PERSPECTIVES

MICROBIOME
into question13–16. We would argue that
OPINION
the concept of a pathogen is still valid for
studying communicable diseases, such
The germ-organ theory of as tuberculosis, but it does not provide
a theoretical framework that adequately
non-communicable diseases captures the contribution of microbial
communities to non-communicable diseases
(BOX 2). In other words, rather than omitting
Mariana X. Byndloss and Andreas J. Bäumler the term ‘pathogen’, the core concept of the
germ theory, we should be mindful of using
Abstract | Gut dysbiosis is associated with many non-communicable human it in the context of communicable diseases
diseases, but the mechanisms maintaining homeostasis remain incompletely while independently developing a new
understood. Recent insights suggest that during homeostasis, epithelial hypoxia theory to guide studies on the microbial
limits oxygen availability in the colon, thereby maintaining a balanced microbiota origin of non-communicable diseases. In
that functions as a microbial organ, producing metabolites contributing to host this Opinion article, we discuss the most
recent findings of microbiota research in
nutrition, immune education and niche protection. Dysbiosis is characterized by a search of a useful theoretical framework
shift in the microbial community structure from obligate to facultative anaerobes, that adequately captures the contribution
suggesting oxygen as an important ecological driver of microbial organ of gut-associated microbial communities
dysfunction. The ensuing disruption of gut homeostasis can lead to non- to human non-communicable diseases
communicable disease because microbiota-derived metabolites are either and its implication for guiding future
research efforts.
depleted or generated at harmful concentrations. This Opinion article describes
the concept that host control over the microbial ecosystem in the colon is critical The microbial organ of the large bowel
for the composition and function of our microbial organ, which provides a More than 2 millennia have passed
theoretical framework for linking microorganisms to non-communicable diseases. since Hippocrates (circa 460–370 bce)
commented “all disease begins in the gut”,
but in light of recent data revealing how
The germ theory triggered a scientific and even neurological disorders11,12. microorganisms that inhabit our body
revolution during the time of Robert However, research into the microbial origin affect human health, this quote feels more
Koch (1843–1910) and Louis Pasteur of non-communicable human diseases timely than ever 17. The gut microbiota, the
(1822–1895) by guiding efforts to identify differs in one important aspect from the largest microbial community inhabiting
bacterial pathogens as leading causes studies performed at the end of the 19th our body, fulfils many functions important
of communicable diseases (BOX 1). By century: the identification of pathogens as for human physiology, including nutrition,
developing culture-dependent methods for causes of communicable human diseases development of the immune system and
studying microorganisms, Louis Pasteur was guided by Louis Pasteur’s germ theory, niche protection against enteric pathogens18
laid the foundation for a new discipline, whereas contemporary studies on the role (FIG. 1). The microbiota of the large bowel
bacteriology. Similarly, recent advances in of the gut microbiota in health and disease is of particular importance and forms the
culture-independent methods for analysing lack an overarching theoretical framework focus of this Opinion article because in the
microorganisms have opened up a new for guidance (BOX 2) and therefore colon, bacterial density reaches a staggering
field of study, microbiota research, that remain heavily focused on technological 1011 organisms per gram, making this
parallels the remarkable advances made development, census taking and predicting bacterial community the principal source of
during the ‘golden age’ of microbiology physiological potential. microbial metabolites in the human body.
in the 19th century. Whereas the initial The germ theory influenced initial In turn, these metabolites are responsible for
use of culture-dependent methods by studies on the microbial origin of non- the beneficial effects of the gut microbiota
Louis Pasteur and his contemporaries communicable diseases as investigators on host physiology 18. It was thus proposed
established a microbial cause for many used 16S ribosomal RNA gene profiling in 2005 that the gut-associated microbial
communicable human diseases, recent to identify pathogens linked to an illness, community of the large bowel represents
advances in culture-independent methods which is understandable given the focus a ‘microbial organ’ (REF. 19) because this
have made it possible to associate on the contribution of microorganisms collection of microorganisms resides as a
gut-associated microbial communities with to disease. However, the realization that structural unit and has a common function,
many non-communicable human diseases, most non-communicable diseases are not which is the production of metabolites
including obesity 1,2, arthritis3, asthma4,5, associated with a pathogen resulted in a at concentrations that promote health.
cardiovascular disease6, cancer7–9, diabetes10 conceptual crisis, calling the term ‘pathogen’ Although this idea is generally accepted17,

