Adhesion to the yeast cell surface as a mechanism for trapping pathogenic bacteria by Saccharomyces probiotics

Transcription

1 Journal of Medical Microbiology (2012), 61, DOI /jmm Adhesion to the yeast cell surface as a mechanism for trapping pathogenic bacteria by Saccharomyces probiotics F. C. P. Tiago, 1 F. S. Martins, 1 E. L. S. Souza, 1 P. F. P. Pimenta, 2 H. R. C. Araujo, 2 I. M. Castro, 3 R. L. Brandão 3 and Jacques R. Nicoli 1 Correspondence Jacques R. Nicoli jnicoli@icb.ufmg.br 1 Departamento de Microbiologia, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil 2 Laboratório de Entomologia Médica, Instituto René Rachou, Fiocruz, Belo Horizonte, MG, Brazil 3 Núcleo de Pesquisa em Ciências Biológicas, Universidade Federal de Ouro Preto, Ouro Preto, MG, Brazil Received 2 January 2012 Accepted 7 May 2012 Recently, much attention has been given to the use of probiotics as an adjuvant for the prevention or treatment of gastrointestinal pathology. The great advantage of therapy with probiotics is that they have few side effects such as selection of resistant bacteria or disturbance of the intestinal microbiota, which occur when antibiotics are used. Adhesion of pathogenic bacteria onto the surface of probiotics instead of onto intestinal receptors could explain part of the probiotic effect. Thus, this study evaluated the adhesion of pathogenic bacteria onto the cell wall of Saccharomyces boulardii and Saccharomyces cerevisiae strains UFMG 905, W303 and BY4741. To understand the mechanism of adhesion of pathogens to yeast, cell-wall mutants of the parental strain of Saccharomyces cerevisiae BY4741 were used because of the difficulty of mutating polyploid yeast, as is the case for Saccharomyces cerevisiae and Saccharomyces boulardii. The tests of adhesion showed that, among 11 enteropathogenic bacteria tested, only Escherichia coli, Salmonella Typhimurium and Salmonella Typhi adhered to the surface of Saccharomyces boulardii, Saccharomyces cerevisiae UFMG 905 and Saccharomyces cerevisiae BY4741. The presence of mannose, and to some extent bile salts, inhibited this adhesion, which was not dependent on yeast viability. Among 44 cell-wall mutants of Saccharomyces cerevisiae BY4741, five lost the ability to fix the bacteria. Electron microscopy showed that the phenomenon of yeast bacteria adhesion occurred both in vitro and in vivo (in the digestive tract of dixenic mice). In conclusion, some pathogenic bacteria were captured on the surface of Saccharomyces boulardii, Saccharomyces cerevisiae UFMG 905 and Saccharomyces cerevisiae BY4741, thus preventing their adhesion to specific receptors on the intestinal epithelium and their subsequent invasion of the host. INTRODUCTION In the majority of cases, infectious diarrhoea is treated by rehydration and/or the use of antibiotics. However, the World Health Organization has stimulated the search for treatments other than antibiotics, and probiotics have been proposed as an alternative for the prevention or treatment of a number of diarrhoeal diseases (Vieira et al., 2008). According to the currently adopted definition by the Food and Agriculture Organization/World Health Organization, Abbreviations: GPI, glycosylphosphatidylinositol; PIR, proteins with internal repeats; SEM, scanning electron microscopy. probiotics are live micro-organisms which when administered in adequate amounts confer a health benefit to the host (FAO/WHO, 2002). The great advantage of therapy with probiotics is that they have few side effects such as selection of resistant bacteria and disturbance of the gastrointestinal microbial ecosystem, which can occur when antibiotics are used. Although no proof of efficacy of such treatment against intestinal infections has been demonstrated in humans, murine models have indicated that some probiotics may be efficient against bacterial infections (Jain et al., 2008, 2009; Truusalu et al., 2008; Martins et al., 2009). Most probiotics are strains of Lactobacillus and Bifidobacterium and are used as pharmaceutical preparations G 2012 SGM Printed in Great Britain

2 Bacterial trapping on probiotic yeast surface or fermented dairy products. Species from other bacterial genera such as Escherichia, Bacillus and Enterococcus have also been used, but there are concerns surrounding the safety of such probiotics, as these genera contain opportunistic pathogenic species, particularly Enterococcus. Few non-bacterial micro-organisms such as yeasts (Saccharomyces boulardii) have been studied or commercialized as probiotics. Many mechanisms of action have been proposed to explain yeast probiotic protection against bacterial infection, such as modulating the immune system (Buts et al., 1990; Rodrigues et al., 2000), degrading Clostridium difficile toxins A and B and their respective receptors on colonic mucosa (Castagliuolo et al., 1999; Qamar et al., 2001), inhibiting cholera toxin action (Czerucka et al., 1994; Brandão et al., 1998) and modulating the transduction pathway when activated by enteropathogenic bacteria (Czerucka et al., 2000; Dahan et al., 2003; Martins et al., 2010, 2011). Another frequently cited mechanism for probiotic action is the production of diffusible antagonistic compounds inhibiting pathogenic bacterial growth. In contrast to probiotic bacteria (particularly lactic acid bacteria), probiotic yeasts rarely show this antagonistic ability, at least in vivo (Rodrigues et al., 1996). Finally, a study conducted by Gedek (1999) reported the adhesion of Salmonella Typhimurium, enteropathogenic Escherichia coli and enterohaemorrhagic E. coli to the probiotic yeast Saccharomyces boulardii. Pérez-Sotelo et al. (2005) also noted that the yeast Saccharomyces cerevisiae strain Sc47 binds to strains of Salmonella, particularly those that express type I fimbriae. These observations suggest that the fixation of enteropathogenic bacteria onto the surface of Saccharomyces cells could in part explain the protective effect of some yeast probiotics. Until now, Saccharomyces boulardii has been the only yeast commercialized worldwide as a probiotic for humans, but some authors have suggested the use of other yeast species or genera based essentially on in vitro assays and a few clinical trials (Kovacs & Berk, 2000; Kumura et al., 2004; van der Aa Kühle et al., 2005). We demonstrated previously that Saccharomyces cerevisiae strain UFMG 905 has potential as a probiotic due to its ability to protect mice against Salmonella enterica serovar Typhimurium and Clostridium difficile infection in a murine model (Martins et al., 2005),andtoinhibitbacterial translocation and modulate both local and systemic immunity of mice (Martins et al., 