-infected. ** P < 0.001.
Fig. 3. Predosing with B. subtilis spores (1 × 109 spores/mouse) of suckling mice infected with C. rodentium (1 × 103 CFUs/mouse): effects on enteropathogen colonization. (A) colon; (B) liver; (C) MLN. There were six mice per group and the experiment was repeated two times with analogous results. In the control group (ctr) mice were non-infected. ** P < 0.001.
deed, we counted 48 dead mice out of 56 (86%) without spores and 21 dead mice out of 56 (40%) after spore predosing. Interestingly, administration of B. subtilis spores did not affect the C. rodentium-induced high number of intestinal CD4+ cells (Fig. 4A); in accordance with this finding, the transcript levels of IFNγ detected in infected mice further increased following spore predosing (Fig. 4B). 4. Discussion C. rodentium is the etiological agent of transmissible murine colonic hyperplasia, a naturally occurring disease of laboratory mice characterized by epithelial cell hyperproliferation in the descending colon [34]. Ultrastructural analysis of C. rodentiuminfected colonic tissue reveals AE pedestal formation beneath adherent bacteria [33], indistinguishable from that caused by the human EPEC [31] and EHEC E. coli [10]. T-cell infiltration, a highly polarized Th1 immune response and epithelial cell pro-
liferation have been observed following C. rodentium infection [16]. It is noteworthy that these features resemble those seen in mouse models of inflammatory bowel diseases (IBD) [24,39]. Therefore, the murine model of C. rodentium infection could represent a useful tool for studies aimed at identification of the immunological basis of different gut diseases in humans, not exclusively due to infection. Normally, infection in adult mice is subclinical and selflimiting, producing little morbidity or mortality [3]. In contrast, suckling and adult animals of some inbred strains [22], as well as transgenic lines [27], have been shown to be more susceptible to infection associated with clinical signs of the disease. Among factors that determine whether a significant inflammatory response will develop are host genetic background, diet [2] and indigenous microbiota [20]. In this work, we initially tested adult NIH Swiss mice for Citrobacter inoculation on the basis of literature data [2]. In accordance with these findings, we were able to demonstrate colon colonization and crypt hyperplasia following infection. However, previous reports which described disease onset had used a lower number of bacteria than that used by us (1 × 1010 CFUs) [3,21]; one possible explanation for this discrepancy can be found in the different housing and dietary conditions of the colonies that influence intestinal microbiota and, consequently the bacterial load able to induce mucosal lesions. Accordingly,
896
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
germ-free mice may develop gut inflammation earlier than nongerm-free mice [21]. It is also significant that diet itself can influence the baseline colon morphology [2] and presumably, the presentation of clinical disease. Further studies are needed to better address this issue. In sucklings, C. rodentium administration at 1 × 107 CFUs/ mouse caused a high mortality rate. Interestingly, high rates were maintained also at very low doses of the pathogen, with surviving mice showing clear signs of enteropathy. It is noteworthy that C. rodentium colon colonization in suckling mice infected with 1 × 103 CFUs reached very high levels. The stage of development of the mucosal immune system represents an important parameter to be considered, influencing the onset of infection and its progression: in the absence of a mature immune system 1 × 103 CFUs are manifestly sufficient for massive expansion in the colon and internal organs, but also for inducing severe disease and mortality. On the other hand, activation of the immune response in surviving mice was highlighted by increased CD4+ cell number and INFγ expression in the distal colon, in accordance with literature data [16,37]. In line with the perspective of identifying strategies aimed at preventing the onset of infection-mediated enteropathy and death, we focused our attention on the use of B. subtilis spores. The use of spores for this purpose represents an interesting alternative approach, as it is generally accepted that protection from pathogens depends upon administration of vaccines. However, microbial spores have several advantages over vegetative cells and molecular antigens used as oral vaccines, the most obvious being their higher resistance to gastric acidity, toxicity of bile salts and hydrolytic enzyme activities in the gut. Spores of the Gram-positive soil bacterium B. subtilis, a nonpathogenic microorganism, are currently used as probiotics for both humans and animals, and particularly in the treatment of diarrhea in humans [18,35]. However, in spite of extensive commercial use, rigorous studies addressing their mode of action and controlled trials to prove their efficacy have not yet been performed. We thus began to analyze the efficacy of spore predosing to prevent C. rodentium infection in mice. Initially, we evaluated the potential of administering increasing amounts of spores in adult NIH Swiss mice before infection; we found that B. subtilis spores were unable to suppress the various examined aspects of C. rodentium infection, with the exception of a statistically significant decrease in colon colonization at 1 × 1011 spores/mouse. On the basis of these findings, we hypothesized that the ability of B. subtilis spores to reduce enteropathy might be strictly dependent on the dose of C. rodentium used to infect mice. Interestingly, in suckling mice infected with 1 × 103 CFUs, predosing with 1 × 109 spores was effective in drastically decreasing colon colonization; more importantly, spore predosing was successful in preventing enteropathy: reduced crypt length, no epithelial damage and normal vascularization were all observed. Accordingly, a significant reduction in the mortality rate was seen. In contrast, C. rodentium colonization of liver and MLN was not reduced by spore predosing, suggesting that the low levels of internal organ colonization that
we detected did not play a major role in the outcome of the disease induced by C. rodentium in this model. It has been shown that CD4+ cells infiltrating the colonic lamina propria and epithelium are associated with C. rodentium infection [24]. A central role for CD4+ T cells in raising immunity to C. rodentium has been clearly shown, as mice depleted of this cell subset were found to be highly susceptible to infection [37]. In accordance with these data, we detected a significant increase in CD4+ cells in the colonic lamina propria both in adult and suckling mice; interestingly, spore predosing did not significantly influence CD4+ cell recruitment at the various examined doses. It is noteworthy that the levels of IFNγ mRNA, assessed by real-time PCR, remained high following spore predosing, in analogy with recent observations in mice predosed with a mixture of lactobacilli [23]. Hence, the number of residual enteropathogens, drastically reduced by spore predosing, was still sufficient to elicit an immune response. Alternatively, we speculate that the intrinsic immunogenicity of B. subtilis spores contributed to the highlighted immune response. Recently, assessment of the immunogenicity of B. subtilis spores in mice showed induction of both systemic and mucosal humoral immunity, not exclusively against spores but also against vegetative bacilli [11]. Further studies are required to better address this issue. Taken together, our immunological data suggest also that the ability of the spores to reduce the infection could be better explained by competitive exclusion rather than by immune exclusion mechanisms. In conclusion, we established a new model of C. rodentium infection by identifying the minimal dose of CFUs required for inducing enteropathy and death in suckling mice. Using this model, we demonstrated that B. subtilis spore predosing was effective in reducing the infection and cognate enteropathy, leaving unaltered the immune response to the pathogen. The availability of such a model will contribute to highlighting other important parameters related to the mechanism of action of B. subtilis spores. From this point of view, future work will be aimed at better defining the individual contributions of competitive exclusion and immuno-enhancing activity exerted by B. subtilis spores in prophylaxis against infections by enteropathogens. Acknowledgements We thank Drs. Emilia Mauriello and Angela Cordone (Federico II University of Naples) for their help in preparing B. subtilis spores. This work was supported by a grant from the European Commission (contract no. CE QLK5 2001 01729 Sporebiotics). References [1] T.M. Barbosa, C.R. Serra, R.M. La Ragione, M.J. Woodward, A.O. Henriques, Screening for Bacillus isolates in the broiler gastrointestinal tract, Appl. Environ. Microbiol. 71 (2005) 968–978. [2] S.W. Barthold, G.W. Osbaldiston, A.M. Jonas, Dietary, bacterial, and host genetic interactions in the pathogenesis of transmissible murine colonic hyperplasia, Lab. Anim. Sci. 27 (1977) 938–945.