NATURE REVIEWS | MICROBIOLOGY VOLUME 16 | FEBRUARY 2018 | 103

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PERSPECTIVES

Box 1 | The germ theory of communicable diseases mammalian gut because the diversity of
this microbial community in adults does
Girolamo Fracastoro is credited with proposing in 1546 that microorganisms might cause human not suggest selection for any particular
diseases. However, this concept did not gain acceptance because a connection could not be bacterial species20. There is evidence that a
established until the invention of culture-dependent methods for isolating microorganisms, which mammalian host directs the composition of
was initiated by the development of pasteurization by Louis Pasteur in 1864 (REF. 99). This
the gut microbiota during infancy through
technological advance made it possible to associate microorganisms with communicable disease,
a link first demonstrated for an illness of silkworms by Louis Pasteur in 1870 (REF. 100). The first milk oligosaccharides28 and to some degree
human diseases for which an association with microbial pathogens was proposed were leprosy101 throughout life by epithelial release of
and amoebic dysentery in 1875 (REF. 102), anthrax in 1877 (REF. 103), typhoid in 1880 (REF. 104) and antimicrobials, such as defensins29,30. It is
pneumococcal pneumonia in 1881 (REFS 105,106). However, 1.5 decades after the invention of also well established that anaerobiosis is
pasteurization, evidence for a microbial origin of human diseases remained correlative. Similarly, preserved in the mammalian large bowel, but
1.5 decades after the advent of next-generation sequencing, contemporary microbiota research the idea that this global ecosystem control is
has mostly linked the gut microbiota with human diseases rather than proving causation17. Thus, imposed by the host emerged only recently.
the innovations of culture-dependent and culture-independent methods were followed by an We now know that mammalian hosts restrict
initial phase of intense research that remained largely correlative. oxygen availability in the lumen of the large
In 1882, Robert Koch proposed a game-changing approach ascertaining cause and effect,
bowel, thereby driving the composition
termed ‘Koch’s postulates’, to conclusively establish the microbial origin of communicable disease,
which he first applied to tuberculosis, the leading cause of human mortality at the time107. The of the microbial community towards a
ability to prove causation spawned a flurry of seminal contributions over the following 2 decades, dominance of obligate anaerobes, which is
including the identification of microbial causes for cholera108,109, typhoid110, diphtheria111, critical for maintaining gut homeostasis31.
tetanus112, bubonic plague113,114 and bacillary dysentery115, the identification of the first virulence Epithelial cells are maintained in a state of
factors (diphtheria toxin116 and Shiga toxin117,118) and the development of a treatment for hypoxia (<1% oxygen) at the colonic surface,
diphtheria119,120 and vaccines against anthrax121, rabies122 and typhoid123. The discoveries made which limits the diffusion of oxygen from
during the golden age of microbiology firmly established the germ theory, the idea that a specific the colonic mucosa into the intestinal lumen,
microbial pathogen causes a discrete communicable human disease. This concept set the germ an attribute that preserves anaerobiosis in the
theory apart from competing ideas, which held that microorganisms would be transformed into large bowel32. In turn, anaerobiosis ensures
pathogenic forms through exposure to miasma, a type of poisonous vapour. The identification of
a dominance of obligate anaerobic bacteria
pathogens facilitated vaccine development and made it possible to diagnose infected individuals,
thereby advancing public health measures to curb the spread of disease. It also identified bacterial that belong to the Bacteroidetes phylum and
pathogens as a target for treatment, which eventually led to the development of antibiotics. Clostridia class33. This feature is conserved
between individuals, although microbial
communities tolerate an intimidating
it has never become an alternative to the they are present as quantitatively minor degree of species diversity 20. Constraining
germ theory. One aspect that has curbed components of the gut microbiota25. The oxygen availability is critical because
the use of the microbial organ concept as a quest to identify pathobionts by profiling under anaerobic conditions, our microbial
guide for microbiota research is the daunting microbial communities in health and organ self-assembles into a community of
complexity of our gut-associated microbial disease follows in the footsteps of the microorganisms that stably coexist34 by
community. Comparison of the gut search for pathogens during the golden forming an anaerobic trophic network35. Each
microbiota composition between different age of microbiology (BOX 1), illustrating position in this trophic network represents
individuals reveals very little overlap on the the marked influence that the germ theory a nutrient niche that can be occupied by
species level20, thus making it problematic continues to exert on microbiota research. any microorganism that discharges the
to define what a ‘balanced microbial However, although a focus on individual required metabolic functions. Each occupant
community’ or a functional microbial organ microorganisms is appropriate for studying of a nutrient niche is predicted to be able
should look like. As a result of limited pathogens that cause communicable diseases, to use a few limiting resources better than
progress in understanding how a balanced this viewpoint is too narrow to fully capture any other member within the microbial
microbial community is maintained during the contribution of microbial community community, and the abundance of these
gut homeostasis, the microbial organ ecology to health and disease (BOX 2). Recent limiting resources determines the abundance
concept has long remained elusive, thus insights into how the host controls the gut of the respective occupant, a concept known
curtailing its use as a conceptual framework. ecosystem to ensure homeostasis suggest as the nutrient niche hypothesis36. The
Our poor understanding of gut that it is time to revisit the usefulness of the formation of this anaerobic trophic network
homeostasis has shifted the focus to microbial organ concept as a comprehensive generates metabolite profiles that are similar
aspects that are easier to grasp, such as the guide for microbiota research. among individuals, regardless of the species
association of disease with the presence Invertebrate models reveal a striking composition of the microbial community.
or absence of individual microorganisms. control of microbial community composition During gut homeostasis, our microbial
Specifically, the concept that disease by the host, as species-specific antimicrobial organ hydrolyses and ferments complex
can result from a loss of beneficial peptides are believed to be the selective force dietary carbohydrates (fibre) that escape
commensal bacteria or an overgrowth of behind the different bacterial communities degradation by host enzymes in the
resident bacteria that have the potential associated with closely related species of the upper gastrointestinal tract into smaller
to cause disease, termed ‘pathobionts’, is a cnidarian Hydra26, whereas the light organ of compounds, with the short-chain fatty
commonly used guide for contemporary Euprymna scolopes is associated with a single acids acetate, propionate and butyrate being
microbiota research21–24. The term ‘keystone bacterial species, Aliivibrio fischeri 27. At first the most abundant metabolites produced
species’ has been proposed for pathobionts glance, host control over the composition during gut homeostasis37 (FIG. 1). On the
that instigate inflammation even when of the gut microbiota is less evident in the basis of their main metabolic functions,

104 | FEBRUARY 2018 | VOLUME 16 www.nature.com/nrmicro

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 PERSPECTIVES

obligate anaerobic bacteria that degrade
complex carbohydrates can be grouped into
butyrate-producing bacteria, saccharolytic
bacteria, sulfate-reducing bacteria and
acetate-producing bacteria (acetogens)38.
Saccharolytic bacteria produce hydrogen,
which is consumed by hydrogenotrophic
microorganisms, including sulfate-reducing
bacteria, acetogens and methanogens39.
Obligate anaerobic microorganisms that
belong to these metabolic groupings
produce a multitude of metabolites in
addition to short-chain fatty acids, as
has been elegantly reviewed elsewhere40.
Importantly, the majority of microbiota-
derived short-chain fatty acids is absorbed
by the host 41 and contributes to host
nutrition42. Evolutionarily, anaerobiosis
could be viewed as a host-driven strategy
to limit the ability of the microbiota to
compete nutritionally with the host for
compounds produced by the microbial
fermentation of fibre. In the presence of
oxygen, facultative anaerobic bacteria can
catabolize fermentation products to carbon
dioxide. By limiting oxygen availability,
the host ensures that obligate anaerobes
contribute to its own nutrition by producing
metabolites through the fermentation
of fibre while at the same time limiting
catabolism of fermentation products to
carbon dioxide by facultative anaerobic
bacteria. Through this mechanism, the
host maintains control over the microbial
ecosystem without imposing constraints
on the species composition of the
microbial community.
Microbiota-derived short-chain
fatty acids also contribute to immune
development 43–48 and gut homeostasis3,31,49,50.
The latter process requires a continuous
supply of microbiota-derived short-chain
fatty acids to stimulate the intracellular
butyrate sensor peroxisome proliferator-
activated receptor‑γ (PPARγ) in colonic
epithelial cells31 and to promote maturation
and expansion of regulatory T cells in the
colonic mucosa45–48,50. Regulatory T cells
and epithelial PPARγ signalling cooperate
to drive the metabolism of the colonic
surface towards mitochondrial β-oxidation31,
a process that consumes large quantities
of oxygen, thereby rendering epithelial
cells at the colonic surface hypoxic32.
Hence, epithelial hypoxia ensures that
obligate anaerobes produce short-chain
fatty acids, which in turn induce epithelial
hypoxia, thereby driving a virtuous cycle
that maintains gut homeostasis and
balances the composition of the microbial
community 31 (FIG. 1).
The picture emerging from these lumen, which gives rise to the most
studies is that during gut homeostasis, consistent and robust ecological pattern
the host controls its microbial organ in the observed during gut dysbiosis, a shift in
large bowel through mechanisms that the microbial community structure from
include limiting oxygen availability in this obligate to facultative anaerobes51,52. An
ecosystem. The latter provides a benefit by expansion of facultative anaerobic bacteria
curbing the ability of facultative anaerobic of the phylum Proteobacteria has been
microorganisms to compete with the host described in individuals consuming a
for nutrients and by ensuring the production Western-style diet 53,54, undergoing antibiotic
of microbial fermentation products that therapy 55 or suffering from irritable
promote nutrition, immune development bowel syndrome56,57, inflammatory bowel
and gut homeostasis. disease58, metabolic syndrome59, necrotizing
enterocolitis60 or colorectal cancer 61. An
Microbial organ dysfunction expansion of Proteobacteria is also observed
Based on the above model of microbial organ in animal models of chemically induced
function (FIG. 1), we propose that dysbiosis colitis62,63, genetically induced colitis62,64,
is a state of microbial organ dysfunction, colorectal cancer 61, infection-induced
a condition in which the gut-associated colitis62,65,66 or antibiotic-mediated loss of
microbial community becomes a liability colonization resistance67–69. The regularity
because the host no longer maintains proper with which a disruption of gut homeostasis
control over the ecosystem. Microbial organ produces an expansion of Proteobacteria
dysfunction can produce disease because suggests that this is a microbial signature
metabolites are either depleted or produced of gut dysbiosis70. Below, we discuss
at a harmful level21–24. As the gut microbiota recent data supporting the concept that
can produce many different metabolites increased availability of oxygen or other
that are capable of acting both locally exogenous electron acceptors constitutes
and on cells in distant tissues40, microbial the main ecological driver of expansion
organ dysfunction can result in metabolite of Proteobacteria.
imbalances that affect numerous organs, Increased oxygen availability in the
thereby possibly explaining the broad colon can ensue from disruption of
spectrum of non-communicable diseases the gut microbiota, which can be triggered
associated with dysbiosis17. by either antibiotic therapy or intestinal
Importantly, disruption of the virtuous inflammation (FIG. 2). Depletion of gut
cycle that maintains anaerobiosis in the microorganisms during antibiotic therapy
colon31 (FIG. 1) impairs the ability of the host reduces the production of microbiota-
to limit the flow of oxygen into the gut derived metabolites, as indicated by a
Box 2 | Communicable diseases versus non-communicable diseases
Communicable diseases received their name because they are transmitted from one person or
animal to another, either by direct contact, by exposure to bodily discharge or indirectly through
fomites or vectors, such as arthropods. Causes of communicable diseases are infectious agents
that fit the classic definition of a microbial pathogen, such as Mycobacterium tuberculosis, the
causative agent of tuberculosis; Plasmodium falciparum, which causes malaria; or HIV, an infection
that eventually progresses to AIDS. The notion that pathogens are the cause of communicable
diseases is known as the germ theory (BOX 1). At the centre of this concept is the idea that frank
pathogens can overcome host defences in individuals with an intact immune system. Thus,
communicability is not due to an underlying defect in the host but is driven by virulence factors of
the microbial pathogen. Worldwide, communicable diseases continue to be the leading causes
of human mortality.
However, in high-income countries, the leading causes of death are non-communicable
diseases, such as cardiovascular diseases, cancer, chronic respiratory diseases and diabetes.
Non-communicable diseases were originally defined as medical conditions that are not caused by
an infectious agent. Surprisingly, research over the past 14 years has linked the resident microbial
community that inhabits our large bowel to many of those non-communicable human diseases.
The germ theory, the leading concept on the microbial origin of human disease, does not apply to
non-communicable diseases, because they are not due to infection with communicable infectious
agents but rather are driven by underlying defects in host physiology that are not communicable.
Thus, while the germ theory provides a theoretical framework for communicable diseases, there is
a need to develop a new theoretical framework that explains how host physiology balances the
gut-associated microbial community and how defects in host physiology disrupt this balance,
thereby linking microorganisms to non-communicable human diseases.

