Mutualism Humans represent a scaffold on which diverse microbial disease during colonization. Surprisingly, the normal
A symbiotic association in ecosystems are established. Immediately after birth, all microbiota also contains microorganisms that have been
which both members benefit mammals are initiated into a life-long process of colo- shown to induce inflammation under particular con-
from the relationship.
nization by foreign microorganisms that inhabit most ditions. Therefore, the microbiota has the potential to
Pathogen environmentally exposed surfaces (such as the skin, exert both pro- and anti-inflammatory responses, and
An opportunistic organism mouth, gut and vagina)1,2. Shaped by millennia of evo- the composition of the bacterial communities in the gut
that rarely comes into contact lution, some hostbacterial associations have developed may be intimately linked to the proper functioning of
with the host, but causes acute into beneficial relationships, creating an environment the immune system.
or chronic disease following
infection. Derived from the
for mutualism. A key example of such an environment The immune system is responsible for recognizing,
Greek word pathos, which is provided by the vast numbers and diversity of bac- responding and adapting to countless foreign and self
means suffering. teria that are found in the lower gastrointestinal tract molecules and is therefore important during condi-
of mammals1,35. By young adulthood, both humans tions of both health and disease. Although the immune
Microbiome
and other mammals support one of the most complex system is classically thought to have evolved to protect
The collective genomes of
a microbiota. microbial ecosystems on the planet, with over 100 tril- from infection by microbial pathogens, animals peace-
lion bacteria in the distal gut 6,7. Symbiotic bacteria of fully coexist with a vast and complex microbiota, which
the mammalian gut have long been appreciated for the extensively interacts with the immune system. In this
benefits they provide to the host: they supply essential Review, we discuss recent evidence suggesting that a
nutrients, metabolize indigestible compounds, defend beneficial partnership has evolved between symbiotic
against colonization by opportunistic pathogens and bacteria and the immune system. The molecular inter-
Division of Biology,
even contribute to the development of the intestinal actions seem to direct the development of immune
California Institute of
Technology, Pasadena, architecture 8. Moreover, it seems that certain basic responses, and in turn the immune system shapes the
California 91125, USA. developmental features and functions of the mamma- composition of the microbiota. We highlight seminal
Correspondence to S.K.M. lian immune system depend on interactions with the examples of microorganisms that have a role in pre-
email: sarkis@caltech.edu human microbiome9. Unlike opportunistic pathogens, venting inflammatory bowel disease (IBD) and discuss the
doi:10.1038/nri2515
Published online 3 April 2009;
which elicit immune responses that result in tissue beneficial immune responses they elicit during protec-
corrected after print damage during infection, some symbiotic bacterial tion. Furthermore, technological advances now allow
17 July 2009 species have been shown to prevent inflammatory a more detailed understanding of the alterations of the
Microbiota microbial population of the gut during IBD. If some Together with various morphological tissue defects
The amalgam of bacteria are actively shaping a healthy immune system, that are observed in the intestines of germ-free animals,
microorganisms that make does the absence of these organisms lead to disease? It it seems that the entire ultrastructural development of
up a complex and diverse has recently been proposed that the total information the gut is intimately connected to intestinal bacteria.
community living within a given
anatomical niche (usually an
encoded by the mammalian genome is not sufficient For example, intestinal epithelial cells (IeCs), which line
environmentally exposed to carry out all functions that are required to maintain the gut and form a physical barrier between lumenal
surface of the body). health and that products of our microbiome are crucial contents (including the microbiota) and the under-
for protection from various diseases10. It is possible that lying cells of the immune system, have altered patterns
Inflammatory bowel disease
alterations in the development or composition of the of microvilli formation and decreased rates of cell turn-
A chronic condition of the
intestine that is characterized
microbiota (known as dysbiosis) disturb the partnership over in germ-free animals compared with wild-type
by severe inflammation and between the microbiota and the human immune system, animals13. Furthermore, gut bacteria have been shown
mucosal destruction. The most ultimately leading to altered immune responses that may to direct the glycosylation of lumenally exposed surface
common forms in humans are underlie various human inflammatory disorders. proteins of IeCs14.
ulcerative colitis and Crohns
disease.
IeCs have many immunological functions (TABLE 1):
Insights gained from germ-free mice they secrete and respond to various cytokines and express
Gut-associated Developmental defects in germ-free mice. Several impor- molecules that directly interact with lymphocytes (such as
lymphoid tissue tant effects of the microbiota on the host immune sys- MHC molecules). expression and localization of pattern
Lymphoid structures and
tem have been determined by studies of gnotobiology, recognition receptors (such as Toll-like receptors; TlRs)
aggregates associated with the
intestinal mucosa, specifically
which is the selective colonization of germ-free (sterile) by the epithelium is influenced by bacterial coloniza-
the tonsils, Peyers patches, animals, the immune responses of which have not been tion of the gut, and the expression of defensins and other
lymphoid follicles, appendix or influenced by interactions with molecules of pathogenic antimicrobial proteins is defective in germ-free animals
caecal patch and mesenteric and beneficial microorganisms. Germ-free animals show (TABLE 1). Consistent with this notion, the Gram-negative
lymph nodes. They are
extensive defects in the development of gut-associated commensal organism Bacteroides thetaiotaomicron, but
enriched in conventional and
unconventional lymphocytes lymphoid tissues9,11 and in antibody production (BOX 1), not the Gram-positive microorganism Bifidobacterium
and specialized dendritic cell and have fewer and smaller Peyers patches and mesenteric longum, induces the expression of the antimicrobial
and macrophage subsets. They lymph nodes (Mlns) compared with animals housed peptide ReG3 (regenerating islet-derived 3) by spe-
provide the first line of defence
under specific pathogen free (SPF) conditions (TABLE 1). cialized IeCs known as Paneth cells15,16. Intriguingly, the
against entry of pathogens
across the mucosal barrier.
A recent report has also shown that germ-free animals specificity of ReG3 is directed towards certain Gram-
show impaired development and maturation of isolated positive bacteria. It is therefore tempting to speculate that
lymphoid follicles (IlFs)12. These inducible structures symbiotic bacteria direct innate immune responses of the
seem to form normally following the introduction of gut gut in an effort to protect their environment. Collectively,
bacteria, suggesting a dynamic relationship between the the observations of developmental defects in germ-free
immune system and the microbiota. mice at the tissue, cellular and molecular levels suggest
that normal immune function may be impaired in the
absence of the gut microbiota.