2007; Generoso et al., 2010). Recent immunological data demonstrated that both Saccharomyces boulardii and Saccharomyces cerevisiae UFMG 905 decreased the levels of pro-inflammatory cytokines and modulated the activation of mitogenactivated protein kinases (p38 and JNK, but not ERK1/2), NF-kB and AP-1 signalling pathways, which are involved in the transcriptional activation of pro-inflammatory mediators induced by Salmonella infection (Martins et al., 2010, 2011). Preliminary results obtained from electron microscopy revealed that these probiotic effects could be due, at least in part, to the binding of Salmonella to the yeast cells. The objective of the present study was to evaluate in vitro and in vivo the adhesion of enteropathogenic bacteria onto the surface of yeast cells as a trapping mechanism, which could explain in part the protective effect of probiotics. To understand the mechanism of adhesion of pathogens to yeast, we also used cell-wall mutants of the parental strain Saccharomyces cerevisiae BY4741. METHODS Micro-organisms, growth conditions and conservation. Four yeast strains (Saccharomyces boulardii 17 and Saccharomyces cerevisiae strains UFMG 905, BY4741 and W303) were used as well as 44 cell-wall mutants of Saccharomyces cerevisiae BY4741. Saccharomyces boulardii 17 and Saccharomyces cerevisiae UFMG 905 and BY4741 were selected for their ability to adhere to some pathogenic bacteria, whereas Saccharomyces cerevisiae W303 was used as a negative control for such adhesion. Saccharomyces boulardii 17 (Floratil) was provided by Merck. Saccharomyces cerevisiae UFMG 905 was from the Yeast Bank of Dr Carlos Augusto Rosa (Laboratório de Ecologia e Biotecnologia de Leveduras, Departamento de Microbiologia, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil). The identity of the strain was first determined as Saccharomyces cerevisiae using the computer program YEASTCOMPARE (Ciriello & Lachance, 2001) and retrospectively confirmed by sequencing the D1/D2 variable domains of the large-subunit rrna gene as described by Lachance et al. (1999). Saccharomyces cerevisiae BY4741 and the 44 cell-wall deletion strains came from the European Saccharomyces cerevisiae Archive for Functional Analysis (EUROSCARF) Collection (Table 1, and Saccharomyces cerevisiae W303 was provided by Dr Ieso Miranda Castro and Dr Rogelio Lopes Brandão (Núcleo de Pesquisa em Ciências Biológicas, Universidade Federal de Ouro Preto, Ouro Preto, Brazil). These mutants were not derived from Saccharomyces cerevisiae UFMG 905 and Saccharomyces boulardii due to the difficulty of mutating these polyploid yeasts. Yeast strains were maintained in medium containing 1 % yeast extract, 2 % peptone and 25 % glycerol, and stored at 280 uc. For all experiments, yeast cells were grown in YPG medium (1 % yeast extract, 2 % peptone, 2 % glucose) for 24 h under constant agitation (150 r.p.m.) at 37 uc for Saccharomyces cerevisiae UFMG 905 and Saccharomyces boulardii, and at 28 uc for the yeast-mating BY4741 collection and Saccharomyces cerevisiae W303. The yeast cells were then concentrated by centrifugation at 7000 g to give a concentration of ~10 8 c.f.u. ml 21. Salmonella enterica serovar Typhimurium and Vibrio cholerae, of human origin, were obtained from the Departamento de Microbiologia, Instituto de Ciências Biológicas, UFMG, Belo Horizonte, Brazil. Strains of Bacillus cereus (ATCC 11778), Clostridium difficile (ATCC 9689) Clostridium perfringens type A (ATCC 13124), E. coli (ATCC 25723), Listeria monocytogenes (ATCC 15313), Salmonella enterica serovar Typhi (ATCC 19430), Salmonella enterica serovar Typhimurium (ATCC 14028), Shigella sonnei (ATCC 11060) and Enterococcus faecalis (ATCC 19433) were provided by the Laboratório de Materiais de Referência, Instituto Nacional de Controle de Qualidade em Saúde da Fundação Oswaldo Cruz

3 F. C. P. Tiago and others Table 1. Yeast mutant strains from EUROSCARF used in this study Associated gene Systematic name General description AGA2 YGL032C Adhesion subunit of a-agglutinin BGL2 YGR282C Main protein involved in cell-wall maintenance CCW12 YLR110C Cell-wall mannoprotein CKA2 YOR061W Cell-wall flocculation factor via carbohydrate protein interaction CIS3 YJL158C Cell-wall glycoprotein of the proteins with internal repeats (PIR) family CRH1 YGR189C Cell-wall protein involved in chitin-b-1,6 glucan transfer and induced under stress conditions CRR1 YLR213C Protein produced during sporulation CWH41 YGL027C Involved in cell-wall b-1,6-glucan CWP2 YKL096W-A Protein covalently bound to cell-wall mannoprotein; involved in low-ph resistance DCW1 YKL046C Associated with mannosidase; membrane protein required for cell-wall biosynthesis DFG5 YMR238W Associated with mannosidase; membrane protein required for cell-wall biosynthesis and cell budding DSE2 YHR143W Protein secreted during division; suppressed by camp EXG1 YLR300W Main cell-wall exo-1,3-b-glucanase EXG2 YDR261C Main cell-wall exo-1,3-b-glucanase FIG2 YCR089W Cell-wall adhesin FIT1 YDR534C Mannoprotein incorporated in cell wall via a glycosylphosphatidylinositol (GPI) anchor FIT2 YOR382W Mannoprotein incorporated in cell wall via a GPI anchor FKS3 YMR306W Protein important for spore wall FLO1 YAR050W Cell-wall flocculation factor via carbohydrate protein interaction FLO8 YER109C Cell-wall flocculation factor via carbohydrate protein interaction FLO10 YKR102W Cell-wall flocculation factor via carbohydrate protein interaction GSC2 YGR032W Catalytic subunit for b-1,3-glucan synthesis HKR1 YDR420W Protein involved in synthesis regulation of cell-wall b-1,3-glucan KNH1 YDL049C Protein involved in synthesis regulation of cell-wall b-1,6-glucan KRE6 YPR159W Protein involved in cell-wall b-1,6-glucan synthesis KTR1 YOR99W a-1,2-mannosyltransferase involved in binding of glycosylated proteins to membrane proteins LAS21 YJL062W Cytoplasmic membrane protein involved in GPI synthesis MNT2 YGL257C Mannosyltransferase involved in the addition of mannose to glucose PIR3 YKL163W Glycosylated protein important for cell-wall stability PST1 YDR055W Cell-wall protein containing a binding site to GPI ROT2 YBR229C Glucosidase II catalytic subunit needed for cell-wall synthesis SCW4 YGR279C Cell-wall protein similar to glucanase SCW10 YMR305C Cell-wall protein similar to glucanase SCW11 YGL028C Cell-wall protein similar to glucanase SHE10 YGL228W Associated with protein anchored to GPI; unknown function SKN1 YGR143W Protein involved with sphingolipid synthesis SMK1 YPR054W Kinase protein activated specifically by sporulation SPI1 YER150W Cell-wall protein of unknown function SPR1 YOR190W Protein specific for sporulation SRL1 YOR247W Mannoproteins needed