Documentos de Académico
Documentos de Profesional
Documentos de Cultura
Feelings of hunger and satiety dominate the life of all interactions10. However, the role of gut bacteria in the
animals and are the main involuntary motivations for regulation of host appetite has, until recently, received
feeding-related behaviour. Both feelings are of visceral little attention.
origin; that is, they involve the gastrointestinal tract Understanding how the body generates and reg-
secreting molecules that reach the nervous system. The ulates appetite is of fundamental importance for the
brain integrates peripheral hunger-related and satiety- treatment of chronic pathological conditions such as
related signals to generate motivated behaviour neces- obesity, anorexia nervosa and cachexia, in which the
sary for obtaining, ingesting and digesting food. The normal regulation is lost. The current obesity epidemic
gut of all animals (except experimental germ-free ani- and the growing numbers of individuals with eating
mals) contains numerous bacteria, which are dependent disorders reflect the lack of efficient treatments and
on host feeding behaviour for receiving the nutrients highlight our insufficient understanding of the under-
necessary for maintenance of their population. The lying causes of altered appetite11,12. Conversely, several
gut, therefore, represents a stable ecological niche for studies have revealed dysbioses of the gut microbiota
its inhabiting bacteria, which rely on host physiology to in obesity and anorexia nervosa13,14. Such associations
maintain their basic biological processes such as feeding might reflect not only microbiota modifications sec-
Nutrition, Gut & Brain
Laboratory, Inserm UMR
and reproduction1–3. The complex interactions that ondary to malnutrition, but suggest a causal role of
1073, University of Rouen occur between gut bacteria and their host have been gut bacteria in altered regulation of energy metabo-
Normandy, 22 Boulevard conceptualized as symbiotic4–6. Indeed, bacterial contri- lism and possibly feeding behaviour 15,16. Indeed, a 2016
Gambetta, 76183 Rouen, butions to host physiology vary from digestion of die- study revealed that gut bacterial proteins can influence
France.
tary fibre and vitamin production to proper functioning host control of appetite dependent on the bacterial
serguei.fetissov@univ-rouen.
fr of the immune system, and even supplying ribosomal growth phase17.
machinery for protein synthesis7–9. Accumulating data This Review discusses the evolving conceptual model
doi:10.1038/nrendo.2016.150
Published online 12 Sep 2016 on the effects of the gut microbiota on brain function of appetite control supported by recent data, presents
Corrected online 18 Nov 2016 and behaviour represents another intrinsic part of such the symbiotic arrangement viewed from both sides (the
Inclination of this balance towards low food intake and control are functionally interrelated, the short-term con-
high energy expenditure must trigger appetite. The trol is responsible for the cyclic alternation of hunger
model assumes that the energy balance controlling and satiety on a daily basis. By contrast, the long-term
hypothalamic centres must be able to measure availa- control concerns appetite changes over extended periods
ble energy on both short-term and long-term scales, in to meet new energy demands.
order to meet immediate physiological energy needs and The current understanding of the mechanisms
to maintain the body weight set-point, respectively. This responsible for short-term control of appetite implies
homeostatic model does not, however, take into account that the nutritional-status-dependent release of hunger
the energy needs of gut bacteria that are distinct from and satiety hormones from the gastrointestinal tract
those of the host and, hence, might independently affect underlies the corresponding sensations and activates
host control of appetite. the brain pathways regulating feeding behaviour via
both humoral and neuronal pathways. Accordingly,
Homeostatic versus hedonic control food intake is proposed to be triggered by increased
Whereas the homeostatic model explains increased plasma concentrations of ghrelin, which occurs before
appetite and food intake as triggered by energy short- a meal, leading to activation of NPY and AgRP neu-
age, the feeding-associated feeling of pleasure, involving rons that express growth hormone secretagogue recep-
activation of the brain reward system, might underlie tor 1 (GHSR1)71. Next, the meal-induced increase in
the hedonic reason for eating. Although reward percep- plasma levels of PYY leads to inhibition of NPY and
tion is a normal part of all types of motivated behav- AgRP neurons via direct binding of PYY (residues
iour, including food intake, an abnormal hedonic drive 3–36) to the neuropeptide Y receptor type 2 (Y2R)72.
for eating can override homeostatic signals and might This inhibition of NPY and AgRP neurons removes
be present in bulimia nervosa51 or potentially drive their inhibitory GABA-ergic input on the ARC POMC
increased intake of palatable energy-rich foods leading neurons, thus activating the melanocortin satiety path-
to obesity 52,53. The key molecular pathways leading to way 73. Furthermore, the vagal afferents in the gut also
feeding reward involve the mesolimbic dopamine system express GHSR1 and Y2R and can be directly inhibited
that originates in the ventral tegmental area (VTA). The or stimulated by ghrelin and PYY (residues 3 − 36),
homeostatic appetite signals interact with the dopamine respectively 74–76. These vagal afferents also mediate
system both at its source and in target areas of the brain, satietogenic effects of cholecystokinin77.
including the nucleus accumbens and the hypothalamus; The cyclic changes in gut-hormone production and
the latter is involved in both homeostatic and hedonic release associated with food intake, thus, seem to be key
eating responses54–58. As the VTA dopaminergic neurons signalling events in the brain for short-term regulation
are protected by the blood–brain barrier, the possible of appetite. However, an open question remains on the
effects of gut bacteria-derived circulating molecules nature of mechanisms underlying the specific dynamics
on this dopamine system should by relayed via other of gut hormone release in both short-term and long-
neurons, such as those in the PBN59. term control of feeding. Another question concerns the
inconsistency of the homeostatic model with the short-
Intestinal signalling to the brain term regulation of appetite in regularly fed animals and
Both the homeostatic and the hedonic brain systems humans who continue to perfectly alternate their hun-
controlling appetite are activated by hormones derived ger and satiety feelings under conditions of stable and
from tissues and organs that signal various metabolic often excessive energy supply. As gut-bacteria-derived
processes, including energy storage and nutritional signals might access the intestinal hormone signalling
status60. The latter mainly involves mucosal entero pathways directly in the gut, and possibly also in the
endocrine cells in the gastrointestinal tract secreting systemic circulation, gut bacteria could conceivably use
several peptide hormones into the systemic circu- these pathways to influence host control of appetite in
lation61–63. Ghrelin, the only known orexigenic hor- order to sustain bacterial population size.
mone so far, is produced by the stomach64. Among
the anorexigenic hormones, produced mainly in the Gut bacterial population dynamics
small and large intestine, are cholecystokinin, which Nutrient-induced bacterial growth
contributes to satiation, and peptide tyrosine tyrosine Similar to the molecular pathways regulating energy
(PYY), which induces postprandial satiety 65–67. GLP1, metabolism in the host, the gut microbiota is also sub-
a product of the proglucagon gene expressed in jected to both short-term and long-term modifications.
enteroendocrine cells, acts as a satietogenic hormone The short-term daily changes include fluctuations
and further acts as an incretin, lowering postpran- in bacterial numbers whereas the long-term changes
dial glucose levels by stimulating insulin secretion68. mainly concern microbiota composition. Potentially,
Insulin has a major role in tissue glucose utilization a functional link might exist between changes in the
and also inhibits ghrelin production69. The gut-derived number and composition of gut bacteria and the short-
peptide hormones are involved in both short-term and term and long-term regulation of host appetite, respec-
long-term appetite control. For instance, while ghre- tively. In this section, some bacterial and host-related
lin secretion peaks before a regular meal, long-term factors involved in the regulation of bacterial popula-
negative energy balance stimulates a sustained increase tion dynamics are discussed, as they potentially relate
in its baseline levels70. Although both types of appetite to appetite control.
a Decline c
phase
Bacterial biomass
Figure 1 | Host factors influencing gut bacterial growth. a | Typical growth dynamics of a large versus small bacterial
Nature Reviews | Endocrinology
population illustrate the different durations of the exponential growth phase (Exp). b | Key host-related factors influencing
the balance between stimulation and inhibition of the bacterial cell number in the gut. The role of the immune system is
not shown, but it contributes by stabilizing the autochthonic community and neutralizing pathogenic invaders.
c | Presence of chemical and digestive factors and the transit time along the gastrointestinal tract might underlie the
increasing rostro–caudal gradient of bacterial content225. In the upper gut, the transit time is, apparently, shorter than the
time necessary for the bacterial population to reach the stationary growth phase (Stat), as calculated using the formula: t
(min) = G (generation time, 20 min, assumed based on in vitro experiments and in vivo infusions) × 3.3 log (minimal bacterial
number in the Stat phase, that is, 109) / bacterial number before multiplication (for example, 103 in the duodenum). Figure
1a modified from Cell Metab. 23 (2), Breton, J. et al. Gut commensal E. coli proteins activate host satiety pathways following
nutrient-induced bacterial growth. 324–334 © (2016), with permission from Elsevier.
Bacteria that live and thrive in communities are increase in bacterial numbers after nutrient provision
capable of intrinsically regulating their own growth in vitro17. This phenomenon has also been reproduced
rate and population size. Bacterial duplication results in in rats, where the stationary phase of gut bacteria was
exponential growth with the division rate determined induced 20 min after nutrient infusion into the large
by the time necessary for DNA replication and cell intestine17. An earlier study (in 1983) showed that the
division. Under conditions of regular nutrient supply, doubling time of E. coli implanted into the small or large
which keeps cell-division machinery in a ‘ready’ state78, intestine of mice was 2 h with an initial bacterial con-
it takes about 20 min for bacteria such as Escherichia coli centration of 2.4 × 104 cells per ml81. Nutrient-induced
to duplicate after provision of nutrients. After bacterial bacterial growth, thus, displays specific dynamics that
multiplication to a level of ~109 cells per ml, the bacte- are defined by the intrinsic properties of a bacterial
rial population enters a stationary growth phase that is population (including the gut microbiota).
characterized by maintenance of bacterial numbers over
several hours. Accordingly, the larger the initial bacterial Host control
population, the shorter the time it takes to get to the In contrast to in vitro conditions, where bacterial
stationary phase and vice versa (FIG. 1a) The stationary growth is regulated only by supplied nutrients and
phase is followed by a decline phase that is character- quorum sensing, in the gut, besides this bacteria-
ized by a progressive decrease in the number of bacteria intrinsic regulation, the growth of bacterial popula-
owing to the natural process of cell lysis79. The molecu- tions is constantly influenced by diet 82 and limited by
lar mechanisms responsible for onset of the stationary the host production of chemical and physical factors.
phase involve the production of autoinducers, which are A balance between these positive and negative influences
bacterial signals used in quorum sensing 80 (BOX 1). determines the relative stability of a bacterial population
Depending on growth dynamics, the production at each part of the gastrointestinal tract (FIG. 1b). Among
of 109 new bacterial cells in 20 min requires a starting the chemical factors that cause bacterial lysis and prevent
population of at least 109 cells, as only one division of mucosal adhesion83 are digestive juices containing gas-
each bacteria is possible in the time frame of 20 min. tric and bile acids and various digestive enzymes (FIG. 1c).