Box 1 | Antibody responses in germ-free animals
One of the first immunological defects observed in germ-free mice was a marked Defects in immune responses in germ-free mice. Germ-
reduction in the levels of secretory IgA found in the intestine94. Association of mice free animals are more susceptible to infection by certain
with a specific bacterial species leads to increased IgA expression. As numerous bacteria, viruses and parasites. When challenged with
studies have shown that secretory IgA coats commensal (and pathogenic) bacteria,
the Gram-negative enteric pathogen Shigella flexneri,
some have speculated that IgA is involved in limiting the penetration of bacteria into
host tissues. Studies from activation-induced cytosine deaminase (AID)-deficient
germ-free animals showed decreased immune resist-
animals (which cannot undergo class switching to IgA), have shown that these mice ance to infection and increased mortality compared
display lymphoid hyperplasia of the gut and an altered microbiota, thereby favouring with conventionally colonized animals17. Prior coloni-
the outgrowth of specific classes of bacteria95. Although some studies have shown zation with specific commensal bacteria antagonized
that IgA is involved in protection from infection by some enteric bacteria and viruses96, S. flexneri infection, whereas colonization with con-
other studies suggest that IgA deficiency does not cause increased prevalence of trol species such as Escherichia coli did not, implying
disease in animals and that IgA-deficient individuals are generally healthy. However, a that some members of the microbiota provide protec-
role for symbiotic bacteria in actively shaping the production of secretory IgA is now tion against intestinal bacterial pathogens18. Infection
emerging. Dendritic cells (DCs) that have acquired gut commensal bacteria migrate to by the Gram-positive intracellular pathogen Listeria
mesenteric lymph nodes, where they induce the production of IgA from naive B cells97.
monocytogenes results in decreased bacterial clearance
This process is required to control the penetration of commensal bacteria through the
gut epithelial cell barrier98. Recently, it has been shown that sensing of symbiotic
in germ-free animals compared with colonized ani-
bacterial products by Toll-like receptors on intestinal epithelial cells results in mals19. The mechanism for this increased susceptibility
localized class switching to IgA2 (rEf. 99), a process that potentially affects has been attributed to a T cell trafficking defect to the
interactions between the immune system and the microbiota100. The discovery that site of L. monocytogenes infection in germ-free animals.
symbiotic bacteria direct the function of a specialized mucosal DC population Specifically, germ-free mice that have been infected
that induces IgA class switching also implicates a role for the microbiota in shaping with L. monocytogenes have decreased accumulation of
intestinal immune responses101. Furthermore, IgA responses were recently shown CD44+CD62l+ T cells (CD44 and CD62l (also known
to be involved in maintaining hostbacterial mutualism by limiting innate immune as l-selectin) are known to be involved in the homing
responses to a specific gut symbiont102. The renewed interest in the biological of lymphocytes to sites of inflammation), resulting in
functions of intestinal secretory IgA promises to provide important clues on the
increased bacterial burden compared with SPF ani-
molecular communication between the immune system and the microbiota.
mals20. Finally, Salmonella enterica subspecies enterica
cytokine receptor -chain); cytolysis; and targeting the parasite burden. Previous work has suggested that TlR
maturation of dendritic cells (DCs) through cell surface signalling is important for maintaining gut homeo-
expression of molecules such as cytotoxic T lymphocyte stasis42, and these recent findings extend this obser-
antigen 4 and lymphocyte activation gene 3 (rEf. 35). A vation to suggest that a molecular dialogue between
population of intestinal DCs expressing the cell surface immune receptors and microbial molecules confers
antigen CD103 has recently been shown to be instru- resistance to enteric infection.
mental in the development and function of intestinal The contribution of the microbiota to the develop-
FoXP3+ TReg cells. CD103+ DCs, but not CD103 DCs, ment of TReg cells remains unclear, as conflicting obser-
can promote the conversion of CD4+FoXP3 T cells vations have been reported. An initial study43 observed
into CD4+FoXP3+ TReg cells in a TGF- and retinoic a reduction in the percentage of FoXP3+ cells within
acid-dependent manner 36, which indicates that special- the CD4+CD25+ T cell subset in the Mlns of germ-free
ized mechanisms exist in the intestine to promote the mice compared with conventionally colonized animals.
induction and maintenance of TReg cells. Foxp3 mRnA levels were also lower in CD4+CD25+
The importance of TReg cells in the regulation of intes- cells isolated from the lymph nodes of germ-free mice.
tinal homeostasis is best illustrated by the finding that Consistent with these findings, another study 45 observed
these cells can prevent the induction of experimental lower levels of Foxp3 mRnA in CD4+CD62l T cells
colitis following transfer into diseased hosts37. The ability from germ-free mice. In addition, it has been reported
of TReg cells to secrete Il-10 and Il-35 has been reported that TReg cells from germ-free animals are not as effective
to be important during protection. Indeed, TReg cells as TReg cells from conventionally colonized animals at
that are deficient in either of the two subunits of Il-35 suppressing CD4+ T cell proliferation in vitro44. Indeed,
(epsteinBarr virus-induced protein 3 and Il-12) can- populations of TReg cells from germ-free animals pro-
not provide protection from induction of experimental duced less Il-10 and could not prevent disease in a trans-
colitis38. In addition, animals in which Il-10 has been fer model of experimental colitis45. In contrast to these
specifically ablated from CD4+FoXP3+ T cells succumb observations, recent studies have reported no change
to inflammatory disease of the intestine (as well as of the in the percentage of CD4+FoXP3+ T cells in the colon
skin and lungs), but show no signs of autoimmunity. So, lamina propria of germ-free mice46, and in fact another
it seems that cytokine production by TReg cells might be study 39 reported increased percentages of CD4+FoXP3+
Crohns disease an important protective mechanism that limits uncon- in the small intestine39.
A form of chronic inflammatory trolled immune responses at environmentally exposed These conflicting observations might be due to the
bowel disease that can affect surfaces such as the gut. particular subsets of TReg cells that were analysed, dif-
the entire gastrointestinal tract, Recent studies have begun to reveal the mechanisms ferences in experimental methodologies and/or the
but is most common in the
colon and terminal ileum. It is
of intestinal immune modulation by the microbiota. tissues from which TReg cells were harvested. Alternatively,
characterized by transmural A recent study 39 has shown that germ-free animals the particular diet given to the animal might influence
inflammation, strictures and have defective T H17 cell development in the small TReg cell subsets in the intestine, as most animal food
granuloma formation, and is intestine and that the reduction in Il-17 production might have varying amounts of microbial molecules
believed to result from an
is associated with a reciprocal increase in the number (such as TlR ligands), even if autoclaved. However,
abnormal T cell-mediated
immune response to of CD4+FoXP3+ TReg cells in the colon of these mice. these data collectively suggest that intestinal bacteria
commensal bacteria. Reconstitution of these animals with a complex and interact with the mammalian immune system to direct
diverse microbiota that does not contain the promi- the differentiation of both pro- and anti-inflammatory
Ulcerative colitis nent phyla Bacteroidetes does not restore proper T cell populations. Therefore, induction of effector T cell
A chronic disease that
is characterized by
immune balance, suggesting that distinct organisms responses and modulation of TReg cell function by the
inflammation of the mucosa might have the capacity to modulate pro- and anti- microbiota may be a crucial component of diseases such
and sub-mucosa of the large inflammatory responses in the gut. The identity of spe- as IBD. It is possible that different classes (or even spe-
intestine. cific bacterial species and bacterial molecules that are cies) of bacteria induce distinct immunological functions.
involved in regulating the balance between TH cell and Therefore, the equilibrium between inflammation and
Regulatory T (TReg) cell
A specialized type of CD4+ TReg cell subsets in the gut remain unknown. However, homeostasis in the gut could be due to the composition
T cell that can suppress the several common bacterial products are known to have of the microbiota.