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
[3] S.W. Barthold, G.L. Coleman, R.O. Jacoby, E.M. Livestone, A.M. Jonas, Transmissible murine colonic hyperplasia, Vet. Pathol. 15 (1978) 223– 236. [4] G. Casula, S.M. Cutting, Bacillus probiotics: Spore germination in the gastrointestinal tract, Appl. Environ. Microbiol. 68 (2002) 2344–2352. [5] H.D. Chen, G. Frankel, Enteropathogenic Escherichia coli: Unravelling pathogenesis, FEMS Microbiol. Rev. 29 (2005) 83–98. [6] P. Chomczynski, N. Sacchi, Single-step method isolation by acid guanidium thiocyanate-phenol-chloroform extraction, Anal. Biochem. 162 (1987) 156–159. [7] S. Cutting, P.B. Vander Horn, Genetic analysis, in: C. Harwood, S. Cutting (Eds.), Molecular Biological Methods for Bacillus, Wiley, Chichester, UK, 1990, pp. 27–74. [8] D. Czerucka, S. Dahan, B. Mograbi, B. Rossi, P. Rampal, P. Saccharomyces boulardii preserves the barrier function and modulates the signal transduction pathway induced in enteropathogenic Escherichia coliinfected T84 cells, Infect. Immun. 68 (2000) 5998–6004. [9] D. Czerucka, P. Rampal, Experimental effects of Saccharomyces boulardii on diarrheal pathogens, Microbes Infect. 4 (2002) 733–739. [10] M.S. Donnenberg, S. Tzipori, M.L. McKee, A.D. O’Brien, J. Alroy, J.B. Kaper, The role of the eae gene of enterohemorrhagic Escherichia coli in intimate attachment in vitro and in a porcine model, J. Clin. Invest. 92 (1993) 1418–1424. [11] L.H. Duc, H.A. Hong, S.M. Cutting, Germination of the spore in the gastrointestinal tract provides a novel route for heterologous antigen delivery, Vaccine 21 (2003) 4215–4224. [12] L.H. Duc, H.A. Hong, N.Q. Uyen, S.M. Cutting, Intracellular fate and immunogenicity of B. subtilis spores, Vaccine 22 (2004) 1873–1885. [13] L.H. Duc, H.A. Hong, T.M. Barbosa, A.O. Henriques, S.M. Cutting, Characterization of Bacillus probiotics available for human use, Appl. Environ. Microbiol. 70 (2004) 2161–2171. [14] P.G. Falk, L.V. Hooper, T. Midvedt, J.I. Gordon, Creating and maintaining the gastrointestinal ecosystem: What we know add need to know for gnotobiology, Microbiol. Mol. Biol. Rev. 62 (1998) 1157–1170. [15] J.M. Hamilton-Miller, The role of probiotics in the treatment and prevention of Helicobacter pylori infection, Int. J. Antimicrob. Agents 22 (2003) 360–366. [16] L.M. Higgins, G. Frankel, G. Douce, G. Dougan, T.T. MacDonald, Citrobacter rodentium infection in mice elicits a mucosal Th1 cytokine response and lesions similar to those in murine inflammatory bowel disease, Infect. Immun. 67 (1999) 3031–3039. [17] T.T. Hoa, L.H. Duc, R. Isticato, L. Baccigalupi, E. Ricca, P.H. Van, S.M. Cutting, Fate and dissemination of Bacillus subtilis spores in a murine model, Appl. Environ. Microbiol. 67 (2001) 3819–3823. [18] H.A. Hong, L.H. Duc, S.M. Cutting, The use of bacterial spore formers as probiotics, FEMS Microbiol. Rev. 29 (2005) 813–835. [19] L.V. Hooper, L. Bry, P.G. Falk, J.I. Gordon, Host–microbial symbiosis in the mammalian intestine: Exploring an internal ecosystem, Bioessays 20 (1998) 336–343. [20] K. Itoh, K. Maejima, K. Ueda, K. Fujiwara, Effect of intestinal flora on megaenteron of mice, Microbiol. Immunol. 22 (1978) 661–672. [21] K. Itoh, K. Ueda, K. Fujiwara, Susceptibility of germ-free mice to infectious megaenteron, Microbiol. Immunol. 24 (1980) 281–290. [22] K. Itoh, T. Matsui, K. Tsuji, T. Mitsuoka, K. Ueda, Genetic control in the susceptibility of germfree inbred mice to infection by Escherichia coli O115a,c:K(B), Infect. Immun. 56 (1988) 930–935. [23] K.C. Johnson-Henry, M. Nadjafi, Y. Avitzur, D.J. Mitchell, B.Y. Ngan, E. Galindo-Mata, N.L. Jones, P.M. Sherman, Amelioration of the effects of Citrobacter rodentium infection in mice by pretreatment with probiotics, J. Infect. Dis. 191 (2005) 2106–2117.