This cell-number generation serves as a critical mass for The main physical factors are the intestinal peristalsis and
induction of the stationary phase, as higher initial num- contractions of the colon and rectum, which lead to regu-
bers of E. coli still result in the same ~109 cells per ml lar elimination of bacteria by defecation. As an example,
Box 1 | Quorum sensing: a coordinated behaviour of bacterial populations Although influenced by diurnal rhythms and by short-
term effects of different foods or medicines, the com-
Autoinducers position of the gut microbiota is fairly stable over long
Signalling molecules produced by bacteria that act on neighbouring bacteria. periods of time, thereby demonstrating the ability of the
Exponential growth phase gut microbiota to maintain population homeostasis90–92.
Rapid increase in the number of cells, where each new cell further divides. Duration of Nevertheless, the gut microbiota also undergoes long-
the exponential growth phase of a bacterial population depends on the initial number term changes characteristic of age, sex, nutritional habits
of bacteria, as cell division stops once the number of cells reaches a specific threshold. and geographical location93–96. Moreover, various chronic
Stationary growth phase pathological conditions are characterized by dysbioses of
Maintenance of the size of a population by equilibrium between dividing and lysing the gut microbiota and are often accompanied by altered
cells. Duration of the stationary phase under in vitro conditions depends mainly on the appetite and food intake, which, in turn, can worsen the
availability of nutrients. In viscera, the stationary phase is shortened by host-regulated underlying disease, such as is the case for obesity, dia-
bacteria-lysing factors (FIG. 1). betes mellitus and cancer, as well as liver, kidney and
Growth dynamics cardiac diseases15,97–100. Although the specific bacterial
The growth dynamics of a bacterial population are regulated by both the availability of species that might contribute to altered appetite and
nutrients and by quorum sensing. Growing bacteria release autoinducers that act on body weight in these different diseases remain to be dis-
the same bacteria by inducing arrest of cell division and initiating the stationary phase. covered, a decreased bacterial diversity has been found in
Three main types of autoinducers have been described: specific oligopeptides;
both obesity 101 and anorexia nervosa102. Moreover, stud-
acylhomoserine lactones; and the S‑ribosylhomocysteine lyase (LuxS; also known as
ies of microbiome composition have found associations
autoinducer‑2 production protein LuxS and AI‑2 synthesis protein) inducer
4,5‑dihydroxy 2,3‑pentanedione. Production of these autoinducers is characterized by between the gut microbiota and BMI and adiposity 103.
high, medium and low energy cost, respectively, and their action is inversely related to For instance, increased presence of E. coli correlates
their specificity80. In addition to regulating the same bacterial species, autoinducers can inversely with BMI after bariatric surgery, as well as in
also act on other neighbouring bacterial species, thus coordinating changes in the patients with anorexia or obesity 104,105. An increased pres-
complex microbiota222. Furthermore, they can also act on host cells, for instance, ence of Enterobacteriaceae has been linked to low body
AI‑2 from E. coli interferes with cytokine production223 and some quorum-sensing weight in severely or chronically malnourished chil-
peptides are able to cross the blood–brain barrier224. Whether autoinducers influence dren88,106, and the presence of Akkermansia muciniphila
host pathways involved in appetite control remains to be determined. is associated with anti-obesogenic effects107. An initial
finding of increased ratios of Firmicutes to Bacteroidetes
in obesity 108 has, however, not been reproduced in other
a daily faecal output of a standard white man contains studies109. The causality between changes in the micro-
~15 g of bacteria, which corresponds to 55% of the dry biota composition and host feeding behaviour remains
weight of faeces84. The increasing rostro–caudal gas- to be established. Although changes in body weight
trointestinal tract gradient of bacterial numbers might, gain following gut microbiota transplantation in germ-
therefore, reflect the decreasing antibacterial power of free mice support this link, these experiments were not
both chemical and physical host-derived factors (FIG. 1c). always accompanied by immediate modification of feed-
In fact, the colon, which contains the highest density of ing behaviour in recipients. For instance, while germ-
bacteria, does not secrete digestive enzymes and has no free mice receiving gut microbiota from hyperphagic
peristalsis. Inversely, the transit time of ingested nutri- toll-like receptor 5 (TLR5)-deficient mice, indeed dis-
ents in the upper gut is shorter than the time necessary played increased food intake110, the hyperphagic trait
for bacteria to enter the stationary phase after nutrient- was not transmittable by microbiota transfer from
induced bacterial growth (FIG. 1c). Spontaneous bacterial leptin-deficient ob/ob mice13. Similarly, transplanta-
lysis is also present in the large intestine, and has been tion of gut microbiota from malnourished children
estimated to occur at a rate of approximately three to impaired normal weight gain in recipient mice with-
one cells in the rat 85. Notably, the colonic transit time is out causing notable modification of food intake111.
highly variable whereas decreases in the transit time are Another example comes from bariatric surgery,
associated with increased microbial richness and diver- showing that its therapeutic effects of decreasing
sity 86. The constant dynamics of bacterial growth, lysis appetite and body weight in obesity can be due to
and elimination, thus, keep the gut bacterial population modifications of the gut microbiota after the proce-
at a fairly stable level, which indicates that both bacteria dure112. Furthermore, growth dynamics of gut bacteria
and the host contribute to the maintenance of energy can be associated with altered metabolic control, even
homeostasis in a gut-bacterial population. The immune without notable changes in their relative abundance;
system also influences gut-bacterial homeostasis via both for example, an increased cell division rate of E. coli
innate and adaptive immunity, for example by producing correlates with occurrence of type 2 diabetes mellitus113.
secretory antibacterial IgA87–89.
Diurnal rhythms of gut microbiota and host feeding.
Links with microbiota composition Bacterial growth and host-related rhythmic factors,
Host feeding and energy metabolism including regular nutrient supply, underlie the diurnal
Gut microbiota and long-term changes in energy fluctuations in bacterial composition that have been
metabolism. Long-term regulation of appetite serves described in the past few years as circadian rhythms
the physiological needs of the organism, which vary of the gut microbiota 114–117. Indeed, metagenomic
with age, sex, reproductive status and the environment. analysis revealed that all gut bacteria in both mice
and humans undergo rhythmic changes over the day Bacterial and host energy homeostasis
with ~20% of species exhibiting diurnal fluctuation Energy needs of gut bacteria
in their relative abundance118. A bi‑directional link The gut microbiota, as well as bacteria inhabiting other
between hypothalamic clock mechanisms and micro- niches in the body, depend fully on nutrients and water
biota composition suggests an effect of sunlight on gut provided by their host. Some ingested nutrients can be
bacterial physiology 116,117. Bacterial rhythms can also used directly by gut bacteria, such as in the upper gut,
be driven by the prevailing daily distribution of feed- but the majority of intestinal bacteria (located in the
ing; changing the feeding pattern between the dark and large intestine) receive food-remnant nutrients that have
the light phases shifts the diurnal distribution of bac- already been digested and absorbed (with the exception
terial abundance114,115. It is of a practical interest that of non-digestible fibre). When feeding starts and food
imposed alterations in circadian rhythms are accom- reaches the oral cavity, nutrients in the circulation and
panied by obesogenic changes in gut microbiota114,119. tissues, and water, are released into the intestinal lumen
Inversely, obesity is characterized by increased eating via the cephalic phase response126, which indicates that
in the resting phase. For instance, whereas increased animals feed their gut microbiota by secreting nutrients
light-phase resting eating underlies hyperphagia in into the gut as often as they eat and, even under peri-
obese Zucker rats120, nocturnal hyperphagia is often ods of fast, the body continues to secrete these factors.
observed in humans with obesity and is called the Considering the remarkable long-term stability of the
‘night-eating syndrome’ (REF. 121). An involvement of gut microbial community, gut bacteria should be able to
dysbiotic microbiota in such mistimed overeating is regulate their own energy balance, that is, to use the host
suspected, on the basis of improved metabolic param- mechanisms for their energy needs.
eters in obese rodents after targeting their gut bacteria Gut bacteria compose a metabolically active organ
using antibiotics122,123 or in humans with obesity after weighing between 1 kg and 2 kg in adult humans. By
bariatric surgery 124,125. Associations between diurnal calculating how much energy this organ might con-
rhythms of eating and microbiota composition might, sume, one can determine its contribution to the daily
thus, underlie different metabolic requirements dur- needs of the energy intake of the host. Nutrient-induced
ing the active and resting states, which is a longer-term bacterial growth in a large population, such as in the
parameter than alternations in hunger and satiety. The colon, results in simple duplication of bacteria. Bacterial
peak amplitude during 24 h changes in relative bacte- duplication is energetically costly, requiring 1 mol of
rial abundance in human faecal samples is ~50%114, ATP per 10.3 g of cells127. As the complete hydrolysis
which points to a high turnover rate of the gut bacterial of ATP yields ΔG° = −10.9 kcal/mol, it can be roughly
population. estimated that 1 g of gut bacteria needs 1 kcal (4.18 kJ) of
energy for duplication. Accordingly, to duplicate 1–2 kg
of gut bacteria, the energy needs would be 1,000–2,000
kcal. The latter numbers represent about half of the
Energy intake as food
recommended daily energy intake for a healthy indi-
vidual, depending on age, sex and level of physical
Used by gut Used by host activity. Luckily, because of quorum sensing, bacterial
bacteria division is limited, and the total weight of 109 cells per
During starvation, ml of newly formed bacteria in a 1,000 ml volume of
gut bacteria receive the large intestine is only ~1.0 g. Bacterial growth in the
energy only from
host energy stores colon after each meal-induced nutrient provision will,
therefore, consume ~1 kcal. However, for the daily
Bacterial growth Metabolic needs energy consumption by gut bacteria, this value should
be increased by including the energy needs of dividing
Energy dissipation bacteria in other parts of the gut and multiplied by the
number of meals, snacks and also presence of non-food
Host energy use:
related factors.