NATURE REVIEWS | MICROBIOLOGY VOLUME 16 | FEBRUARY 2018 | 105

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PERSPECTIVES

Secondary Niche
Clostridium
bile acids protection difficile
Microbial organ
Diet (obligate Metabolites
anaerobes)
Butyrate Propionate Acetate
Mature
colonocyte Anaerobiosis Nutrition Lumen
Immune
development
O2 CO2
O2 O2 gradient
Crypt
Blood vessel
Lactate Regulatory T cells
Dividing
Glucose
colonocyte
Figure 1 | The host controls the gut ecosystem to maintain homeostasis. Hypoxia of the colonic
epithelial cells (colonocytes) lining the mucosal surface maintains luminal anaerobiosis,
Nature which drives
Reviews | Microbiology
the composition of the gut microbiota towards a dominance of obligate anaerobes. Butyrate produced
by this microbial organ is oxidized by the colonic epithelium to carbon dioxide (CO2) to preserve
epithelial hypoxia (<1% oxygen (O2)), thereby maintaining gut homeostasis. Dividing colonocytes at
the bottom of crypts obtain energy through anaerobic glycolysis (that is, the fermentation of glucose
to lactate) and are thus not hypoxic. A healthy diet ensures that the microbial organ produces
metabo­lites that function in nutrition, immune development and niche protection. For simplicity,
only examples of metabolites are shown. The O2 concentration is indicated by the red colouring of
the schematic.
marked drop in the concentration of As oxygen originates from the host
short-chain fatty acids that are crucial for the epithelium, an expansion of Proteobacteria
maintenance of gut homeostasis71. Intestinal could be viewed as a microbial signature of
inflammation induced by enteric pathogens, epithelial dysfunction31.
such as Salmonella serovars (phylum A second mechanism that can lead
Proteobacteria), causes transepithelial to increased oxygen availability in the
migration of neutrophils that deplete colon is induction of excessive epithelial
Clostridia72,73, thereby lowering short-chain repair mechanisms (FIG. 3). Ulcerative
fatty acid levels in the colon74. The depletion colitis is associated with an expansion
of short-chain fatty acids by antibiotics of Proteobacteria70 that is indicative of a
or host neutrophils shifts the energy disruption in anaerobiosis51. Consistent with
metabolism of colonic epithelial cells from this idea, recent work from animal models
mitochondrial β‑oxidation to anaerobic of ulcerative colitis, including dextran
glycolysis31,75. As the latter pathway does not sulfate sodium (DSS)-induced colitis and
consume oxygen76, epithelial oxygenation Citrobacter rodentium-induced colitis,
increases, which in turn increases the levels suggests that the mechanism driving
of oxygen emanating from the epithelial expansion of Proteobacteria is increased
surface. An antibiotic-mediated increase oxygen availability in the colon77,78. Epithelial
in oxygen availability may explain why damage induced during DSS treatment or
treatment of mice with streptomycin C. rodentium infection triggers colonic crypt
increases the redox potential in the caecum hyperplasia, an excessive epithelial repair
to levels that approximate an aerobic broth response that results in crypt elongation,
culture71. The consequent disruption a reduction of goblet cell numbers and the
in anaerobiosis drives an expansion of appearance of undifferentiated epithelial
facultative anaerobic Proteobacteria cells at the luminal surface79,80. The presence
regardless of their pathogenic potential31,74. of undifferentiated epithelial cells increases
oxygenation of the colonic surface78 because
this cell type obtains energy through
anaerobic glycolysis76. The consequent
disruption in anaerobiosis drives an
expansion of Proteobacteria in the colon
through aerobic respiration78. Importantly, as
discussed below, dysbiosis can lead to disease
by mechanisms that are not linked to the
expansion of Proteobacteria6,81,82. However,
an expansion of this taxon is relevant because
it indicates an increased bioavailability of
host-derived respiratory electron acceptors,
which points to an impaired ability of the
host to control the gut ecosystem as an
important driver of dysbiosis.
Although mild intestinal inflammation
increases epithelial oxygenation31,74 (FIGS 2,3),
severe intestinal inflammation can render
0.1%
the mucosal surface hypoxic. During severe
intestinal inflammation, necrosis of the upper
1% mucosa leads to loss of the epithelial surface
and formation of a pseudomembrane that
is composed of necrotic tissue, neutrophils
10% and fibrin83. This pseudomembrane becomes
hypoxic because neutrophils covering the
surface undergo a respiratory burst that
depletes oxygen84. Although the mucosal
surface is hypoxic, these pathological changes
drive an expansion of Proteobacteria,
such as Salmonella enterica65 and Yersinia
enterocolitica85, as the neutrophil respiratory
burst generates electron acceptors for
anaerobic bacterial respiration, such as
tetrathionate, which promote growth of
these enteric pathogens. Thus, depletion of
short-chain fatty acids (FIG. 2), colonic crypt
hyperplasia (FIG. 3) and pseudomembrane
formation each drive an expansion of
Proteobacteria by increasing the availability
of host-derived exogenous respiratory
electron acceptors in the lumen of the
large bowel.
It is important to keep in mind that
although increased availability of respiratory
electron acceptors contributes to the
expansion of Proteobacteria, it may not
be the only driver. For instance, antibiotic
treatment increases the availability
of monosaccharides, which have an
important role in the observed expansion
of Proteobacteria following antibiotic
treatment86,87. Furthermore, whereas some
Proteobacteria are pathogens or pathobionts,
most members of this taxon are commensal
microorganisms, and the expansion of such
commensals does not drive disease but
does represent a biomarker for dysbiotic
ecosystem disruption. In other words,
although an increase in the abundance of
Proteobacteria may be observed during
dysbiosis, disease might be caused by