responses of other T cells. immunomodulatory effects. For example, ATP gen-
These cells provide a crucial erated by intestinal bacteria specifically increases the IBD and a breakdown in tolerance to gut bacteria. IBD
mechanism for the
maintenance of peripheral
production of Il-17 (but not IFn) in the colon40; con- involves a shift from a regulated intestinal immune
self tolerance and are sistent with this, germ-free animals have reduced Il-17 response to one that is driven by unrestrained immune
characterized by expression and ATP levels in the colon. In addition, DnA from cell activation and pro-inflammatory cytokine produc-
of CD25 (the -chain of the commensal bacteria triggers TlR9 signalling and con- tion4749. The cause of this increase in immune stimulation
interleukin-2 receptor) and the
fers resistance to the enteric parasite Encephalitozoon is of great interest, and several lines of evidence indicate a
transcription factor forkhead
box P3 (fOXP3). cuniculi 41. Accordingly, antibiotic treatment of ani- fundamental role for commensal bacteria in the progres-
mals to eliminate gut bacteria results in increased sion of disease50. Patients with IBD respond favourably
Parasite susceptibility to infection by this parasite. Treatment to antibiotic treatment and faecal diversion and have
An opportunistic organism that of parasite-infected mice with DnA isolated from the higher antibody titres against indigenous bacteria than
maintains a prolonged and
close association with the host,
intestinal microbiota led to the upregulation of T H1 unaffected individuals5153. In addition, inflammatory
which benefits the parasite at and TH17 cell responses, which coincided with sup- lesions are more pronounced in areas of the intestine
the expense of the host. pression of TReg cell activity; this resulted in decreased that contain the highest number of bacteria.
VSL#3 obese people is lower than that of lean people, suggest- Beneficial gut bacteria promote homeostasis
A mixture of bacteria ing that alterations in the normal human microbiota may The evidence described above implicates the microbiota
consisting of four strains of affect disease. Similarly to IBD, obesity seems to have a in shaping immune responses during health and disease,
Lactobacillus (L. casei, strong genetic component 71. The complex interplay but it is still not known which particular organisms are
L. plantarum, L. acidophilus
and L. delbrueckii subspecies
between host genotype and its effects on the microbiota mediating these beneficial responses and, more impor-
bulgaricus), three strains of is an area worthy of investigation, and such studies may tantly, how this is achieved. Here we review the mecha-
Bifidobacterium (B. longum, provide further support for an important role for dysbiosis nisms by which the microbiota affects the mammalian
B. breve and B. infantis) and in metabolic disorders. immune system and the implications for the prevention
Streptococcus salivarius
Similar to the findings in animal studies, dysbiosis has or treatment of IBD.
subspecies thermophilus.
been implicated in IBD in humans, with several stud- In the early 1900s, Ilya Mechnikov was the first to
ies showing a significant alteration in the microbiota of propose the use of live microorganisms to maintain
patients with IBD7274. A recent metagenomic study 74 bowel health and prolong life. now, the term probi-
compared the microbiota of patients with IBD with that otic is used to describe dietary microorganisms that
of non-IBD controls and revealed a statistically signifi- are beneficial to the health of the host 76. As shown in
cant difference in their composition. Specifically, the TABLE 2 , many individual or combinations of bacte-
microbiota of IBD patients showed abnormal microbial rial species have been shown to ameliorate the symp-
composition that was characterized by depletion of two toms of IBD in humans and mouse models. Although
phyla of bacteria, the Firmicutes and Bacteriodetes, which many of these probiotic strains decrease toxic micro-
are both prominently represented in non-IBD controls. bial metabolic activities, more recent evidence shows
longitudinal studies are required to determine whether that these organisms can modulate intestinal immune
this particular profile (that is, the loss of certain classes of responses. The common feature of almost all bacte-
bacteria) can be used as a diagnostic tool to identify people rial species that are used as probiotics is their ability
with a greater likelihood of developing IBD. Although our to control inflammation. Bacterial species can act on
understanding of how dysbiosis might affect IBD is still several cell types (epithelial cells, DCs and T cells), but
at its infancy, new sequencing technologies provide the recent evidence suggests that the induction of TReg cells
means to analyse the microbiomes of numerous healthy by these microorganisms is crucial to their ability to
and diseased individuals75. With increased knowledge of limit inflammation and disease. Treatment of mice
species-specific alterations during disease, the molecular with colitis with the probiotic cocktail VSL#3 increased
mechanisms that link dysbiosis of the microbiota to intes- the production of Il-10 and the percentage of TGF-
tinal inflammation can systematically be explored in both expressing T cells 77. More importantly, transfer of
animal and human studies. lamina propria mononuclear cells from vSl#3-treated
Symbiont mice prevented colitis in recipient mice, indicating that subset abolished the ability of the treated DCs to protect
An organism that lives in the vSl#3 cocktail can initiate the generation of a pro- against disease, suggesting that L. rhamnosus-treated
association with a host (usually tective population of cells. Depletion of TGF-bearing DCs can initiate TReg cell activity 79. It has recently been
for a lifetime) without obvious CD4+ T cells from mice treated with probiotic bacteria shown that some patients with Crohns disease have a
benefit or harm to either
member.
before the transfer of lamina propria cells abolished specific reduction in a prominent gut microorganism,
the protective capacity of these cells77. More recently, Faecalibacterium prausnitzii 80. Intriguingly, oral adminis-
in a model of pathogen-induced inflammation, treat- tration of F. prausnitzii or supernatant from F. prausnitzii
ment of mice with Bifidobacteria infantis led to a reduc- cultures increased the production of Il-10 by periph-
tion in intestinal inflammation and an increase in the eral blood mononuclear cells, reduced the production of
number of CD4+CD25+ TReg cells78. Adoptive transfer of TnF in the colon and ameliorated intestinal disease in
the CD4+CD25+ TReg cell population from mice fed with mice80. This seminal study therefore provides evidence
B. infantis inhibited inflammation-induced activation of for a direct link between a decrease in a particular spe-
nuclear factor-B (nF-B) in recipient mice. cies from the human microbiota and the development of
naive CD4+ T cells can adopt a regulatory pheno- intestinal disease, suggesting that symbiotic microorgan-
type by interacting with intestinal DCs36. Foligne et al.79 isms might have a direct role in maintaining a healthy
show that bone-marrow-derived DCs (BMDCs) internal- gut. The specific molecules produced by these bacterial
ize Lactobacillus rhamnosus, but maintain an immature species remain unknown.
phenotype. Transfer of BMDCs incubated with L. rham- The first demonstration that a single molecule made
nosus could protect against the induction of intestinal by a commensal microorganism could promote ben-
disease79. Moreover, depletion of the CD4+CD25+ T cell eficial immune responses was provided by the identi-
fication of polysaccharide A (PSA), which is produced
by the human symbiont Bacteroides fragilis (fIG. 2) .
