Interaction of Saccharomyces boulardii with Salmonella
enterica Serovar Typhimurium Protects Mice and
Modifies T84 Cell Response to the Infection
Flaviano S. Martins1,3,8,10¤a, Guillaume Dalmasso1,3¤b, Rosa M. E. Arantes9, Anne Doye2,3, Emmanuel
Lemichez2,3,6, Patricia Lagadec1,3, Veronique Imbert1,3, Jean-François Peyron1,3,4,5, Patrick Rampal7,
Jacques R. Nicoli8, Dorota Czerucka1,3*
1 Team 4: Inflammation, Cancer, Cancer Stem Cells, Unité INSERM U895, C3M: Centre Méditerranéen de Médecine Moléculaire, Nice, France, 2 Team 6: Toxines
microbiennes dans la relation hôte-pathogènes, Unité INSERM U895, C3M: Centre Méditerranéen de Médecine Moléculaire, Nice, France, 3 Université de Nice-Sophia
Antipolis, UFR Médecine, IFR50, Faculté de Médecine, Nice, France, 4 Service de Pédiatrie, Centre Hospitalier Universitaire de Nice, Hôpital de l’Archet, Nice, France, Centre
Hospitalier Universitaire de Nice, Hôpital de l’Archet, Nice, France, 5 Service d’Hématologie Clinique, Centre Hospitalier Universitaire de Nice, Hôpital de l’Archet, Nice,
France, 6 Service de Bactériologie, Centre Hospitalier Universitaire de Nice, Hôpital de l’Archet, Nice, France, 7 Centre Hospitalier Princesse Grace, Service d’Hépato-GastroEntérologie, Monaco, Monaco, 8 Departamento de Microbiologia, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais,
Brazil, 9 Departamento de Patologia Geral, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil, 10 Departamento de
Pediatria, Faculdade de Medicina, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil
Abstract
Background: Salmonella pathogenesis engages host cells in two-way biochemical interactions: phagocytosis of bacteria by
recruitment of cellular small GTP-binding proteins induced by the bacteria, and by triggering a pro-inflammatory response
through activation of MAPKs and nuclear translocation of NF-kB. Worldwide interest in the use of functional foods
containing probiotic bacteria for health promotion and disease prevention has increased significantly. Saccharomyces
boulardii is a non-pathogenic yeast used as a probiotic in infectious diarrhea.
Methodology/Principal Findings: In this study, we reported that S. boulardii (Sb) protected mice from Salmonella enterica
serovar Typhimurium (ST)-induced death and prevented bacterial translocation to the liver. At a molecular level, using T84
human colorectal cancer cells, we demonstrate that incubation with Sb before infection totally abolished Salmonella
invasion. This correlates with a decrease of activation of Rac1. Sb preserved T84 barrier function and decreased ST-induced
IL-8 synthesis. This anti-inflammatory effect was correlated with an inhibitory effect of Sb on ST-induced activation of the
MAPKs ERK1/2, p38 and JNK as well as on activation of NF-kB. Electron and confocal microscopy experiments showed an
adhesion of bacteria to yeast cells, which could represent one of the mechanisms by which Sb exerts its protective effects.
Conclusions: Sb shows modulating effects on permeability, inflammation, and signal transduction pathway in T84 cells
infected by ST and an in vivo protective effect against ST infection. The present results also demonstrate that Sb modifies
invasive properties of Salmonella.
Citation: Martins FS, Dalmasso G, Arantes RME, Doye A, Lemichez E, et al. (2010) Interaction of Saccharomyces boulardii with Salmonella enterica Serovar
Typhimurium Protects Mice and Modifies T84 Cell Response to the Infection. PLoS ONE 5(1): e8925. doi:10.1371/journal.pone.0008925
Editor: Stefan Bereswill, Charité-Universitätsmedizin Berlin, Germany
Received December 4, 2009; Accepted December 30, 2009; Published January 27, 2010
Copyright: ß 2010 Martins et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was supported by BIOCODEX S.A. (Gentilly, France), the Region Provence-Alpes Cote dAzur, the Conseil General des Alpes Maritimes, the
Faculte de Medecine de l’Universite de Nice-Sophia Antipolis, and the Centre Hospitalier Regional de Nice. This work was financed in part by an institutional grant
from INSERM. F. S. Martins is recipient of a doctoral fellowship from CAPES, Brazil. The funders had no role in study design, data collection and analysis, decision to
publish, or preparation of the manuscript.
Competing Interests: D. Czerucka has acted as a consultant or scientific advisor to the company interested in developing therapeutic interventions for
intestinal infections and inflammatory disorders; this company is: BIOCODEX S.A. (Gentilly, France). She has a financial competing interest including salaries and
travel grant. The content of this article was neither influenced nor constrained by this fact. P. Rampal acts as an expert for Biocodex with non financial competing
interest. The content of this article was neither influenced nor constrained by this fact. There are not relevant patents or products in development or marketed
product related to this research. It does not alter these authors’ adherence to all the PLoS ONE policies on sharing data and materials.
* E-mail: czerucka@unice.fr
¤a Current address: Division of Gastroenterology, Faculty of Medicine and Dentistry, University of Alberta. Katz Group Rexall Center. Edmonton, Alberta, Canada.
¤b Current address: Division of Digestive Diseases, Department of Medicine, Emory University School of Medicine, Atlanta, Georgia, United States of America
adheres to and then induces its own uptake into intestinal
epithelial cells through a specialized mechanism involving
injection of virulence factors into host cells by a type III protein
secretion system (TTSS). This activates host signaling pathways
involved in actin cytoskeleton rearrangements leading to bacterial
Introduction
In human, Salmonella spp. is responsible for over one billion
infections annually, with consequences ranging from self-limiting
gastroenteritis to typhoid fever. To initiate disease, Salmonella first
PLoS ONE | www.plosone.org
1
January 2010 | Volume 5 | Issue 1 | e8925
Anti-Infectious Effect of S. b
uptake [1,2]. Indeed, Salmonella-induced epithelial cell signaling
leads to reorganization of the host cytoskeleton, membrane ruffles
at the point of bacterium-cell contact leading to internalization of
bacteria [3]. Besides, the hijacking of host signaling pathways by
Salmonella triggers an inflammatory response and the release of
inflammatory mediators, such as the interleukin (IL)-8 chemokine
[4] that is responsible for recruitment and transepithelial migration
of polymorphonuclear leucocytes (PMN) [5–7], a specific clinical
feature of salmonellosis. Maximal IL-8 amounts are generated by a
combination of three different mechanisms: derepression of the
gene promoter, transcriptional activation by nuclear factor kB
(NF-kB) and c-jun-NH2 terminal kinase (JNK) pathways and
stabilization of the mRNA by the p38 mitogen-activated protein
kinase (MAPK) pathway [8,9].
In recent years, worldwide interest for the use of functional
foods containing probiotic microorganisms for health promotion
and disease prevention has increased significantly. According to
the Food and Agriculture Organization and the World Health
Organization, a probiotic is ‘‘a live microorganism which, when
administered in adequate amounts, confers a health benefit to the
host’’ [10]. Lyophilized Saccharomyces boulardii is a probiotic yeast
used worldwide for the prevention and treatment of a variety of
diarrheal diseases [11]. In the case of infectious diarrhea,
administration of S. boulardii to animals provides protection against
intestinal lesions caused by several diarrheal pathogens [12]. The
mechanisms by which S. boulardii exerts its protective effects are
diverse, including proteolytic cleavage of Clostridium difficile toxins
A and B [13,14], inhibition of cholera toxin stimulated cAMP
production [15], binding and elimination of cholera toxin [16],
and interference on bacterial-stimulated cellular signaling pathways [17–20].
