Available online at www.sciencedirect.com
n
arbeen an immense obstacle for mechanistic studies. Sim-
plified model systems and recent advances in analytic
methodology have fuelled significant progress. However,
it should be kept in mind that no single study has so far
been able to monitor all relevant parameters (some of
which might still be unknown) of the complex mamma-
lian gut ecosystem in parallel. This may leave room for
alternative interpretations and explains the field’s keen
mouse colony microbiota. However, effects of the species
composition should be carefully addressed in each case.
General functions of the microbiota
In most cases the beneficial functions of the microbiota
outweigh potentially harmful side effects. The micro-
biota provides digestive functions, modulates host metab-
olism and stimulates development of lymphatic tissue
Current Opinion in Microbiology 2011, 14:82–91 www.sciencedirect.comMechanisms controlling pathoge
Ba¨rbel Stecher1 and Wolf-Dietrich H
The intestinal microbiota can protect efficiently against
colonization by many enteric pathogens (‘colonization
resistance’, CR). This phenomenon has been known for
decades, but the mechanistic basis of CR is incompletely
defined. At least three mechanisms seem to contribute, that is
direct inhibition of pathogen growth by microbiota-derived
substances, nutrient depletion by microbiota growth and
microbiota-induced stimulation of innate and adaptive immune
responses. In spite of CR, intestinal infections are well known to
occur. In these cases, the multi-faceted interactions between the
microbiota, the host and the pathogen are shifted in favor of the
pathogen. We are discussing recent progress in deciphering the
underlying molecular mechanisms in health and disease.
Addresses
1 Max von Pettenkofer Institut, Pettenkoferstrasse 9a, 80336 Mu¨nchen,
Germany
2 Institute of Microbiology, ETH Zu¨rich, Switzerland
Corresponding authors: Stecher, Ba¨rbel
(Stecher@mvp.uni-muenchen.de) and Hardt, Wolf-Dietrich
(hardt@micro.biol.ethz.ch)
Current Opinion in Microbiology 2011, 14:82–91
This review comes from a themed issue on
Host–microbe interactions: bacteria
Edited by Brett Finlay and Ulla Bonas
Available online 28th October 2010
1369-5274/$ – see front matter
# 2010 Elsevier Ltd. All rights reserved.
DOI 10.1016/j.mib.2010.10.003
To understand the whole it is necessary to understand the
parts. To understand the parts, it is necessary to under-
stand the whole. Such is the circle of understanding [1].
A word of caution
The mammalian gut is a highly complex ecosystem
shaped by the host, a complex microbial community
called microbiota and profoundly affected by interactions
with the outside environment, for example, the intake of
food or infection by pathogens [2]. This complexity hascolonization of the gut
dt2
interest in reproducible animal models and comprehen-
sive analytical tools.
The composition of the gastrointestinal
microbiota
The composition of the microbiota and its collective
genome, the microbiome, has been intensely studied.
Early cultivation based studies suggested that the human
intestinal microbiota harbors at least 400 different, mostly
obligate anaerobic bacterial species [3,4]. This was con-
firmed by modern, culture-independent approaches esti-
mating that an individual’s microbiota comprises not
more than 500 species [5]. Two predominant phyla were
observed, the Firmicutes and the Bacteroidetes. Other
phyla such as as Proteobacteria, Actinobacteria, Fusobac-
teria, Verrucomicrobia, and Cyanobacteria are minor con-
stituents. Of note, the murine and the human microbiota
are remarkably similar [6], suggesting that mouse models
can be used to study basic functional principles of
relevance for human health.
Nevertheless, data generated by novel culture-indepen-
dent approaches (i.e. 16S rRNA gene and metagenomic
sequence analyses) have to be interpreted with care as
certain species may be missed or underestimated in terms
of their abundance due to differential bacterial lysis
efficiency, primer bias and variable 16S rRNA gene copy
numbers [7,8]. Moreover, PCR on mixed templates can
give rise to significant ‘noise’ which leads to overestima-
tion of ecosystem complexity [5,9].
Deep microbiome sequencing analyses have revealed a
wide array of shared genes (‘core’ microbiome) but also
demonstrated a considerable degree of species diversity
among the microbiota, even between closely related hosts
[5]. Interestingly, despite this species diversity, different
microbiomes are functionally highly conserved as shown
by the abundance of COG categories [10]. Therefore,
besides all variability, some general ordering principles do
exist and lead to functionally widely conserved micro-
biota properties. This explains why key functional prop-
erties of the microbiome are consistently observed in
different laboratories worldwide, in spite of significant
lab-to-lab variations in the species composition of the

strated recently that the microbiota can in fact be ‘trans-
Mechanisms controlling pathogen colonization Stecher and Hardt 83
Glossary
COG: cluster of orthologous groups of proteins used for phylogenetic
classification of proteins encoded in complete genomes
Colonization resistance (CR): characteristic of the intestinal
microbiota to block colonization of pathogens
Defensin: peptide with antimicrobial activity
Gnotobiotic: colonized with bacteria of known identity
IBD: inflammatory bowel disease (e.g. Crohn’s disease and
Ulcerative Colitis)
0