Intestinal lumen H. hepaticus
Colonization of germ-free mice with B. fragilis or treat-
B. fragilis Pro-inflammatory ment with purified PSA directs the development of
signals the immune system, including the expansion and dif-
+++++
PSA - - - - - Epithelial cell ferentiation of splenic CD4+ T cells81. PSA has several
immunomodulatory activities in germ-free mice that
are colonized by B. fragilis, and these include correcting
systemic T cell deficiencies, restoring balance between
TH cell subsets and directing lymphoid organogenesis.
The importance of B. fragilis in maintaining a healthy
immune response was recently illustrated by the find-
+++
- -
ing that colonization of germ-free mice by B. fragilis
DC or treatment with purified PSA can protect against the
TH17 cell
IL-10 induction of experimental IBD66. Moreover, mice that
MHC class II
IL-17, IL-23 are colonized by a mutant form of this microorganism
and TNF that lacks expression of PSA (B. fragilis PSA) are no
TCR TH1 cell
longer protected from the disease. Furthermore, oral
treatment of mice with purified PSA protects recipi-
++
- -
- -
++
Symbiosis For many years, Il-10-producing regulatory T cells absence of these symbiotic organisms and their ben-
A constant and intimate (known as TR1 cells) were considered to be distinct from eficial molecules? In other words, if symbiosis factors
relationship that occurs naturally occurring, thymic-derived CD4+CD25+FoXP3+ actively maintain health, can dysbiosis compromise
between dissimilar species, TReg cells82. It is now apparent that there is overlap bacterium-mediated immune regulation and lead to
which was originally defined
as living together. Although
between these two populations and that Il-10-producing inflammation? Recent epidemiological and clinical
it is often used to describe a TR1 cells can be found in the FoXP3+ TReg cell subset reports have described marked increases in the inci-
beneficial relationship, and are crucial for the control of experimental colitis. dence of several immune-mediated disorders, such
symbiosis does not necessarily Although TR1 cell clones that are specific for caecal as IBD, asthma, atopy (affecting the skin, respiratory
imply that either partner gains
bacterial contents have been generated83, the ability of tract and gut), rheumatoid arthritis, type 1 diabetes and
an advantage.
a molecule from symbiotic bacteria to regulate FoXP3+ multiple sclerosis, in Western populations, including
TReg cell differentiation and function awaits further vali- those of western european nations, the United States
dation. nevertheless, current evidence supports that idea and Japan. The rapidity of the increase in disease
that certain beneficial bacteria have evolved molecules rates does not support a solely genetic basis for these
(known as symbiosis factors) that induce protective intes- observations84. Instead, this could be due to changes
tinal immune responses. Knowledge of which beneficial in hostmicroorganism interactions caused by the
species of bacteria can prevent or cure disease, and har- implementation of antimicrobial strategies (including
nessing the potent immunosuppressive potential of sym- vaccination, sanitation, Western diets and antibacterial
biosis factors will be important steps towards designing therapeutics) that do not discriminate between infec-
new and natural therapeutics for IBD. tious and non-infectious microorganisms (fIG. 3).
If improvements in hygiene and health care have
The microbiota hypothesis and human disease altered the process by which a healthy microbiota is
Does harbouring certain strains of bacteria predispose assembled and maintained, then patients with these
an individual to disease or protect from it? B. fragilis has diseases in developed countries should display signs
been shown to protect its host from inflammatory dis- of dysbiosis. This indeed seems to be the case, at least
ease caused by subsequent infection with H. hepaticus in according to a growing number of studies that are now
an animal model of experimental colitis66. As symbiotic linking these diseases (which are mostly prevalent in
bacteria seem to have evolved mechanisms to provide Western populations) to alterations in the microbiota.
protection from colonization by pathobionts that are The bacterial composition of the intestines of patients
present in the microbiota, does disease result from the with IBD is known to differ from that of healthy con-
trols74. However, studies that aimed to identify the spe-
cific pathogenic organism (or organisms) that triggers
Host genetics Lifestyle Early colonization Medical practices inflammation have repeatedly identified reactions only
Mutations in Diet Birth in hospitals Vaccination use to intestinal bacteria that are shared by all humans
NOD2, IL23R, Stress Altered exposure Antibiotic
ATG16L and IGRM to microbes Hygiene healthy and ill. So far, no infectious organisms have been
conclusively shown to be the causative agents of IBD.
This raises the possibility that the targets of inflamma-
tion in IBD are not pathogens and instead are pathobi-
Dysbiosis onts that are overrepresented during dysbiosis (fIG. 1).
Indeed, in 1999 an investigation of the role of intestinal
bacteria in the development of asthma concluded that
Disease Health allergic children from Sweden and estonia had lower
TH1, TH2 and TH17 cells TReg cells levels of colonization by Bacteroides spp. and higher lev-
els of colonization by aerobic microorganisms than non-
Figure 3 | Proposed causes of dysbiosis of the microbiota. We propose that the allergic children from either region85. epidemiological
composition of the microbiota can shape a healthy immune Nature Reviews
response | Immunology
or predispose studies have provided evidence for a link between altered
to disease. Many factors can contribute to dysbiosis, including host genetics, lifestyle,
intestinal microbiota to other allergic disorders, such as
exposure to microorganisms and medical practices. host genetics can potentially
influence dysbiosis in many ways. An individual with mutations in genes involved atopic eczema and rheumatoid arthritis8688. Although it
in immune regulatory mechanisms or pro-inflammatory pathways could lead to is not clear whether dysbiosis is the cause or an effect of
unrestrained inflammation in the intestine. It is possible that inflammation alone disease, it seems that deviations in the composition of
influences the composition of the microbiota, skewing it in favour of pathobionts. the gut microbiota may be one factor underlying the
Alternatively, a host could select or exclude the colonization of particular organisms. development of disease in genetically predisposed
This selection can be either active (as would be the case of an organism recognizing a individuals.
particular receptor on the host) or passive (the host environment is more conducive to on these basis of these recent studies, investiga-
fostering the growth of select organisms). Selection of pathobionts by the host could tors are now turning their attention to understanding
tip the balance in favour of inflammation. Diet and stress also have the potential to how (and, more importantly, why) mammals harbour
influence the microbiota103. Birth in the sterile environment of hospitals can protect
multitudes of symbiotic bacteria. As discussed above,
from exposure to dangerous pathogens, but can also prevent early exposure to
health-promoting bacteria. overuse of vaccination and antibiotics, which do not the effects of the microbiota on the immune system
distinguish between pathogenic or symbiotic microorganisms, could adversely alter are becoming increasingly evident. It is therefore of
the microbiota. ATG16L, autophagy-related gene 16-like; IGRM, immunity-related great interest that the immune-mediated disorders,
GTPase family, M; IL23R, interleukin-23 receptor; NOD2, nucleotide-binding the incidence of which has increased in Western coun-
oligomerization domain 2; Th, T helper; TReg, regulatory T. tries, seem to involve reduced TReg cell activity. Indeed,
studies in animal models and humans indicate that alter the outcome of immune development and poten-
deficiencies in TReg cell populations or function underlie tially predispose individuals to various inflammatory
asthma, IBD, rheumatoid arthritis, type 1 diabetes and diseases later in life (fIG. 3).