106 | FEBRUARY 2018 | VOLUME 16 www.nature.com/nrmicro

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 PERSPECTIVES

additional metabolite imbalances that are not
linked to electron acceptor availability or the
presence of Proteobacteria. The exact nature
of the metabolite imbalance that develops
during dysbiosis depends on the factors
that trigger the disruption of the ecosystem.
For example, the antibiotic-mediated
disruption of the microbiota drives an
expansion of Proteobacteria, but antibiotic-
associated colitis is caused by an expansion
of toxin-producing Clostridium difficile, an
opportunistic pathogen that is an obligate
anaerobe88. Antibiotic treatment lowers
colonization resistance against C. difficile by
depleting other obligate anaerobic members
of the microbiota that are producers of
secondary bile acids89,90, which are
metabolites that inhibit growth of vegetative
C. difficile cells91 (FIG. 1). Restoration
of a balanced community of obligate
anaerobic bacteria by faecal microbiota
transplantation can restore microbial organ
function and thus provides an effective
treatment for antibiotic-associated colitis92,93.
Indeed, inoculation of the secondary bile
acid-producing Clostridium scindens was
shown to restore levels of secondary bile
acids in antibiotic-treated mice, thereby
re-establishing colonization resistance
against C. difficile81.
Another example is consumption of a
Western-style diet, which is associated with
theoretical framework to describe
microbial organ physiology. During gut
homeostasis, the host controls the microbial
ecosystem97 by limiting luminal oxygen
availability in the colon, an ecological
input that is critical for maintaining
a balanced microbial community 31. This
balanced microbial community functions
as a microbial organ, producing metabolites
that maintain a virtuous cycle to preserve
epithelial hypoxia (FIG. 1). Microbial organ
dysfunction represents an ecosystem
disturbance, commonly involving a shift in
the microbial community from obligate to
facultative anaerobes, which is indicative
of a defect in the ability of the host to limit
the flow of oxygen and other exogenous
electron acceptors into the gut lumen.
This loss of host control disrupts the
microbial ecosystem, thereby contributing
to non-communicable diseases because
a microbiota-derived metabolite is either
depleted or produced at harmful levels. We
propose the term ‘germ-organ theory’ to
describe the concept that host control over
the microbial ecosystem in the large bowel is
critical for the composition and function of
te anaero
iga
be
l
the microbial organ, whereas disruption
of host control triggers microbial organ
dysfunction, which provides a link between
the gut microbiota and non-communicable
diseases. The focus on host control over
the microbial ecosystem sets the concept
of the germ-organ theory apart from the
germ theory, which pictures the pathogen
in control (BOX 2).
Conclusions
Broadening the concepts guiding
microbiota research towards an ecological
perspective of microbial organ function
puts the spotlight on questions that are
currently just ‘wallflowers’ in contemporary
microbiota research. For instance, the
germ-organ theory suggests that non-
communicable diseases associated with
an expansion of Proteobacteria can be
linked to impaired host control over the
gut ecosystem. One example of impaired
host control is epithelial dysfunction, which
leads to a disruption of anaerobiosis in the
large bowel31. We do not know whether
depleting short-chain fatty acids (FIG. 2)
or inducing excessive epithelial repair
Antibiotics

Butyrate
Ob

an expansion of Proteobacteria53,54 but is

s

Disruption
linked to disease because dietary inputs of the
alter the metabolite profile produced by ro microbiota Propionate
t e o b c t e ri
P

the gut microbiota. A Western-style diet a

is associated with obesity, a condition that Acetate
increases the risk of atherosclerosis94. A link Mature
colonocyte O2
between diet and cardiovascular disease is Lumen
the consumption of l‑carnitine and choline,
both of which are abundant in red meat,
a common component of a Western-style O2
diet, and are converted by the gut microbiota Glucose Lactate
to trimethylamine (TMA)6,95. TMA is
O2 gradient
absorbed by the host and oxidized by human Neutrophils
O2
hepatic flavin-containing monooxygenases
Crypt 0.1%
to TMA N‑oxide (TMAO), a metabolite
that increases the risk of atherosclerosis6,82.
Pathways to convert l‑carnitine and choline Blood vessel 1%
into TMA are present in bacteria that Inflammation
belong to various phyla6,95,96. Thus, although Dividing
antibiotic treatment and the consumption colonocyte Regulatory T cell 10%
of a Western-style diet both disrupt gut
homeostasis, as indicated by an expansion Figure 2 | Microbiota disruption causes microbial organ dysfunction. Antibiotic treatment or the
of Proteobacteria, each is linked to disease migration of neutrophils into the intestinal lumen during severe intestinal
Natureinflammation can cause
Reviews | Microbiology
the disruption of the microbiota, which leads to a depletion of short-chain fatty acids, thereby shift-
through a different metabolite imbalance.
ing the metabolism of mature colonic epithelial cells (colonocytes) from mitochondrial β‑oxidation to
anaerobic glycolysis (that is, the fermentation of glucose to lactate). The resulting increase in epithelial
The germ-organ theory oxygenation disrupts anaerobiosis, which drives an expansion of facultative anaerobic Proteobacteria,
Recent findings on the mechanisms that a microbial signature of dysbiosis. Neutrophils that encounter bacteria (for example, during trans­
maintain or disrupt gut homeostasis epithelial migration into the lumen) undergo a respiratory burst and become hypoxic (<1% oxygen
can be assembled into the following (O2))124. The O2 concentration is indicated by the red colouring of the schematic.