Given the growing interest in the field of probiotic in medical
areas, the purpose of this study was to investigate, using an in vivo
murine model of infection, the protective effect of S. boulardii and,
to dissect in vitro on T84 human colorectal cancer cells the
molecular mechanisms mediating S. boulardii protection. Our study
revealed that S. boulardii increased survival of S. Typhimurium
infected mice and prevented bacterial translocation to the liver.
Cellular studies demonstrated that S. boulardii decreased the ability
of S. Typhimurium to invade cells and prevented the secretion of
IL-8 through inhibition of the MAPK and NF-kB signaling
pathways.
Figure 1. S. boulardii decreases S. Typhimurium induced death
of mice. Survival (%) of mice treated daily (Sb+ST) or not (ST) with S.
boulardii during 10 days before challenged with 104 S. Typhimurium. (C)
control (not-treated, not-challenged mice), (Sb) mice treated with S.
boulardii without bacterial challenge. N = 10. *Indicates statistical
difference in relation to control (not treated) group. d.p.i., days post
infection.
doi:10.1371/journal.pone.0008925.g001
evidenced by the higher cellularity, presence of inflammatory cells
and, degenerative changes of hepatocytes. In the Sb-treated group,
the pathogenic bacterial infection did not induce significant
inflammatory changes in the liver parenchyma, which preserves its
normal lobular architecture. This observation was confirmed by
measure of the bacterial load in the liver: 5.360.9 log colony
forming units (CFU) g21 of organ in the ST-group, while no
bacteria could be detected in the group previously treated with Sb
(Fig. 3E).
Results
Lyophilized S. boulardii Decreases Mortality and Prevents
Bacterial Translocation to the Liver in Mice Challenged
with S. Typhimurium
Results presented in figure 1 show that an infection of NIH mice
with S. Typhimurium (ST) killed 60% of mice by 16 days. Oral
treatment with S. boulardii (Sb) significantly increased (P,0.05)
survival from 40% (ST group) to 70% (Sb-treated + ST group).
This protective effect was next assessed at the histological level.
Mice were sacrificed and tissue samples from intestine were
prepared for histological analysis. In control mice (Fig. 2, A and B)
and in the Sb + ST group of mice (Fig. 2, E and F) we clearly
observed the structural integrity of surface colonocytes contrasting
with the loss of integrity of epithelial layer and reactive changes of
the colonocytes in ST group (Fig. 2, C and D). Using a germ-free
mice model, we have observed that this anti-inflammatory effect is
not due to a decrease of ST number by Sb or an ability of the yeast
to kill the bacteria (Supporting Figure S1).
Data presented in Figure 3 showed that ST group animals
presented evidences of bacterial translocation to liver. This is
PLoS ONE | www.plosone.org
Figure 2. S. boulardii protects colon against S. Typhimurium
lesions. Representative histopathological aspects of large intestine
sections in control (A, B); ST (C, D), and Sb + ST (E, F) groups. Notice the
integrity of surface colonocytes in B and F (large arrows) contrasting
with edema (C), the loss of integrity of epithelial layer and reactive
changes of the colonocytes (D, large arrows). Shortening of mucosa
length (C) and decrease in goblet cells (D) were observed in the ST
group, which were preserved in control (A, B) and in Sb + ST (E, F, thin
arrows). H&E, Bar = 20 mm (A, C, E) and 2 mm (B, D, F). C (control), ST (S.
Typhimurium), Sb (S. boulardii).
doi:10.1371/journal.pone.0008925.g002
2
January 2010 | Volume 5 | Issue 1 | e8925
Anti-Infectious Effect of S. b
by ST. Altogether these data show that Sb actively protects
monolayer barrier function during infection by ST.
Effect of Lyophilized S. boulardii on S. Typhimurium
Growth, Adhesion and Invasion of T84 Cells
In order to decipher the protective effect of Sb, we first analyzed
whether the yeast could inhibit the growth of bacteria. As
presented in Table 1 the growth of ST was not affected by Sb
(1.6260.796108 CFU well21 in ST group, 1.8160.186108 CFU
well21 in Sb + ST group, and 1.8560.426108 CFU well21 in Sb
overnight + ST group). Moreover, measurement of Salmonella fecal
content of gnotobiotic mice either untreated or treated with Sb
show that ST is not affected by Sb (Supporting Figure S1). As
Salmonella adhesion to host cells is the first necessary step for
invasion, this parameter was investigated on T84 cells. Data
presented in Table 1 show that Sb did not significantly change the
number of adherent bacteria to T84 cells. Then, using a classical
gentamicin protective assay, we evaluated T84 cell invasion. The
number of intracellular bacteria recovered in T84 cells infected by
ST alone was 0.360.036106 CFU well21 corresponding to 11.7%
of total bacteria. When Sb was applied together with ST the
number of intracellular bacteria significantly decreased to a value
of 0.260.0136106 CFU well21 representing 5.12% of total
bacteria. Overnight pre-incubation with Sb further increased the
effect
as
the
number
of
bacteria
dropped
to
0.01660.0076106 CFU well21, corresponding to 0.37% of total
bacteria. These data clearly demonstrate that Sb can interfere with
host cell invasion by ST.
Analysis of Interactions between S. boulardii and
Salmonella
To investigate whether ST might bind to yeast cells, electron
and confocal microscopy techniques were used. Scanning electron
micrograph of T84 cells infected by ST in the presence of Sb
shows yeasts on the surface of cell monolayer and ST adhering
specifically to yeast cells (Fig. 5A). Sb did not totally coat the
monolayer, and thus may explain the fact that the yeast did not
totally prevent ST penetrating the T84 cells. Transmission
electronic micrograph confirmed that the bacteria can bind
intimately to the yeast (Fig. 5B). To investigate more precisely this
interaction we used ruthenium red coloration. Ruthenium red is a
polyvalent cation that binds a range of highly charged polyanions
such as glycoproteins, and thus we could underline adhesion
between bacterial pili and the cell wall of Sb (Fig. 5, C and D).
Confocal micrographs also confirm binding of ST to Sb cell wall
(data not shown). Since Salmonella piliated strain adhesion to
Saccharomyces cerevisiae has been shown to be mannose sensitive [21],
we analyzed the effects of mannose in Sb-ST interaction. As
shown in Figure 6, agglutination between Sb and ST was inhibited
by a-D-mannose suggesting that adhesion is mediated by type I
fimbriae.
Figure 3. S. boulardii protects liver against S. Typhimurium
lesions. Representative aspects of liver in ST (A, C) and Sb + ST (B, D)
groups showing the portal spaces (*) and its surrounding parenchyma.
In A and C there are diffuse small foci of inflammatory cells (large
arrows) and degenerative changes (thin arrow) contrasting with
architectural preservation and absence of significant changes in liver
parenchyma of treated group (B, D). H&E, Bar = 20 mm (A, B) and 10 mm
(C, D). (E) S. boulardii prevents S. Typhimurium translocation to the liver.
Translocation of S. Typhimurium to liver in experimental (Sb + ST) or
control (ST) conventional mice challenged intragastrically with 4.0 log
CFU of the bacteria. Determination was performed ,10 days after
challenged (beginning of mortality). Vertical bars represent standard
deviations of the means. N = 10. ST (S. Typhimurium), Sb (S. boulardii).
doi:10.1371/journal.pone.0008925.g003
Lyophilized S. boulardii Preserves Barrier Function of S.