for cell-wall stability SUN4 YNL066W Cell-wall protein related to glucanase involved with cell septation UTR2 YEL040W Cell-wall protein involved in chitin-b-1,6-glucan binding YPS1 YLR120C Aspartic protease bound to cytoplasmic membrane through GPI YPS3 YLR121C Aspartic protease bound to cytoplasmic membrane through GPI (FIOCRUZ), Rio de Janeiro, Brazil. The following bacteria were used as components of indigenous intestinal microbiota: E. coli Nissle 1917 (isolated from Mutaflor); Bacteroides fragilis (ATCC 25285); Bifidobacterium longum A5 (isolated from a healthy infant) and Lactobacillus delbrueckii H2B20 (isolated from a healthy newborn). The aerobic and facultatively anaerobic pathogenic bacterial strains were cultured in brain heart infusion broth (BHI; Difco) for 24 h at 37 uc. The anaerobic bacteria were grown for h at 37 uc in BHI-S broth (BHI supplemented with 0.5 % yeast extract, 0.1 % menadione and 0.1 % haemin) in an anaerobic chamber (Forma Scientific) containing 85 % N 2, 10% H 2 and 5 % CO 2. For 1196 Journal of Medical Microbiology 61

4 Bacterial trapping on probiotic yeast surface preservation of each sample, 1 ml of a culture in BHI or BHI-S containing 20 % glycerol was stored in a 2 ml cryogenic tube at 280 uc. Adhesion and sedimentation assay. The method described by Pérez-Sotelo et al. (2005) was used with some modifications to evaluate the ability of the yeast to agglutinate and deposit with the tested pathogens. Yeasts were grown in YPG medium for 24 h and then diluted with sterile saline to a concentration of 10 8 c.f.u. ml 21. One millilitre of this yeast suspension was added to a 1.5 ml microcentrifugetubewith500ml (10 9 c.f.u.) of the bacteria to be tested. The resulting suspension was homogenized and incubated at room temperature for 120 min. At intervals of 0, 15, 30, 60 and 120 min, a 100 ml aliquot of supernatant was plated onto specific agar medium supplemented with 0.1 % cycloheximide to count the number of pathogenic bacteria [expressed as log (c.f.u. ml 21 )]. The control groups contained only a suspension of the bacteria to be tested. Adhesion visualization. To visualize the ability of the yeast to agglutinate bacterial pathogens, the method described by Pérez-Sotelo et al. (2005) was used. One millilitre of yeast culture grown in YPG medium at 37 uc for 24 h under constant agitation (150 r.p.m.) and containing 10 8 c.f.u. ml 21 was mixed with 500 ml of a suspension containing 10 9 c.f.u. ml 21 of the bacteria to be tested in a six-well plate. The bacteria were grown for h at 37 uc in BHI broth. The presence or absence of agglutination was observed macroscopically after 15 min and 1 h, and then after a maximum time of 3 h. A sample was taken from all wells presenting agglutination, Gram stained and analysed under an optical microscope. Evaluation of factors influencing adhesion. ph. The adhesion and sedimentation assays described above were performed with ph values adjusted to 4.0, 5.0, 6.0 and 8.0 with 1 M HCl or 1 M NaOH. Carbohydrates. The adhesion and sedimentation assays described above were performed in the presence of mannose, galactose, glucose or maltose, as described by Pérez-Sotelo et al. (2005). Yeasts were diluted in 1 ml buffered saline containing 5 % (w/v) mannose, glucose, maltose or galactose to a concentration of 10 8 c.f.u. ml 21, and an equal volume of bacterial culture containing 10 9 c.f.u. ml 21 was added. The presence or absence of adhesion was observed as described above to evaluate the influence of each carbohydrate tested. Bile salts. To determine the occurrence of adhesion between yeasts and pathogenic bacteria in the presence of bile salts (Oxgall; Difco), cultures were treated as described by Guglielmetti et al. (2009), with some modifications. The yeasts and bacteria were grown in YPG medium and BHI, respectively, supplemented with 1 or 3 g Oxgall l 21 for 24 h. In a second experiment, the solutions containing 1 or 3 g Oxgall l 21 were added after growth. Yeast viability. The yeasts were killed using moist heat (121 uc for 15 min) and submitted to adhesion and sedimentation assays as described above. Growth stage. Adhesion of pathogenic bacteria was evaluated using yeasts harvested in two phases of growth, exponential and stationary, using the assays described above for adhesion and sedimentation. All yeasts showed the same growth-curve profile in YPG medium at 37 uc, reaching the stationary phase at about 20 h. Samples representing the stationary phase were harvested after 24 h. For the exponential phase, yeast cells were harvested 1 h after resuspension of a 24 h culture in YPG medium at 37 uc. Scanning electron microscopy (SEM). For in vitro SEM, a sample of the yeast/bacteria suspension was fixed using 2.5 % glutaraldehyde in 0.1 M sodium cacodylate buffer (ph 7.2) at room temperature. After three washes with PBS, the cells were incubated on glass coverslips pre-treated with poly-l-lysine for 1 h, and two drops of 1 % osmium tetroxide plus 0.2 % potassium ferrocyanide were placed on each coverslip and the preparation left for 30 min protected from light. The coverslip was washed with PBS to remove the osmium tetroxide and dehydrated with increasing acetone concentrations ( %) for 30 min for each concentration. Drying was then performed using a critical point device with CO 2 (Silva et al., 2008) and the coverslips were glued on SEM stubs and coated with gold particles by sputtering (Pimenta & De Souza, 1986). Finally, the samples were observed in a JEOL JSM-5600 scanning electron microscope. Mice, and treatment and infection procedures. Germ-free day-old NIH mice (Taconic) were used in this study. The animals were housed in flexible plastic isolators (Standard Safety Equipment Company) and handled according to established procedures (Pleasants, 1974). Experiments with gnotobiotic mice were carried out in micro-isolators (UNO Roestvaststaal). Water and a commercial autoclavable diet (Nuvital) were sterilized by steam and administered ad libitum, and the animals were maintained in an open animal house with controlled lighting (12 h light/12 h dark). All experimental procedures were carried out according to the standards set out by COBEA (2006). The study was approved by the Ethics Committee in Animal Experimentation of the Federal University of Minas Gerais (CETEA/UFMG, protocol no. 197/ 2007). For experiments of probiotic pathogen adhesion, germ-free mice received a single dose of 0.1 ml containing 9.0 log (c.f.u. ml 21 )of the yeast by oral gavage, 24 h before infection. The monoxenic animals were inoculated intragastrically with 0.1 ml Salmonella Typhimurium suspension containing 7.0 log (c.f.u. ml 21 ). Control mice received only sterile water by oral gavage 24 h before the pathogenic challenge. After 2 h, the mice were sacrificed by cervical dislocation and the small intestine tissueswerefixedand processed for SEM. Intestinal tissues were fixed overnight at room temperature with 2.5 % glutaraldehyde in 0.1 M cacodylate buffer (ph6.8)containing7%sucrose.theywerethentreatedwith2% osmium tetroxide solution plus 1.6 % potassium ferrocyanide in the same buffer for 3 h. After dehydration with increasing acetone concentrations ( %), the tissues were dried using a critical point device with CO 2 (Silva et al., 2008). The tissue samples were then mounted on SEM stubs, coated with gold particles by sputtering and analysed in a JEOL JSM 5600 scanning electron microscope. Statistical analysis. The results were expressed as the mean of at least two independent experiments. Data were analysed using Fisher s exact test or Student s t-test at a probability level of P,0.05. Statistical analyses were performed using the program Sigma Stat version 1.0 (Jandel Scientific Software). RESULTS Demonstration of adhesion between yeasts and pathogenic bacteria The adhesion of a panel of pathogenic bacteria to the surface of various yeasts was determined as aggregation by optical and electron microscopic examination and by sedimentation to determine bacterial counts in supernatants of

5 F. C. P. Tiago and others Table 2. Adhesion of enteropathogenic bacteria to live or dead yeast cells determined by macroscopic and microscopic visualization and by bacterial count in the supernatant after sedimentation +, Adhesion observed in 15 min; ++, adhesion observed after 1 h; 2, absence of adhesion after 3 h. All bacterial counts in the supernatants of yeast bacteria associations were significantly different from those of the counterpart control without yeast (Student s t-test, P,0.05). Indicator strain Adhesion/bacterial count [log (c.f.u. ml 1 )] Saccharomyces cerevisiae UFMG 905 (dead) Saccharomyces boulardii (dead) Saccharomyces cerevisiae W303 (live) Saccharomyces cerevisiae UFMG 905 (live) Control Saccharomyces boulardii (live) Salmonella Typhimurium ATCC / / / /5.68 Salmonella Typhimurium (human origin) / / / /5.71 Escherichia coli ATCC / / / /5.54 Salmonella Typhi ATCC /6.72 2/ /5.61 +/5.43 Shigella sonnei ATCC Enterococcus faecalis ATCC Listeria monocytogenes ATCC Bacillus cereus ATCC Vibrio cholerae (human origin) Clostridium difficile ATCC Clostridium perfringens ATCC suspensions containing the bacteria alone or associated with the yeasts. Table 2 shows that both live and heatkilled yeast cells of Saccharomyces boulardii and Saccharomyces cerevisiae UFMG 905 were able to fix Salmonella Typhimurium (from human origin, as well as ATCC 14028) and E. coli ATCC onto their surface. Flocculation between the yeasts and Salmonella Typhi or E. coli was also observed (except for live cells of Saccharomyces cerevisiae UFMG 905 and Salmonella Typhi), but the phenomenon was slower when compared with Salmonella Typhimurium. Flocculation was not observed when Saccharomyces cerevisiae W303 was used. Counts of pathogenic bacteria in the supernatant of the adhesion/sedimentation assays confirmed the visual results.atenfoldreductioninbacterialnumberinthe supernatant was observed when the yeasts were added, and this bacterial population reduction was statistically significant in all situations (P,0.05). The adhesion phenomenon was confirmed as being independent of yeast viability. Table 3 shows that adhesion, as determined by sedimentation assays, was not observed between the yeasts and some components of the indigenous digestive microbiota (some of which are used as probiotics), except for Saccharomyces cerevisiae UFMG 905 and Bacteroides fragilis. Fig. 1 shows a representative microscopic visualization of the various positive assays observed for adhesion. Each yeast cell was surrounded by numerous bacteria in a phenomenon similar to flocculation, which also occurred between the yeast cells themselves. Fig. 2 shows scanning electron micrographs from in vitro adhesion assays with different yeasts and pathogenic bacteria. Adhesion was clearly observed when Salmonella Typhimurium ATCC (Fig. 2a, c) or E. coli ATCC (Fig. 2b) was associated with Saccharomyces boulardii (Fig. 2a) or Saccharomyces cerevisiae UFMG 905 (Fig. 2b, c). In contrast, absence of adhesion was clearly noted when Salmonella Typhimurium ATCC was associated with Saccharomyces cerevisiae W303 (Fig. 2d). Flocculation between yeast cells was again observed, and this phenomenon was noted more frequently for yeast strains showing adhesion with bacteria. Additionally, the bacteria, which were uniformly distributed in the observation field when there was no adhesion (Fig. 2d), seemed to be attracted to the yeast surface in the presence of Saccharomyces boulardii or Saccharomyces cerevisiae UFMG 905 (Fig. 2a c), suggesting a possible chemotaxis phenomenon. Influence of various other factors on adhesion To characterize some aspects of the adhesion between bacteria and yeasts, the influence of ph, the growth phase of the yeast and the presence of various carbohydrates and bile salts was evaluated. Adhesion assays were performed over a ph range of , and no influence of this factor was observed (data not shown). Table 4 shows the influence of yeast growth phase on the adhesion 1198 Journal of Medical Microbiology 61

6 Bacterial trapping on probiotic yeast surface Table 3. Adhesion of some bacteria of the indigenous human faecal microbiota to live yeast cells as determined by bacterial count in the supernatant after sedimentation Strain YGL032C is a BY4741 mutant positive for adhesion to pathogens. Association Bacteria count [log (c.f.u. ml 1 )] Escherichia coli Nissle ±0.40 +Saccharomyces boulardii 6.74±0.54 +Saccharomyces cerevisiae UFMG ±0.62 +Saccharomyces cerevisiae BY ±0.43 +Mutant YGL032C 7.11±0.45 +Mutant YMR306W 6.77±0.45 Bacteroides fragilis ATCC ±0.60 +Saccharomyces boulardii 7.19±0.08 +Saccharomyces cerevisiae UFMG ±0.22* +Saccharomyces cerevisiae BY ±0.37 +Mutant YGL032C 7.35±0.20 +Mutant YMR306W 7.11±0.36 Bifidobacterium longum A5 6.79±0.77 +Saccharomyces boulardii 6.77±0.72 +Saccharomyces cerevisiae UFMG ±0.97 +Saccharomyces cerevisiae BY ±0.19 +Mutant YGL032C 6.60±1.21 +Mutant YMR306W 7.13±0.68 Lactobacillus delbrueckii H2B ±0.49 +Saccharomyces boulardii 7.08±0.57 +Saccharomyces cerevisiae UFMG ±0.21 +Saccharomyces cerevisiae BY ±0.68 +Mutant YGL032C 6.70±0.14 +Mutant YMR306W 6.86±0.27 *Significant difference from the counterpart control without yeast (Student s t-test, P,0.05). to pathogenic bacteria. Greater adhesion was generally observed when the