NATURE REVIEWS | MICROBIOLOGY VOLUME 16 | FEBRUARY 2018 | 107

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PERSPECTIVES

service facilities, mechanistic follow‑up

studies aimed at understanding trophic
te anaero networks, the influence of host physiology
iga
on the microbial ecosystem and the role

be
l
Ob

s
Metabolites Lumen that microbiota-derived metabolites
have in health and disease will require
ro
t e o b a c t e ri expertise in both bacterial physiology and

Epithelial Dividing
damage colonocyte
Epithelial
repair
response
Crypt
hyperplasia
Figure 3 | Excessive epithelial repair triggers microbial organ dysfunction. Epithelial damage
triggers epithelial repair responses, which are characterized by cryptNature
dividing colonocytes at the epithelial surface. These immature dividing colonocytes obtain energy
through anaerobic glycolysis (that is, the fermentation of glucose to lactate), which renders the colonic
surface normoxic (3–10% oxygen (O2)). The resulting disruption of anaerobiosis drives an expansion
of facultative anaerobic Proteobacteria, a microbial signature of dysbiosis. The O2 concentration is
indicated by the red colouring of the schematic.
(FIG. 3) are the only mechanisms increasing
epithelial oxygenation in the colon. More
research is needed to unravel the causes
of epithelial dysfunction for the various
non-communicable diseases associated
with expansion of Proteobacteria. The
focus on changes in host physiology that
impair control over the gut ecosystem
represents a paradigm shift from the
current focus on identifying pathobionts
or microbial biomarkers of disease within
the gut microbiota. Identifying causes of
epithelial dysfunction is relevant because
the germ-organ theory points to host
signalling pathways as a potential target for
intervention strategies. Such approaches
would be aimed at rebalancing the gut
microbiota by limiting the availability of
exogenous electron acceptors and restoring
epithelial hypoxia, which could be used to
complement faecal microbiota transfer or
to develop alternative treatment strategies.
Additionally, the germ-organ theory
advocates for obtaining a more complete
understanding of the composition and
function of our microbial organ. We know
embarrassingly little about the identity of
limiting resources that define each nutrient
a
Butyrate Propionate Acetate
O2
O2
Lactate
Glucose
O2
O2 gradient
Blood vessel
Regulatory T cells
Reviews
elongation | Microbiology
in the presence of
niche within our microbial organ or the
metabolic functions that make a specific
microorganism a suitable occupant of a
particular nutrient niche. Shedding light
on these crucial knowledge gaps will be
necessary to incorporate an improved
understanding of trophic networks into the
germ-organ theory, thereby increasing its
power to predict additional mechanisms
through which the host controls the
ecosystem. Ecosystem drivers in addition
to oxygen levels might explain dysbiosis in
settings where expansion of Proteobacteria
is not observed, but the identities of such
drivers remain unknown. Tackling this
question will require a ‘new bacteriology’
(REF. 98) to conduct mechanistic studies on
microbial physiology in its natural context
of a gut-associated microbial community.
As we venture forward, it is helpful
to reflect on the skill sets best suited to
tackle challenges that lie ahead. The
germ-organ theory instructs a shift from
microbial community profiling towards
understanding host-mediated control
of microbial organ ecology. Although
the profiling approaches required to
tackle this question can be performed by
immunology, which can be obtained only
by attracting scientists from these fields
to enter microbiota research. Researchers
in the field of bacterial pathogenesis
seem particularly well prepared to enter
this arena as their research exposes them
to bacterial physiology as well as to the
interaction of pathogens with the immune
system, thus providing an ideal background
for mechanistic microbiota research.
Alternatively, collaboration between
0.1% scientists engaged in bioinformatics,
bacterial physiology and ecology,
cell biology and immunology could
1% accelerate decryption of the mechanistic
basis for microbial contributions to
non-communicable diseases. Although
10%
funding agencies have not provided major
incentives for bacterial physiologists
and immunologists to enter microbiota
research, history suggests that this might
be a move in the right direction as we stand
at the threshold of a ‘second golden age
of microbiology’.
Mariana X. Byndloss and Andreas J. Bäumler are at
the Department of Medical Microbiology and
Immunology, School of Medicine, University of
California, Davis, California 95616, USA.
Correspondence to A.J.B.
ajbaumler@ucdavis.edu
doi:10.1038/nrmicro.2017.158
Published online 8 Jan 2018
1. Turnbaugh, P. J. et al. An obesity-associated gut
microbiome with increased capacity for energy
harvest. Nature 444, 1027–1031 (2006).
2. Henao-Mejia, J. et al. Inflammasome-mediated
dysbiosis regulates progression of NAFLD and
obesity. Nature 482, 179–185 (2012).
3. Maslowski, K. M. et al. Regulation of inflammatory
responses by gut microbiota and chemoattractant
receptor GPR43. Nature 461, 1282–1286 (2009).
4. Russell, S. L. et al. Early life antibiotic-driven
changes in microbiota enhance susceptibility to
allergic asthma. EMBO Rep. 13, 440–447 (2012).
5. Trompette, A. et al. Gut microbiota metabolism of
dietary fiber influences allergic airway disease and
hematopoiesis. Nat. Med. 20, 159–166 (2014).
6. Wang, Z. et al. Gut flora metabolism of
phosphatidylcholine promotes cardiovascular disease.
Nature 472, 57–63 (2011).
7. Castellarin, M. et al. Fusobacterium nucleatum
infection is prevalent in human colorectal carcinoma.
Genome Res. 22, 299–306 (2012).
8. Kostic, A. D. et al. Genomic analysis identifies
association of Fusobacterium with colorectal
carcinoma. Genome Res. 22, 292–298 (2012).
9. Yoshimoto, S. et al. Obesity-induced gut microbial
metabolite promotes liver cancer through
senescence secretome. Nature 499, 97–101
(2013).
10. Cani, P. D. et al. Changes in gut microbiota control
metabolic endotoxemia-induced inflammation in
high-fat diet-induced obesity and diabetes in mice.
Diabetes 57, 1470–1481 (2008).