Typhimurium-Infected T84 Cells
To evaluate the cellular effects of Sb on intestinal epithelium
resistance, we infected human T84 cell monolayer by ST. Results
in figure 4A show that ST infection diminished gradually the
electrical resistance (TER) of monolayers. A significant decrease in
TER (25%, P,0.05) was already recorded at 3 h after infection
when compared to control monolayers. At 8 h post-infection, the
resistance dropped almost to 50% (Fig. 4A). When Sb was added
either before or at the moment of ST infection, the TER was
maintained at the level of uninfected monolayers. Overnight
challenge with yeast alone had no effect on TER. The
permeability activity of T84 monolayers was also evaluated by
measuring the paracellular diffusion of FITC-dextran across T84
monolayers. Data presented in Figure 4B show that at 7 hours
after infection, the permeability of T84 monolayers to FITCdextran increased in the condition of monolayers infected by ST
alone. This phenomenon was significantly reduced (P,0.05) in the
presence of Sb added either before or at the moment of infection
PLoS ONE | www.plosone.org
Effects of Lyophilized S. boulardii on Signaling Pathways
Induced upon S. Typhimurium Infection
The Rac1 GTPase is important for cytoskeleton remodeling
during bacteria infections. Kinetics studies performed with ST
strain ATTC 14028 did not show significant activation of Rac1 in
T84 or Hela cells (data not shown). For this reason the more
invasive SL1344 strain was used in these series of experiment.
SL1344 strain induced a rapid activation of Rac1 in HeLa cells
measured as the amount of its GTP-bound form. Activation was
detectable 15 min after the beginning of infection and further
increased after 3 h of infection (9 fold, Fig. 7A) The Escherichia coli
3
January 2010 | Volume 5 | Issue 1 | e8925
Anti-Infectious Effect of S. b
Figure 4. S. boulardii preserves barrier function of S. Typhimurium-infected T84 cells. (A) Transepithelial electrical resistance (TER) were
measured at different time in: (Control) control T84 monolayers, (S.b. ON) monolayers incubated with S. boulardii alone (overnight), (ST) monolayers
infected with ST alone, (ST + S.b.) monolayers infected with ST in the presence of S. boulardii added at the same time, or monolayers incubated
overnight with S. boulardii prior the infection (S.b. ON + ST). The TER values are displayed as percentages of initial values. (n = 5). An asterisk denotes
significantly different versus control cells (P,0.05). (B) Permeability of T84 monolayers to FITC-dextran was determined in monolayers exposed to ST
alone or to ST and S. boulardii. Graph represents percentages of variation of permeability as compared to initial value (t = 30 min) of each monolayers.
(n = 2). An asterisk denotes significantly different versus ST infected cells (P,0.05). The data are expressed as means 6 SEMs.
doi:10.1371/journal.pone.0008925.g004
(29.362.7 pg/ml) (Fig. 8B). Sb alone did not induce significant IL8 release (31.565.7 pg/ml). Expression of IL-8 mRNA started one
hour after ST infection (78.5 fold) and reached a plateau after 2
and 3 hours (respectively 1110.5 and 1106.4 folds) (Fig. 8C).
Simultaneous addition of Sb and ST for 1, 2 or 3 h did not modify
IL-8 mRNA content when compared to cells infected by ST alone.
However, when cells were pre-incubated with yeast before ST
infection, a significant decrease in ST-stimulated IL-8 synthesis
was observed (9.5, 4.8 and 2.7 fold decrease, after 1, 2 and 3 h of
ST infection, respectively) (Fig. 8C).
CNF1 toxin, used as a positive control [22], strongly activated
Rac1 (13 fold). We next compared two modes of Sb treatment,
either by simultaneous addition (curative protocol) or by overnight
pre-incubation (preventive protocol). While in the preventive
protocol, Sb could interfere with ST-induced Rac1 activation (5.4
vs 13.5 fold, Fig. 7B), only a slight inhibitory effect was detected
using the curative protocol (10.3 vs 13.5 fold).
Cell fractionation experiments were performed to visualize
activated Rac1 that locates to membranes. ST infection induced a
strong membrane localization of Rac1 (1.6 fold) that was
significantly decreased in cells pre-incubated with yeast (1.06
fold). Of note, this decrease in membrane associated Rac1 was
mirrored by an increase of Rac1 in cytosolic fractions. These two
approaches demonstrated that pre-incubation of cells with Sb
strongly interfered with Rac1 activation by ST.
Effect of Lyophilized S. boulardii on S. TyphimuriumInduced MAPK Activation and NF-kB Signaling Pathway
We next investigated the effect of Sb on different signaling
pathways that are known to be involved in IL-8 production in
kinetics experiments using phospho-specific antibodies. Blots were
probed with antibodies recognizing the non-phosphorylated form
of MAPKs to rule out any variation of the total protein amount
during the infection procedure. The phosphorylated active-forms
of ERK1/2 (p42 and p44), p38 and JNK (p46 and p54) were
nearly undetectable in control cells and in cells treated with Sb
alone (Fig. 9A). Activation of p38 and ERK1/2 were detectable
after 1 h of ST infection and increased after 2 and 3 h. Activation
of JNK was not detectable until 2 h of infection. When Sb was
simultaneously added with bacteria the level of phosphorylation of
Effect of Lyophilized S. boulardii on IL-8 Secretion and
Synthesis in S. Typhimurium Infected Cells
We next examined the ability of Sb to reduce or prevent key
inflammatory responses following ST infection. Kinetic studies
showed that a significant increase in IL-8 production was observed
3 h after ST infection (266.14641.3 pg/ml versus 4.0661.56 pg/
ml for uninfected cells) (Fig. 8A). Addition of Sb together with ST
significantly lowered IL-8 production (86.7617.6 pg/ml). Moreover, overnight pre-incubation almost abolished IL8 production
Table 1. Effect of S. boulardii (Sb) on adhesion and invasion of S. Typhimurium in T84 cells.
Cell treatment
Mean no. 6 SEM of
% Invasion
Total bacteria (x 108 CFU/
well)
Cell-associated bacteria (x
106 CFU/well)
Intracellular bacteria (x
106 CFU/well)
S. Typhimurium
1.6260.79
2.2560.21
0.3060.030
11.7860.90
Sb + S. Typhimurium
1.8160.18
3.8060.52
0.2060.013
5.1260.64*
Sb (ON) + S. Typhimurium
1.8560.42
4.5660.10
0.01660.007
0.3760.20*
T84 cells were infected for three hours with ST ATCC 14028 (108 CFU/well) in the presence (or not) of the yeast (107 CFU/well). Invasion was assessed by the gentamicin
protection method.
*Indicates statistical difference in relation to Salmonella alone infected cells (P,0.05).
(n = 3).
doi:10.1371/journal.pone.0008925.t001
PLoS ONE | www.plosone.org
4
January 2010 | Volume 5 | Issue 1 | e8925
Anti-Infectious Effect of S. b
Figure 5. Binding of S. Typhimurium (ST) to S. boulardii (Sb) cell wall. (A) Scanning electron micrograph of T84 cells infected by ST in the
presence of Sb. (B) Transmission electron microscopy. (C, D) Transmission electron micrograph after red ruthenium staining. N = 3. Black arrows show
the binding of bacteria to the yeast.
doi:10.1371/journal.pone.0008925.g005
the three MAPKs was similar to the levels measured in cells
infected by ST-alone. In contrast, pre-incubation of cells with Sb
prior to ST infection decreased activation of ERK1/2 and JNK
kinases (Fig. 9A). The effect of Sb on p38 activation appeared less
pronounced although no more activation could be detected at
3 hs.