897
[24] R. Kuhn, J. Lohler, D. Rennick, K. Rajewsky, W. Muller, Interleukin-10deficient mice develop chronic enterocolitis, Cell 75 (1993) 263–274. [25] R.M. La Ragione, M.J. Woodward, Competitive exclusion by Bacillus subtilis spores of Salmonella enterica serotype Enteritidis and Clostridium perfringens in young chickens, Vet. Microbiol. 94 (2003) 245–256. [26] Y.K. Lee, S. Salminen, The coming age of probiotics, Trends Food Sci. Technol. 6 (1995) 241–245. [27] L. Maggio-Price, K.L. Nicholson, K.M. Kline, T. Birkebak, I. Suzuki, D.L. Wilson, D. Schauer D, P.J. Fink, Diminished reproduction, failure to thrive, and altered immunologic function in a colony of T-cell receptor transgenic mice: Possible role of Citrobacter rodentium, Lab. Anim. Sci. 48 (1998) 145–155. [28] W.L. Nicholson, P. Setlow, Sporulation, germination and outgrowth, in: C. Harwood, S. Cutting (Eds.), Molecular Biological Methods for Bacillus, Wiley, Chichester, UK, 1990, pp. 391–450. [29] E. Nurmi, M. Rantala, New aspects of Salmonella infection in broiler production, Nature 241 (1973) 210–211. [30] S. Resta-Lenert, K.E. Barrett, Live probiotics protect intestinal epithelial cells from the effects of infection with enteroinvasive Escherichia coli (EIEC), Gut 52 (2003) 988–997. [31] R.J. Rothbaum, J.C. Partin, K. Saalfield, A.J. McAdams, An ultrastructural study of enteropathogenic Escherichia coli infection in human infants, Ultrastruct. Pathol. 4 (1983) 291–304. [32] D.B. Schauer, S. Falkow, Attaching and effacing locus of a Citrobacter freundii biotype that causes transmissible murine colonic hyperplasia, Infect. Immun. 61 (1993) 2486–2492. [33] D.B. Schauer, S. Falkow, The eae gene of Citrobacter freundii biotype 4280 is necessary for colonization in transmissible murine colonic hyperplasia, Infect. Immun. 61 (1993) 4654–4661. [34] D.B. Schauer, B.A. Zabel, I.F. Pedraza, C.M. O’Hara, A.G. Steigerwalt, D.J. Brenner, Genetic and biochemical characterization of Citrobacter rodentium sp. nov, J. Clin. Microbiol. 33 (1995) 2064–2068. [35] S. Senesi, Bacillus spores as probiotic products for human use, in: E. Ricca, A.O. Henriques, S.M. Cutting (Eds.), Bacterial Spore Formers, Horizon Bioscience, Norfolk, UK, 2004, pp. 131–141. [36] S. Senger, D. Luongo, F. Maurano, F.M. Mazzeo, R.A. Siciliano, C. Gianfrani, C. David, R. Troncone, S. Auricchio, M. Rossi, Intranasal administration of a recombinant alpha-gliadin down-regulates the immune response to wheat gliadin in DQ8 transgenic mice, Immunol. Lett. 88 (2003) 127–134. [37] C.P. Simmons, S. Clare, M. Ghaem-Maghami, T.K. Uren, J. Rankin, A. Huett, R. Goldin, D.J. Lewis, T.T. MacDonald, R.A. Strugnell, G. Frankel, G. Dougan, Central role for B lymphocytes and CD4+ T cells in immunity to infection by the attaching and effacing pathogen Citrobacter rodentium, Infect. Immun. 71 (2003) 5077–5086. [38] N.K. Tam, N.Q. Uyen, H.A. Hong, H. Duc le, T.T. Hoa, C.R. Serra, A.O. Henriques, S.M. Cutting, The intestinal life cycle of Bacillus subtilis and close relatives, J. Bacteriol. 188 (2006) 2692–2700. [39] M. Watanabe, Y. Ueno, T. Yajima, S. Okamoto, T. Hayashi, M. Yamazaki, Y. Iwao, H. Ishii, S. Habu, M. Uehira, H. Nishimoto, H. Ishikawa, J. Hata, T. Hibi, Interleukin 7 transgenic mice develop chronic colitis with decreased interleukin 7 protein accumulation in the colonic mucosa, J. Exp. Med. 187 (1998) 389–402. [40] S. Wiles, S. Clare, J. Harker, A. Huett, D. Young, G. Dougan, G. Frankel, Organ specificity, colonization and clearance dynamics in vivo following oral challenges with the murine pathogen Citrobacter rodentium, Cell. Microbiol. 6 (2004) 963–972. [41] P. Youngman, J.B. Perkins, R. Losick, A novel method for the rapid cloning in Escherichia coli of Bacillus subtilis chromosomal DNA adjacent to Tn917 insertion, Mol. Gen. Genet. 195 (1984) 424–433.
Bacillus subtilis spores reduce susceptibility to Citrobacter rodentium-mediated enteropathy in a mouse model Rossana D’Arienzo a , Francesco Maurano a , Giuseppe Mazzarella a , Diomira Luongo a , Rosita Stefanile a , Ezio Ricca b , Mauro Rossi a,∗ a Istituto di Scienze dell’Alimentazione, CNR, via Roma 52, 83100 Avellino, Italy b Dipartimento di Biologia Strutturale e Funzionale, Università Federico II, via Cinthia complesso Monte Sant’Angelo, 80126 Naples, Italy
Received 6 March 2006; accepted 5 June 2006 Available online 26 September 2006
Abstract The present work was aimed at investigating whether Bacillus subtilis spores, widely used in probiotic as well as pharmaceutical preparations for mild gastrointestinal disorders, can suppress enteric infections. To address this issue, we developed a mouse model of infection using the mouse enteropathogen Citrobacter rodentium, a member of a family of human and animal pathogens which includes the clinically significant enteropathogenic (EPEC) and enterohemorrhagic (EHEC) Escherichia coli strains. This group of pathogens causes transmissible colonic hyperplasia by using attaching and effacing (A/E) lesions to colonize the host colon. Because of its similarities to human enteropathogens, C. rodentium is now widely used as an in vivo model for gastrointestinal infections. Swiss NIH mice were orally administered B. subtilis spores one day before infection with C. rodentium. Mice were sacrificed on day 15 after infection, and distal colon, liver and mesenteric lymph nodes were removed for bacteria counts, morphology, immunohistology and IFNγ mRNA analysis. We observed that spore predosing was effective in significantly decreasing infection and enteropathy in suckling mice infected with a dose of C. rodentium sufficient to cause colon colonization, crypt hyperplasia and high mortality rates. Moreover, in mice predosed with spores, the number of CD4+ cells and IFNγ transcript levels remained high. These results thus indicate that our newly established model of C. rodentium infection is a suitable system for analyzing the effects of probiotic bacteria on enteroinfections and that B. subtilis spores are efficient at reducing C. rodentium infection in mice, leaving unaltered the immune response against the pathogen. © 2006 Elsevier Masson SAS. All rights reserved. Keywords: Citrobacter rodentium; Bacillus subtilis; Enteropathy; Mucosal immunity; Mesenteric lymph nodes; Spores
1. Introduction The human gastrointestinal tract (GIT) is normally colonized by a large variety of microorganisms, forming the so-called endogenous microflora and including at least 400 microbial species. In this symbiotic interaction, microbes form a first line of defense against invasion by pathogenic organisms and provide the host with essential products, such as vitamins, amino acids and organic acids. These contributions are reciprocated by stable conditions of temperature, pH, osmolarity and food supply for the microorganisms [14,19]. * Corresponding author.