multiple sclerosis89 and that CD4+CD25+FoXP3+ TReg on the basis of clinical, epidemiological and immuno-
cells can prevent, and in some cases treat, these disor- logical evidence, it seems possible that changes in the intes-
ders in laboratory animals. In addition, the observed tinal microbiota may be an essential factor in the incidence
reduction in the numbers and function of certain TReg of numerous inflammatory disorders. It is conceivable
cell populations 45,43, together with numerous other that the absence of beneficial microorganisms (owing to
immunological defects that may precipitate disease in dysbiosis) that promote the appropriate development of
germ-free animals, implicates a role for the microbiota the immune system leads to the induction of inflamma-
in actively supporting health. After millions of years of tory responses and immune-mediated disease. Recent
co-evolution, have societal advances paradoxically and studies have shown that at least for experimental IBD,
adversely affected human health by reducing our exposure spontaneous disease occurs when immune suppression
to health-promoting bacteria? is defective; thus, inflammation seems to be a default
immunological state in the absence of regulation 93.
Concluding remarks Pathogenic bacteria clearly induce local inflammation
Although it has been known for decades that we har- during acute infections, but have symbiotic bacteria
bour millions of commensal bacteria, recent studies evolved to regulate the inflammatory processes that
have only just begun to reveal the extraordinary com- are harmful to the host (and therefore, harmful to the
plexity and diversity of the human microbiota. This con- existence of the symbiont)? Research has implicated
sortium of bacteria contains tenfold more cells than the innate and adaptive immune suppression during the
human body, 100 times the number of genes than control of disorders such as IBD, autoimmunity, asthma
the human genome and has the metabolic capacity of the and allergy, cancer and infectious diseases. According
human liver 90,91. How is such a complex microbial net- to numerous recent studies, we propose that there is a
work assembled after birth? A recent study 92 on devel- vast, intricate and unexpected level of interdependence
opment of the intestinal microbiota in infants revealed between beneficial bacteria and the immune system.
that in the first few days to weeks of life, the microbiota It is possible that genetic and habitual factors shape the
of newborns is highly variable and subject to waves of composition of the microbiota, which in turn shapes
temporal fluctuations (possibly representing a time the immune system of individuals that are predisposed to
of sampling or trial and error) to coordinately assem- inflammatory disease (fIG. 3). The recent identification of
ble a stable microbiota. The first years of life are also symbiotic bacteria with potent anti-inflammatory prop-
a time of great post-natal development of the immune erties, and their correlative absence during disease,
system. As the microbiota has marked influences on the suggests that certain aspects of human health may
immune system, deviations from the normal develop- depend on the status of the microbiota. The medi-
ment of the microbiota (through modern strategies cal and social reconsideration of the microbial world
such a caesarean section, formula-based diet, hygiene, may have profound consequences for the health of our
vaccination and use of antimicrobials in infants) may future generations.
1. Ley, R. E., Peterson, D. A. & Gordon, J. I. Ecological 11. Falk, P. G., Hooper, L. V., Midtvedt, T. & Gordon, J. I. 16. Sonnenburg, J. L., Chen, C. T. & Gordon, J. I. Genomic
and evolutionary forces shaping microbial diversity Creating and maintaining the gastrointestinal and metabolic studies of the impact of probiotics on a
in the human intestine. Cell 124, 837848 (2006). ecosystem: what we know and need to know model gut symbiont and host. PLoS Biol. 4, e413
2. Dethlefsen, L., McFall-Ngai, M. & Relman, D. A. from gnotobiology. Microbiol. Mol. Biol. Rev. (2006).
An ecological and evolutionary perspective on 62, 11571170 (1998). Using co-colonization of germ-free mice with
humanmicrobe mutualism and disease. Nature 12. Bouskra, D. et al. Lymphoid tissue genesis induced B. thetaiotaomicron (a symbiont) and B. longum (a
449, 811818 (2007). by commensals through NOD1 regulates intestinal probiotic), this study shows that B. longum can
3. Hooper, L. V. Bacterial contributions to mammalian homeostasis. Nature 456, 507510 (2008). increase the diversity of polysaccharides that can
gut development. Trends Microbiol. 12, 129134 This study shows that peptidoglycans from be degraded by B. thetaiotaomicron, demonstrating
(2004). Gram-negative bacteria induce the generation that distinct intestinal bacterial species can affect
4. Mazmanian, S. K. & Kasper, D. L. The lovehate of ILFs through the recognition of NOD1. In the each others function.
relationship between bacterial polysaccharides absence of ILF formation, marked changes in 17. Sprinz, H. et al. The response of the germfree guinea
and the host immune system. Nature Rev. Immunol. the composition of the microbiota occur. pig to oral bacterial challenge with Escherichia coli and
6, 849858 (2006). 13. Abrams, G. D., Bauer, H. & Sprinz, H. Influence of Shigella flexneri. Am. J. Pathol. 39, 681695 (1961).
5. Peterson, D. A., Frank, D. N., Pace, N. R. & the normal flora on mucosal morphology and cellular 18. Maier, B. R. & Hentges, D. J. Experimental Shigella
Gordon, J. I. Metagenomic approaches for defining renewal in the ileum. A comparison of germ-free and infections in laboratory animals. I. Antagonism by
the pathogenesis of inflammatory bowel diseases. conventional mice. Lab. Invest. 12, 355364 (1963). human normal flora components in gnotobiotic mice.
Cell Host Microbe 3, 417427 (2008). 14. Bry, L., Falk, P. G., Midtvedt, T. & Gordon, J. I. Infect. Immun. 6, 168173 (1972).
6. Frank, D. N. & Pace, N. R. Gastrointestinal A model of hostmicrobial interactions in an open 19. Zachar, Z. & Savage, D. C. Microbial interference and
microbiology enters the metagenomics era. mammalian ecosystem. Science 273, 13801383 colonization of the murine gastrointestinal tract by
Curr. Opin. Gastroenterol. 24, 410 (2008). (1996). Listeria monocytogenes. Infect. Immun. 23, 168174
7. Ley, R. E. et al. Evolution of mammals and their gut 15. Cash, H. L., Whitham, C. V., Behrendt, C. L. & (1979).
microbes. Science 320, 16471651 (2008). Hooper, L. V. Symbiotic bacteria direct expression 20. Inagaki, H., Suzuki, T., Nomoto, K. & Yoshikai, Y.
8. Hooper, L. V. & Gordon, J. I. Commensal hostbacterial of an intestinal bactericidal lectin. Science 313, Increased susceptibility to primary infection with
relationships in the gut. Science 292, 11151118 11261130 (2006). Listeria monocytogenes in germfree mice may be due
(2001). These authors show that microbial colonization to lack of accumulation of L-selectin+ CD44+ T cells in
9. Macpherson, A. J. & Harris, N. L. Interactions between of germ-free mice induces the production of sites of inflammation. Infect. Immun. 64, 32803287
commensal intestinal bacteria and the immune REG3, a secreted C-type lectin. REG3 is shown (1996).
system. Nature Rev. Immunol. 4, 478485 (2004). to have antimicrobial activity by binding to 21. Nardi, R. M., Silva, M. E., Vieira, E. C., Bambirra, E. A.