We next tried to characterize the mode of Sb’s action on
MAPK. Before incubation on cells, Sb received different
treatments: heat inactivation (HT), glass beads (GB) treatment or
PBS washed (Sb were pre-incubated with T84 cells and before
Salmonella-infection, yeast was removed by several PBS washings).
Figure 9B shows that, HT and PBS washing totally abolished the
inhibitory effect of Sb on MAPK activation. Thus Sb must be alive
and present during infection. By contrast, the inhibitory effect of
Sb was maintained when an extract obtained after disruption of
the yeast with GB was used. This result suggests that components
of yeast cell structure were implicated in the inhibitory effect
observed.
We next examined the action of Sb on NF-kB activation. For
this, we first evaluated phosphorylation of IkB-a inhibitory
subunit, which results in IkB-a proteasomal degradation and
translocation of active NF-kB into the nucleus. In T84 cells, ST
induced phosphorylation of IkB-a that occurred 1 h after the
beginning of infection, increased at 2 h and remained elevated
over the course of 3 h (Fig. 10A). ST-induced phosphorylation of
IkB-a decreased when cells were pre-incubated overnight with Sb
but not when Sb and bacteria were simultaneously applied An
EMSA experiment showed a strong NF-kB binding activity 1 h
after the beginning of ST infection that remained elevated over
3 h (Fig. 10A). Again, overnight pre-incubation with Sb, but not
simultaneous addition completely prevented NF-kB activation.
Flagellin represents a major pro-inflammatory determinant of ST
that is able to activate NF-kB. As shown on Fig 10B, Sb did not
Figure 6. Phase-contrast microscopy showing agglutination
between S. boulardii (Sb) and S. Typhimurium (ST) in the
presence and absence of a-D-mannose. ST (S. Typhimurium), Sb (S.
boulardii), Man (a-D-mannose).
doi:10.1371/journal.pone.0008925.g006
PLoS ONE | www.plosone.org
5
January 2010 | Volume 5 | Issue 1 | e8925
Anti-Infectious Effect of S. b
Figure 7. S. boulardii prevents Rho-GTPase Rac1 signaling pathway in HeLa cells. (A) Activation of Rac1 in cells infected for various time
with S. Typhimurium (ST) strain SL1344 in the presence or absence of S. boulardii (Sb) (representative of five independent experiments). (B) Activation
of Rac1 in cells infected for 3 h by ST alone or ST and Sb added at the same time and in cells exposed overnight to Sb prior infection (representative
of two independent experiments). At indicated time, cells were lysed and GTP-bound Rac1 were precipitated from lysates by incubation with GSTPAK, resolved by SDS-PAGE and analyzed by immunoblotting using anti-Rac1 antibody. The control lane contained uninfected cells. CNF1 is a toxin
produced by uropathogenic E. coli that activate Rac1 [34] and it is used in this study as a positive control. (C) Immunoblots showing the distribution
of Rac proteins between membrane (M) and cytosolic (C) fractions in HeLa cells infected with ST for 3 h alone or in the presence of Sb (representative
of two independent experiments). Immunoblots were performed with RhaGDI as loading control for cytosolic fraction and human Transferrin
Receptor (hTR) was used as loading control for membrane fraction. Densitometry was performed using Multi Gauge software.
doi:10.1371/journal.pone.0008925.g007
Among the five cell lineages present in the intestinal epithelium
(i.e. enterocytes, M cells, goblet cells, enterochromaffin, Paneth
cells), M cells and enterocytes are generally thought to be the most
important for bacterial invasion and transcytosis [26]. The relative
importance of these two lineages as an entry portal for ST is a
matter of debate. After experimental infection with high doses of
ST, most bacteria are observed within the M cells in the small
intestine, indicating that these cells have a higher capacity to
internalize ST as compared to surrounding enterocytes. On the
other hand, a lower uptake efficiency of enterocytes for ST may be
compensated by its greater numbers, as they greatly outnumber M
cells in the small intestine (by .100:1). Furthermore, the colon is
markedly devoid of M cells, yet the colonic mucosa can be
efficiently colonized by ST. Cell culture studies on the interaction
between ST and intestinal epithelial cells have generally been
performed with model that mimic enterocytes, i.e. the human
colonic epithelial T84 cells.
Histological data presented in this study show that enterocytes
architecture in the colon of infected mice was preserved in the
presence of Sb during ST infection. Study performed on filter
grown T84 cells confirmed this observation. Translocation of ST
across the intestinal epithelium to the liver is greatly attenuated in
blocks flagellin-induced NF-kB activation. The presence of
mannose during infection did not interfere with the inhibitory
effect of Sb on ST-induced NF-kB activation (Fig. 10C). This
result suggests that inhibition of NF-kB binding activity is
independent of ST adhesion on Sb.
Discussion
A wide range of antibiotics are used to treat human
salmonellosis. However, genetic mutations and selective pressure
have pushed Salmonella spp., as well as other bacteria, to become
resistant or multi-resistant to antibiotics [23,24]. Development of
alternative natural processes for the treatment and prevention of
gastrointestinal disorders, such as probiotics, has become attractive
therapeutic options. In the present work we investigate the effects
of the non pathogenic yeast S. boulardii on Salmonella infection. A
protective effect of lyophilized Sb against ST infection has been
already described in conventional and gnotobiotic mice [25]. In
the present study we have investigated the protective effect of
lyophilized Sb against Salmonella infection at cellular and molecular
levels. Two primordial aspects were identified: Sb affects the
invasion property of ST, and it exerts an anti-inflammatory effect.
PLoS ONE | www.plosone.org
6
January 2010 | Volume 5 | Issue 1 | e8925
Anti-Infectious Effect of S. b
Figure 8. S. boulardii prevents IL-8 secretion and synthesis in S. Typhimurium infected cells. Pro-inflammatory response of T84
monolayers in response to S. Typhimurium infection in the absence (A) or presence (B) of S. boulardii. (A) At indicated time of infection, medium was
collected and IL-8 content was estimated by ELISA. The asterisk indicates that the value is significantly different from the value for uninfected control
cells (P,0.01), as determined by the Bonferroni-Dunn tests. (B) IL-8 content was estimated in the supernatant of T84 cells after 3 h of infection with
ST in the presence (curative and preventive protocol) or absence of S. boulardii. Errors bars indicate standard deviations. An asterisk denotes
significantly different versus ST-infected cells and two asterisks denotes significantly different versus control cells (P,0.05) when compared by the
Student’s t-test. ON, overnight. The data in A and B are expressed as means 6 SEMs (n = 8). (C) IL-8 mRNA content was estimated in T84 cells by real
time PCR after 1, 2 and 3 h of infection with ST in the presence (curative and preventive protocol) or absence of S. boulardii. Errors bars show the
SEMs. An asterisk denotes significantly different versus ST-infected cells (P,0.05, n = 6) when compared by the Student’s t-test. ON, overnight.
doi:10.1371/journal.pone.0008925.g008
mice receiving Sb. Incubation of T84 cells with Sb maintained the
TER of ST-infected monolayer, suggesting that the presence of the
yeast decreased the paracellular permeability of cells exposed to
ST.