yeasts were harvested in the stationary phase for the test compared with harvesting in the exponential phase, except when Salmonella Typhi was used as the pathogenic bacterium, when the opposite result was noted. The adhesion phenomenon was clearly inhibited by mannose in all situations, as shown in Table 5. Similar inhibition was also noted for glucose and maltose but only for Saccharomyces boulardii with Salmonella Typhimurium from human origin and Salmonella Typhi. Table 6 shows that there was a marked inhibition of adhesion ability when Saccharomyces cerevisiae UFMG 905 was grown in medium containing 0.1 % bile salts before the test. The effect was not so marked when the bile salts were added after growth. However, at a higher concentration (0.3 %), bile salts inhibited the adhesion phenomenon for all associations. Fig. 1. Optical microphotography of in vitro adhesion between Salmonella Typhimurium ATCC and Saccharomyces cerevisiae UFMG 905. The arrows show Gram-stained yeast cells surrounded by bacteria. Magnification, ¾1000. Table 7 shows that Saccharomyces cerevisiae B4741 presented the same adhesion capacity as Saccharomyces boulardii and Saccharomyces cerevisiae UFMG 905 when tested with the four pathogenic bacteria. Among its 44 cell-wall mutants, five had lost this adhesion capacity, YJL158C,

7 F. C. P. Tiago and others (a) (b) (c) (d) Fig. 2. Scanning electron micrograph of in vitro adhesion between Saccharomyces boulardii and Salmonella Typhimurium ATCC (a), Saccharomyces cerevisiae UFMG 905 and Escherichia coli ATCC (b), Saccharomyces cerevisiae UFMG 905 and Salmonella Typhimurium ATCC (c) and Saccharomyces cerevisiae W303 and Salmonella Typhimurium ATCC (d). The magnification is indicated in each panel. YKL096W-A, YMR306W, YKL163W and YGR279C, as demonstrated by sedimentation assays (Table 8). In vivo evaluation of adhesion using gnotobiotic mice Demonstration and characterization of the adhesion between bacteria and yeasts as described above were determined by in vitro assays. However, it is important to demonstrate that the phenomenon can also occur in vivo. The probable contribution of this mechanism would be difficult to demonstrate in the presence of a complex intestinal microbial ecosystem. For this reason, gnotobiotic mice were used, which provide a simplified in vivo system that allows the observation of ecological interactions in the digestive tract between a few microbial strains inoculated in this ecosystem. As can be seen in Fig. 3, in Salmonella monoxenic mice, the bacteria covered the epithelium in an evenly distributed way (Fig. 3a). When the mice were previously mono-associated with the yeast, binding between the yeast and bacteria was clearly observed for both Saccharomyces boulardii and Saccharomyces cerevisiae UFMG 905 (Fig. 3b, c). As observed in Fig. 2(d), in the presence of the probiotic Saccharomyces, the bacterial cells that were initially uniformly distributed on the intestinal epithelial surfaces seemed to be attracted to the yeast cell surface. DISCUSSION Adhesion to host tissue surfaces is an initial step of fundamental importance for the success of an infectious process, particularly on the gut epithelium. Adherence of microbial cells has been suggested to be the result of two essentially different mechanisms: specific and nonspecific binding. Non-specific binding involves electrostatic and hydrophobic interactions of lower affinity than for specific binding. The localization and mechanisms of binding to the epithelial surface have been studied previously in enteropathogenic bacteria such as E. coli and Salmonella, and type I fimbriae are frequently involved in this adhesion, generally using mannose as a 1200 Journal of Medical Microbiology 61

8 Bacterial trapping on probiotic yeast surface Table 4. Influence of growth phase on adhesion of enteropathogenic bacteria to live yeast cells, determined by bacterial counts in the supernatant after sedimentation SP, Stationary phase; EP, exponential phase. Association Bacterial count [log (c.f.u. ml 1 )] Salmonella Typhimurium ATCC ±0.15 +Saccharomyces boulardii (SP) 5.62±0.21* +Saccharomyces boulardii (EP) 6.80±0.17 Salmonella Typhimurium ATCC ±0.15 +Saccharomyces cerevisiae UFMG 905 (SP) 5.78±0.15* +Saccharomyces cerevisiae UFMG 905 (EP) 6.87±0.26 Salmonella Typhimurium (human origin) 6.43±0.26 +Saccharomyces boulardii (SP) 5.77±0.10* +Saccharomyces boulardii (EP) 6.38±0.82 Salmonella Typhimurium (human origin) 6.43±0.26 +Saccharomyces cerevisiae UFMG 905 (SP) 5.87±0.08* +Saccharomyces cerevisiae UFMG 905 (EP) 6.39±0.09 Salmonella Typhi ATCC ±0.26 +Saccharomyces boulardii (SP) 6.72±0.21 +Saccharomyces boulardii (EP) 5.54±0.91* Salmonella Typhi ATCC ±0.26 +Saccharomyces cerevisiae UFMG 905 (SP) 6.51±0.38 +Saccharomyces cerevisiae UFMG 905 (EP) 5.77±0.31* Escherichia coli ATCC ±0.14 +Saccharomyces boulardii (SP) 5.45±0.33* +Saccharomyces boulardii (EP) 5.79±0.39 Escherichia coli ATCC ±0.14 +Saccharomyces cerevisiae UFMG 905 (SP) 5.83±0.13* +Saccharomyces cerevisiae UFMG 905 (EP) 6.26±0.14 *Significant difference from the counterpart control without yeast (Student s t-test, P,0.05). receptor (Kline et al., 2009). A number of different approaches have been taken over the years to inhibit adhesion of pathogenic bacteria, and addition of exogenous sugars and inhibition of fimbria assembly are two strategies that have been studied (Cusumano & Hultgren, 2009). The cell walls of yeasts are known to be rich in mannose, and we have hypothesized that probiotic yeasts may exert their protective effect by the trapping of pathogenic bacteria on their surface, as we observed previously in a cell-culture model for Saccharomyces boulardii (Martins et al., 2010). This is a reasonable hypothesis, as this could explain many of the probiotic effects observed during infection such as the inhibition of signalling transduction pathways activation and translocation (Martins et al., 2010, 2011). The present study showed that Saccharomyces boulardii and Saccharomyces cerevisiae UFMG 905 aggregated a limited range of Gram-negative enteropathogenic bacteria. The observation of agglutination between yeast and Gramnegative bacteria (E. coli and Salmonella) is not a new observation and was used initially as a model to simulate and study the adhesion of pathogenic bacteria to host mammalian cells (Korhonen, 1979; Korhonen et al., 1981). Later, a study conducted by Gedek (1999) reported the adhesion of Gram-negative enteropathogenic bacteria to Saccharomyces boulardii. The location of the binding sites of these enteric bacteria on the cell wall of Saccharomyces was studied using a lectin histochemical method and electron microscopy. The study showed that type I fimbriae of E. coli were more strongly linked to the surface of Saccharomyces boulardii than to other strains of Saccharomyces cerevisiae. Pérez-Sotelo et al. (2005) also noted that the yeast Saccharomyces cerevisiae Sc47 bound to strains of Salmonella, particularly those expressing type I fimbriae. In this case, the affinity was mediated by cell-wall mannose residues, and the binding capacity was influenced by the growth conditions of the Salmonella, such as temperature and ph. Interestingly, although Saccharomyces boulardii has been demonstrated to be able to fix cholera toxin onto its surface (Brandão et al., 1998), its adhesion to V. cholerae toxin-producing cells was not observed here. Previous