Direct Indirect
To exactly calculate the energy needs of gut bacteria,
Figure 2 | Distribution of Nature
food-derived
Reviewsenergy
| Endocrinology additional studies are required, focusing on their daily
between the host and gut bacteria. Energy derived turnover as ~15 g of bacteria excreted daily in human
from ingested nutrients is directly and indirectly available faeces should represent only a fraction of lost bacteria
to both gut bacteria and the host. Gut bacteria use this beside those lysed in the gastrointestinal tract. A 27%
energy for bacterial multiplication, which results in decrease in metabolic rate (oxygen consumption) shown
population maintenance. In turn, bacteria generate energy in germ-free mice128 might be indicative of an approx-
as a part of bacterial catabolism and nutrient processing,
imate contribution of the gut microbiota to the total
which releases energy that is available to the host. The
host, therefore, receives energy derived both directly from
energy balance in mice. The example of the constitu-
nutrient digestion and indirectly from gut bacteria to meet tive role of gut bacteria in energy consumption, thus,
its metabolic needs. Part of the energy generated by both illustrates that it should be taken into account in the
gut bacteria and the host is dissipated as heat and waste. energy balance of the host (FIG. 2). Moreover, it implies
Alternatively, during starvation, gut bacteria receive that alterations in the dynamics of bacterial growth, for
energy only from host energy stores. example, those resulting in bacterial overgrowth, will
be accompanied by increased energy demands of gut However, an absence of the gut microbiota in mice
bacteria, which suggests that it might lead to increased does not always protect them from gaining weight and
appetite in the host. adiposity on different types of HFD, which suggests a
role for diet composition in energy assimilation131,132.
Gut bacteria as a source of energy Hyperphagia, which can be a compensatory response to
Energy exchange through the food chain represents a inefficient energy assimilation, is present in patients after
universal link between all organisms129. The energy (in small bowel resection133,134. The bacterial role in host
the form of ATP) produced by gut bacteria is used not energy extraction is further highlighted by an increased
only for their own growth but also becomes available gut-mucosal surface in obesity that does not, however,
to the host. Bacteria release enzymes and metabolites prevent hyperphagia135. Moreover, the bacterial capacity
that help digestion of nutrients, including fibre, that are of energy extraction depends on the individual compo-
not digestible by the host. For instance, E. coli proteins sition of the gut microbiota, as has been shown to be
released after nutrient-induced bacterial growth can increased in obesity 13.
synthesize ATP that is used by the host 17. Moreover, the During host starvation, gut bacteria receive nutrients
continuously renewing bacterial biomass, upon bacterial only from the host, slowly depleting energy reserves
lysis, liberates in the gut all their structural molecules, in fat, liver and muscle. Although host starvation (for
including proteins constituting ~24% of the bacterial example in anorexia nervosa) leads to an altered compo-
cell. This indirect bacteria-mediated supply of energy sition of gut microbiota102,136, gut bacteria continue their
and nutrients to the host seems a more efficient way of metabolic activities and maintain their community at
energy provision than direct extraction from nutrients the expense of the host. Changes in the composition of
by the host digestive system. Indeed, germ-free mice the gut microbiota during starvation favours species that
fed a Western diet display increased food intake but are better adapted to low energy supply, which, in turn,
low adiposity with otherwise similar body weight, and might help the host to survive energy deprivation137.
can be resistant to such high-fat diets (HFDs)117,128,130. Gut bacteria might, thus, supply the host with energy
more efficiently than the host extracting it directly from
nutrients using its own metabolic processes. The fact
a Satiety perception b E. coli and colonic bacteria growth
that germ-free mice can regulate food intake to achieve
the body weight set-point indicates that gut bacteria are
Bacterial biomass
Plasma levels
Prandial phase Postprandial phase Preprandial phase be considered with caution, as although the circadian
fluctuations of bacterial faecal content correlate with
↑ Satiety signalling feeding behaviour, the direct effect of a single spontane-
Meal to the host New meal
ous meal on bacterial growth still remains to be shown.
The molecular mechanisms supporting this model are
Stat discussed in the next section.
dynamics in the colon
ClpB and other bacterial proteins have been detected in neuropeptide expression as observed in both obesity,
the gut mucosa17,179 and, hence, might act directly on the anorexia nervosa and cachexia188. Lipopolysaccharide-
intestinal epithelium, including enteroendocrine cells. induced low-grade inflammation might also interfere
Remarkably, colonic infusion in rats of E. coli proteins with the long-term regulation of energy metabolism
from the stationary, but not exponential, growth phase via increased production of endocannabinoids, such as
increased portal levels of PYY17. By contrast, proteins anandamide189. These lipid mediators are elevated in the
from the exponential phase increased portal levels hypothalamus of obese rodents and are known for their
of GLP1 (REF. 17). orexigenic effects by activating cannabinoid CB1 recep-
Such differential effects of bacterial proteins sug- tors expressed in inhibitory presynaptic GABA terminals
gest their involvement in regulation of the sequential in orexigenic neurons190,191.
release of satiety hormones; acute release of GLP1 with Other products of bacterial metabolism relevant to
an incretin function during the first 20 min of the bacte- appetite control and sensed by the host include ATP and
rial exponential growth phase, and a prolonged release lactate, which are ubiquitous energy substrates. Beside
of PYY initiated after the onset of the stationary phase17. its key role as an energy substrate, circulating ATP might
Although these effects of bacterial proteins might not also signal the state of energy balance192. Catabolism of all
only depend on ClpB, this protein seems to have a key macronutrients results in increased production of ATP,
role in the host appetite control effects of E. coli. Indeed, whereas anabolism consumes ATP and results in a rela-
intragastric gavage of ClpB-deficient E. coli failed to tive increase of AMP. Changes in the ATP to AMP ratio
reduce food intake and body weight in mice, effects regulate 5ʹ‑AMP-activated protein kinase, which is linked
observed with the wild-type ClpB-expressing strain of to central appetite pathways193. ATP receptors are present
E. coli 176. The types of melanocortin receptor (MCR) on nerve cells, including the NPY/AgRP neurons194. The
that might possibly transmit ClpB action on the gut efficient production of ATP by E. coli proteins in vitro
satiety pathway remain to be clarified. In fact, although suggests that nutrient-induced increases in bacterial bio-
L‑cells express MC4R180, a 14‑amino acid fragment of mass in the gut, including the enzymes necessary for ATP
ClpB containing the αMSH-like epitope was shown to production, might increase both local and systemic
activate MC1R but not MC4R181. The gut epithelium ATP levels and lead to modulation of the activity of cells
abundantly expresses functional MC1R, which has a constituting appetite-regulating pathways.
protective role against inflammation182. However, while Lactate, the preferred substrate for neurons195,196,
MC1R is probably present on enterocytes, it might not be is abundantly produced in the gut by Lactobacilli, and
expressed by the enteroendocrine cells180. In addition to also by Enterobacteriaceae and Bifidobacteria, from fer-
the demonstrated effects of gut bacterial proteins on the mentable carbohydrates197. Lactate levels are increased
intestinal satiety pathway, future studies should explore postprandially in the portal blood, whereas portal infu-
their effects on the hunger pathway, that is, the release of sion of lactate reduces meal size, thus supporting a role
ghrelin from the stomach. In fact, ghrelin production has for gut-derived lactate in satiation198,199. SCFAs can also
been associated with presence of Helicobacter pylori in the participate in long-term regulation of energy metabo-
gastrointestinal tract183. lism; systemic administration of acetate in mice failed
to stimulate secretion of PYY and GLP1 but decreased
Gut bacteria in long-term control of appetite. The caus- appetite through central effects200, whereas butyrate and
ative associations between host energy metabolic pheno- propionate stimulated intestinal gluconeogenesis201.
types and composition of the gut microbiota, including An immunoassay for Enterobacteriaceae ClpB has
its long-term stability, point to the existence of molecular been developed and used to show that this bacterial pro-
links between gut bacteria and the hypothalamic pon- tein is naturally present at variable levels in plasma of
derostatic network. To directly influence this neuronal both rodents and humans17,202. Although plasma levels
circuitry, bacterial components should be present in the of ClpB in rats remained stable after colonic infusion of
circulation where they might reflect the relative abun- nutrients or E. coli proteins, plasma ClpB levels correlated
dance of a gut bacterial population. Although bacterial directly with ClpB DNA levels in the gut microbiota17.
DNA has been detected in many organs and tissues184, data These results, coupled with ClpB activation of ARC
on potential functional effects of bacterial components POMC neurons17, support a ClpB-mediated humoral
circulating in the host are limited. link between the gut microbiota and hypothalamic neuro-
Lipopolysaccharide is probably the only bacterial peptidergic circuitry. Furthermore, E. coli proteins in the
fragment that has been relatively well studied con- ClpB-rich stationary phase increased c‑Fos in the ARC
cerning its systemic effects; plasma lipopolysaccharide POMC, VMN and CeA neurons, demonstrating activa-
levels increase postprandially after a high-fat meal tion of both the sensory and integrative centres of the ano-
and are associated with increased energy intake185,186. rexigenic circuitry in the brain. Such anorexigenic effects
Although high doses (≥0.1 mg/kg) of lipopolysaccha- of E. coli proteins are in agreement with data associating
ride are known to decrease food intake in animals, inde- the relative abundance of Enterobacteriaceae with human
pendent from inducing fever 187, its physiological role in metabolic phenotypes, including obesity and anorexia
the regulation of long-term energy balance is unclear. nervosa. Nevertheless, a 2016 study in patients with eating
Lipopolysaccharide, via its downstream messenger disorders found that plasma levels of ClpB did not corre-
IL‑1β induces a chronic inflammatory state that includes late significantly with BMI, but did correlate with other
hypothalamic inflammation, with a similar pattern of feeding-related parameters such as bulimia nervosa202.