E-mail address: [email protected] (M. Rossi). 0923-2508/$ – see front matter © 2006 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.resmic.2006.06.001
A normally balanced microbiota is important in preventing intestinal disorders as well as microbial infections and, for these reasons, there has been much interest in developing probiotic products containing live bacteria able to survive inside the GIT and display positive effects on human health [26]. Although the use of probiotics is not new, with some products licensed for human use and marketed in the late ’50s [35], it is only recently that this field has begun to receive an increasing level of scientific interest. Lactic acid bacteria (e.g., Lactobacillus spp. and Bifidobacterium spp.) are the microorganisms most commonly used as probiotics. These bacteria are normal colonizers of the human and animal GIT and their use in probiotic products is thought to contribute to restoring the natural microflora of the gut. Mi-
892
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
croorganisms not normally present in the GIT are also widely used as probiotics. The Gram-positive spore-forming bacteria, and normally members of the genus Bacillus, belong to this group [18]. Although these bacteria have long been considered soil microorganisms, a number of recent reports have pointed out that they are commonly found associated with the GIT of humans and animals [1,13]. In addition, molecular as well as immunological data have convincingly shown that within the animal GIT Bacillus spores are able to carry on a complete life cycle, germinating in response to water availability and nutritional stimuli, proliferating and resporulating [4,11,12,17,38]. Starting with the pioneering work of Nurmi and Rantala [29], the concept of using probiotics to suppress or reduce pathogenic infections has been elaborated as a potentially protective approach [15]. A series of studies has demonstrated that a mixture of Streptococcus thermophilus and Lactobacillus acidophilus protects epithelial cells from enteroinvasive Escherichia coli strains [30], that Saccharomyces boulardii is effective in preventing the occurrence of Clostridium difficile [9] and Escherichia coli [8] infections, that Bacillus subtilis spores may be successfully used as competitive exclusion agents in poultry [25]. To expand those studies and to determine whether sporebased probiotic products can be successfully used to prevent infections by enteric pathogens, we began developing a new animal model. As the model pathogen, we used Citrobacter rodentium, a member of a family of human and animal pathogens that use attaching and effacing (A/E) lesions to colonize the host colon. The epithelial lesions caused by C. rodentium are indistinguishable from those of enteropathogenic (EPEC) [31] and enterohemorrhagic (EHEC) [10] E. coli strains. Following oral infection of mice, bacteria colonize the distal colon, causing crypt hyperplasia, mucosal thickening and uneven epithelial surface. A feature common to all these bacteria is the production of A/E lesions [3,33], characterized by localized destruction (effacement) of brush border microvilli and attachment of the bacterium to the host cell membrane, followed by the formation of an underlying pedestal-like structure in the host cell [5,40]. 2. Materials and methods 2.1. Mice NIH Swiss mice were obtained from Harlan-Nossan (Milan, Italy). A colony of pathogen-free mice was then developed at our animal facility. Mice were kept in sterilized cages and fed autoclaved food and water. The protocols employed were in direct accordance with guidelines drafted by the Veterinary Department of the Italian Health Ministry. 2.2. Bacterial strains and growth C. rodentium (formerly Citrobacter freundii strain biotype 4280) strain DBS100 [32] was grown in rich (LB) medium at 37 ◦ C in aerobic conditions and the number of cells (CFUs) determined prior to mouse inoculation. Early stationary phase
cells were always used to inoculate mice. A spontaneous nalidixic-acid-resistant (nalr ) strain was obtained by plating 5 × 109 cells of DBS100 on LB plates supplemented with 100 µg/ml of nalidixic acid. Resistant colonies grown on the selective plate were purified and used to inoculate mice. B. subtilis strain PY79 [41] was used to obtain spores by the exhaustion method [7]. In brief, cells were grown at 37 ◦ C in aerobic conditions in Difco sporulation medium (DSM; Difco Laboratories) for 36 h and spores were collected by centrifugation, washed several times and purified by lysozyme treatment to eliminate vegetative cells, as previously described [28]. The number of purified spores obtained was measured by direct counting with a Burker chamber under an optical microscope (Olympus BH-2 with 40× lenses). 2.3. Bacterial administration in mice Mice were infected by oral gavage with 0.1 ml and 20 µl broth for adult (21-day-old) and suckling (7-day-old) mice, respectively, containing different CFUs of C. rodentium (day 0). In spore predosing experiments, mice were orally administered different amounts of B. subtilis spores in the same volumes used for C. rodentium one day before infection. Mice were sacrificed on day 15 after infection; following careful dissection the distal colon (4 cm proximal to the anal verge), liver and mesenteric lymph nodes (MLN) were aseptically removed for analysis, as described below. 2.4. Animal studies 2.4.1. Mortality rate assessment Following infection, mice were monitored every day and when any mouse became moribund, it was immediately sacrificed. Data were expressed as percentages of the initial number. 2.4.2. Bacterial counts Colonic tissue, liver and MLN were resuspended in 4 ml PBS and homogenized on ice. Serial dilutions of the homogenates were plated onto LB-agar plates containing 100 µg/ml nalidixic acid and incubated at 37 ◦ C. Bacterial colonies were enumerated the following day. 2.4.3. Mucosal hyperplasia score Full-thickness colonic tissues were fixed in 10% neutral buffered formalin. Sections (3 µm) were cut and stained with hematoxylin and eosin. Crypt height was measured by micrometry with 10 measurements for each mouse by counting only well-oriented crypts. 2.4.4. Immunohistochemistry Cryostat sections (5 µm) were air-dried at room temperature and fixed in acetone for 10 min. All sections were repeatedly washed at room temperature in Tris-buffered saline (TBS), incubated in normal rabbit serum and diluted 1:100 in TBS for 30 min. Sections were then incubated with the monoclonal antibody YTS 191 (anti-CD4; BD Biosciences, Heidelberg, Germany) for 1 h at room temperature. Next, sections were incu-