As demonstrated using the T84 cells, Sb did not affect the
number of cell-attached Salmonella but significantly decreased the
number of intracellular bacteria. The main finding of our study is
the demonstration of a physical interaction of Sb with ST that
could account for the observed interference with bacterial
invasion. In case of cells incubated with yeast and bacteria, the
number of ‘‘cell associated’’ bacteria is slightly more elevated that
in cells incubated with ST alone. This increase can be explained
by observations made using scanning electronic microscopy which
revealed few yeasts attached to the monolayer and many bacteria
attached to the yeast cell wall. Yeast and adherent bacteria formed
aggregates attached to cell monolayer and this probably
contributed to the increased number of bacteria depicted in our
assay as adherent bacteria. Previous reports have shown that
Salmonella strains that express type I fimbriae have some affinity to
Sb. This affinity is mediated by mannan oligosaccharides of yeast
cell wall, and can be inhibited by mannose complexes [27]. The
aggregation between yeast and bacteria described in our study was
also inhibited by mannose. Transmission electronic micrograph
showed that the bacterial pili are in contact with yeast.
Microscopic study revealed that the yeast attracts and binds many
PLoS ONE | www.plosone.org
of the bacteria explaining why a low number of ST alone was
observed adhered to the monolayer of T84 cells. Once the yeast
attracted the bacteria on its surface, a diminished or inhibited
signaling cascade was generated by a lower number of bacteria in
direct contact with the host cell. Another consequence of Sb
addition during infection was the decrease of Salmonella’s cell
invasion by more than 50% (5.1% versus 11.8% in Salmonella alone
infected cells). The invasive properties of Salmonella were drastically
decreased (and reach a level of 0.3% of the control) in cells that
were incubated with yeast prior to infection. However, as depicted
on the scanning electron micrograph, Salmonella that are in contact
with T84 cells can also be visualized in that monolayers exposed to
yeast before infection. This indicates that besides adhesion of
bacteria to yeast, Sb decreased the invasion property of ST by an
additional mechanism.
Salmonella internalization is closely associated to injection through
the TTSS of effectors proteins that activate the Rho family of
GTPases. Recent evidence suggests that only Rac1 and RhoG are
indispensable for actin remodeling events that are generated by
Salmonella spp. during host cell entry [28]. We thus investigated the
effect of Sb on Rac1 activation in ST infected cells. In Salmonellainfected HeLa cells, activation of Rac1 started within 30 min after
infection and increased over 3 h of infection. The active form of
Rac1 was significantly decreased when cells were treated overnight
with Sb before ST infection. This observation was supported by
7
January 2010 | Volume 5 | Issue 1 | e8925
Anti-Infectious Effect of S. b
Figure 9. S. boulardii influences S. Typhimurium-induced MAPK activation (A). (B) Effect of S. boulardii (Sb ON), Sb heat treated (Sb HT), Sb
glass beads treated (Sb GB), and Sb PBS washed (Sb washed) on MAPK (p-38 and ERK1/2) activation induced by ST in infected T84 cells. Cells were
lysed at indicated times after infection. Samples were resolved by SDS-PAGE and analyzed by immunoblotting with anti-phospho-ERK1/2, antiphospho-p38, and anti-phospho-JNK antibodies. Total ERK, p38 and JNK are shown as loading controls. Control lanes correspond to uninfected T84
cells and T84 cells with Sb. ON, overnight. (C) Samples were resolved by SDS-PAGE and analyzed by immunoblotting with anti-phospho-IkB or antiIkB antibodies. The control lane corresponds to uninfected T84 cells. Densitometry was performed using Image J software [33]. Western blots shown
here are representative of 5 independent experiments.
doi:10.1371/journal.pone.0008925.g009
another set of experiments on Rac-GTP recruitment to the
membrane. As demonstrated in this study, the recruitment of the
active Rac1 to the membrane is decreased in cells infected in the
presence of yeast. This decrease is correlated with an increase of the
inactive Rac1 found in the cytosolic fraction. Altogether, these
results indicate that the yeast interferes with Salmonella induced
signaling pathways that are implicated in bacteria internalization. A
recent study reported that Sb did not reduce host cell invasion by
Shigella [20]. This difference may be due in part to the different
PLoS ONE | www.plosone.org
relationships between these bacteria and the host cell. Salmonella spp.
utilizes TTSS to directly activate Cdc42 and Rac1, whereas for
Shigella spp., the TTSS effector proteins do not engage the GTPase
directly, but rather they generate signals through activation of src
family tyrosine kinase [29].
The fact that Sb reduces Salmonella invasion, and induced
activation of host Rho GTPase, is of primordial importance in the
prevention of pathogenesis, since activation of Rho family is not
only implicated in actin cytoskeleton rearrangement but also in
8
January 2010 | Volume 5 | Issue 1 | e8925
Anti-Infectious Effect of S. b
EHEC [19], C. difficile [17] and, recently in the case of Shigella [20].
Reduction of IL-8 levels was in each case correlated with inhibition
of signaling pathways that play a role in IL-8 production. In the case
of C. difficile and Shigella infection, Sb supernatant also exhibits an
anti-inflammatory effect suggesting that soluble factor(s) produced by
the yeast is (are) implicated. A fraction containing a small (,1 kDa)
heat-stable and water soluble anti-inflammatory molecule, termed
Saccharomyces anti-inflammatory factor (SAIF), has been identified in
the yeast supernatant [30]. This fraction inhibited NF-kB activation
by LPS, IL-1b and TNF-a. As reported in our study, the inhibitory
effect disappeared after heat treatment of yeast but was present in the
supernatant containing yeast cell wall. These results suggest that a
heat labile molecule present in the cell wall of the yeast mediates the
inhibition of MAPKs.
Concluding, the results presented here clearly demonstrate that
S. boulardii modifies the invasive properties of Salmonella. Addition
of yeast during infection exerts a beneficial effect both by reducing
by 50% invasion by Salmonella and maintaining the barrier
function of epithelial monolayer. However, exposure to yeast
before infection, significantly amplifies the beneficial effect. The
invasion of Salmonella is completely abolished and the inflammatory process is stopped. Additionally, exposure to Sb before the
infection modulates activation of Rac1, MAPK and NF-kB during
infection.
Materials and Methods
Microorganisms
Two strains of Salmonella enterica serovar Typhimurium (ST) were
used in this study. The strain ATCC 14028 was kindly provided by
Oswaldo Cruz Foundation (FIOCRUZ), Rio de Janeiro, Brazil.
The strain SL1344 was kindly provided by Stéphane Meresse,
Centre d’Immunologie de Marseille-Luminy, CNRS-INSERM
Université de la Méditerranée, Marseille, France. Bacteria were
stored in Luria-Bertani (LB) medium plus 15% glycerol at –80uC
and grown in LB broth overnight at 37uC without shaking.
Cultures of S. boulardii (Sb) were obtained by inoculating a
commercial lyophilized preparation of the yeast (Ultra-LevureH,
BIOCODEX, France) and growing overnight at 37uC, with
shaking, in Halvorston minimal medium with 2% glucose.
Figure 10. S. boulardii influences on S. Typhimurium-induced
IkB-a phosphorylation and NF-kB signaling pathway. (A)
Samples were resolved by SDS-PAGE and analyzed by immunoblotting
with anti-phospho-IkB antibodies. Determination of NF-kB binding
activity was analyzed by EMSA. (B) S. boulardii did not decrease NF-kB
binding activity induced by 1 or 10 mg/ml of flagellin. (C) S. boulardii
induced decrease of NF-kB binding activity is not modified in the
presence of a-D-mannose. Western blots and EMSA shown are
representative of 5 independent experiments.
doi:10.1371/journal.pone.0008925.g010
modulation of tight junction and induction of inflammatory
responses.