10. Zaneveld, J. et al. Hostbacterial coevolution and the peptidoglycans, suggesting that microbial species & Nicoli, J. R. Intragastric infection of germfree and
search for new drug targets. Curr. Opin. Chem. Biol. actively shape the intestinal environment to their conventional mice with Salmonella typhimurium.
12, 109114 (2008). advantage. Braz. J. Med. Biol. Res. 22, 13891392 (1989).
22. Stecher, B. et al. Salmonella enterica serovar 47. De Winter, H., Cheroutre, H. & Kronenberg, M. These authors show that the microbiome from
typhimurium exploits inflammation to compete with Mucosal immunity and inflammation. II. The yin obese mice has an increased capacity for energy
the intestinal microbiota. PLoS Biol. 5, e244 (2007). and yang of T cells in intestinal inflammation: harvest. Transfer of the microbiota to non-obese
These authors show that the intestinal pathogen pathogenic and protective roles in a mouse colitis mice increases their mean fat body weight,
S. Typhimurium uses inflammation to disturb the model. Am. J. Physiol. 276, G1317G1321 suggesting that a change in the microbiota can
commensal microbiota to induce disease. (1999). induce obesity.
23. Nardi, R. M. et al. Bacteriological and immunological 48. Simpson, S. J., de Jong, Y. P., Comiskey, M. & 70. Ley, R. E., Turnbaugh, P. J., Klein, S. & Gordon, J. I.
aspects of conventional and germfree mice infected Terhorst, C. Pathways of T cell pathology in models Microbial ecology: human gut microbes associated
with Salmonella typhimurium. Rev. Latinoam. of chronic intestinal inflammation. Int. Rev. Immunol. with obesity. Nature 444, 10221023 (2006).
Microbiol. 33, 239243 (1991). 19, 137 (2000). 71. Pomp, D., Nehrenberg, D. & Estrada-Smith, D.
24. Podolsky, D. K. The current future understanding of 49. Elson, C. O. et al. Monoclonal anti-interleukin 23 Complex genetics of obesity in mouse models.
inflammatory bowel disease. Best Pract. Res. Clin. reverses active colitis in a T cell-mediated model in Annu. Rev. Nutr. 28, 331345 (2008).
Gastroenterol. 16, 933943 (2002). mice. Gastroenterology 132, 23592370 (2007). 72. Lepage, P. et al. Biodiversity of the mucosa-associated
25. Shanahan, F. Crohns disease. Lancet 359, 6269 50. Sartor, R. B. The influence of normal microbial microbiota is stable along the distal digestive tract in
(2002). flora on the development of chronic mucosal healthy individuals and patients with IBD. Inflamm.
26. Targan, S. R. & Karp, L. C. Defects in mucosal inflammation. Res. Immunol. 148, 567576 Bowel Dis. 11, 473480 (2005).
immunity leading to ulcerative colitis. Immunol. Rev. (1997). 73. Scanlan, P. D., Shanahan, F., OMahony, C. &
206, 296305 (2005). 51. Macpherson, A., Khoo, U. Y., Forgacs, I., Philpott- Marchesi, J. R. Culture-independent analyses of
27. Bouma, G. & Strober, W. The immunological and Howard, J. & Bjarnason, I. Mucosal antibodies in temporal variation of the dominant fecal microbiota
genetic basis of inflammatory bowel disease. Nature inflammatory bowel disease are directed against and targeted bacterial subgroups in Crohns disease.
Rev. Immunol. 3, 521533 (2003). intestinal bacteria. Gut 38, 365375 (1996). J. Clin. Microbiol. 44, 39803988 (2006).
28. Kullberg, M. C. et al. IL-23 plays a key role in 52. Elson, C. O. Commensal bacteria as targets in 74. Frank, D. N. et al. Molecular-phylogenetic
Helicobacter hepaticus-induced T cell-dependent Crohns disease. Gastroenterology 119, 254257 characterization of microbial community imbalances in
colitis. J. Exp. Med. 203, 24852494 (2006). (2000). human inflammatory bowel diseases. Proc. Natl Acad.
29. Hue, S. et al. Interleukin-23 drives innate and T cell- 53. Tannock, G. W. Exploring the relationships between Sci. USA 104, 1378013785 (2007).
mediated intestinal inflammation. J. Exp. Med. 203, intestinal microflora and inflammatory conditions of This study indicates that the intestinal microbial
24732483 (2006). the human bowel and spine. Antonie Van Leeuwenhoek populations in patients with IBD and non-IBD
30. Duerr, R. H. et al. A genome-wide association study 81, 529535 (2002). patients differ greatly.
identifies IL23R as an inflammatory bowel disease 54. Kent, T. H., Summers, R. W., DenBesten, L., Swaner, 75. Turnbaugh, P. J. et al. The human microbiome project.
gene. Science 314, 14611463 (2006). J. C. & Hrouda, M. Effect of antibiotics on bacterial Nature 449, 804810 (2007).
31. Schmechel, S. et al. Linking genetic susceptibility to flora of rats with intestinal blind loops. Proc. Soc. Exp. 76. Sartor, R. B. Therapeutic manipulation of the enteric
Crohns disease with Th17 cell function: IL-22 serum Biol. Med. 132, 6367 (1969). microflora in inflammatory bowel diseases: antibiotics,
levels are increased in Crohns disease and correlate 55. Videla, S. et al. Role of intestinal microflora in chronic probiotics, and prebiotics. Gastroenterology 126,
with disease activity and IL23R genotype status. inflammation and ulceration of the rat colon. Gut 35, 16201633 (2004).
Inflamm. Bowel Dis. 14, 204212 (2008). 10901097 (1994). 77. Di Giacinto, C., Marinaro, M., Sanchez, M.,
32. Kobayashi, T. et al. IL23 differentially regulates the 56. Taurog, J. D. et al. The germfree state prevents Strober, W. & Boirivant, M. Probiotics ameliorate
Th1/Th17 balance in ulcerative colitis and Crohns development of gut and joint inflammatory disease recurrent Th1-mediated murine colitis by inducing
disease. Gut 57, 16821689 (2008). in HLA-B27 transgenic rats. J. Exp. Med. 180, IL-10 and IL-10-dependent TGF--bearing
33. Sartor, R. B. Microbial influences in inflammatory 23592364 (1994). regulatory cells. J. Immunol. 174, 32373246 (2005).
bowel diseases. Gastroenterology 134, 577594 57. Rath, H. C. Role of commensal bacteria in chronic 78. OMahony, C. et al. Commensal-induced regulatory
(2008). experimental colitis: lessons from the HLA-B27 T cells mediate protection against pathogen-
34. Fontenot, J. D. & Rudensky, A. Y. A well adapted transgenic rat. Pathobiology 70, 131138 (2002). stimulated NF-B activation. PLoS Pathog. 4,
regulatory contrivance: regulatory T cell development 58. Sellon, R. K. et al. Resident enteric bacteria are e1000112 (2008).
and the forkhead family transcription factor Foxp3. necessary for development of spontaneous colitis 79. Foligne, B. et al. A key role for dendritic cells in
Nature Immunol. 6, 331337 (2005). and immune system activation in interleukin-10- probiotic functionality. PLoS ONE 2, e313 (2007).