108 | FEBRUARY 2018 | VOLUME 16 www.nature.com/nrmicro

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 PERSPECTIVES

11. Hsiao, E. Y. et al. Microbiota modulate behavioral
and physiological abnormalities associated with
neurodevelopmental disorders. Cell 155, 1451–1463
(2013).
12. Minter, M. R. et al. Antibiotic-induced perturbations in
gut microbial diversity influences neuro-inflammation
and amyloidosis in a murine model of Alzheimer’s
disease. Sci. Rep. 6, 30028 (2016).
13. Methot, P. O. & Alizon, S. What is a pathogen? Toward
a process view of host-parasite interactions. Virulence
5, 775–785 (2014).
14. Casadevall, A. & Pirofski, L. A. Microbiology: Ditch the
term pathogen. Nature 516, 165–166 (2014).
15. Altmann, D. & Boyton, R. Nomenclature: Replace
‘pathogens’ with ‘perceptogens’. Nature 518, 35
(2015).
16. Sultana, S., Sarker, S. A. & Brussow, H. What happened
to Koch’s postulates in diarrhea? Environ. Microbiol.
http://dx.doi.org/10.1111/1462-2920.13787 (2017).
17. Cani, P. D. Gut microbiota — at the intersection of
everything? Nat. Rev. Gastroenterol. Hepatol. 14,
321–322 (2017).
18. Lopez, C. A., Kingsbury, D. D., Velazquez, E. M. &
Baumler, A. J. Collateral damage: microbiota-derived
metabolites and immune function in the antibiotic era.
Cell Host Microbe 16, 156–163 (2014).
19. O’Hara, A. M. & Shanahan, F. The gut flora as a
forgotten organ. EMBO Rep. 7, 688–693 (2006).
20. Tap, J. et al. Towards the human intestinal microbiota
phylogenetic core. Environ. Microbiol. 11,
2574–2584 (2009).
21. Shreiner, A., Huffnagle, G. B. & Noverr, M. C.
The “Microflora Hypothesis” of allergic disease.
Adv. Exp. Med. Biol. 635, 113–134 (2008).
22. Packey, C. D. & Sartor, R. B. Commensal bacteria,
traditional and opportunistic pathogens, dysbiosis
and bacterial killing in inflammatory bowel diseases.
Curr. Opin. Infect. Dis. 22, 292–301 (2009).
23. Round, J. L. & Mazmanian, S. K. The gut microbiota
shapes intestinal immune responses during health and
disease. Nat. Rev. Immunol. 9, 313–323 (2009).
24. Bjorksten, B. in Microbial–Host Interaction: Tolerance
versus Allergy. 64th Nestlé Nutrition Institute
Workshop, Pediatric Program (eds Brandtzaeg, P.,
Isolauri, E. & Prescott, S. L.) 11–22 (Sydney, 2008).
25. Hajishengallis, G., Darveau, R. P. & Curtis, M. A. The
keystone-pathogen hypothesis. Nat. Rev. Microbiol.
10, 717–725 (2012).
26. Franzenburg, S. et al. Distinct antimicrobial peptide
expression determines host species-specific bacterial
associations. Proc. Natl Acad. Sci. USA 110,
E3730–E3738 (2013).
27. Ruby, E. G. Lessons from a cooperative, bacterial-
animal association: the Vibrio fischeri-Euprymna
scolopes light organ symbiosis. Annu. Rev. Microbiol.
50, 591–624 (1996).
28. Garrido, D., Dallas, D. C. & Mills, D. A. Consumption
of human milk glycoconjugates by infant-associated
bifidobacteria: mechanisms and implications.
Microbiology 159, 649–664 (2013).
29. Salzman, N. H. et al. Enteric defensins are essential
regulators of intestinal microbial ecology. Nat. Immunol.
11, 76–83 (2010).
30. Hayase, E. et al. R-Spondin1 expands Paneth cells and
prevents dysbiosis induced by graft-versus-host
disease. J. Exp. Med. http://dx.doi.org/10.1084/
jem.20170418 (2017).
31. Byndloss, M. X. et al. Microbiota-activated PPAR‑γ-
signaling inhibits dysbiotic Enterobacteriaceae
expansion. Science 357, 570–575 (2017).
32. Furuta, G. T. et al. Hypoxia-inducible factor
1‑dependent induction of intestinal trefoil factor
protects barrier function during hypoxia. J. Exp. Med.
193, 1027–1034 (2001).
33. Eckburg, P. B. et al. Diversity of the human intestinal
microbial flora. Science 308, 1635–1638 (2005).
34. Faith, J. J. et al. The long-term stability of the human
gut microbiota. Science 341, 1237439 (2013).
35. Rivera-Chavez, F. & Baumler, A. J. The pyromaniac
inside you: Salmonella metabolism in the host gut.
Annu. Rev. Microbiol. 69, 31–48 (2015).
36. Freter, R., Brickner, H., Fekete, J., Vickerman, M. M.
& Carey, K. E. Survival and implantation of Escherichia
coli in the intestinal tract. Infect. Immun. 39,
686–703 (1983).
37. Roy, C. C., Kien, C. L., Bouthillier, L. & Levy, E.
Short-chain fatty acids: ready for prime time? Nutr. Clin.
Pract. 21, 351–366 (2006).
38. Fischbach, M. A. & Sonnenburg, J. L. Eating for two:
how metabolism establishes interspecies interactions
in the gut. Cell Host Microbe 10, 336–347 (2011).
39. Nakamura, N., Lin, H. C., McSweeney, C. S.,
Mackie, R. I. & Gaskins, H. R. Mechanisms of microbial
hydrogen disposal in the human colon and implications
for health and disease. Annu. Rev. Food Sci. Technol. 1,
363–395 (2010).
40. Nicholson, J. K. et al. Host-gut microbiota metabolic
interactions. Science 336, 1262–1267 (2012).
41. den Besten, G. et al. The role of short-chain fatty acids
in the interplay between diet, gut microbiota, and host
energy metabolism. J. Lipid Res. 54, 2325–2340
(2013).
42. Velazquez, O. C., Lederer, H. M. & Rombeau, J. L.
Butyrate and the colonocyte. Production, absorption,
metabolism, and therapeutic implications. Adv. Exp.
Med. Biol. 427, 123–134 (1997).
43. Millard, A. L. et al. Butyrate affects differentiation,
maturation and function of human monocyte-derived
dendritic cells and macrophages. Clin. Exp. Immunol.
130, 245–255 (2002).
44. Wang, B., Morinobu, A., Horiuchi, M., Liu, J. &
Kumagai, S. Butyrate inhibits functional differentiation
of human monocyte-derived dendritic cells.
Cell. Immunol. 253, 54–58 (2008).
45. Atarashi, K. et al. Induction of colonic regulatory T cells
by indigenous Clostridium species. Science 331,
337–341 (2011).
46. Arpaia, N. et al. Metabolites produced by commensal
bacteria promote peripheral regulatory T‑cell
generation. Nature 504, 451–455 (2013).
47. Furusawa, Y. et al. Commensal microbe-derived
butyrate induces the differentiation of colonic
regulatory T cells. Nature 504, 446–450 (2013).
48. Smith, P. M. et al. The microbial metabolites, short-
chain fatty acids, regulate colonic Treg cell
homeostasis. Science 341, 569–573 (2013).
49. Chang, P. V., Hao, L., Offermanns, S. & Medzhitov, R.
The microbial metabolite butyrate regulates intestinal
macrophage function via histone deacetylase
inhibition. Proc. Natl Acad. Sci. USA 111, 2247–2252
(2014).
50. Singh, N. et al. Activation of Gpr109a, receptor for
niacin and the commensal metabolite butyrate,
suppresses colonic inflammation and carcinogenesis.
Immunity 40, 128–139 (2014).