The stimulation of Cdc42 triggers several MAPKs pathways
including ERK1/2, JNK and p38, which results in the activation of
the transcription factors AP-1 and NF-kB. These transcription
factors then direct the production of pro-inflammatory cytokines
such as IL-8, which stimulate PMN transmigration and the
inflammatory response leading to diarrhea. Kinetic studies presented
in this report show that after 3 h of infection, Salmonella stimulated
IL-8 secretion. Secretion of IL-8 was significantly decreased in cells
exposed to Sb during infection, but the IL-8 mRNA level was not
affected by the presence of the yeast. However, when cells were preincubated with yeast before infection, both IL-8 secretion and IL-8
mRNA level were decreased to basal levels. These results suggest that
Sb acts at two levels. When added simultaneously with infection, Sb
directly acts on the cytokines or the secretory process without
affecting the transcription machinery. When added before infection,
Sb modifies IL-8 gene transcription. These results suggest an
interference of Sb with MAPKs pathways and NF-kB activation.
However, data presented in this study reveal that addition of yeast
during infection did not modify activation of any MAPKs or NF-kB
activation. Conversely, incubation of cells with yeast before infection
inhibited activation of ERK1/2 and JNK pathways and, induced a
decrease in p38 phosphorylation and IkB. The anti-inflammatory
properties of Sb have been reported in the case of infection caused by
PLoS ONE | www.plosone.org
Mice
Conventional 21-day-old NIH mice (Taconic, USA) were used
in this work. Water and commercial autoclavable diet (Nuvital,
Brazil) were sterilized by steam and administered ad libitum, and
animals were maintained in an open animal house with controlled
lighting. All experimental procedures were carried out according
to the standard procedures [31]. The study was approved by the
Ethics Committee in Animal Experimentation of the Federal
University of Minas Gerais (protocol Nu 11/2007).
Determination of S. boulardii Protective Effect In Vivo
To investigate the effect of Sb on the mortality induced by ST
infection, mice were orally treated, during 10 days, with 0.1 ml
containing 108 colony forming units (CFU) of viable Sb or 0.1 ml
vehicle (control) and then challenged with 104 CFU ST.
Cumulative mortality was accompanied during 28 days post
infection (d.p.i).
Determination of S. Typhimurium Translocation In Vivo
To investigate the effect of Sb on ST translocation to liver, mice
were treated, during 10 days, with 0.1 ml containing 108 CFU of
viable Sb or 0.1 ml vehicle (control) and then challenged with
9
January 2010 | Volume 5 | Issue 1 | e8925
Anti-Infectious Effect of S. b
104 CFU ST. Bacteria translocation was determined after 10 days
of challenge, as previously determined [32].
sterile PBS. Cells were then trypsinized and lysed in water
containing 0.1% bovine serum albumin (BSA). The cell lysates
contained ‘‘cell-associated bacteria’’ corresponding to adherent as
well as intracellular bacteria. For the determination of invasion,
after PBS washes, monolayers were incubated for an additional
hour with DMEM/F-12 containing 100 mg of gentamicin per ml.
Since gentamicin was not concentrated in epithelial cells,
intracellular bacteria survived to the incubation, while adherent
and extracellular bacteria were killed. The monolayers were then
washed with sterile PBS, and epithelial cells with intracellular
bacteria were detached by trypsin and lysed as described
elsewhere. The percentage of invasion was calculated as follows:
Histological Analysis
Tissue samples from intestines of sacrificed mice were fixed in
buffered 4% formaldehyde and processed for paraffin embedding.
The histopathological sections were stained with hematoxylineosin (H&E) and examined by a single pathologist, who was
unaware of the experimental conditions.
Cell Lines and Growth Conditions
The human T84 colonic cell line and HeLa cells were obtained
from the European Collection of Animal Cell Cultures (Salisbury,
England). The T84 culture medium contained a 1:1 mixture of
Dulbecco-Vogt modified Eagle medium and Ham’s-F12 medium
(DMEM/F12) supplemented with 50 mg ml21 penicillin, 50 mg
ml21 streptomycin (Sigma, France), and 4% fetal bovine serum
(Hyclone, France). HeLa cells were maintained in DMEM
medium with antibiotics and 10% fetal bovine serum (Hyclone,
France).
percent invasion~
Electron Microscopy
For transmission electron microscopy, T84 cells were fixed in
situ with 1.6% glutaraldehyde with or without 0.075% ruthenium
red in 0.1 M cacodylate buffer pH 7.5 at room temperature. Cells
were washed five times in the same buffer, i.e. with or without
0.075% ruthenium red and post-fixed for 1 h at room temperature
in 1% osmium tetroxyde with or without 0.075% ruthenium red in
cacodylate buffer. Cells were then rinsed with distilled H2O,
dehydrated with ethanol and embedded in Epon (Embed 812).
Blocks were cut conventionally, stained and examined with a
Philips CM12 microscope operating under standard conditions.
For scanning electron microscopy cells were fixed in situ with
1.6% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.5) at room
temperature. Samples were washed with the same buffer,
dehydrated with increasing ethanol series, treated with hexamethyldisilazane and air dried. Cells were coated by with 3 nm of
gold-palladium and observed at low voltage with a Jeol 7400F
scanning electron microscope.
Infection Procedure In Vitro
For infection, T84 cells were seeded into six-well tissue culture
plates at 106 cells per well and grown to confluence. Prior to
infection, culture medium was changed to medium without serum
and antibiotics. Bacteria, grown overnight into LB broth medium,
were pelleted by centrifugation, re-suspended in DMEM/F12
medium, and added to cells (108 bacteria well21). After
determined times of infection, bacteria were eliminated by several
washes with cold sterile phosphate-buffered saline (PBS) and cell
monolayers were frozen in liquid nitrogen and conserved at
280uC for protein determination. When infection was performed
in the presence of yeast (107 yeasts well21), cells were preincubated overnight with Sb and then infected by ST (preventive
protocol) or received the yeast simultaneously (co-infection) with
the infection (curative protocol). The yeast-to-bacterium ratio did
not modify intestinal cell viability. For infection of filter-grown
cells, bacteria (107 bacteria well21) and yeast (106 bacteria well21)
were added to the apical compartment.
Measure of Rac Protein Activation by Effector-Binding
Pull-Down
HeLa cells were seeded in 100 mm Petri dishes. At 70–90%
confluence, cells were depleted overnight in serum-and antibioticsfree medium supplemented with 0.1% BSA (Sigma). Infections
were carried out in this medium with 36108 bacteria in the
presence (or not) of 36107 yeast overnight. At indicated time cell
monolayers were washed in PBS at 4uC and lysed at 4uC in cell
lysis buffer (25 mM Tris-HCl pH 7.5, 150 mM NaCl, 5 mM
MgCl2, 0.5% Triton X100, 4% glycerol, 10 mM NaF) supplemented extemporaneously with 1 mM PMSF, 2 mM Na3VO4,
2 mM DTT, 20 mM b-glycerophosphate. Lysates were centrifuged 10 min at 10,000 g at 4uC. An aliquot of 50 ml was collected
(Total Rho protein input) and mixed with Laemmli blue buffer
(4% SDS, 20% glycerol, 10% 2-mercaptoethanol, 0.004%
bromophenol blue, 0.125 M Tris-HCL). Lysates were incubated
with 30 mg of GST-PAK70–106 bound to glutathione-agarose beads
(Sigma-Aldrich, France) 45 min at 4uC on a rotating shaker. Beads
were washed twice with 1 ml of lysis buffer. Beads were mixed
with 30 ml of Laemmli blue buffer, and proteins were resolved on a
12 SDS-PAGE and transferred onto PVDF transfer membrane.
Immunoblots were performed using monoclonal antibodies
incubated overnight at 4uC at a dilution of 1:1000 for anti-Rac1
(clone 102; Transduction Laboratories, France) antibody, followed
by a peroxidase-conjugated sheep anti-mouse (Amersham Biosciences). The presence of antibodies was revealed with the ECL
Determination of Epithelial Monolayer Resistance and
Permeability
T84 cells were grown on collagen-coated, 0.33-cm2 porous filter
membranes (3-mm-diameter pores; Costar; Poly Lab. Paul Block,
Strasbourg, France). Transmonolayer electrical resistance (TER)
was measured with a Millicell-ERS apparatus (Millipore, Molsheim, France). The permeability of FITC-dextran (MW 70 kDa,
Invitrogen, Cergy Pontoise, France) through cell monolayers was
determined in the apical-to-basolateral directions. An FITCdextran aliquot of 0.5 mg ml21 was added to the apical side of
culture at the beginning of infection. The basolateral side was
sampled after 30 min (initial value) and 7 h of infection. Levels of
FITC-dextran in the samples were determined with a fluoroscan
(FLUOstar OPTIMA) using an excitation wavelength of 485 nm
and a detection emission at 538 nm. The permeability of each
monolayer was normalized to the initial value determined 30 min
after FITC-dextran addition.