35. Vignali, D. A., Collison, L. W. & Workman, C. J. deficient mice. Infect. Immun. 66, 52245231 80. Sokol, H. et al. Faecalibacterium prausnitzii is an
How regulatory T cells work. Nature Rev. Immunol. (1998). anti-inflammatory commensal bacterium identified by
8, 523532 (2008). 59. Cahill, R. J. et al. Inflammatory bowel disease: an gut microbiota analysis of Crohn disease patients.
36. Coombes, J. L. et al. A functionally specialized immunity-mediated condition triggered by bacterial Proc. Natl Acad. Sci. USA 105, 1673116736 (2008).
population of mucosal CD103+ DCs induces Foxp3+ infection with Helicobacter hepaticus. Infect. Immun. These authors show that F. prausnitzii is
regulatory T cells via a TGF- and retinoic acid- 65, 31263131 (1997). specifically reduced in the intestine of patients
dependent mechanism. J. Exp. Med. 204, 60. Kullberg, M. C. et al. Induction of colitis by a CD4+ with Crohns disease. In addition, this bacterium
17571764 (2007). T cell clone specific for a bacterial epitope. Proc. Natl is shown to have an anti-inflammatory capacity
37. Powrie, F. & Maloy, K. J. Immunology. Regulating the Acad. Sci. USA 100, 1583015835 (2003). and to protect animals from disease when given
regulators. Science 299, 10301031 (2003). 61. Barnich, N. et al. CEACAM6 acts as a receptor for orally, suggesting that symbiotic microorganisms
38. Collison, L. W. et al. The inhibitory cytokine IL-35 adherent-invasive E. coli, supporting ileal mucosa may be directly involved in maintaining health.
contributes to regulatory T-cell function. Nature 450, colonization in Crohn disease. J. Clin. Invest. 117, 81. Mazmanian, S. K., Liu, C. H., Tzianabos, A. O. &
566569 (2007). 15661574 (2007). Kasper, D. L. An immunomodulatory molecule of
39. Ivanov, II. et al. Specific microbiota direct the 62. Hampe, J. et al. Association between insertion symbiotic bacteria directs maturation of the host
differentiation of IL-17-producing T-helper cells in the mutation in NOD2 gene and Crohns disease in immune system. Cell 122, 107118 (2005).
mucosa of the small intestine. Cell Host Microbe 4, German and British populations. Lancet 357, 82. Roncarolo, M. G. & Battaglia, M. Regulatory T-cell
337349 (2008). 19251928 (2001). immunotherapy for tolerance to self antigens and
40. Atarashi, K. et al. ATP drives lamina propria TH17 cell 63. Kim, S. C., Tonkonogy, S. L., Karrasch, T., Jobin, C. & alloantigens in humans. Nature Rev. Immunol. 7,
differentiation. Nature 455, 808812 (2008). Sartor, R. B. Dual-association of gnotobiotic Il-10/ 585598 (2007).
41. Hall, J. A. et al. Commensal DNA limits regulatory mice with 2 nonpathogenic commensal bacteria 83. Cong, Y. et al. Generation of antigen-specific, Foxp3-
T cell conversion and is a natural adjuvant of induces aggressive pancolitis. Inflamm. Bowel Dis. expressing CD4+ regulatory T cells by inhibition of
intestinal immune responses. Immunity 29, 637649 13, 14571466 (2007). APC proteosome function. J. Immunol. 174,
(2008). 64. OHara, A. M. & Shanahan, F. The gut flora as a 27872795 (2005).
42. Rakoff-Nahoum, S., Paglino, J., Eslami-Varzaneh, F., forgotten organ. EMBO Rep. 7, 688693 (2006). 84. Noverr, M. C. & Huffnagle, G. B. Does the microbiota
Edberg, S. & Medzhitov, R. Recognition of 65. Ley, R. E., Knight, R. & Gordon, J. I. The human regulate immune responses outside the gut? Trends
commensal microflora by Toll-like receptors is microbiome: eliminating the biomedical/environmental Microbiol. 12, 562568 (2004).
required for intestinal homeostasis. Cell 118, dichotomy in microbial ecology. Environ. Microbiol. 85. Bjorksten, B. The environmental influence on
229241 (2004). 9, 34 (2007). childhood asthma. Allergy 54, S17S23 (1999).
This is one of the first studies to suggest that 66. Mazmanian, S. K., Round, J. L. & Kasper, D. L. 86. Penders, J. et al. Gut microbiota composition and
recognition of TLR ligands from commensal A microbial symbiosis factor prevents intestinal development of atopic manifestations in infancy:
bacteria by the host is important for the inflammatory disease. Nature 453, 620625 the KOALA Birth Cohort Study. Gut 56, 661667
maintenance of intestinal homeostasis. (2008). (2007).
43. Ostman, S., Rask, C., Wold, A. E., Hultkrantz, S. & 67. Glimcher, L. H. Trawling for treasure: tales of T-bet. 87. Kalliomaki, M. & Isolauri, E. Pandemic of atopic
Telemo, E. Impaired regulatory T cell function in Nature Immunol. 8, 448450 (2007). diseases a lack of microbial exposure in early
germ-free mice. Eur. J. Immunol. 36, 23362346 68. Garrett, W. S. et al. Communicable ulcerative colitis infancy? Curr. Drug Targets. Infect. Disord. 2,
(2006). induced by T-bet deficiency in the innate immune 193199 (2002).
44. Ishikawa, H. et al. Effect of intestinal microbiota on the system. Cell 131, 3345 (2007). 88. Kalliomaki, M. & Isolauri, E. Role of intestinal flora in
induction of regulatory CD25+ CD4+ T cells. Clin. Exp. This study shows that mice that lack T-bet the development of allergy. Curr. Opin. Allergy Clin.
Immunol. 153, 127135 (2008). expression in innate immune cells develop Immunol. 3, 1520 (2003).