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
893
bated for 30 min with rabbit anti-mouse serum (Dako, Milan, Italy) and for a further 30 min with peroxidase/anti-peroxidase (PAP) complex (Dako). Slides were washed in TBS for 10 min and subsequently developed by a final incubation of five min with 2 amino-9 ethyl-carbazole (Sigma). As control, nonimmune mouse IgGs of the same isotype were used. Sections were finally stained with Mayer’s hematoxylin and mounted. CD4+ cell number in the colonic lamina propria (LP) compartment was determined by counting the number of stained cells within a total area of 1 mm2 of LP using a microscope with a calibrated ocular aligned parallel to the muscularis mucosae. Counts were independently analyzed in a blinded manner by two observers. 2.4.5. Real-time PCR analysis of IFNγ transcript levels Total RNA was extracted from colonic tissue using a single step guanidine isothiocianate-phenol-chloroform isolation method [6]. Reverse transcription of 1 µg RNA was primed using oligo-(dT)12–18 and different aliquots of cDNA were then used for PCR. Real-time quantitative PCR was performed and analyzed using the iCycler (BIO-RAD Laboratories Inc., Hercules, CA, USA). The PCR reaction buffer contained 1× SYBR Green supermix (BIO-RAD) and 500 nM of each primer. The temperature profile of the amplification consisted of 50 cycles of 30 s denaturation at 95 ◦ C, 30 s annealing at 54.6 ◦ C (L32, housekeeping gene) or 56.4 ◦ C (IFNγ) and 40 s extension at 72 ◦ C. Negative controls were performed by omitting RNA from the cDNA synthesis and specific PCR amplification. The oligonucleotides used for amplification of IFNγ were previously described [36]. The primers used for amplification of the housekeeping gene L32 were designed to avoid amplification from contaminating genomic DNA by using Beacon designer Software (BIO-RAD). Their sequences are: 5 -AGCAGAGCTGGAGTCGCTTT (sense); 5 -GGAGCTGCCATCCAAAAGATACTA (anti-sense). The amount of IFNγ transcript was normalized to L32 according IFNγ to the formula: 2-CT where CT = CT − CL32 T . Normalized IFNγ mRNA levels were expressed relative to the expression in control mice according to the formula: 2-CT where sample CT = CT − Ccontrol . Thus, 1 arbitrary unit (AU) T corresponds to the relative levels of IFNγ in control mice. 2.5. Statistical analysis Student’s t test was used to compare results from two different experimental groups. Results were expressed as mean ± SD. Statistical significance was reached when P values were less than 0.05. 3. Results 3.1. Spore predosing in adult NIH Swiss mice In order to test the effect of B. subtilis spores at preventing or reducing enteroinfections, we decided to establish an infection model using the mouse enteropathogen C. rodentium. In preliminary experiments, adult NIH Swiss mice (21-days-old)
Fig. 1. Predosing with B. subtilis spores of adult mice. 21-day-old mice were orally administered different amounts of spores one day before infection with C. rodentium (1 × 1010 CFUs). On day 15 after infection, mice were sacrificed and distal colon removed for bacteria counts (A), evaluation of crypt length (B) and infiltrating CD4+ cells in the LP (C). There were six mice per group and the experiment was repeated three times with analogous results. In the control group (ctr) mice were non-infected. * P < 0.05. ** P < 0.001.
were orally challenged with a nalr derivative of the DBS100 strain at different doses and it was found that the optimal dose for inducing colon colonization and enteropathy was 1 × 1010 CFUs (data not shown). However, no dead mice were observed after oral challenge. Based on these results, we decided to test three doses of B. subtilis spores, 1 × 109 , 1 × 1010 and 1 × 1011 spores/mouse, to evaluate their effectiveness in reducing infection caused by C. rodentium. Purified spores were administered one day before infection and mice sacrificed 15 days after infection to assess colon colonization and mucosal hyperplasia. As shown in Fig. 1A, a significant (P < 0.05) reduction in colonic colonization was observed only in mice predosed with the highest amounts of spores. However, no differences in crypt length (Fig. 1B) nor in the number of CD4+ cells infiltrating the
894
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
Fig. 2. Predosing with B. subtilis spores (1 × 109 spores) of suckling mice infected with C. rodentium (1 × 103 CFUs): effects on enteropathy. (A) diagram of crypt length; (B) hematoxylin- and eosin-stained colonic section of mock-infected (sterile broth) mouse showing normal colonic architecture; (C) colonic section of C. rodentium-infected mouse showing crypt hyperplasia, loss of surface epithelial cells and increased vascularization of LP; (D) colonic section of infected mouse pretreated with spores showing absence of enteropathy. There were six mice per group and the experiment was repeated three times with analogous results. In the control group (ctr) mice were non-infected. Panels B–D are at an identical level of magnification (100×). * P < 0.05. ** P < 0.001.