51. Rigottier-Gois, L. Dysbiosis in inflammatory bowel
diseases: the oxygen hypothesis. ISME J. 7,
1256–1261 (2013).
52. Rivera-Chavez, F., Lopez, C. A. & Baumler, A. J.
Oxygen as a driver of gut dysbiosis. Free Radic. Biol.
Med. 105, 93–101 (2017).
53. Devkota, S. et al. Dietary-fat-induced taurocholic acid
promotes pathobiont expansion and colitis in Il10−/−
mice. Nature 487, 104–108 (2012).
54. Martinez-Medina, M. et al. Western diet induces
dysbiosis with increased E coli in CEABAC10 mice,
alters host barrier function favouring AIEC
colonisation. Gut 63, 116–124 (2014).
55. Vollaard, E. J., Clasener, H. A. & Janssen, A. J.
Co‑trimoxazole impairs colonization resistance in
healthy volunteers. J. Antimicrob. Chemother. 30,
685–691 (1992).
56. Carroll, I. M., Ringel-Kulka, T., Siddle, J. P. & Ringel, Y.
Alterations in composition and diversity of the
intestinal microbiota in patients with diarrhea-
predominant irritable bowel syndrome.
Neurogastroenterol. Motil. 24, 521–530 (2012).
57. Krogius-Kurikka, L. et al. Microbial community analysis
reveals high level phylogenetic alterations in the overall
gastrointestinal microbiota of diarrhoea-predominant
irritable bowel syndrome sufferers. BMC Gastroenterol.
9, 95 (2009).
58. Morgan, X. C. et al. Dysfunction of the intestinal
microbiome in inflammatory bowel disease and
treatment. Genome Biol. 13, R79 (2012).
59. Ferreira, R. B. et al. The intestinal microbiota plays a
role in Salmonella-induced colitis independent of
pathogen colonization. PLoS ONE 6, e20338 (2011).
60. Normann, E., Fahlen, A., Engstrand, L. & Lilja, H. E.
Intestinal microbial profiles in extremely preterm
infants with and without necrotizing enterocolitis.
Acta Paediatr. 102, 129–136 (2013).
61. Arthur, J. C. et al. Intestinal inflammation targets
cancer-inducing activity of the microbiota. Science
338, 120–123 (2012).
62. Lupp, C. et al. Host-mediated inflammation disrupts
the intestinal microbiota and promotes the overgrowth
of Enterobacteriaceae. Cell Host Microbe 2, 204
(2007).
63. Winter, S. E. et al. Host-derived nitrate boosts growth
of E. coli in the inflamed gut. Science 339, 708–711
(2013).
64. Garrett, W. S. et al. Enterobacteriaceae act in concert
with the gut microbiota to induce spontaneous and
maternally transmitted colitis. Cell Host Microbe 8,
292–300 (2010).
65. Winter, S. E. et al. Gut inflammation provides a
respiratory electron acceptor for Salmonella. Nature
467, 426–429 (2010).
66. Haag, L. M. et al. Intestinal microbiota shifts towards
elevated commensal Escherichia coli loads abrogate
colonization resistance against Campylobacter jejuni
in mice. PLoS ONE 7, e35988 (2012).
67. Bohnhoff, M., Drake, B. L. & Miller, C. P. Effect of
streptomycin on susceptibility of intestinal tract to
experimental Salmonella infection. Proc. Soc. Exp.
Biol. Med. 86, 132–137 (1954).
68. Saito, K. Studies on the habitation of pathogenic
Escherichia coli in the intestinal tract of mice. I.
Comparative experiments on the habitation of each
type of resistant pathogenic Escherichia coli under
an administration of streptomycin [Japanese].
Paediatr. Jpn. 65, 385–393 (1961).
69. Saito, K. Studies on the habitation of pathogenic
Escherichia coli in the intestinal tract of mice. II.
Experimental inoculation of type 055 Escherichia coli
after long-term administration of streptomycin
[Japanese]. Paediatr. Jpn. 65, 394–399 (1961).
70. Shin, N. R., Whon, T. W. & Bae, J. W. Proteobacteria:
microbial signature of dysbiosis in gut microbiota.
Trends Biotechnol. 33, 496–503 (2015).
71. Meynell, G. G. Antibacterial mechanisms of the
mouse gut. II. The role of Eh and volatile fatty acids
in the normal gut. Br. J. Exp. Pathol. 44, 209–219
(1963).
72. Gill, N. et al. Neutrophil elastase alters the murine gut
microbiota resulting in enhanced Salmonella
colonization. PLoS ONE 7, e49646 (2012).
73. Deatherage Kaiser, B. L. et al. A multi-omic view of
host-pathogen-commensal interplay in -mediated
intestinal infection. PLoS ONE 8, e67155 (2013).
74. Rivera-Chavez, F. et al. Depletion of butyrate-
producing Clostridia from the gut microbiota drives an
aerobic luminal expansion of Salmonella. Cell Host
Microbe 19, 443–454 (2016).
75. Kelly, C. J. et al. Crosstalk between microbiota-derived
short-chain fatty acids and intestinal epithelial HIF
augments tissue barrier function. Cell Host Microbe
17, 662–671 (2015).
76. Fan, Y. Y. et al. A bioassay to measure energy
metabolism in mouse colonic crypts, organoids, and
sorted stem cells. Am. J. Physiol. Gastrointest. Liver
Physiol. 309, G1–G9 (2015).
77. Hughes, E. R. et al. Microbial respiration and formate
oxidation as metabolic signatures of inflammation-
associated dysbiosis. Cell Host Microbe 21, 208–219
(2017).
78. Lopez, C. A. et al. Virulence factors enhance
Citrobacter rodentium expansion through aerobic
respiration. Science 353, 1249–1253 (2016).
79. Ahmed, I. et al. Critical roles of Notch and Wnt/β-
catenin pathways in the regulation of hyperplasia
and/or colitis in response to bacterial infection. Infect.
Immun. 80, 3107–3121 (2012).
80. Chandrakesan, P. et al. Novel changes in NF-κB
activity during progression and regression phases of
hyperplasia: role of MEK, ERK, and p38. J. Biol. Chem.
285, 33485–33498 (2010).
81. Buffie, C. G. et al. Precision microbiome reconstitution
restores bile acid mediated resistance to Clostridium
difficile. Nature 517, 205–208 (2015).
82. Bennett, B. J. et al. Trimethylamine-N‑oxide, a
metabolite associated with atherosclerosis, exhibits
complex genetic and dietary regulation. Cell Metab.
17, 49–60 (2013).
83. Tsolis, R. M., Adams, L. G., Ficht, T. A. & Baumler, A. J.
Contribution of Salmonella typhimurium virulence
factors to diarrheal disease in calves. Infect. Immun.
67, 4879–4885 (1999).
84. Campbell, E. L. et al. Transmigrating neutrophils shape
the mucosal microenvironment through localized
oxygen depletion to influence resolution of
inflammation. Immunity 40, 66–77 (2014).
85. Kamdar, K. et al. Genetic and metabolic signals during
acute enteric bacterial infection alter the microbiota
and drive progression to chronic inflammatory disease.
Cell Host Microbe 19, 21–31 (2016).
86. Ng, K. M. et al. Microbiota-liberated host sugars
facilitate post-antibiotic expansion of enteric
pathogens. Nature 502, 96–99 (2013).
87. Faber, F. et al. Host-mediated sugar oxidation
promotes post-antibiotic pathogen expansion. Nature
534, 697–699 (2016).