Adhesion and Invasion Assays
Bacterial adhesion to T84 cells was quantified in the presence of
the yeast or not using the plate dilution method as previously
described [18]. After 3 h of infection, bacteria and yeasts present
in the culture medium were eliminated by extensive washes with
PLoS ONE | www.plosone.org
number of intracellular bacteria
adherent and intracellular bacteria
10
January 2010 | Volume 5 | Issue 1 | e8925
Anti-Infectious Effect of S. b
detection system. The density of bands was quantified using a
Multi Gauge software (Fuji Film Global).
described [19]. The protein concentration of the supernatant was
determinated using BioRad DC reagents. Equal amounts (50 mg)
of whole cell lysates were subjected to 12 SDS-polyacrylamide gel.
The proteins were transferred onto a polyvinylidene fluoride
membrane (PVDF Hybond-P, Amersham, Orsay, France), and
incubated overnight at 4uC with anti-phospho-ERK1/2, antiphospho-p38, anti-phospho-JNK, anti-phospho-IkB-a, anti-IkB-a
rabbit antibodies (Cell Signaling Technology), or anti-ERK2, antip38, anti-JNK (Santa Cruz Biotechnology) and HRP-conjugated
anti-rabbit antibodies (New England Biolabs). The presence of
antibodies was revealed with the Enhanced Chemiluminescence
detection system (ECL, Amersham).
Recruitment of Rac Protein to Cellular Membrane
HeLa cells were seeded in 100 mm Petri dishes and infected as
described above. After 3 h of infection, cell dishes were chilled on
ice and rinsed twice with cold PBS. Further steps were performed
at 4uC. Plates were scraped in 5 ml of cold PBS, centrifuged 5 min
at 1,000 rpm, and pellets were homogenized in 0.25 ml of cold
BSI buffer (3 mM imidazole, pH 7.4, 250 mM sucrose) supplemented extemporaneously with 1 mM PMSF. The mixture was
transferred to a 1.5 ml tube, and cells were lysed by passing 20
times through a 1-ml syringe (U-100 Insulin, Terumo). Nuclei
were removed by centrifugation 10 min at 10,000 g at 4uC. Postnuclear supernatants (PNS) were centrifuged 1 h at 100,000 g at
4uC. Supernatants (cytosolic fractions) were transferred in a next
1.5 ml tube, and pellets (membrane fractions) were homogenized
in an equal volume of BSI. Pellets and supernatants were mixed
with Laemmli blue buffer and proteins were resolved on a 12%
SDS-PAGE and transferred onto PVDF transfer membrane.
Immunoblots were performed using monoclonal antibodies
incubated overnight at 4uC at a dilution of 1:1000 for anti-Rac1
(clone 102; Transduction Laboratories, France), 1:3000 for antiRhoGDI polyclonal antibodies (A-20 Santa Cruz) and 1:2000 for
anti-Human Transferrin Receptor monoclonal antibodies
(ZYMED Laboratories). The density of bands was quantified with
Multi Gauge software (Fuji Film Global).
Electrophoretic Mobility Shift Assay (EMSA)
Cells were washed and infected with S. Typhimurium in the
presence of the yeast (overnight or at the same time of bacterial
infection) or not. At the indicated times, the infected cells were
washed with cold PBS. NF-kB DNA binding activities were
analyzed in total cleared cellular extracts prepared in Totex buffer
(20 mM HEPES pH 7.9, 350 mM NaCl, 20% glycerol, 1% NP40,
1 mM MgCl2, 0.5 mM EDTA, 0.1 mM EGTA, 1 mM DTT,
1 mM PMSF, 2 mg ml21 aprotinin). Samples (10 mg) were
incubated for 25 min at 25uC with radiolabeled double-stranded
oligonucleotide containing the kB site (59-GATCCAAGGGGACTTTCCATG-39). The specificity of the complexes was
analyzed by incubation with an excess of unlabeled kB
oligonucleotides. Complexes were separated by electrophoresis
on a 6% non-denaturing polyacrylamide gel in 0.5 X TBE as
previously described [19]. The dried gels were autoradiographed
(Amersham).
IL-8 Assay
IL-8 assays were performed on monolayers grown in six-well
tissue culture plates. Cells were incubated with the yeast (overnight
or not) and infected with ST. At the indicated times, the culture
supernatants were centrifuged for 10 min at 10,000 rpm to pellet
residual bacteria. The IL-8 concentration was determined with the
Quantikine Human IL-8 Immunoassay (R&D System, Abington,
U.K.), according to the manufacturer’s instructions.
Statistical Analysis
All the experiments were repeated at least three times. Results
are presented as the mean 6 the standard error of the mean
(SEM). Statistical significance was determined by analysis of
variance with the StatView program for MacIntosh, followed by
post hoc comparison with the Bonferroni and Dunn tests. The
level of significance was set at P,0.05.
Real-Time Quantitative Polymerase Chain Reaction
Briefly, after extraction of total RNA from cell culture by the
acid guanidium thiocyanate-phenol-chloroform method, the RNA
was converted to complementary DNA (cDNA) using oligo (dT)18
in 20 mL reverse-transcription reaction solution (Fermentas,
France) and then used for polymerase chain reaction. Reverse
transcribed cDNA were then amplified using the SYBR Green
PCR Core Reagents Kit (Eurogentec, France) in special optical
96-well microtiter plates (Applied Biosystems, Courtaboeuf,
France) in an ABI PRISM 7000 Sequence Detection System
(Applied Biosystems), according to the manufacturer’s instructions.
The primers used were as follow: IL-8 sense, 59-AAGGAACCATCTCACTGTGTGTAAAC-39; IL-8 antisense, 59-ATCAGGAAGGCTGCCAAGAC-39. The reference housekeeping gene
RPLP0 (encoding human acidic ribosomal phosphoprotein P0)
was used for all calculations as it showed no change in expression
upon treatment of cells.
Supporting Information
Figure S1 Fecal populations of S. Typhimurium in gnotobiotic
NIH mice treated (Sb + ST) or not (ST) with S. boulardii for 10 days
before the challenge with the bacteria. Fecal population numbers
of S. boulardii (Sb). Arrow indicates the day of pathogenic
challenge. N = 3 animals in each group.
Found at: doi:10.1371/journal.pone.0008925.s001 (2.51 MB
DOC)
Acknowledgments
We thank S. Meresse for providing us the Salmonella strain SL1344 and A.
Darfeuille Michaud for providing us flagellin purified from SL1344 strain
and P Gounon for electronic microscopy analysis.