mucosa (Fig. 1C) were observed between predosed and nonpredosed infected mice at any dose. A possible interpretation of these results is that spore administration was potentially effective at reducing enteropathy, as suggested by the reduction in colon colonization, but that the high dose of pathogen used to induce the disease in adult mice masked this effect. 3.2. Suckling NIH Swiss mice are a sensitive infection model for C. rodentium To test whether our interpretation of data was correct, we decided to develop a more sensitive infection model that would enable us to use a lower dose of pathogen for detecting clear symptoms of infection. With this aim, we tested suckling (7 days old) NIH Swiss mice with decreasing doses of C. rodentium, from 1 × 107 to 1 × 103 CFUs. We registered high mortality rates, ranging between 58 and 78%, at all examined doses (data not shown). Mice infected with 1 × 103 CFUs of the nalr strain of C. rodentium that survived the challenge were sacrificed 15 days after infection and were found to have increased crypt length (Fig. 2A), high levels of colon colonization (Fig. 3A) and increased numbers of CD4+ cells (Fig. 4A). Colonic sections of infected and control non-infected mice were hematoxylin- and eosin-stained and microscopically analyzed. As shown in panels B and C of Fig. 2, infected mice showed clear signs of
enteropathy with crypt hyperplasia, loss of surface epithelial cells and increased vascularization of LP. 3.3. Spore predosing in suckling NIH Swiss mice On the basis of these findings, we decided to test whether B. subtilis spores were able to reduce enteropathy in suckling mice infected with 1 × 103 CFUs of C. rodentium. We predosed mice with a single dose of 1 × 109 /mouse, since the small volume of the inoculum used with suckling mice (20 µl) did not allow us to use higher amounts of spores. Purified spores were administered one day before infection with 1 × 103 CFUs of C. rodentium, and mice were sacrificed 15 days after infection to assess colonization of internal organs and mucosal hyperplasia. In a parallel experiment, groups of mice were followed for 28 days after infection to assess the mortality rate. We observed that spore predosing was successful at preventing enteropathy, as indicated by the analysis of various morphological parameters characterizing C. rodentium infection, such as loss of epithelial cell surface and increased vascularization of LP. In comparison with infected mice (Fig. 2C), absence of epithelial damage, normal vascularization (Fig. 2D) and reduced crypt length (Fig. 2A) were observed in spore-predosed mice. In addition, spore predosing significantly reduced colon colonization (Fig. 3A); however, the low levels of colonization of internal organs detected in infected mice resulted unchanged by spore predosing (Fig. 3B–3C). Most important, in this model, spore predosing significantly decreased the mortality rate: in-
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
895
Fig. 4. Modulation of the immune response following B. subtilis spore predosing. Mice treated as described in the legend of Fig. 3 were analyzed for CD4+ cell number (A) and IFNγ transcript expression in the colon (B). For panel A, there were six mice per group and the experiment was repeated two times with analogous results. For panel B, reported results are representative of three independent experiments. In the control group (ctr) mice were non-infected. ** P < 0.001.
Fig. 3. Predosing with B. subtilis spores (1 × 109 spores/mouse) of suckling mice infected with C. rodentium (1 × 103 CFUs/mouse): effects on enteropathogen colonization. (A) colon; (B) liver; (C) MLN. There were six mice per group and the experiment was repeated two times with analogous results. In the control group (ctr) mice were non-infected. ** P < 0.001.
deed, we counted 48 dead mice out of 56 (86%) without spores and 21 dead mice out of 56 (40%) after spore predosing. Interestingly, administration of B. subtilis spores did not affect the C. rodentium-induced high number of intestinal CD4+ cells (Fig. 4A); in accordance with this finding, the transcript levels of IFNγ detected in infected mice further increased following spore predosing (Fig. 4B). 4. Discussion C. rodentium is the etiological agent of transmissible murine colonic hyperplasia, a naturally occurring disease of laboratory mice characterized by epithelial cell hyperproliferation in the descending colon [34]. Ultrastructural analysis of C. rodentiuminfected colonic tissue reveals AE pedestal formation beneath adherent bacteria [33], indistinguishable from that caused by the human EPEC [31] and EHEC E. coli [10]. T-cell infiltration, a highly polarized Th1 immune response and epithelial cell pro-
liferation have been observed following C. rodentium infection [16]. It is noteworthy that these features resemble those seen in mouse models of inflammatory bowel diseases (IBD) [24,39]. Therefore, the murine model of C. rodentium infection could represent a useful tool for studies aimed at identification of the immunological basis of different gut diseases in humans, not exclusively due to infection. Normally, infection in adult mice is subclinical and selflimiting, producing little morbidity or mortality [3]. In contrast, suckling and adult animals of some inbred strains [22], as well as transgenic lines [27], have been shown to be more susceptible to infection associated with clinical signs of the disease. Among factors that determine whether a significant inflammatory response will develop are host genetic background, diet [2] and indigenous microbiota [20]. In this work, we initially tested adult NIH Swiss mice for Citrobacter inoculation on the basis of literature data [2]. In accordance with these findings, we were able to demonstrate colon colonization and crypt hyperplasia following infection. However, previous reports which described disease onset had used a lower number of bacteria than that used by us (1 × 1010 CFUs) [3,21]; one possible explanation for this discrepancy can be found in the different housing and dietary conditions of the colonies that influence intestinal microbiota and, consequently the bacterial load able to induce mucosal lesions. Accordingly,
896
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
germ-free mice may develop gut inflammation earlier than nongerm-free mice [21]. It is also significant that diet itself can influence the baseline colon morphology [2] and presumably, the presentation of clinical disease. Further studies are needed to better address this issue. In sucklings, C. rodentium administration at 1 × 107 CFUs/ mouse caused a high mortality rate. Interestingly, high rates were maintained also at very low doses of the pathogen, with surviving mice showing clear signs of enteropathy. It is noteworthy that C. rodentium colon colonization in suckling mice infected with 1 × 103 CFUs reached very high levels. The stage of development of the mucosal immune system represents an important parameter to be considered, influencing the onset of infection and its progression: in the absence of a mature immune system 1 × 103 CFUs are manifestly sufficient for massive expansion in the colon and internal organs, but also for inducing severe disease and mortality. On the other hand, activation of the immune response in surviving mice was highlighted by increased CD4+ cell number and INFγ expression in the distal colon, in accordance with literature data [16,37]. In line with the perspective of identifying strategies aimed at preventing the onset of infection-mediated enteropathy and death, we focused our attention on the use of B. subtilis spores. The use of spores for this purpose represents an interesting alternative approach, as it is generally accepted that protection from pathogens depends upon administration of vaccines. However, microbial spores have several advantages over vegetative cells and molecular antigens used as oral vaccines, the most obvious being their higher resistance to gastric acidity, toxicity of bile salts and hydrolytic enzyme activities in the gut. Spores of the Gram-positive soil bacterium B. subtilis, a nonpathogenic microorganism, are currently used as probiotics for both humans and