9 F. C. P. Tiago and others Table 5. Influence of carbohydrates (5 %, w/v) on adhesion of enteropathogenic bacteria to live or dead yeast cells, determined by macroscopic and microscopic visualization +, Adhesion observed in 15 min; ++, adhesion observed after 1 h; 2, absence of adhesion up to 3 h. Association Carbohydrate Galactose Glucose Maltose Mannose Salmonella Typhimurium ATCC Saccharomyces boulardii (live) Saccharomyces boulardii (dead) Saccharomyces cerevisiae UFMG 905 (live) Saccharomyces cerevisiae UFMG 905 (dead) Salmonella Typhimurium (human origin) +Saccharomyces boulardii (live) Saccharomyces boulardii (dead) Saccharomyces cerevisiae UFMG 905 (live) Saccharomyces cerevisiae UFMG 905 (dead) Salmonella Typhi ATCC Saccharomyces boulardii (live) Saccharomyces boulardii (dead) Saccharomyces cerevisiae UFMG 905 (live) Saccharomyces cerevisiae UFMG 905 (dead) Escherichia coli ATCC Saccharomyces boulardii (live) Saccharomyces boulardii (dead) Saccharomyces cerevisiae UFMG 905 (live) Saccharomyces cerevisiae UFMG 905 (dead) reports have shown that Salmonella strains that express type I fimbriae have some affinity to Saccharomyces boulardii. This affinity is mediated by mannan oligosaccharides of the yeast cell wall and can be inhibited by mannose complexes (Gedek, 1999). The aggregation between yeast and bacteria described in our study was also inhibited by mannose. In previous studies, transmission electron micrographs showed that the bacterial pili are in contact with yeast. Once the yeast fixed the bacteria on its surface, a diminished or inhibited signalling cascade was generated by a lower number of bacteria in direct contact with the host cell (Martins et al., 2010, 2011). Table 6. Influence of bile salts (resuspended or grown with 0.1 or 0.3 % Oxgall) on adhesion of enteropathogenic bacteria to live yeast cells, determined by bacterial counts [log (c.f.u. ml 1 )] in the supernatant after sedimentation Association 0.1 % Oxgall 0.3 % Oxgall Resuspended Grown Resuspended Grown Salmonella Typhimurium ATCC ± ± ± ±0.46 +Saccharomyces boulardii 6.20±0.11* 6.74±0.23* 6.49± ±0.24 +Saccharomyces cerevisiae UFMG ±0.60* 6.84± ± ±0.48 Salmonella Typhimurium (human origin) 7.03± ± ± ±0.64 +Saccharomyces boulardii 7.17± ±0.22* 6.40± ±0.57 +Saccharomyces cerevisiae UFMG ±0.55* 6.49± ±0.20* 5.91±0.12 Salmonella Typhi ATCC ± ± ± ±0.45 +Saccharomyces boulardii 6.55±0.36* 5.89± ±0.33* 5.31±0.20 +Saccharomyces cerevisiae UFMG ±0.32* 5.94± ±0.29* 5.40±0.34 Escherichia coli ATCC ± ± ± ±0.55 +Saccharomyces boulardii 6.79± ±0.26* 5.87±0.36* 5.88±0.50 +Saccharomyces cerevisiae UFMG ±0.19* 6.43± ±0.42* 6.04±0.21 *Significant difference from the counterpart control without yeast (Student s t-test, P,0.05) Journal of Medical Microbiology 61

10 Bacterial trapping on probiotic yeast surface Table 7. Adhesion of enteropathogenic bacteria to Saccharomyces cerevisiae BY4741 and its cell-wall mutants, determined by macroscopic and microscopic visualization +, Adhesion observed in 15 min; ++, adhesion observed after 1 h; 2, absence of adhesion up to 3 h. Mutants that had lost their ability to adhere are shown in bold. Access no. Systematic name Escherichia coli ATCC Salmonella Typhimurium ATCC Salmonella Typhimurium human origin Salmonella Typhi ATCC BY4741 Parental Y04400 YGL032C Y05934 YGR282C Y07107 YLR110C Y01837 YOR061W Y01267 YJL158C Y04819 YGR189C Y04162 YLR213C Y04395 YGL027C Y07026 YKL096W-A Y04895 YKL046C Y00824 YMR238W Y02836 YHR143W Y05210 YLR300W Y03620 YDR261C Y07200 YCR089W Y04368 YDR534C Y01679 YOR382W Y06450 YMR306W Y06870 YAR050W Y06107 YER109C Y07106 YKR102W Y06979 YGR032W Y04256 YDR420W Y03746 YDL049C Y05574 YPR159W Y02355 YOR099W Y01361 YJL062W Y04624 YGL257C Y05013 YKL163W Y03991 YDR055W Y03369 YBR229C Y05931 YGR279C Y00894 YMR305C Y04396 YGL028C Y04595 YGL228W Y04773 YGR143W Y05473 YPR054W Y06147 YER150W Y02446 YOR190W Y02503 YOR247W Y07214 YNL066W Y00281 YEL040W Y02731 YLR120C Y02732 YLR121C

11 F. C. P. Tiago and others Table 8. Adhesion of enteropathogenic bacteria to the five cell-wall mutants of Saccharomyces cerevisiae BY4741, determined by bacterial count [log (c.f.u. ml 1 )] in the supernatant after sedimentation None of the bacterial counts in the supernatants of the yeast bacteria associations were significantly different from those of the counterpart control without yeast (Student s t-test, P.0.05). Association Bacterial count Salmonella Typhimurium ATCC ±0.11 +Mutant YKL096W 7.92±0.42 +Mutant YJL158C 8.32±0.64 +Mutant YKL163W 6.99±0.27 +Mutant YGR279C 6.91±0.51 +Mutant YMR306W 7.40±0.45 Salmonella Typhimurium (human origin) 7.36±0.33 +Mutant YKL096W 7.09±1.06 +Mutant YJL158C 7.15±1.19 +Mutant YKL163W 6.88±0.19 +Mutant YGR279C 7.53±0.17 +Mutant YMR306W 7.34±0.19 Salmonella Typhi ATCC ±0.85 +Mutant YKL096W 7.73±0.65 +Mutant YJL158C 8.20±0.20 +Mutant YKL163W 8.41±0.36 +Mutant YGR279C 7.89±0.06 +Mutant YMR306W 7.60±0.16 Escherichia coli ATCC ±0.12 +Mutant YKL096W 7.28±0.79 +Mutant YJL158C 6.56±0.74 +Mutant YKL163W 7.04±0.57 +Mutant YGR279C 7.31±0.39 +Mutant YMR306W 7.48±0.27 Guglielmetti et al. (2009) found that bile salts reduced the adhesion of Bifidobacterium bifidum strain MIMBb75 to intestinal epithelial cells, particularly when the bacterial cells were grown in medium containing 0.3 % Oxgall. These results agree with those of Gómez-Zavaglia et al. (2002), who showed that growth of two Bifidobacterium bifidum strains in bile altered their adhesion to Caco-2 cells more markedly than when the same bacteria were shocked with bile after having been grown in a bile-free medium. The same authors correlated the changes in adhesion with a decrease in surface hydrophobicity and an increase in surface potential. Moreover, they proposed that changes in surface and adhesion properties may correlate with changes in sugar components induced by bile during growth (Gómez-Zavaglia et al., 2002). With regard to the influence of bile salts on