Long-term regulation of appetite by gut bacteria might According to this model, the central sites responsive
also involve the immune system. Diet-induced changes in to short-term and long-term signals controlling appetite
microbiota and microbial antigens has profound effects might integrate both host-derived and bacteria-derived
on gut mucosal immunity stimulating maturation of signals, that is, they will respond to the homeostatic needs
the immune system and antibody production9,203–206. of both host and gut bacteria. Integration of short-term
Immunoglobulins are among the most stable biomol- control might occur in the gut, where gut sensory systems
ecules; the 26 day half-life of IgG207 is a characteristic are activated by nutrients released by the host in response
suitable for long-term effects. Circulating immunoglob- to nutrient ingestion, as well as by signals derived from
ulins that react with several appetite-regulating peptide gut bacteria during their different growth phases. The
hormones, including αMSH and ghrelin, are present hypothalamus and other central sites might integrate
in healthy individuals and their plasma levels correlate the circulating long-term signals of host energy storage
with behavioural traits208–212. In particular, αMSH-reactive and bacterial community energy status, in part assisted
IgG and IgM are associated with eating disorder-related by antibodies against gut bacterial proteins that are
traits, which suggests that they have a role in modulation crossreactive with hunger and satiety hormones.
of αMSH signalling via MC4R209. Such a modulatory This new model could help to resolve some con-
role of IgG is not necessarily inhibitory but might, con- ceptual problems inherited from the model of a closed,
versely, reduce the threshold of MC4R activation213,214. feedback-driven self-regulatory system. For instance, the
Furthermore, circulating IgG might protect highly unsta- apparent inability of the central sites to homeostatically
ble peptide hormones from degradation by functioning respond to host-derived signals during chronic altera-
as a peptide carrier. Such a role has been shown for tions of host energy balance is observed in obesity. In fact,
ghrelin-reactive IgG, which has slightly increased micro increased appetite and food intake in obesity cannot be
molar affinity for ghrelin in individuals with obesity, explained by energy shortage as, on the contrary, excess
thereby enhancing the orexigenic effects of ghrelin215. energy is available in the body. A constitutive role of gut
Altered affinity kinetics of IgG for ghrelin might bacteria in the regulation of host energy homeostasis
be a response of the immune system to changes in might explain such inconsistencies. In conditions such
the composition of the gut microbiota induced by as obesity, bacteria-derived signals might dominate over
an obesogenic diet; however, bacterial ghrelin-like host-derived signals to enable gut bacteria to maintain
antigens have yet to be identified216. Nevertheless, their own energy homeostasis despite discrepancy with
the identification of ClpB as an antigenic E. coli- host energy needs. For instance, bacterial signals from the
derived protein responsible for production of immu- gut might compete with increased plasma levels of ano-
noglobulins crossreactive with αMSH provides an rexigenic homeostatic signals in obesity such as leptin219,
example of molecular mimicry underlying a role of whereas modification of the gut microbiota by prebiotics
the gut microbiota at the origin of peptide-hormone- in obese mice improves leptin sensitivity 220. This possibil-
reactive IgG176. Furthermore, an altered immune response ity is in line with a theory of genetic competition between
to some bacterial antigens might also be involved in the the host and its microbiota for food sources221.
pathophysiology of eating disorders. For instance,
the crossreactivity of antibodies against ClpB and αMSH Conclusions
could alter αMSH-mediated signalling of satiety and This Review has focused on the molecular pathways of
anxiety, and possibly lead to eating disorders176,202,209,217. hunger and satiety in the context of gut bacterial biology,
Specific bacterial antigens in the gut microbiota are, and proposes a new approach to an old basic physiolog-
thus, plausible long-term factors regulating appetite ical question concerning the mechanisms of appetite
control under both normal and pathological conditions. control. A new model is justified by the accumulating
data implicating the gut microbiota in host metabolic
Integration of gut bacterial growth phenotypes, including eating behaviour. The key con-
From the point of view of co‑evolution of animals ceptual change is the suggestion that nutrient-induced
and their microbiota218, animals have probably phylo growth dynamics of the gut microbiota might partici-
genetically developed their molecular pathways of pate, on a regular daily basis, in the host regulation of
hunger and satiety as an adaptation to bacterial chem- appetite so as to maintain bacterial homeostasis. In fact,
ical signals. In this symbiotic relationship, interactions the energy necessary for gut bacterial multiplication
between the microbiota and their host should not be could constitute an important part of the daily energy
in conflict, but instead, underlie a supraorganismal requirements of a healthy person and might shed light
physiology of energy metabolism regulation. The cur- on the mechanisms underlying increased food intake
rent homeostatic model of alternation of hunger and during obesity. These molecular mechanisms include
satiety is based on activation of orexigenic and ano- bacteria-derived chemical signals that might transmit to
rexigenic pathways in the brain that depend on the the host the energy status of the gut microbiota that has
energy status of the body. Considering data linking a constitutive role in short-term regulation of appetite.
gut bacteria to host regulation of feeding behaviour, a Moreover, the microbiota composition might be a deter-
new integrative model of appetite control can be pro- mining factor in fine tuning the regulation of host energy
posed that incorporates gut bacterial growth dynamics metabolism (specifically in the long term), including sspe-
with the host molecular pathways for homeostatic cific immune changes. This new model implies that pres-
control of appetite (FIG. 6). ence of a ‘healthy’ gut microbiota is necessary to ensure
Fullness
PVN: OT, CRH Meal Meal
Food VMN: BDNF
ingestion
Hunger
Hypothalamus 20 min 5h
NPY
ARC and/or 2 POMC NTS
AgRP
TH
Nodose ganglion
Fat
Figure 6 | Bacteria–host integrative homeostatic model of appetite control. Hypothetical model of appetite control
that integrates gut bacteria-derived signals into host molecular pathways that control energyNature Reviews | Endocrinology
homeostasis. According to
this model, activation of host intestinal satiety pathways by nutrients is integrated with nutrient-induced dynamics of gut
bacterial growth (1). For example, glucagon-like peptide 1 (GLP1) is stimulated during the exponential growth phase (Exp)
and functions as an incretin via stimulation of insulin; conversely, peptide tyrosine tyrosine (PYY) is stimulated during the
stationary growth phase (Stat), which occurs 20 min after nutrient supply and leads to activation of the anorexigenic
circuitry. The contribution of the gut microbiota to long-term control of appetite might involve systemic effects of
bacterial components whose plasma levels depend on the composition of the microbiota. For example, caseinolytic
peptidase B (protein homologue ClpB), an E. coli-derived antigen-mimetic protein of α-melanocyte stimulating hormone
(αMSH), can activate POMC ARC neurons in a similar way to leptin, which suggests that the hypothalamic peptidergic
network might integrate both host and gut microbiota energy states (2). Furthermore, while immunoglobulins modulate
long-term stability and the functional activity of hunger and satiety hormones, such as ghrelin and αMSH, their plasma
levels and affinity are influenced by molecular mimicry of gut bacterial antigens, such as between ClpB and αMSH (3).
These natural antibodies might serve as molecular ‘bridges’ in communication between the gut microbiota and host appe-
tite-controlling pathways as well as for other homeostatic functions. AgRP, agouti-related protein; ARC, arcuate nucleus;
BDNF, brain-derived neurotrophic factor; CeA, central nucleus of amygdala; CGRP, calcitonin gene-related peptide; CRH,
corticotropin-releasing factor; EEC, enteroendocrine cell; MC4R, melanocortin receptor type 4; NPY, neuropeptide
tyrosine; NTS, nucleus of the solitary tract; OT, oxytocin; PBN, parabrachial nucleus; POMC, pro-opiomelanocortin; PVN,
paraventricular nucleus; TH, tyrosine hydroxylase; VMN, ventromedial nucleus.
optimal regulation of host appetite both for short-term help guide future studies on the involvement of bacterial
and long-term goals. Determining the key bacterial spe- signals in the regulation of host energy homeostasis. The
cies characteristic of ‘healthy’ versus dysbiotic microbiota, identification of additional bacterial molecules involved in
and the underlying mechanisms of their action, is a hot appetite control should help to create the knowledge nec-
topic of research. The integrative model of appetite control essary for improved prevention and treatment of appetite
presented here might provide a conceptual framework to problems and their metabolic consequences.
1. Turnbaugh, P. J. et al. The human microbiome project. 29. Elmquist, J. K., Maratos-Flier, E., Saper, C. B. & 53. Denis, R G. P. et al. Palatability can drive feeding
Nature 449, 804–810 (2007). Flier, J. S. Unraveling the central nervous system independent of AgRP neurons. Cell Metab. 22,
2. Sekirov, I., Russell, S. L., Antunes, L. C. & Finlay, B. B. pathways underlying responses to leptin. 646–657 (2015).
Gut microbiota in health and disease. Physiol. Rev. Nat. Neurosci. 1, 445–450 (1998). 54. Avena, N. M., Rada, P. & Hoebel, B. G. Underweight
90, 859–904 (2010). 30. Luquet, S., Perez, F. A., Hnasko, T. S. & Palmiter, R. D. rats have enhanced dopamine release and blunted
3. Dethlefsen, L., McFall-Ngai, M. & Relman, D. A. An NPY/AgRP neurons are essential for feeding in adult acetylcholine response in the nucleus accumbens while
ecological and evolutionary perspective on human– mice but can be ablated in neonates. Science 310, bingeing on sucrose. Neuroscience 156,
microbe mutualism and disease. Nature 449, 683–685 (2005). 865–871 (2008).
811–818 (2007). 31. Bumaschny, V. F. et al. Obesity-programmed mice are 55. Jerlhag, E. et al. Ghrelin administration into tegmental
4. Gilbert, S. F., Sapp, J. & Tauber, A. I. A symbiotic view rescued by early genetic intervention. J. Clin. Invest. areas stimulates locomotor activity and increases
of life: we have never been individuals. Q. Rev. Biol. 122, 4203–4212 (2012). extracellular concentration of dopamine in the nucleus
87, 325–341 (2012). 32. Loh, K., Herzog, H. & Shi, Y.‑C. Regulation of energy accumbens. Addict. Biol. 12, 6–16 (2007).