Author Contributions
Western Blotting
Conceived and designed the experiments: JFP PR JN DC. Performed the
experiments: FdSM GD RMEA PL DC. Analyzed the data: FdSM RMEA
AD EL VI. Contributed reagents/materials/analysis tools: AD EL VI.
Wrote the paper: FdSM EL VI JFP DC.
At the indicated times, the infected cells were washed with PBS
and scraped at 4uC in lysis buffer, solubilized for 30 min at 4uC,
then centrifuged at 14,000 rpm for 20 min at 4uC, as previously
References
1. Galán JE, Bliska JB (1996) Cross-talk between bacterial pathogens and their host
cells. Annu Rev Cell Dev Biol 12: 221–255.
PLoS ONE | www.plosone.org
2. Galán JE, Collmer A (1999) Type III secretion machines: bacterial devices for
protein delivery into host cells. Science 284: 1322–1328.
11
January 2010 | Volume 5 | Issue 1 | e8925
Anti-Infectious Effect of S. b
3. Wallis TS, Galyov EE (2000) Molecular basis of Salmonella-induced enteritis. Mol
Microbiol 36: 997–1005.
4. Hobbie S, LM Chen LM, Davis RJ, Galán JE (1997) Involvement of mitogenactivated protein kinase pathways in the nuclear responses and cytokine
production induced by Salmonella typhimurium in cultured intestinal epithelial cells.
J Immunol 159: 5550–5559.
5. Giannella RA (1979) Importance of the intestinal inflammatory reaction in
Salmonella-mediated intestinal secretion. Infect Immun 23: 140–145.
6. Jung HC, Eckmann L, Yang SK, Panja A, Fierer J, et al. (1995) A distinct array
of proinflammatory cytokines is expressed in human colon epithelial cells in
response to bacterial invasion. J Biol Investig 95: 647–665.
7. McCormick BA, Colgan SP, Delp-Archer C, Miller SI, Madara JL (1993)
Salmonella typhimurium attachment to human intestinal epithelial monolayers:
transcellular signaling to subepithelial neutrophils. J Cell Biol 123: 895–907.
8. Hoffmann E, Dittrich-Breiholz O, Holtmann H, Kracht M (2002) Multiple
control of interleukin-8 gene expression. J Leukoc Biol 72: 847–855.
9. Mukaida N, Okamoto SI, Ishikawa Y, Matsushima K (1994) Molecular
mechanism of interleukin-8 gene expression. J Leukoc Biol 56: 554–558.
10. FAO/WHO (2002) Guidelines for the Evaluation of Probiotics in Food,
London, Ontario, Canada.
11. Czerucka D, Piche T, Rampal P (2007) Review article: yeast as probioticsSaccharomyces boulardii. Aliment Pharmacol Ther 26: 767–778.
12. Czerucka D, Rampal P (2002) Experimental effects of Saccharomyces boulardii on
diarrheal pathogens. Microbes Infect 4: 733–739.
13. Castagliuolo I, Lamont JT, Nikulasson ST, Pothoulakis C (1996) Saccharomyces
boulardii protease inhibits Clostridium difficile toxin A effects in the rat ileum. Infect
Immun 64: 5225–5232.
14. Pothoulakis C, Kelly CP, Joshi MA, Gao N, O’Keane CJ, et al. (1993)
Saccharomyces boulardii inhibits Clostridium difficile toxin A binding and enterotoxicity in rat ileum. Gastroenterology 104: 1108–1115.
15. Czerucka D, Roux I, Rampal P (1994) Saccharomyces boulardii inhibits
secretagogue-mediated adenosine 39, 59-cyclic monophosphate induction in
intestinal cells. Gastroenterology 106: 65–72.
16. Brandão RL, Castro IM, Bambirra EA, Amaral SC, Fietto LG, et al. (1999)
Intracellular signal triggered by cholera toxin in Saccharomyces boulardii and
Saccharomyces cerevisiae. Appl Environ Microbiol 64: 564–568.
17. Chen X, Kokkotou EG, Mustafa N, Bhaskar KR, Sougioultzis S, et al. (2006)
Saccharomyces boulardii inhibits ERK1/2 mitogen-activated protein kinase
activation both in vitro and in vivo, and protects against Clostridium difficile toxininduced enteritis. J Biol Chem 281: 24449–24459.
18. Czerucka D, Dahan S, Mograbi B, Rossi B, Rampal P (2000) Saccharomyces
boulardii preserves the barrier function and modulates the transduction pathway
PLoS ONE | www.plosone.org
View publication stats
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
12
induced in enteropathogenic Escherichia coli-infected T84 cells. Infect Immun 68:
5998–6004.
Dahan S, Dalmasso G, Imbert V, Peyron JF, Rampal P, et al. (2003)
Saccharomyces boulardii interferes with enterohemorrhagic Escherichia coli-induced
signaling pathways in T84 cells. Infect Immun 71: 766–773.
Mumy KL, Chen X, Kelly CP, McCormick BA (2008) Saccharomyces boulardii
interferes with Shigella pathogenesis by postinvasion signaling events. Am J Physiol
Gastrointest Liver Physiol 294: G599–G609.
Kornonen TK, Leffler H, Svanborg Eden C (1981) Binding specificity of piliated
strains of Escherichia coli and Salmonella typhimurium to epithelial cells, Saccharomyces
cerevisiae cells, and erythrocytes. Infect Immun 32: 796–804.
Doye A, Boyer L, Mettouchi A, Lemichez E (2002) Ubiquitin-mediated
proteasomal degradation of Rho proteins by the CNF1 toxin. Meth Enzymol
406: 447–456.
Hopkins KL, Day M, Threlfall EJ (2008) Plasmid mediated quinolone resistance
in Salmonella enterica. Emerg Infect Dis 14: 340–342.
Whichard JM, Gay K, Stevenson JE, Joyce KJ, Cooper KL, et al. (2007) Human
Salmonella and concurrent decreased susceptibility to quinolones and extendedspectrum cephalosporins. Emerg Infect Dis 13: 1681–1688.
Rodrigues ACP, Nardi RM, Bambirra EA, Vieira EC, Nicoli JR (1996) Effect of
Saccharomyces boulardii against experimental oral infection with Salmonella
typhimurium and Shigella flexneri in conventional and gnotobiotic mice. J Appl
Bacteriol 81: 251–256.
Jones BD, Falkow S (1996) Salmonellosis: host immune responses and bacterial
virulence determinants. Ann Rev Immunol 14: 533–561.
Gedek BR (1999) Adherence of Escherichia coli serogroup O 157 and the
Salmonella typhimurium mutant DT 104 to the surface of Saccharomyces boulardii.
Mycoses 42: 261–264.
Patel JC, Galan JE (2006) Differential activation and function of RhoGTPases
during Salmonella-host cell interactions. J Cell Biol 175: 453–463.
Tran Van Nhieu G, Enninga J, Sansonetti P, Grompone G (2005) Tyrosine
kinase signaling and type III effectors orchestrating Shigella invasion. Curr Opin
Microbiol 8: 16–20.
Sougioultzis S, Simeonidis S, Bhaskar KR, Chen X, Anton PM, et al. (2006)
Saccharomyces boulardii produces a soluble anti-inflammatory factor that inhibits
NF-kB-mediated IL-8 gene expression. Biochem Biophys Res Commun 343:
69–76.
National Research Council (1996) Guide for the Care and Use of Laboratory
Animals. Washington, DC: National Academy Press .
Martins FS, Rodrigues ACP, Tiago FCP, Penna FJ, Rosa CA, et al. (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: 352–359.
January 2010 | Volume 5 | Issue 1 | e8925