the adhesion ability of yeast, no report has been found in the literature. In relation to the influence of growth phase on the adhesion phenomenon, it is well known that yeasts frequently flocculate in the stationary phase, and this characteristic has a high practical importance in the brewing industry (Stratford, 1993). Flocculation of yeast cells involves lectin-like proteins so-called flocculins that stick out of the cell walls of flocculent cells and selectively bind mannose residues present in the cell walls of adjacent yeast cells. Flocculation is inhibited by mannose in the growth medium, presumably because free mannose occupies the flocculin-binding sites so that they can no longer bind the mannose residues of other cells. However, the results of the present study did not allow us to determine whether, in addition to mannose, flocculin might also be involved. Another interesting aspect of the present results was that yeast viability was not necessary for the adhesion phenomenon. This property could open the possibility of using nonviable yeast cells in immunocompromised patients, in which treatment with live probiotics is not recommended due to the possible occurrence of opportunistic infection (although rarely observed) (Boyle et al., 2006). Although the adhesion phenomenon has been clearly demonstrated and visualized in vitro in the present study as well as in previous preliminary reports (Gedek, 1999; Pérez-Sotelo et al., 2005; Martins et al., 2010, 2011), it is not known whether it occurs in vivo. Forthispurpose,an animal model using germ-free mice was used to confirm and visualize the interaction between the yeast and the pathogenic bacterium on the intestinal epithelium. The gnotobiotic animal provides a simplified model that allows the observation of interrelationships between a few microbial strains inoculated into the gastrointestinal ecosystem without the interference of the complex indigenous microbiota. The data obtained here clearly showed adhesion between the enteropathogenic bacteria and the yeast on the intestinal epithelium. The results of both in vitro and in vivo visualization of adhesion also suggested that a chemotaxis phenomenon could be involved in these interactions between the bacteria and yeasts. The use of cell-wall deletion mutants of Saccharomyces cerevisiae BY4741 showed that five of the 44 mutants tested lost their ability to adhere to pathogenic bacteria. The mutant strains were YJL158C, YKL096W-A, YMR306W, YKL163W and YGR279C. Mutation of YJL158C (gen CIS3) is related to a cell-wall mannose-containing glycoprotein, which is a member of the PIR family. YKL096W- A (CWP2) has a mutation of a covalently linked cell-wall mannoprotein, a major constituent of the cell wall, which plays a role in cell-wall stabilization and acid resistance. For YMR306W (gen FKS3), the mutation involves a protein acting in spore wall assembly, which has similarity to the 1,3-b-D-glucan synthase catalytic subunits FKS1p and Gsc2p. Mutation of YKL163W (gen PIR3) is related to a glycosylated covalently bound cell-wall protein required for cell-wall stability. Its expression is cell-cycle regulated, peaking in M/G 1, and is also subjected to regulation by the cell-wall integrity pathway. Finally, for YGR279C (gen SCW4), the mutation involves a cell-wall 1204 Journal of Medical Microbiology 61

12 Bacterial trapping on probiotic yeast surface (a) (b) their effect on adhesion. However, various other mutant strains involving mannose did not lose their adhesion ability. For the last three strains, the influence of the mutation on the loss of bacterial adhesion is less clear. In conclusion, the present study showed that some enteropathogenic bacteria were captured on the surface of Saccharomyces boulardii, Saccharomyces cerevisiae UFMG 905 and Saccharomyces cerevisiae BY4741, and this could be one mechanism by which probiotic yeasts prevent pathogen adhesion to specific receptors on the intestinal epithelium and subsequent invasion of the host. However, this trapping mechanism was limited to some specific Gram-negative enteropathogens (Salmonella and E. coli), and the search for a yeast probiotic presenting a wider range of adhesion ability could be of interest. Another approach to improve the use of this possible probiotic property involves a better genetic and mechanistic understanding of the adhesion phenomenon, which could lead to its superexpression. ACKNOWLEDGEMENTS The authors are grateful to Clelia Nunes da Silva for valuable technical help and to Antônio M. Vaz for the animal care. This work was supported by grants from Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG). F. S. M. is the recipient of a postdoctoral fellowship from CNPq. (c) Fig. 3. Scanning electron micrograph of the intestinal epithelium of gnotobiotic mice associated with Salmonella Typhimurium ATCC only (a), Saccharomyces cerevisiae UFMG 905 plus Salmonella Typhimurium ATCC (b) and Saccharomyces boulardii plus Salmonella Typhimurium ATCC (c). The magnification is indicated in each panel. protein with similarity to glucanases. As described above, only the first two strains had a more direct participation of mannose in their mutations, which could explain REFERENCES Boyle, R. J., Robins-Browne, R. M. & Tang, M. L. K. (2006). Probiotic use in clinical practice: what are the risks? Am J Clin Nutr 83, Brandão, R. L., Castro, I. M., Bambirra, E. A., Amaral, S. C., Fietto, L. G., Tropia, M. J. M., Neves, M. J., Dos Santos, R. G., Gomes, N. C. M. & Nicoli, J. R. (1998). Intracellular signal triggered by cholera toxin in Saccharomyces boulardii and Saccharomyces cerevisiae. Appl Environ Microbiol 64, Buts, J. P., Bernasconi, P., Vaerman, J. P. & Dive, C. (1990). Stimulation of secretory IgA and secretory component of immunoglobulins in small intestine of rats treated with Saccharomyces boulardii. Dig Dis Sci 35, Castagliuolo, I., Riegler, M. F., Valenick, L., LaMont, J. T. & Pothoulakis, C. (1999). Saccharomyces boulardii protease inhibits the effects of Clostridium difficile toxins A and B in human colonic mucosa. Infect Immun 67, Ciriello, C. J. & Lachance, M. A. (2001). YEASTCOMPARE. London, ON, Canada: University of Western Ontario. COBEA (2006). Colegio Brasileiro de Experimentação Animal. LegislaçãoeÉtica. Cusumano, C. K. & Hultgren, S. J. (2009). Bacterial adhesion a source of alternate antibiotic targets. IDrugs 12, Czerucka, D., Roux, I. & Rampal, P. (1994). Saccharomyces boulardii inhibits secretagogue-mediated adenosine 39,59-cyclic monophosphate induction in intestinal cells. Gastroenterology 106,