animals, and particularly in the treatment of diarrhea in humans [18,35]. However, in spite of extensive commercial use, rigorous studies addressing their mode of action and controlled trials to prove their efficacy have not yet been performed. We thus began to analyze the efficacy of spore predosing to prevent C. rodentium infection in mice. Initially, we evaluated the potential of administering increasing amounts of spores in adult NIH Swiss mice before infection; we found that B. subtilis spores were unable to suppress the various examined aspects of C. rodentium infection, with the exception of a statistically significant decrease in colon colonization at 1 × 1011 spores/mouse. On the basis of these findings, we hypothesized that the ability of B. subtilis spores to reduce enteropathy might be strictly dependent on the dose of C. rodentium used to infect mice. Interestingly, in suckling mice infected with 1 × 103 CFUs, predosing with 1 × 109 spores was effective in drastically decreasing colon colonization; more importantly, spore predosing was successful in preventing enteropathy: reduced crypt length, no epithelial damage and normal vascularization were all observed. Accordingly, a significant reduction in the mortality rate was seen. In contrast, C. rodentium colonization of liver and MLN was not reduced by spore predosing, suggesting that the low levels of internal organ colonization that
we detected did not play a major role in the outcome of the disease induced by C. rodentium in this model. It has been shown that CD4+ cells infiltrating the colonic lamina propria and epithelium are associated with C. rodentium infection [24]. A central role for CD4+ T cells in raising immunity to C. rodentium has been clearly shown, as mice depleted of this cell subset were found to be highly susceptible to infection [37]. In accordance with these data, we detected a significant increase in CD4+ cells in the colonic lamina propria both in adult and suckling mice; interestingly, spore predosing did not significantly influence CD4+ cell recruitment at the various examined doses. It is noteworthy that the levels of IFNγ mRNA, assessed by real-time PCR, remained high following spore predosing, in analogy with recent observations in mice predosed with a mixture of lactobacilli [23]. Hence, the number of residual enteropathogens, drastically reduced by spore predosing, was still sufficient to elicit an immune response. Alternatively, we speculate that the intrinsic immunogenicity of B. subtilis spores contributed to the highlighted immune response. Recently, assessment of the immunogenicity of B. subtilis spores in mice showed induction of both systemic and mucosal humoral immunity, not exclusively against spores but also against vegetative bacilli [11]. Further studies are required to better address this issue. Taken together, our immunological data suggest also that the ability of the spores to reduce the infection could be better explained by competitive exclusion rather than by immune exclusion mechanisms. In conclusion, we established a new model of C. rodentium infection by identifying the minimal dose of CFUs required for inducing enteropathy and death in suckling mice. Using this model, we demonstrated that B. subtilis spore predosing was effective in reducing the infection and cognate enteropathy, leaving unaltered the immune response to the pathogen. The availability of such a model will contribute to highlighting other important parameters related to the mechanism of action of B. subtilis spores. From this point of view, future work will be aimed at better defining the individual contributions of competitive exclusion and immuno-enhancing activity exerted by B. subtilis spores in prophylaxis against infections by enteropathogens. Acknowledgements We thank Drs. Emilia Mauriello and Angela Cordone (Federico II University of Naples) for their help in preparing B. subtilis spores. This work was supported by a grant from the European Commission (contract no. CE QLK5 2001 01729 Sporebiotics). References [1] T.M. Barbosa, C.R. Serra, R.M. La Ragione, M.J. Woodward, A.O. Henriques, Screening for Bacillus isolates in the broiler gastrointestinal tract, Appl. Environ. Microbiol. 71 (2005) 968–978. [2] S.W. Barthold, G.W. Osbaldiston, A.M. Jonas, Dietary, bacterial, and host genetic interactions in the pathogenesis of transmissible murine colonic hyperplasia, Lab. Anim. Sci. 27 (1977) 938–945.
R. D’Arienzo et al. / Research in Microbiology 157 (2006) 891–897
[3] S.W. Barthold, G.L. Coleman, R.O. Jacoby, E.M. Livestone, A.M. Jonas, Transmissible murine colonic hyperplasia, Vet. Pathol. 15 (1978) 223– 236. [4] G. Casula, S.M. Cutting, Bacillus probiotics: Spore germination in the gastrointestinal tract, Appl. Environ. Microbiol. 68 (2002) 2344–2352. [5] H.D. Chen, G. Frankel, Enteropathogenic Escherichia coli: Unravelling pathogenesis, FEMS Microbiol. Rev. 29 (2005) 83–98. [6] P. Chomczynski, N. Sacchi, Single-step method isolation by acid guanidium thiocyanate-phenol-chloroform extraction, Anal. Biochem. 162 (1987) 156–159. [7] S. Cutting, P.B. Vander Horn, Genetic analysis, in: C. Harwood, S. Cutting (Eds.), Molecular Biological Methods for Bacillus, Wiley, Chichester, UK, 1990, pp. 27–74. [8] D. Czerucka, S. Dahan, B. Mograbi, B. Rossi, P. Rampal, P. Saccharomyces boulardii preserves the barrier function and modulates the signal transduction pathway induced in enteropathogenic Escherichia coliinfected T84 cells, Infect. Immun. 68 (2000) 5998–6004. [9] D. Czerucka, P. Rampal, Experimental effects of Saccharomyces boulardii on diarrheal pathogens, Microbes Infect. 4 (2002) 733–739. [10] M.S. Donnenberg, S. Tzipori, M.L. McKee, A.D. O’Brien, J. Alroy, J.B. Kaper, The role of the eae gene of enterohemorrhagic Escherichia coli in intimate attachment in vitro and in a porcine model, J. Clin. Invest. 92 (1993) 1418–1424. [11] L.H. Duc, H.A. Hong, S.M. Cutting, Germination of the spore in the gastrointestinal tract provides a novel route for heterologous antigen delivery, Vaccine 21 (2003) 4215–4224. [12] L.H. Duc, H.A. Hong, N.Q. Uyen, S.M. Cutting, Intracellular fate and immunogenicity of B. subtilis spores, Vaccine 22 (2004) 1873–1885. [13] L.H. Duc, H.A. Hong, T.M. Barbosa, A.O. Henriques, S.M. Cutting, Characterization of Bacillus probiotics available for human use, Appl. Environ. Microbiol. 70 (2004) 2161–2171. [14] P.G. Falk, L.V. Hooper, T. Midvedt, J.I. Gordon, Creating and maintaining the gastrointestinal ecosystem: What we know add need to know for gnotobiology, Microbiol. Mol. Biol. Rev. 62 (1998) 1157–1170. [15] J.M. Hamilton-Miller, The role of probiotics in the treatment and prevention of Helicobacter pylori infection, Int. J. Antimicrob. Agents 22 (2003) 360–366. [16] L.M. Higgins, G. Frankel, G. Douce, G. Dougan, T.T. MacDonald, Citrobacter rodentium infection in mice elicits a mucosal Th1 cytokine response and lesions similar to those in murine inflammatory bowel disease, Infect. Immun. 67 (1999) 3031–3039. [17] T.T. Hoa, L.H. Duc, R. Isticato, L. Baccigalupi, E. Ricca, P.H. Van, S.M. Cutting, Fate and dissemination of Bacillus subtilis spores in a murine model, Appl. Environ. Microbiol. 67 (2001) 3819–3823. [18] H.A. Hong, L.H. Duc, S.M. Cutting, The use of bacterial spore formers as probiotics, FEMS Microbiol. Rev. 29 (2005) 813–835. [19] L.V. Hooper, L. Bry, P.G. Falk, J.I. Gordon, Host–microbial symbiosis in the mammalian intestine: Exploring an internal ecosystem, Bioessays 20 (1998) 336–343. [20] K. Itoh, K. Maejima, K. Ueda, K. Fujiwara, Effect of intestinal flora on megaenteron of mice, Microbiol. Immunol. 22 (1978) 661–672. [21] K. Itoh, K. Ueda, K. Fujiwara, Susceptibility of germ-free mice to infectious megaenteron, Microbiol. Immunol. 24 (1980) 281–290. [22] K. Itoh, T. Matsui, K. Tsuji, T. Mitsuoka, K. Ueda, Genetic control in the susceptibility of germfree inbred mice to infection by Escherichia coli O115a,c:K(B), Infect. Immun. 56 (1988) 930–935. [23] K.C. Johnson-Henry, M. Nadjafi, Y. Avitzur, D.J. Mitchell, B.Y. Ngan, E. Galindo-Mata, N.L. Jones, P.M. Sherman, Amelioration of the effects of Citrobacter rodentium infection in mice by pretreatment with probiotics, J. Infect. Dis. 191 (2005) 2106–2117.