5. Bukharin, O. V. & Perunova, N. B. Microsymbiocenosis homeostasis by the NPY system. Trends Endocrinol. 56. Meguid, M. M. et al. Hypothalamic dopamine and
(UrDepart RAS, 2014). Metab. 26, 125–135 (2015). serotonin in the regulation of food intake. Nutrition 16,
6. Bäckhed, F., Ley, R. E., Sonnenburg, J. L., 33. Ollmann, M. M. et al. Antagonism of central 843–857 (2000).
Peterson, D. A. & Gordon, J. I. Host–bacterial melanocortin receptors in vitro and in vivo by agouti- 57. Legrand, R., Lucas, N., Breton, J., Déchelotte, P. &
mutualism in the human intestine. Science 307, related protein. Science 278, 135–138 (1997). Fetissov, S. O. Dopamine release in the lateral
1915–1920 (2005). 34. Hahn, T. M., Breininger, J. F., Baskin, D. G. & hypothalamus is stimulated by α‑MSH in both the
7. Fergus, C., Barnes, D., Alqasem, M. & Kelly, V. The Schwartz, M. W. Coexpression of Agrp and NPY in anticipatory and consummatory phases of feeding.
queuine micronutrient: charting a course from fasting-activated hypothalamic neurons. Psychoneuroendocrinology 56, 79–87 (2015).
microbe to man. Nutrients 7, 2897–2929 (2015). Nat. Neurosci. 1, 271–272 (1998). 58. Barson, J. R., Morganstern, I. & Leibowitz, S. F.
8. Sonnenburg, E. D. et al. Diet-induced extinctions in 35. Tatemoto, K. Neuropeptide Y — a novel brain peptide Similarities in hypothalamic and mesocorticolimbic
the gut microbiota compound over generations. with structural similarities to peptide YY and pancreatic circuits regulating the overconsumption of food and
Nature 529, 212–215 (2016). polypeptide. Nature 296, 659–660 (1982). alcohol. Physiol. Behav. 104, 128–137 (2011).
9. Hooper, L. V., Littman, D. R. & Macpherson, A. J. 36. Fan, W., Boston, B. A., Kesterson, R. A., Hruby, V. J. & 59. Norgren, R., Hajnal, A. & Mungarndee, S. S. Gustatory
Interactions between the microbiota and the immune Cone, R. D. Role of melanocortinergic neurons in reward and the nucleus accumbens. Physiol. Behav.
system. Science 336, 1268–1273 (2012). feeding and the agouti obesity syndrome. Nature 89, 531–535 (2006).
10. Cryan, J. F. & Dinan, T. G. Mind-altering 385, 165–168 (1997). 60. Broberger, C. Brain regulation of food intake and
microorganisms: the impact of the gut microbiota on 37. Poggioli, R., Vergoni, A. V. & Bertolini, A. ACTH-(1–24) appetite: molecules and networks. J. Intern. Med.
brain and behaviour. Nat. Rev. Neurosci. 13, and α-MSH antagonize feeding behavior stimulated by 258, 301–327 (2005).
701–712 (2012). κ opiate agonists. Peptides 7, 843–848 (1986). 61. Mutt, V. in Advances in Metabolic Disorders Vol. 11 (eds
11. Gautron, L., Elmquist, J. K. & Williams, K. W. Neural 38. Harris, J. I. & Lerner, A. B. Amino-acid sequence of the Luft, R. & Levine, R.) 1–545 (Academic Press, 1988).
control of energy balance: translating circuits to α-melanocyte-stimulating hormone. Nature 179, 62. Wren, A. M. & Bloom, S. R. Gut hormones and appetite
therapies. Cell 161, 133–145 (2015). 1346–1347 (1957). control. Gastroenterology 132, 2116–2130 (2007).
12. Aigner, M. Treasure, J., Kaye, W., Kasper, S. & The 39. Cone, R. D. Anatomy and regulation of the central 63. Janssen, S. & Depoortere, I. Nutrient sensing in the
WFSBP Task Force on Eating Disorders. World melanocortin system. Nat. Neurosci. 8, 571–578 gut: new roads to therapeutics? Trends Endocrinol.
Federation of Societies of Biological Psychiatry (2005). Metab. 24, 92–100 (2013).
(WFSBP) guidelines for the pharmacological treatment 40. Broberger, C., Johansen, J., Johansson, C., 64. Kojima, M. et al. Ghrelin is a growth-hormone-
of eating disorders. World J. Biol. Psychiatry 12, Schalling, M. & Hökfelt, T. The neuropeptide Y/agouti releasing acylated peptide from stomach. Nature 402,
400–443 (2011). gene-related protein (AGRP) brain circuitry in normal, 656–660 (1999).
13. Turnbaugh, P. J. et al. An obesity-associated gut anorectic, and monosodium glutamate-treated mice. 65. Feinle-Bisset, C. Upper gastrointestinal sensitivity to
microbiome with increased capacity for energy Proc. Natl Acad. Sci. USA 95, 15043–15048 (1998). meal-related signals in adult humans — relevance to
harvest. Nature 444, 1027–1131 (2006). 41. Zhao, R., Chen, H. & Sharp, B. M. Nicotine-induced appetite regulation and gut symptoms in health,
14. Armougom, F., Henry, M., Vialettes, B., Raccah, D. & norepinephrine release in hypothalamic obesity and functional dyspepsia. Physiol. Behav. 162,
Raoult, D. Monitoring bacterial community of human paraventricular nucleus and amygdala is mediated by 69–82 (2016).
gut microbiota reveals an increase in lactobacillus in N‑methyl-d‑aspartate receptors and nitric oxide in the 66. Degen, L., Matzinger, D., Drewe, J. & Beglinger, C. The
obese patients and methanogens in anorexic patients. nucleus tractus solitarius. J. Pharmacol. Exp. Ther. effect of cholecystokinin in controlling appetite and
PLoS ONE 4, e7125 (2009). 320, 837–844 (2007). food intake in humans. Peptides 22, 1265–1269
15. Rosenbaum, M., Knight, R. & Leibel, R. L. The gut 42. Appleyard, S. M. et al. Proopiomelanocortin neurons (2001).
microbiota in human energy homeostasis and obesity. in nucleus tractus solitarius are activated by visceral 67. Batterham, R. L. et al. Gut hormone PYY(3–36)
Trends Endocrinol. Metab. 26, 493–501 (2015). afferents: regulation by cholecystokinin and opioids. physiologically inhibits food intake. Nature 418,
16. Fetissov, S. O. & Déchelotte, P. The new link between J. Neurosci. 25, 3578–3585 (2005). 650–654 (2002).
gut–brain axis and neuropsychiatric disorders. Curr. 43. Appleyard, S. M. et al. Visceral afferents directly 68. Drucker, D. J. Biologic actions and therapeutic
Opin. Clin. Nutr. Metab. Care 14, 477–482 (2011). activate catecholamine neurons in the solitary tract potential of the proglucagon-derived peptides.
17. Breton, J. et al. Gut commensal E. coli proteins activate nucleus. J. Neurosci. 27, 13292–13302 (2007). Nat. Clin. Pract. Endocrinol. Metab. 1, 22–31 (2005).
host satiety pathways following nutrient-induced 44. Larsen, P. J., Tang-Christensen, M., Holst, J. J. & 69. Flanagan, D. E. et al. The influence of insulin on
bacterial growth. Cell Metab. 23, 1–11 (2016). Ørskov, C. Distribution of glucagon-like peptide‑1 and circulating ghrelin. Am. J. Physiol. Endocrinol. Metab.
18. Thornton, S. N. Thirst and hydration: physiology and other preproglucagon-derived peptides in the rat 284, E313–E316 (2003).
consequences of dysfunction. Physiol. Behav. 100, hypothalamus and brainstem. Neuroscience 77, 70. Müller, T. D. et al. Ghrelin. Mol. Metab. 4, 437–460
15–21 (2010). 257–270 (1997). (2015).
19. Schwartz, M. W. Central nervous system control of 45. Mimee, A., Kuksis, M. & Ferguson, A. V. α‑MSH 71. Nakazato, M. et al. A role for ghrelin in the central
food intake. Nature 404, 661–671 (2000). exerts direct postsynaptic excitatory effects on NTS regulation of feeding. Nature 409, 194–198 (2001).
20. Hökfelt, T., Bartfai, T. & Bloom, F. Neuropeptides: neurons and enhances GABAergic signaling in the 72. Wynne, K. & Bloom, S. R. The role of oxyntomodulin
opportunities for drug discovery. Lancet Neurol. 2, NTS. Neuroscience 262, 70–82 (2014). and peptide tyrosine-tyrosine (PYY) in appetite control.
463–472 (2003). 46. Wang, D. et al. Whole-brain mapping of the direct Nat. Clin. Pract. Endocrinol. Metab. 2, 612–620
21. Carter, M. E., Soden, M. E., Zweifel, L. S. & inputs and axonal projections of pro- (2006).
Palmiter, R. D. Genetic identification of a neural circuit opiomelanocortin (POMC) and agouti-related 73. Cowley, M. A. et al. Leptin activates anorexigenic
that suppresses appetite. Nature 503, 111–114 (2013). peptide (AgRP) neurons. Front. Neuroanat. 9, 40 POMC neurons through a neural network in the
22. Atasoy, D., Betley, J. N., Su, H. H. & Sternson, S. M. (2015). arcuate nucleus. Nature 411, 480–484 (2001).
Deconstruction of a neural circuit for hunger. Nature 47. Cone, R. D. The central melanocortin system and 74. Date, Y. et al. The role of the gastric afferent vagal
488, 172–177 (2012). energy homeostasis. Trends Endocrinol. Metab. 10, nerve in ghrelin-induced feeding and growth hormone
23. Norsted, E., Gömüç, B. & Meister, B. Protein 211–216 (1999). secretion in rats. Gastroenterology 123, 1120–1128
components of the blood–brain barrier (BBB) in the 48. Friedman, J. M. & Mantzoros, C. S. 20 years of leptin: (2002).
mediobasal hypothalamus. J. Chem. Neuroanat. 36, from the discovery of the leptin gene to leptin in our 75. Koda, S. The role of the vagal nerve in peripheral PYY3–
107–121 (2008). therapeutic armamentarium. Metabolism 64, 1–4 36‑induced feeding reduction in rats. Endocrinology
24. Langlet, F. et al. Tanycytic VEGF‑A boosts blood– (2015). 146, 2369–2375 (2005).
hypothalamus barrier plasticity and access of 49. Hillebrand, J. J., de Wied, D. & Adan, R. A. 76. Cui, R. J., Li, X. & Appleyard, S. M. Ghrelin inhibits
metabolic signals to the arcuate nucleus in response Neuropeptides, food intake and body weight visceral afferent activation of catecholamine neurons in
to fasting. Cell Metab. 17, 607–617 (2013). regulation: a hypothalamic focus. Peptides 23, the solitary tract nucleus. J. Neurosci. 31, 3484–3492
25. Schwartz, G. J. Brainstem integrative function in the 2283–2306 (2002). (2011).
central nervous system control of food intake. Forum 50. Bouret, S. G., Draper, S. J. & Simerly, R. B. Trophic 77. Rehfeld, J. F. Cholecystokinin as satiety signal.