13 F. C. P. Tiago and others Czerucka, D., Dahan, S., Mograbi, B., Rossi, B. & Rampal, P. (2000). Saccharomyces boulardii preserves the barrier function and modulates the signal transduction pathway induced in enteropathogenic Escherichia coli-infected T84 cells. Infect Immun 68, Dahan, S., Dalmasso, G., Imbert, V., Peyron, J.-F., Rampal, P. & Czerucka, D. (2003). Saccharomyces boulardii interferes with enterohemorrhagic Escherichia coli-induced signaling pathways in T84 cells. Infect Immun 71, FAO/WHO (2002). Working Group Report on Drafting Guidelines for the Evaluation of Probiotics in Food. London, ON, Canada: Food and Agriculture Organization/World Health Organization. Gedek, B. R. (1999). Adherence of Escherichia coli serogroup O 157 and the Salmonella Typhimurium mutant DT 104 to the surface of Saccharomyces boulardii. Mycoses 42, Generoso, S. V., Viana, M., Santos, R., Martins, F. S., Machado, J. A., Arantes, R. M., Nicoli, J. R., Correia, M. I. & Cardoso, V. N. (2010). Saccharomyces cerevisiae strain UFMG 905 protects against bacterial translocation, preserves gut barrier integrity and stimulates the immune system in a murine intestinal obstruction model. Arch Microbiol 192, Gómez-Zavaglia, A., Kociubinski, G., Pérez, P., Disalvo, E. & De Antoni, G. (2002). Effect of bile on the lipid composition and surface properties of bifidobacteria. J Appl Microbiol 93, Guglielmetti, S., Tamagnini, I., Minuzzo, M., Arioli, S., Parini, C., Comelli, E. & Mora, D. (2009). Study of the adhesion of Bifidobacterium bifidum MIMBb75 to human intestinal cell lines. Curr Microbiol 59, Jain, S., Yadav, H., Sinha, P. R., Naito, Y. & Marotta, F. (2008). Dahi containing probiotic Lactobacillus acidophilus and Lactobacillus casei has a protective effect against Salmonella enteritidis infectioninmice.int J Immunopathol Pharmacol 21, Jain, S., Yadav, H. & Sinha, P. R. (2009). Probiotic dahi containing Lactobacillus casei protects against Salmonella enteritidis infection and modulates immune response in mice. JMedFood12, Kline, K. A., Fälker, S., Dahlberg, S., Normark, S. & Henriques- Normark, B. (2009). Bacterial adhesins in host microbe interactions. Cell Host Microbe 5, Korhonen, T. K. (1979). Yeast agglutination of purified enterobacterial pili. FEMS Microbiol 6, Korhonen, T. K., Leffler, H. & Svanborg Edén, C. (1981). Binding specificity of piliated strains of Escherichia coli and Salmonella Typhimurium to epithelial cells, Saccharomyces cerevisiae cells, and erythrocytes. Infect Immun 32, Kovacs, D. J. & Berk, T. (2000). Recurrent Clostridium difficileassociated diarrhea and colitis treated with Saccharomyces cerevisiae (Baker s yeast) in combination with antibiotic therapy: a case report. J Am Board Fam Pract 13, Kumura, H., Tanoue, Y., Tsukahara, M., Tanaka, T. & Shimazaki, K. (2004). Screening of dairy yeast strains for probiotic applications. J Dairy Sci 87, Lachance, M.-A., Bowles, J. M., Starmer, W. T. & Barker, J. S. F. (1999). Kodamaea kakaduensis and Candida tolerans, two new ascomycetous yeast species from Australian hibiscus flowers. Can J Microbiol 45, Martins, F. S., Nardi, R. M. D., Arantes, R. M. E., Rosa, C. A., Neves, M. J. & Nicoli, J. R. (2005). Screening of yeasts as probiotic based on capacities to colonize the gastrointestinal tract and to protect against enteropathogen challenge in mice. J Gen Appl Microbiol 51, Martins, F. S., Rodrigues, A. C. P., Tiago, F. C. P., Penna, F. J., Rosa, C. A., Arantes, R. M. E., Nardi, R. M. D., Neves, M. J. & Nicoli, J. R. (2007). Saccharomyces cerevisiae strain 905 reduces the translocation of Salmonella enterica serotype Typhimurium and stimulates the immune system in gnotobiotic and conventional mice. J Med Microbiol 56, Martins, F. S., Veloso, L. C., Arantes, R. M. & Nicoli, J. R. (2009). Effects of yeast probiotic formulation on viability, revival and protection against infection with Salmonella enterica ssp. enterica serovar Typhimurium in mice. Lett Appl Microbiol 49, Martins, F. S., Dalmasso, G., Arantes, R. M. E., Doye, A., Lemichez, E., Lagadec, P., Imbert, V., Peyron, J. F., Rampal, P. & other authors (2010). Interaction of Saccharomyces boulardii with Salmonella enterica serovar Typhimurium protects mice and modifies T84 cell response to the infection. PLoS ONE 5, e8925. Martins, F. S., Elian, S. D. A., Vieira, A. T., Tiago, F. C. P., Martins, A. K. S., Silva, F. C. P., Souza, E. L. S., Sousa, L. P., Araújo, H. R. C. & other authors (2011). Oral treatment with Saccharomyces cerevisiae strain UFMG 905 modulates immune responses and interferes with signal pathways involved in the activation of inflammation in a murine model of typhoid fever. Int J Med Microbiol 301, Pérez-Sotelo, L. S., Talavera-Rojas, M., Monroy-Salazar, H. G., Lagunas-Bernabé, S., Cuarón-Ibargüengoytia, J. A., Jimenez, R. M. & Vázquez-Chagoyán, J. C. (2005). In vitro evaluation of the binding capacity of Saccharomyces cerevisiae Sc47 to adhere to the wall of Salmonella spp. Rev Latinoam Microbiol 47, Pimenta, P. F. & De Souza, W. (1986). Cytochemical localization of acid phosphatase in Leishmania mexicana amazonensis. J Submicrosc Cytol 18, Pleasants, J. R. (1974). Gnotobiotics. In Handbook of Laboratory Animal Science, pp Edited by E. C. Melby, Jr & N. H. Altmann. Cleveland, OH: CRC Press. Qamar, A., Aboudola, S., Warny, M., Michetti, P., Pothoulakis, C., LaMont, J. T. & Kelly, C. P. (2001). Saccharomyces boulardii stimulates intestinal immunoglobulin A immune response to Clostridium difficile toxin A in mice. Infect Immun 69, Rodrigues, A. C. P., Nardi, R. M., Bambirra, E. A., Vieira, E. C. & Nicoli, J. R. (1996). Effect of Saccharomyces boulardii against experimental oral infection with Salmonella typhimurium and Shigella flexneri in conventional and gnotobiotic mice. J Appl Bacteriol 81, Rodrigues, A. C., Cara, D. C., Fretez, S. H. G. G., Cunha, F. Q., Vieira, E. C., Nicoli, J. R. & Vieira, L. Q. (2000). Saccharomyces boulardii stimulates siga production and the phagocytic system of gnotobiotic mice. J Appl Microbiol 89, Silva, L. M., Carvalho Botelho, A. C., Nacif-Pimenta, R., Martins, G. F., Alves, L. C., Brayner, F. A., Fortes-Dias, C. L. & Pimenta, P. F. (2008). Structural analysis of the venom glands of the armed spider Phoneutria nigriventer (Keyserling, 1891): microanatomy, fine structure and confocal observations. Toxicon 51, Stratford, M. (1993). Yeast flocculation: flocculation onset and receptor availability. Yeast 9, Truusalu, K., Mikelsaar, R. H., Naaber, P., Karki, T., Kullisaar, T., Zilmer, M. & Mikelsaar, M. (2008). Eradication of Salmonella Typhimurium infection in a murine model of typhoid fever with the combination of probiotic Lactobacillus fermentum ME-3 and ofloxacin. BMC Microbiol 8, Journal of Medical Microbiology 61