897
[24] R. Kuhn, J. Lohler, D. Rennick, K. Rajewsky, W. Muller, Interleukin-10deficient mice develop chronic enterocolitis, Cell 75 (1993) 263–274. [25] R.M. La Ragione, M.J. Woodward, Competitive exclusion by Bacillus subtilis spores of Salmonella enterica serotype Enteritidis and Clostridium perfringens in young chickens, Vet. Microbiol. 94 (2003) 245–256. [26] Y.K. Lee, S. Salminen, The coming age of probiotics, Trends Food Sci. Technol. 6 (1995) 241–245. [27] L. Maggio-Price, K.L. Nicholson, K.M. Kline, T. Birkebak, I. Suzuki, D.L. Wilson, D. Schauer D, P.J. Fink, Diminished reproduction, failure to thrive, and altered immunologic function in a colony of T-cell receptor transgenic mice: Possible role of Citrobacter rodentium, Lab. Anim. Sci. 48 (1998) 145–155. [28] W.L. Nicholson, P. Setlow, Sporulation, germination and outgrowth, in: C. Harwood, S. Cutting (Eds.), Molecular Biological Methods for Bacillus, Wiley, Chichester, UK, 1990, pp. 391–450. [29] E. Nurmi, M. Rantala, New aspects of Salmonella infection in broiler production, Nature 241 (1973) 210–211. [30] S. Resta-Lenert, K.E. Barrett, Live probiotics protect intestinal epithelial cells from the effects of infection with enteroinvasive Escherichia coli (EIEC), Gut 52 (2003) 988–997. [31] R.J. Rothbaum, J.C. Partin, K. Saalfield, A.J. McAdams, An ultrastructural study of enteropathogenic Escherichia coli infection in human infants, Ultrastruct. Pathol. 4 (1983) 291–304. [32] D.B. Schauer, S. Falkow, Attaching and effacing locus of a Citrobacter freundii biotype that causes transmissible murine colonic hyperplasia, Infect. Immun. 61 (1993) 2486–2492. [33] D.B. Schauer, S. Falkow, The eae gene of Citrobacter freundii biotype 4280 is necessary for colonization in transmissible murine colonic hyperplasia, Infect. Immun. 61 (1993) 4654–4661. [34] D.B. Schauer, B.A. Zabel, I.F. Pedraza, C.M. O’Hara, A.G. Steigerwalt, D.J. Brenner, Genetic and biochemical characterization of Citrobacter rodentium sp. nov, J. Clin. Microbiol. 33 (1995) 2064–2068. [35] S. Senesi, Bacillus spores as probiotic products for human use, in: E. Ricca, A.O. Henriques, S.M. Cutting (Eds.), Bacterial Spore Formers, Horizon Bioscience, Norfolk, UK, 2004, pp. 131–141. [36] S. Senger, D. Luongo, F. Maurano, F.M. Mazzeo, R.A. Siciliano, C. Gianfrani, C. David, R. Troncone, S. Auricchio, M. Rossi, Intranasal administration of a recombinant alpha-gliadin down-regulates the immune response to wheat gliadin in DQ8 transgenic mice, Immunol. Lett. 88 (2003) 127–134. [37] C.P. Simmons, S. Clare, M. Ghaem-Maghami, T.K. Uren, J. Rankin, A. Huett, R. Goldin, D.J. Lewis, T.T. MacDonald, R.A. Strugnell, G. Frankel, G. Dougan, Central role for B lymphocytes and CD4+ T cells in immunity to infection by the attaching and effacing pathogen Citrobacter rodentium, Infect. Immun. 71 (2003) 5077–5086. [38] N.K. Tam, N.Q. Uyen, H.A. Hong, H. Duc le, T.T. Hoa, C.R. Serra, A.O. Henriques, S.M. Cutting, The intestinal life cycle of Bacillus subtilis and close relatives, J. Bacteriol. 188 (2006) 2692–2700. [39] M. Watanabe, Y. Ueno, T. Yajima, S. Okamoto, T. Hayashi, M. Yamazaki, Y. Iwao, H. Ishii, S. Habu, M. Uehira, H. Nishimoto, H. Ishikawa, J. Hata, T. Hibi, Interleukin 7 transgenic mice develop chronic colitis with decreased interleukin 7 protein accumulation in the colonic mucosa, J. Exp. Med. 187 (1998) 389–402. [40] S. Wiles, S. Clare, J. Harker, A. Huett, D. Young, G. Dougan, G. Frankel, Organ specificity, colonization and clearance dynamics in vivo following oral challenges with the murine pathogen Citrobacter rodentium, Cell. Microbiol. 6 (2004) 963–972. [41] P. Youngman, J.B. Perkins, R. Losick, A novel method for the rapid cloning in Escherichia coli of Bacillus subtilis chromosomal DNA adjacent to Tn917 insertion, Mol. Gen. Genet. 195 (1984) 424–433.