Nutr. 63, 141–151 (2010). action of leptin on hypothalamic neurons that regulate Int. J. Obes. (Lond.) 5, 465–469 (1981).
26. Waterson, M. J. & Horvath, T. L. Neuronal regulation feeding. Science 304, 108–110 (2004). 78. Wang, J. D. & Levin, P. A. Metabolism, cell growth and
of energy homeostasis: beyond the hypothalamus and 51. Simon, J. J. et al. Neural signature of food reward the bacterial cell cycle. Nat. Rev. Microbiol. 7,
feeding. Cell Metab. 22, 962–970 (2015). processing in bulimic-type eating disorders. Soc. Cogn. 822–827 (2009).
27. Richard, D. Cognitive and autonomic determinants of Affect. Neurosci. http://dx.doi.org/10.1093/scan/ 79. Rice, K. C. & Bayles, K. W. Molecular control of
energy homeostasis in obesity. Nat. Rev. Endocrinol. nsw049 (2016). bacterial death and lysis. Microbiol. Mol. Biol. Rev. 72,
11, 489–501 (2015). 52. Muenzberg-Gruening, H., Qualls-Creekmore, E., Yu, S., 85–109 (2008).
28. Sawchenko, P. E. Toward a new neurobiology of energy Morrison, C. D. & Berthoud, H.‑R. Hedonics act in 80. Keller, L. & Surette, M. G. Communication in bacteria:
balance, appetite, and obesity: the anatomists weigh unison with the homeostatic system to unconsciously an ecological and evolutionary perspective. Nat. Rev.
in. J. Comp. Neurol. 402, 435–441 (1998). control body weight. Front. Nutr. 3, 6 (2016). Microbiol. 4, 249–258 (2006).
81. Freter, R., Brickner, H., Fekete, J., Vickerman, M. M. & 108. Ley, R. E., Turnbaugh, P. J., Klein, S. & Gordon, J. I. 133. Crenn, P. et al. Net digestive absorption and adaptive
Carey, K. E. Survival and implantation of Escherichia Microbial ecology: human gut microbes associated hyperphagia in adult short bowel patients. Gut 53,
coli in the intestinal tract. Infect. Immun. 39, with obesity. Nature 444, 1022–1023 (2006). 1279–1286 (2004).
686–703 (1983). 109. Walters, W. A., Xu, Z. & Knight, R. Meta-analyses of 134. Messing, B. et al. Intestinal absorption of free oral
82. Flint, H. J., Scott, K. P., Louis, P. & Duncan, S. H. The human gut microbes associated with obesity and IBD. hyperalimentation in the very short bowel syndrome.
role of the gut microbiota in nutrition and health. Nat. FEBS Lett. 588, 4223–4233 (2014). Gastroenterology 100, 1502–1508 (1991).
Rev. Gastroenterol. Hepatol. 9, 577–589 (2012). 110. Vijay-Kumar, M. et al. Metabolic syndrome and 135. Monteiro-Sepulveda, M. et al. Jejunal T cell
83. Allweiss, B., Dostal, J., Carey, K. E., Edwards, T. F. & altered gut microbiota in mice lacking toll-like receptor inflammation in human obesity correlates with
Freter, R. The role of chemotaxis in the ecology of 5. Science 328, 228–231 (2010). decreased enterocyte insulin signaling. Cell Metab.
bacterial pathogens of mucosal surfaces. Nature 266, 111. Blanton, L. V. et al. Gut bacteria that prevent growth 22, 113–124 (2015).
448–450 (1977). impairments transmitted by microbiota from 136. Morita, C. et al. Gut dysbiosis in patients with
84. Stephen, A. M. & Cummings, J. H. The microbial malnourished children. Science http://dx.doi. anorexia nervosa. PLoS ONE 10, e0145274 (2015).
contribution to human faecal mass. J. Med. Microbiol. org/10.1126/science.aad3311 (2016). 137. Crawford, P. A. et al. Regulation of myocardial ketone
13, 45–56 (1980). 112. Liou, A. P. et al. Conserved shifts in the gut microbiota body metabolism by the gut microbiota during
85. Kotzé, S. et al. Spontaneous bacterial cell lysis and due to gastric bypass reduce host weight and nutrient deprivation. Proc. Natl Acad. Sci. USA 106,
biofilm formation in the colon of the Cape Dune mole- adiposity. Sci. Transl Med. 5, 178ra141 (2013). 11276–11281 (2009).
rat and the laboratory rabbit. Appl. Microbiol. 113. Korem, T. et al. Growth dynamics of gut microbiota in 138. Labouré, H., Van Wymelbeke, V., Fantino, M. &
Biotechnol. 90, 1773–1783 (2011). health and disease inferred from single metagenomic Nicolaidis, S. Behavioral, plasma, and calorimetric
86. Roager, H. M. et al. Colonic transit time is related to samples. Science 349, 1101–1106 (2015). changes related to food texture modification in men.
bacterial metabolism and mucosal turnover in the gut. 114. Thaiss, Christoph, A. et al. Transkingdom control of Am. J. Physiol. Regul. Integr. Comp. Physiol. 282,
Nat. Microbiol. 1, 16093 (2016). microbiota diurnal oscillations promotes metabolic R1501–R1511 (2002).
87. Kau, A. L., Ahern, P. P., Griffin, N. W., Goodman, A. L. homeostasis. Cell 159, 514–529 (2014). 139. Anini, Y. et al. Comparison of the postprandial release
& Gordon, J. I. Human nutrition, the gut microbiome 115. Zarrinpar, A., Chaix, A., Yooseph, S. & Panda, S. Diet of peptide YY and proglucagon-derived peptides in the
and the immune system. Nature 474, 327–336 and feeding pattern affect the diurnal dynamics of the rat. Pflügers Archiv. 438, 299–306 (1999).
(2011). gut microbiome. Cell Metab. 20, 1006–1017 (2014). 140. Pavlov, I. P. The Work of the Digestive Glands (Charles
88. Kau, A. L. et al. Functional characterization of IgA- 116. Liang, X., Bushman, F. D. & FitzGerald, G. A. Griffin Co. Ltd, 1902).
targeted bacterial taxa from undernourished Rhythmicity of the intestinal microbiota is regulated 141. Gerspach, A. C., Steinert, R. E., Schönenberger, L.,
Malawian children that produce diet-dependent by gender and the host circadian clock. Proc. Natl Graber-Maier, A. & Beglinger, C. The role of the gut
enteropathy. Sci. Transl Med 7, 276ra224 (2015). Acad. Sci. USA 112, 10479–10484 (2015). sweet taste receptor in regulating GLP‑1, PYY, and
89. Slack, E. et al. Innate and adaptive immunity 117. Leone, V. et al. Effects of diurnal variation of gut CCK release in humans. Am. J. Physiol. Endocrinol.
cooperate flexibly to maintain host-microbiota microbes and high-fat feeding on host circadian clock Metab. 301, E317–E325 (2011).
mutualism. Science 325, 617–620 (2009). function and metabolism. Cell Host Microbe 17, 142. Furness, J. B., Rivera, L. R., Cho, H.‑J., Bravo, D. M. &
90. Wu, G. D. et al. Linking long-term dietary patterns 681–689 (2015). Callaghan, B. The gut as a sensory organ. Nat. Rev.
with gut microbial enterotypes. Science 334, 118. Thaiss, C. A., Zeevi, D., Levy, M., Segal, E. & Elinav, E. Gastroenterol. Hepatol. 10, 729–740 (2013).
105–108 (2011). A day in the life of the meta-organism: diurnal rhythms 143. Cox, H. M. et al. Peptide YY is critical for
91. Faith, J. J. et al. The long-term stability of the human of the intestinal microbiome and its host. acylethanolamine receptor Gpr119‑induced activation
gut microbiota. Science 341, 1237439 (2013). Gut Microbes 6, 137142 (2015). of gastrointestinal mucosal responses. Cell Metab. 11,
92. Vrieze, A. et al. Impact of oral vancomycin on gut 119. Mukherji, A. et al. Shifting eating to the circadian rest 532–542 (2010).
microbiota, bile acid metabolism, and insulin phase misaligns the peripheral clocks with the master 144. Perez-Burgos, A. et al. Psychoactive bacteria
sensitivity. J. Hepatol. 60, 824–831 (2014). SCN clock and leads to a metabolic syndrome. Proc. Lactobacillus rhamnosus (JB‑1) elicits rapid frequency
93. Tyakht, A. V. et al. Human gut microbiota community Natl Acad. Sci. USA 112, E6691–E6698 (2015). facilitation in vagal afferents. Am. J. Physiol.
structures in urban and rural populations in Russia. 120. Fetissov, S. O. & Meguid, M. M. Serotonin delivery Gastrointest. Liver Physiol. 304, G211–G220 (2013).
Nat. Commun. 4, 2469 (2013). into the ventromedial nucleus of the hypothalamus 145. Berthoud, H. R., Kressel, M., Raybould, H. E. &
94. Yatsunenko, T. et al. Human gut microbiome viewed affects differently feeding pattern and body weight in Neuhuber, W. L. Vagal sensors in the rat duodenal
across age and geography. Nature 486, 222–227 obese and lean Zucker rats. Appetite 54, 346–353 mucosa: distribution and structure as revealed by in
(2012). (2010). vivo DiI-tracing. Anat. Embryol. (Berl.) 191, 203–212
95. Palmer, C., Bik, E. M., DiGiulio, D. B., Relman, D. A. & 121. Stunkard, A. J., Grace, W. J. & Wolff, H. G. The night- (1995).