2

Lo
CO
N
A
nt
ib
io
tic
-
tr
ea
te
d
LC
M
(S
ch
ae
dle
r)

G
er
m
fre
e
Proteobacteria
Current Opinion in Microbiology
Colonization resistance against oral Salmonella infection in mice with
different types of microbiota. The figure summarizes published and
unpublished data from Salmonella-infected mice. Conventional mice
from different sources (n = 58), streptomycin-pretreated mice (n = 12)
and low-complexity microbiota (LCM) mice (n = 10; all from [29] and
Stecher and Hardt, unpublished) and germfree mice (n = 7; [45]) were
orally infected with Salmonella serovar Typhimurium or serovar
Enteritidis. Intestinal loads in cfu/g cecal content are depicted. Pie plots
illustrate microbiota composition in the respective groups
(Bacteroidetes: yellow; Firmicutes: green; Deferribacteres: purple;
Verrucomicrobia: blue; Proteobacteria: red; other phyla with abundance
<1%: not depicted).and the mucosal immune system. Moreover, it can effi-
ciently limit infection of the gut by pathogenic bacteria.
In fact, during pathogen infection, the microbiota may
have at least three cardinal functions: (i) it may block
growth of the pathogen and thus interfere with the in-
fection right from the beginning. This is termed as
colonization resistance (CR; Box 1) and will be the focus
of this review. (ii) The microbiota may prime the host’s
innate and adaptive immune defenses and prevent path-
ology which would otherwise be caused by the pathogen’s
virulence factors. This has been reviewed recently [11].
(iii) It can help to eliminate the pathogen from the gut
lumen at the end of an infection. This has recently been
demonstrated [12] and will be discussed elsewhere.
Even in the normal, healthy host, microbial products
released from the intestinal microbiota such as peptido-
glycan disseminate to the mesenteric lymph nodes and
even to systemic sites and stimulate immune cell matu-
ration [13]. In some cases, the presence of an intestinal
microbiota contributes to the etiology and progression of
diseases in humans, namely chronic inflammatory bowel
diseases (IBD) [14]. IBD is thought to result if barrier
functions, antimicrobial killing or dampening signaling
Microbiome: collective genome of all present bacteria
Metagenome: collective genome of all organisms present in a given
ecosystem
16S rRNA gene sequencing: The 16SrRNA gene is commonly used
for phylogenetic studies as it is highly conserved between different
species of bacteria and archaea
SPF: specified pathogen freepathways of the innate and/or adaptive immune system
fail [15]. In addition, some bacterial species are associated
with the aggravation of autoimmune diseases and the
incidence of colorectal cancer [16]. Thus, microbiota
manipulation by the introduction or selective elimination
of relevant bacteria might represent future avenues for
cure or prevention. A ‘proof of principle’ study demon-
Box 1 Colonization resistance (CR) has been known for more than
50 years. Anecdotal observations in human patients and animal
studies had shown that disruption of the microbiota by antibiotics
dramatically increases the susceptibility towards enteric infections
[94,95]. This has been confirmed in numerous recent studies
[43,96,97]. Experiments in germ-free mice and ex-germfree mice
colonized with selected bacterial species lent further support to this
[29,45,98]. The mechanisms underlying this natural protection are
still unresolved and a matter of debate.
www.sciencedirect.comFigure 1
Microbiota
composition

4

6

8

10

g 1
0
cf
u/
g
Sa
lm
on
el
la
-
Bacteroidetes
Firmicutes
Deferribacteres
Verrucomicrobia planted’ between two hosts [17].
This review will summarize the recent literature pertain-
ing to the three-way interaction between the microbiota,
the host and enteric pathogens and discuss novel tech-
nologies which may aid analysis of CR in the future.
Mechanisms of CR
What do we know about the mechanisms underlying CR?
Innate immunity, adaptive immunity and bacterial inter-
actions are probably involved in modulating both the
composition of the microbiota and the outcome of infec-
tions. To establish successful infection, all pathogens
need to replicate in the gut lumen in order to reach a
sufficient population density for causing disease. In the
case of S. Typhimurium enterocolitis, this was demon-
strated by infecting mice with defined mixtures of the
wild-type pathogen and avirulent mutants [18]. At least
Current Opinion in Microbiology 2011, 14:82–91