NATURE REVIEWS | MICROBIOLOGY VOLUME 16 | FEBRUARY 2018 | 109

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
PERSPECTIVES

88. Kelly, C. P., Pothoulakis, C. & LaMont, J. T. Clostridium
difficile colitis. N. Engl. J. Med. 330, 257–262
(1994).
89. Sorg, J. A. & Sonenshein, A. L. Bile salts and glycine
as cogerminants for Clostridium difficile spores.
J. Bacteriol. 190, 2505–2512 (2008).
90. Theriot, C. M. et al. Antibiotic-induced shifts in the
mouse gut microbiome and metabolome increase
susceptibility to Clostridium difficile infection.
Nat. Commun. 5, 3114 (2014).
91. Wilson, K. H. Efficiency of various bile salt
preparations for stimulation of Clostridium difficile
spore germination. J. Clin. Microbiol. 18, 1017–1019
(1983).
92. van Nood, E. et al. Duodenal infusion of donor feces
for recurrent Clostridium difficile. N. Engl. J. Med.
368, 407–415 (2013).
93. Weingarden, A. R. et al. Microbiota transplantation
restores normal fecal bile acid composition in
recurrent Clostridium difficile infection.
Am. J. Physiol. Gastrointest. Liver Physiol. 306,
G310–G319 (2014).
94. Fung, T. T. et al. Association between dietary
patterns and plasma biomarkers of obesity and
cardiovascular disease risk. Am. J. Clin. Nutr. 73,
61–67 (2001).
95. Zhu, Y. et al. Carnitine metabolism to trimethylamine
by an unusual Rieske-type oxygenase from human
microbiota. Proc. Natl Acad. Sci. USA 111,
4268–4273 (2014).
96. Rath, S., Heidrich, B., Pieper, D. H. & Vital, M.
Uncovering the trimethylamine-producing bacteria
of the human gut microbiota. Microbiome 5, 54
(2017).
97. Foster, K. R., Schluter, J., Coyte, K. Z. & Rakoff-
Nahoum, S. The evolution of the host microbiome
as an ecosystem on a leash. Nature 548, 43–51
(2017).
98. Pedron, T., Nigro, G. & Sansonetti, P. J. From
homeostasis to pathology: decrypting microbe-host
symbiotic signals in the intestinal crypt. Philos. Trans.
R. Soc. Lond. B. Biol. Sci. http://dx.doi.org/10.1098/
rstb.2015.0500 (2016).
99. Bordenave, G. Louis Pasteur (1822–1895). Microbes
Infect. 5, 553–560 (2003).
100. Pasteur, L. Etudes Sur La Maladie Des Vers A Soie.
Vol. 1,2 (Gauthier-Villars, 1870).
101. Hansen, G. H. A. On the etiology of leprosy.
Br. Foreign Med. Chir. Rev. 55, 459–489 (1875).
102. Losch, F. Massenhafte entwickelung von amöben im
dickdarm. Arch. Pathol. Anat. Physiol. 65, 196–211
(1875).
103. Koch, R. Untersuchungen über Bakterien, V.
Die aetiologie der milzbrandkrankheit, begründet
auf der entwicklungsgeschichte des Bacillus
anthracis. Beiträge Biol. Pflanzen 2, 277–308
(1877).
104. Eberth, K. J. Die organismen in den organen bei
typhus abdominalis [German]. Virchows Arch. 81,
58–74 (1880).
105. Pasteur, L. Note sur la maladie nouvelle provoquee par
la salive d’un enfant mort de la rage [French].
C. R. Acad. Sci. 92, 159–165 (1881).
106. Sternberg, G. M. A fatal form of septicaemia, produced
by the injection of human saliva. An experimental
research. Bull. Nat. Board Health USA 2, 781–783
(1881).
107. Koch, R. Die aetiologie der tuberkulose.
Berliner Klinische Wochenschrift 15, 221–230
(1882).
108. Koch, R. Der zweite bericht der deutschen
cholerakommission. Dt. Med. Wochenschr. 9,
743–744 (1883).
109. Koch, R. Ueber die cholerabakterien. Dt. Med.
Wochenschr. 10, 725–728 (1884).
110. Gaffky, G. Zur ätiologie des abdominaltyphus.
Mitteilungen Kaiserlichen Gesundheitsamt 2,
372–420 (1884).
111. Loeffler, F. Untersuchungen über die bedeutung der
mikroorganismen für die entstehung der diphtherie
beim menschen, bei der taube und beim kalbe.
Mitth. a. d. Kaiserl. Gesundheitsamtes 2, 421–499
(1884).
112. Nicolaier, A. Ueber infectiösen tetanus. Dt. Med.
Wochenschr. 10, 842–844 (1884).
113. Yersin, A. La peste bubonique à Hong-Kong.
Ann. l’Institut Pasteur 8, 662–667 (1894).
114. Kitasato, S. The bacillus of bubonic plague. Lancet 144,
428–430 (1894).
115. Shiga, K. Ueber den dysenterie-bacillus (Bacillus
dysenteriae). Zentralbl Bakteriol Orig 24, 913–918
(1898).
116. Roux, E., Y. A. Contribution à l’étude de la diphtérie
Ann. Inst. Pasteur 2, 421–499 (1888).
117. Conradi, H. Über lösliche, durch aseptische autolyse
erhaltene giftstoffe von ruhr- und typhus-bazillen.
Dtsch. Med. Wochenschr. 29, 26–28 (1903).
118. Neisser, M. & Shiga, K. Ueber freie receptoren
von typhus- und dysenteriebazillen und über das
dysenterietoxin. Dtsch. Med. Wochenschr. 29, 61–62
(1903).
119. von Behring, E. & Kitasato, S. Ueber das
zustandekommen der diphtherie — immunität und der
tetanus — immunität bei tieren. Dt. Med. Wochenschr.
16, 1113–1114 (1890).
120. von Behring, E. Untersuchungen über das
zustandekommen de diphtherie — immunität bei
tieren. Dt. Med. Wochenschr. 16, 1145–1148
(1890).
121. Pasteur, L., Roux, E. & Chamberland, C. Summary
report of the experiments conducted at Pouilly-le‑Fort,
near Melun, on the anthrax vaccination, 1881.
C. R. Acad. Sci. 92, 1378–1383 (1881).
122. Pasteur, L. Methode pour prevenir l rage apres
morsure. C. R. Acad. Sci. 101, 765–772 (1885).
123. Pfeiffer, R. & Kolle, W. Experimentele untersuchungen
zur frage der schutzimpfungen des menschen gegen
den typhus abdominalis. Dtsch. Med. Wochenschr.
22, 735–737 (1896).
124. Jennewein, J. et al. Low-oxygen tensions found in
Salmonella-infected gut tissue boost Salmonella
replication in macrophages by impairing antimicrobial
activity and augmenting Salmonella virulence.
Cell. Microbiol. 17, 1833–1847 (2015).
Acknowledgements
Work in the authors’ laboratory was supported by public
health service grants AI112445, AI112949 and AI096528
(A.J.B.).
Author contributions
M.X.B. and A.J.B. substantially contributed to the discussion
of content and review and editing of the manuscript before
submission.
Competing interests statement
The authors declare no competing interests.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

110 | FEBRUARY 2018 | VOLUME 16 www.nature.com/nrmicro

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.