Brown, P. O. Development of the human infant eating syndrome: a pattern of food intake among 146. Ghatei, M. A., Ratcliffe, B., Bloom, S. R. &
intestinal microbiota. PLoS Biol. 5, e177 (2007). certain obese patients. Am. J. Med. 19, 78–86 (1955). Goodlad, R. A. Fermentable dietary fibre, intestinal
96. Bolnick, D. I. et al. Individual diet has sex-dependent 122. Rajpal, D. K. et al. Selective spectrum antibiotic microflora and plasma hormones in the rat. Clin. Sci.
effects on vertebrate gut microbiota. Nat. Commun. 5, modulation of the gut microbiome in obesity and (Lond.) 93, 109–112 (1997).
4500 (2014). diabetes rodent models. PLoS ONE 10, e0145499 147. Camilleri, M., Madsen, K., Spiller, R., Van
97. Pevsner-Fischer, M. et al. Role of the microbiome in (2016). Meerveld, B. G. & Verne, G. N. Intestinal barrier
non-gastrointestinal cancers. World J. Clin. Oncol. 7, 123. Cani, P. D. et al. Changes in gut microbiota control function in health and gastrointestinal disease.
200–213 (2016). metabolic endotoxemia-induced inflammation in high- Neurogastroenterol. Motil. 24, 503–512 (2012).
98. Haque, T. R. & Barritt, A. S. 4th. Intestinal microbiota fat diet-induced obesity and diabetes in mice. 148. Neunlist, M. et al. The digestive neuronal–glial–
in liver disease. Best Pract. Res. Clin. Gastroenterol. Diabetes 57, 1470–1481 (2008). epithelial unit: a new actor in gut health and disease.
30, 133–142 (2016). 124. de Zwaan, M., Marschollek, M. & Allison, K. C. The Nat. Rev. Gastroenterol. Hepatol. 10, 90–100 (2012).
99. Aron-Wisnewsky, J. & Clement, K. The gut night eating syndrome (NES) in bariatric surgery 149. Hamilton, M. K., Boudry, G., Lemay, D. G. &
microbiome, diet, and links to cardiometabolic and patients. Eur. Eat. Disord. Rev. 23, 426–434 (2015). Raybould, H. E. Changes in intestinal barrier function
chronic disorders. Nat. Rev. Nephrol. 12, 169–181 125. Tremaroli, V. et al. Roux‑en‑Y gastric bypass and and gut microbiota in high-fat diet-fed rats are
(2016). vertical banded gastroplasty induce long-term dynamic and region dependent. Am. J. Physiol.
100. Knip, M. & Siljander, H. The role of the intestinal changes on the human gut microbiome contributing to Gastrointest. Liver Physiol. 308, G840–G851 (2015).
microbiota in type 1 diabetes mellitus. Nat. Rev. fat mass regulation. Cell Metab. 22, 228–238 150. Bruce-Keller, A. J. et al. Obese-type gut microbiota
Endocrinol. 12, 154–167 (2016). (2015). induce neurobehavioral changes in the absence of
101. Cotillard, A. et al. Dietary intervention impact on gut 126. Power, M. L. & Schulkin, J. Anticipatory physiological obesity. Biol. Psychiatry 77, 607–615 (2015).
microbial gene richness. Nature 500, 585–588 regulation in feeding biology: cephalic phase 151. Ukena, S. N. et al. Probiotic Escherichia coli Nissle
(2013). responses. Appetite 50, 194–206 (2008). 1917 inhibits leaky gut by enhancing mucosal
102. Kleiman, S. C. et al. The intestinal microbiota in acute 127. Hempfling, W. P. & Mainzer, S. E. Effects of varying the integrity. PLoS ONE 2, e1308 (2007).
anorexia nervosa and during renourishment: carbon source limiting growth on yield and 152. Ewaschuk, J. B. et al. Secreted bioactive factors from
relationship to depression, anxiety, and eating maintenance characteristics of Escherichia coli in Bifidobacterium infantis enhance epithelial cell barrier
disorder psychopathology. Psychosom. Med. 77, continuous culture. J. Bacteriol. 123, 1076–1087 function. Am. J. Physiol. Gastrointest. Liver Physiol.
969–981 (2015). (1975). 295, G1025–G1034 (2008).
103. Woting, A. & Blaut, M. The intestinal microbiota in 128. Bäckhed, F. et al. The gut microbiota as an 153. Mariadason, J. M., Catto-Smith, A. & Gibson, P. R.
metabolic disease. Nutrients 8, 202 (2016). environmental factor that regulates fat storage. Proc. Modulation of distal colonic epithelial barrier function
104. Furet, J. P. et al. Differential adaptation of human gut Natl Acad. Sci. USA 101, 15718–15723 (2004). by dietary fibre in normal rats. Gut 44, 394–399
microbiota to bariatric surgery-induced weight loss: 129. Yun, A. J., Lee, P. Y., Doux, J. D. & Conley, B. R. A (1999).
links with metabolic and low-grade inflammation general theory of evolution based on energy efficiency: 154. Wall, R. et al. in Microbial Endocrinology: The
markers. Diabetes 59, 3049–3057 (2010). its implications for diseases. Med. Hypoth. 66, Microbiota-Gut–Brain Axis in Health and Disease (eds
105. Million, M. et al. Correlation between body mass 664–670 (2006). Lyte, M. & John Cryan, F.) 221–239 (Springer New
index and gut concentrations of Lactobacillus reuteri, 130. Rabot, S. et al. Germ-free C57BL/6J mice are resistant York, 2014).
Bifidobacterium animalis, Methanobrevibacter smithii to high-fat-diet-induced insulin resistance and have 155. Yano, J. M. et al. Indigenous bacteria from the gut
and Escherichia coli. Int. J. Obes. (Lond.) 37, altered cholesterol metabolism. FASEB J. 24, microbiota regulate host serotonin biosynthesis. Cell
1460–1466 (2013). 4948–4959 (2010). 161, 264–276 (2015).
106. Monira, S. et al. Gut microbiota of healthy and 131. Fleissner, C. K. et al. Absence of intestinal microbiota 156. Reigstad, C. S. et al. Gut microbes promote colonic
malnourished children in Bangladesh. Front. does not protect mice from diet-induced obesity. Br. serotonin production through an effect of short-chain
Microbiol. 2, 228 (2011). J. Nutr. 104, 919–929 (2010). fatty acids on enterochromaffin cells. FASEB J. 29,
107. Everard, A. et al. Cross-talk between Akkermansia 132. Woting, A. et al. Alleviation of high fat diet-induced 1395–1403 (2015).
muciniphila and intestinal epithelium controls diet- obesity by oligofructose in gnotobiotic mice is 157. Lee, H. H., Molla, M. N., Cantor, C. R. & Collins, J. J.
induced obesity. Proc. Natl Acad. Sci. USA 110, independent of presence of Bifidobacterium longum. Bacterial charity work leads to population-wide
9066–9071 (2013). Mol. Nutr. Food Res. 59, 2267–2278 (2015). resistance. Nature 467, 82–85 (2010).
158. Chimerel, C. et al. Bacterial metabolite indole modulates 182. Maaser, C. et al. Crucial role of the melanocortin 207. Mankarious, S. et al. The half-lives of IgG subclasses
incretin secretion from intestinal enteroendocrine L cells. receptor MC1R in experimental colitis. Gut 55, and specific antibodies in patients with primary
Cell Rep. 9, 1202–1208 (2014). 1415–1422 (2006). immunodeficiency who are receiving intravenously
159. Hubbard, T. D. et al. Adaptation of the human aryl 183. Nwokolo, C. U., Freshwater, D. A., O’Hare, P. & administered immunoglobulin. J. Lab. Clin. Med.
hydrocarbon receptor to sense microbiota-derived Randeva, H. S. Plasma ghrelin following cure of 112, 634–640 (1988).
indoles. Sci. Rep. 5, 12689 (2015). Helicobacter pylori. Gut 52, 637–640 (2003). 208. Fetissov, S. O. et al. Autoantibodies against
160. Fetissov, S. O. et al. Expression of hypothalamic 184. Lluch, J. et al. The characterization of novel tissue appetite-regulating peptide hormones and
neuropeptides after acute TCDD treatment and microbiota using an optimized 16S metagenomic neuropeptides: putative modulation by gut
distribution of Ah receptor repressor. Regul. Pept. sequencing pipeline. PLoS ONE 10, e0142334 microflora. Nutrition 24, 348–359 (2008).
119, 113–124 (2004). (2015). 209. Fetissov, S. O. et al. Autoantibodies against
161. Bogunovic, M. et al. Enteroendocrine cells express 185. Amar, J. et al. Energy intake is associated with neuropeptides are associated with psychological
functional Toll-like receptors. Am. J. Physiol. Gastrointest. endotoxemia in apparently healthy men. Am. J. Clin. traits in eating disorders. Proc. Natl Acad. Sci. USA
Liver Physiol. 292, G1770–G1783 (2007). Nutr. 87, 1219–1223 (2008). 102, 14865–14870 (2005).
162. Palazzo, M. et al. Activation of enteroendocrine cells 186. Harte, A. L. et al. High fat intake leads to acute 210. Karaiskos, D. et al. Psychopathological and
via TLRs induces hormone, chemokine, and defensin postprandial exposure to circulating endotoxin in personality features in primary Sjogren’s syndrome
secretion. J. Immunol. 178, 42964303 (2007). type 2 diabetic subjects. Diabetes Care 35, 375–382 — associations with autoantibodies to
163. Beutler, B. & Rietschel, E. T. Innate immune sensing (2012). neuropeptides. Rheumatology 49, 1762–1769
and its roots: the story of endotoxin. Nat. Rev. 187. Huang, Q.‑H., Hruby, V. J. & Tatro, J. B. Role of central (2010).
Immunol. 3, 169–176 (2003). melanocortins in endotoxin-induced anorexia. Am. 211. François, M. et al. Ghrelin-reactive immunoglobulins
164. Mukherji, A., Kobiita, A., Ye, T. & Chambon, P. J. Physiol. Regul. Integr. Comp. Physiol. 276, and anxiety, depression and stress-induced cortisol
Homeostasis in intestinal epithelium is orchestrated R864–R871 (1999). response in adolescents. The TRAILS study. Prog.
by the circadian clock and microbiota cues transduced 188. Dwarkasing, J. T., Marks, D. L., Witkamp, R. F. & van Neuropsychopharmacol. Biol. Psychiatry 59, 1–7
by TLRs. Cell 153, 812–827 (2013). Norren, K. Hypothalamic inflammation and food (2015).