45. Strauch, U. G. et al. Influence of intestinal bacteria spontaneous colitis. Moreover, transfer of the 89. Sakaguchi, S. et al. Foxp3+ CD25+ CD4+ natural
on induction of regulatory T cells: lessons from a microbiota from these mice is shown to induce regulatory T cells in dominant self-tolerance and
transfer model of colitis. Gut 54, 15461552 (2005). disease in wild-type recipient mice. autoimmune disease. Immunol. Rev. 212, 827
46. Zaph, C. et al. Commensal-dependent expression of 69. Turnbaugh, P. J. et al. An obesity-associated gut (2006).
IL-25 regulates the IL-23IL-17 axis in the intestine. microbiome with increased capacity for energy 90. Eckburg, P. B. et al. Diversity of the human intestinal
J. Exp. Med. 205, 21912198 (2008). harvest. Nature 444, 10271031 (2006). microbial flora. Science 308, 16351638 (2005).
91. Gill, S. R. et al. Metagenomic analysis of the human 97. Macpherson, A. J. & Uhr, T. Induction of protective IgA Acknowledgements
distal gut microbiome. Science 312, 13551359 by intestinal dendritic cells carrying commensal We thank members of the Mazmanian laboratory for their
(2006). bacteria. Science 303, 16621665 (2004). critical review of the manuscript. We apologize to those whose
92. Palmer, C., Bik, E. M., Digiulio, D. B., Relman, D. A. & 98. Macpherson, A. J., Geuking, M. B. & McCoy, K. D. work could not be mentioned owing to space limitations and
Brown, P. O. Development of the human infant Immune responses that adapt the intestinal mucosa to the scope of the manuscript, in particular the vast clinical
intestinal microbiota. PLoS Biol. 5, e177 (2007). commensal intestinal bacteria. Immunology 115, data on inflammatory bowel disease. J.L.R is a Merck Fellow
93. Izcue, A. et al. Interleukin-23 restrains regulatory 153162 (2005). of the Jane Coffin Childs Memorial Fund. S.K.M. is a Searle
T cell activity to drive T cell-dependent colitis. 99. He, B. et al. Intestinal bacteria trigger T cell- Scholar. Work in our laboratory is supported by funding from
Immunity 28, 559570 (2008). independent immunoglobulin A2 class switching by the National Institutes of Health, USA, Damon Runyon Cancer
This report shows that FOXP3-deficient T cells inducing epithelial-cell secretion of the cytokine APRIL. Research Foundation and the Crohns & Colitis Foundation of
can induce colitis in IL-23-deficient recipients, Immunity 26, 812826 (2007). America to S.K.M.
suggesting that disease can occur in the absence 100. Cerutti, A. The regulation of IgA class switching.
of regulation. Nature Rev. Immunol. 8, 421434 (2008).
94. Moreau, M. C., Ducluzeau, R., Guy-Grand, D. & 101. Tezuka, H. et al. Regulation of IgA production by DATABASES
Muller, M. C. Increase in the population of duodenal naturally occurring TNF/iNOS-producing dendritic Entrez Gene: http://www.ncbi.nlm.nih.gov/entrez/query.
immunoglobulin A plasmocytes in axenic mice cells. Nature 448, 929933 (2007). fcgi?db=gene
associated with different living or dead bacterial strains 102. Peterson, D. A., McNulty, N. P., Guruge, J. L. & FoxP3 | IFN | IL-10 | IL-17 | IL23R | IL-35 | NOD2 | Tbx21 |
of intestinal origin. Infect. Immun. 21, 532539 (1978). Gordon, J. I. IgA response to symbiotic bacteria as a TGF | TLR9 | TNF
95. Suzuki, K. et al. Aberrant expansion of segmented mediator of gut homeostasis. Cell Host Microbe 2,
filamentous bacteria in IgA-deficient gut. Proc. Natl 328339 (2007). FURTHER INFORMATION
Acad. Sci. USA 101, 19811986 (2004). 103. OMahony, S. M. et al. Early life stress alters behavior, Sarkis K. Mazmanians homepage:
96. Kroese, F. G., de Waard, R. & Bos, N. A. B-1 cells and immunity, and microbiota in rats: implications for http://biology.caltech.edu/Members/Mazmanian
their reactivity with the murine intestinal microflora. irritable bowel syndrome and psychiatric illnesses. All lInkS Are ActIve In the onlIne PDf
Semin. Immunol. 8, 1118 (1996). Biol. Psychiatry 65, 263267 (2009).
1. Mimura, T. et al. Once daily high dose probiotic 1917 is as effective as with standard mesalazine. effect on tumor necrosis factor -induced interleukin-8
therapy (VSL#3) for maintaining remission in recurrent Gut 53, 16171623 (2004). expression. Infect. Immun. 72, 53085314 (2004).
or refractory pouchitis. Gut 53, 108114 (2004). 7. Grabig, A. et al. Escherichia coli strain Nissle 1917 12. Schultz, M. et al. Lactobacillus plantarum 299V in
2. Bibiloni, R. et al. VSL#3 probiotic-mixture induces ameliorates experimental colitis via Toll-like receptor the treatment and prevention of spontaneous colitis
remission in patients with active ulcerative colitis. 2- and Toll-like receptor 4-dependent pathways. in interleukin-10-deficient mice. Inflamm. Bowel Dis.
Am. J. Gastroenterol. 100, 15391546 (2005). Infect. Immun. 74, 40754082 (2006). 8, 7180 (2002).
3. Di Giacinto, C., Marinaro, M., Sanchez, M., Strober, W. 8. Foligne, B. et al. Correlation between in vitro and 13. Peran, L. et al. A comparative study of the
& Boirivant, M. Probiotics ameliorate recurrent in vivo immunomodulatory properties of lactic acid preventative effects exerted by two probiotics,
Th1-mediated murine colitis by inducing IL-10 and bacteria. World J. Gastroenterol. 13, 236243 Lactobacillus reuteri and Lactobacillus fermentum,
IL-10-dependent TGF--bearing regulatory cells. (2007). in the trinitrobenzenesulfonic acid model of rat colitis.
J. Immunol. 174, 32373246 (2005). 9. Amit-Romach, E., Uni, Z. & Reifen, R. Therapeutic Br. J. Nutr. 97, 96103 (2007).
4. Peran, L. et al. A comparative study of the potential of two probiotics in inflammatory 14. Kelly, D. et al. Commensal anaerobic gut bacteria
preventative effects exerted by three probiotics, bowel disease as observed in the trinitrobenzene attenuate inflammation by regulating nuclear-
Bifidobacterium lactis, Lactobacillus casei and sulfonic acid model of colitis. Dis. Colon Rectum cytoplasmic shuttling of PPAR- and RelA. Nature
Lactobacillus acidophilus, in the TNBS model of rat 51, 18281836 (2008). Immunol. 5, 104112 (2004).
colitis. J. Appl. Microbiol. 103, 836844 (2007). 10. Madsen, K. L., Doyle, J. S., Jewell, L. D., 15. Mazmanian, S. K., Round, J. L. & Kasper, D. L.
5. OMahony, C. et al. Commensal-induced regulatory Tavernini, M. M. & Fedorak, R. N. Lactobacillus A microbial symbiosis factor prevents intestinal
T cells mediate protection against pathogen- species prevents colitis in interleukin 10 gene- inflammatory disease. Nature 453, 620625 (2008).
stimulated NF-B activation. PLoS Pathog. 4, deficient mice. Gastroenterology 116, 11071114 16. Sokol, H. et al. Faecalibacterium prausnitzii is an anti-
e1000112 (2008). (1999). inflammatory commensal bacterium identified by gut
6. Kruis, W. et al. Maintaining remission of ulcerative 11. Ma, D, Forsythe, P. & Bienenstock, J. Live microbiota analysis of Crohn disease patients. Proc.
colitis with the probiotic Escherichia coli Nissle Lactobacillus reuteri is essential for the inhibitory Natl Acad. Sci. USA 105, 1673116736 (2008).