165. Zhu, X., He, L. & McCluskey, L. P. Ingestion of bacterial intake regulation during chronic illness. Peptides 77, 212. Garcia, F. D. et al. Anti-neuropeptide Y plasma
lipopolysaccharide inhibits peripheral taste responses 60–66 (2016). immunoglobulins in relation to mood and appetite
to sucrose in mice. Neuroscience 258, 47–61 (2014). 189. Cani, P. D. et al. Endocannabinoids — at the in depressive disorder. Psychoneuroendocrinology
166. Hosoi, T., Okuma, Y., Matsuda, T. & Nomura, Y. Novel crossroads between the gut microbiota and host 37, 1457–1467 (2012).
pathway for LPS-induced afferent vagus nerve metabolism. Nat. Rev. Endocrinol. 12, 133–143 213. Lucas, N. et al. Effects of rabbit anti‑α‑melanocyte-
activation: possible role of nodose ganglion. Autonom. (2016). stimulating hormone (α‑MSH) immunoglobulins on
Neurosci. 120, 104–107 (2005). 190. Di Marzo, V. et al. Leptin-regulated endocannabinoids α‑MSH signaling related to food intake control.
167. Bret-Dibat, J. L., Bluthe, R. M., Kent, S., Kelley, K. W. are involved in maintaining food intake. Nature 410, Neuropeptides 48, 21–27 (2014).
& Dantzer, R. Lipopolysaccharide and interleukin‑1 822–825 (2001). 214. Lucas, N. et al. Anti‑α‑melanocyte-stimulating
depress food-motivated behavior in mice by a vagal- 191. Jo, Y. H., Chen, Y. J., Chua, S. C. Jr, Talmage, D. A. & hormone autoantibodies in patients with eating
mediated mechanism. Brain Behav. Immun. 9, Role, L. W. Integration of endocannabinoid and leptin disorders and melanocortin 4 receptor signaling.
242–246 (1995). signaling in an appetite-related neural circuit. Neuron Eur. Neuropsychopharmacol. 24, S704–S705
168. Topping, D. L. & Clifton, P. M. Short-chain fatty acids 48, 1055–1066 (2005). (2014).
and human colonic function: roles of resistant starch 192. Nicolaidis, S. Metabolic and humoral mechanisms of 215. Takagi, K. et al. Anti-ghrelin immunoglobulins
and nonstarch polysaccharides. Physiol. Rev. 81, feeding and genesis of the ATP/ADP/AMP concept. modulate ghrelin stability and its orexigenic effect
1031–1064 (2001). Physiol. Behav. 104, 8–14 (2011). in obese mice and humans. Nat. Commun. 4, 2685
169. Canfora, E. E., Jocken, J. W. & Blaak, E. E. Short-chain 193. Dzamko, N. L. & Steinberg, G. R. AMPK-dependent (2013).
fatty acids in control of body weight and insulin hormonal regulation of whole-body energy 216. François, M. et al. High-fat diet increases ghrelin-
sensitivity. Nat. Rev. Endocrinol. 11, 577–591 (2015). metabolism. Acta Physiol. 196, 115–127 (2009). expressing cells in stomach, contributing to obesity.
170. Koh, A., De Vadder, F., Kovatcheva-Datchary, P. & 194. Colldén, G., Mangano, C. & Meister, B. P2X2 Nutrition 32, 709–715 (2016).
Bäckhed, F. From dietary fiber to host physiology: purinoreceptor protein in hypothalamic neurons 217. Hamze Sinno, M. et al. Regulation of feeding and
short-chain fatty acids as key bacterial metabolites. associated with the regulation of food intake. anxiety by α‑MSH reactive autoantibodies.
Cell 165, 1332–1345 (2016). Neuroscience 171, 62–78 (2010). Psychoneuroendocrinology 34, 140–149
171. Chambers, E. S., Morrison, D. J. & Frost, G. Control of 195. Zilberter, Y., Zilberter, T. & Bregestovski, P. Neuronal (2009).
appetite and energy intake by SCFA: what are the activity in vitro and the in vivo reality: the role of 218. McFall-Ngai, M. et al. Animals in a bacterial world,
potential underlying mechanisms? Proc. Nutr. Soc. energy homeostasis. Trends Pharmacol. Sci. 31, a new imperative for the life sciences. Proc. Natl
74, 328–336 (2015). 394–401 (2010). Acad. Sci. USA 110, 3229–3236 (2013).
172. Chambers, E. S. et al. Effects of targeted delivery of 196. Mächler, P. et al. In vivo evidence for a lactate 219. Maffei, M. et al. Leptin levels in human and rodent:
propionate to the human colon on appetite regulation, gradient from astrocytes to neurons. Cell Metab. 23, measurement of plasma leptin and ob RNA in obese
body weight maintenance and adiposity in overweight 94–102 (2016). and weight-reduced subjects. Nat. Med. 1,
adults. Gut 64, 1744–1754 (2015). 197. Le Blay, G., Michel, C., Blottière, H. M. & Cherbut, C. 1155–1161 (1995).
173. Samuel, B. S. et al. Effects of the gut microbiota on Prolonged intake of fructo-oligosaccharides induces a 220. Everard, A. et al. Responses of gut microbiota and
host adiposity are modulated by the short-chain fatty- short-term elevation of lactic acid-producing bacteria glucose and lipid metabolism to prebiotics in
acid binding G protein-coupled receptor, Gpr41. Proc. and a persistent increase in cecal butyrate in rats. genetic obese and diet-induced leptin-resistant
Natl Acad. Sci. USA 105, 16767–16772 (2008). J. Nutr. 129, 2231–2235 (1999). mice. Diabetes 60, 2775–2786 (2011).
174. Wanders, A. J. et al. Effects of dietary fibre on 198. Langhans, W. Hepatic and intestinal handling of 221. Alcock, J., Maley, C. C. & Aktipis, C. A. Is eating
subjective appetite, energy intake and body weight: a metabolites during feeding in rats. Physiol. Behav. 49, behavior manipulated by the gastrointestinal
systematic review of randomized controlled trials. 1203–1209 (1991). microbiota? Evolutionary pressures and potential
Obes. Rev. 12, 724–739 (2011). 199. Silberbauer, C. J., Surina-Baumgartner, D. M., mechanisms. BioEssays 36, 940–949
175. Cani, P. D. et al. Gut microbiota fermentation of Arnold, M. & Langhans, W. Prandial lactate infusion (2014).
prebiotics increases satietogenic and incretin gut inhibits spontaneous feeding in rats. Am. J. Physiol. 222. Thompson, J. A., Oliveira, R. A. & Xavier, K. B.
peptide production with consequences for appetite Regul. Integr. Comp. Physiol. 278, R646–R653 Chemical conversations in the gut microbiota.
sensation and glucose response after a meal. (2000). Gut Microbes 7, 163–170 (2016).
Am. J. Clin. Nutr. 90, 1236–1243 (2009). 200. Frost, G. et al. The short-chain fatty acid acetate 223. Jacobi, C. A. et al. Quorum sensing in the probiotic
176. Tennoune, N. et al. Bacterial ClpB heat-shock protein, reduces appetite via a central homeostatic bacterium Escherichia coli Nissle 1917 (Mutaflor)
an antigen-mimetic of the anorexigenic peptide mechanism. Nat. Commun. 5, 3611 (2014). — evidence that furanosyl borate diester (AI‑2) is
α-MSH, at the origin of eating disorders. Transl 201. De Vadder, F. et al. Microbiota-generated metabolites influencing the cytokine expression in the DSS
Psychiatry 4, e458 (2014). promote metabolic benefits via gut–brain neural colitis mouse model. Gut Pathog. 4, 1–10
177. Tennoune, N. et al. Sex-related effects of nutritional circuits. Cell 156, 84–96 (2014). (2012).
supplementation of Escherichia coli: relevance to 202. Breton, J. et al. Elevated plasma concentrations of 224. Wynendaele, E. et al. Quorum sensing peptides
eating disorders. Nutrition 31, 498–507 (2015). bacterial ClpB protein in patients with eating selectively penetrate the blood–brain barrier. PLoS
178. Lee, S. et al. The structure of ClpB: a molecular disorders. Int. J. Eat. Disord. 49, 805–808 ONE 10, e0142071 (2015).
chaperone that rescues proteins from an aggregated (2016). 225. Berg, R. D. The indigenous gastrointestinal
state. Cell 115, 229–240 (2003). 203. Alexander, K. L., Targan, S. R. & Elson, C. O. microflora. Trends Microbiol. 4, 430–435 (1996).
179. Haange, S. B. et al. Metaproteome analysis and Microbiota activation and regulation of innate and 226. Chapelot, D., Aubert, R., Marmonier, C.,
molecular genetics of rat intestinal microbiota reveals adaptive immunity. Immunol. Rev. 260, 206–220 Chabert, M. & Louis-Sylvestre, J. An endocrine and
section and localization resolved species distribution (2014). metabolic definition of the intermeal interval in
and enzymatic functionalities. J. Proteome Res. 11, 204. Garidou, L. et al. The gut microbiota regulates humans: evidence for a role of leptin on the prandial
5406–5417 (2012). intestinal CD4 T cells expressing RORγt and controls pattern through fatty acid disposal. Am. J. Clin.
180. Panaro, B. L. et al. The melanocortin‑4 receptor is metabolic disease. Cell Metab. 22, Nutr. 72, 42–431 (2000).
expressed in enteroendocrine L cells and regulates the 100–112 (2015).
release of peptide YY and glucagon-like peptide 1 in 205. Lindner, C. et al. Diversification of memory B cells Acknowledgements
vivo. Cell Metab. 20, 1018–1029 (2014). drives the continuous adaptation of secretory The author is grateful to all colleagues who participated in
181. Ericson, M. D., Schnell, S. M., Freeman, K. T. & antibodies to gut microbiota. Nat. Immunol. 16, and inspired the reviewed research and to his wife being a
Haskell-Luevano, C. A fragment of the Escherichia coli 880–888 (2015). fervent and pertinent critic.
ClpB heat-shock protein is a micromolar melanocortin 206. Christmann, B. S. et al. Human seroreactivity to gut
1 receptor agonist. Bioorg. Med. Chem. Lett. 25, microbiota antigens. J. Allergy Clin. Immunol. 136, Competing interests
5306–5308 (2015). 1378–1386 (2015). The author is a co‑founder of TargEDys.