Gut Microbiota in Human Metabolic Health and Disease: Yong Fan and Oluf Pedersen

Reviews
Gut microbiota in human metabolic

health and disease
Yong Fan and Oluf Pedersen ✉
Abstract | Observational findings achieved during the past two decades suggest that the
intestinal microbiota may contribute to the metabolic health of the human host and, when
aberrant, to the pathogenesis of various common metabolic disorders including obesity, type 2
diabetes, non-alcoholic liver disease, cardio-metabolic diseases and malnutrition. However, to
gain a mechanistic understanding of how the gut microbiota affects host metabolism, research
is moving from descriptive microbiota census analyses to cause-and-effect studies. Joint analyses
of high-throughput human multi-omics data, including metagenomics and metabolomics data,
together with measures of host physiology and mechanistic experiments in humans, animals
and cells hold potential as initial steps in the identification of potential molecular mechanisms
behind reported associations. In this Review, we discuss the current knowledge on how gut
microbiota and derived microbial compounds may link to metabolism of the healthy host or to
the pathogenesis of common metabolic diseases. We highlight examples of microbiota-targeted
interventions aiming to optimize metabolic health, and we provide perspectives for future basic
and translational investigations within the nascent and promising research field.
From the outcome of recent epidemiological, physiolog- is to have an overall metabolism (as estimated from an
ical and omics-based studies, complemented by cellu- array of various measures of body organ functions, such
lar studies and experiments in animals, it appears that as those of the liver, and intestinal, fat, muscle, heart
a considerable part of the environmental influence on and brain tissues) that in representative epidemiologi-
human health and disease risk may be mediated or mod- cal studies is linked with both desirable life quality and
ified by microbial communities1. These microorganisms, longevity. A considerable part of the world population,
collectively referred to as the microbiota, include a vast however, primarily due to overeating of processed
number of interacting bacteria, archaea, bacteriophages, energy-dense food, urbanization with sedentary occu-
eukaryotic virus and fungi coexisting on human surfaces pations and leisure-time inactivity often combined with
and in all body cavities1. The majority of them are com- smoking addiction, has a suboptimal or even a poor
mensal or mutualistic microorganisms1. The collection metabolic health. The implications are a high prevalence
of all intestinal microbial genes (that is, the microbiome) of metabolic disorders, including obesity, non-alcoholic
in an individual represents a genetic repertoire that is liver disease, type 2 diabetes mellitus (T2D) as well
more than one order of magnitude higher in genes than as cardio-metabolic disease (CMD), and premature
the human genome1. The majority of microorganisms death2–6. In some developing countries, the metabolic
that inhabit humans reside within the intestines and health landscape is quite the opposite — a major part of
are influenced by mode of birth, infant feeding, life- the population is chronically malnourished7.
style, medication and the genetics of the host. The gut Despite the immense differences in pathologies of
microbiome has important roles in the training of host common chronic metabolic disorders, they are asso-
immunity, digesting food, regulating gut endocrine func- ciated with shared and disease-specific abnormali-
Novo Nordisk Foundation tion and neurological signalling, modifying drug action ties in the composition and function of the intestinal
Center for Basic Metabolic and metabolism, eliminating toxins and producing microbiota8–11. Since the start of the microbiota research
Research, Faculty of Health numerous compounds that influence the host1. era, whether disease-associated aberrant microbiota are
and Medical Sciences,
University of Copenhagen,
Here, we review the literature on the intestinal micro- involved in disease causation (that is, predisposition, ini-
Copenhagen, Denmark. biota of the human host with a focus on bacteria and tiation or progression) or are secondary phenomena has
✉e-mail: [email protected] archaea, and their potential importance for human been widely debated. However, one of the first pieces of
https://doi.org/10.1038/ metabolism. In the literature, there is no consensus on evidence for a mechanistic involvement of the gut micro-
s41579-020-0433-9 how to define metabolic health. Yet a tentative definition biota in the regulation of metabolism was provided in
naTure RevIeWS | MIcrobIology volume 19 | January 2021 | 55

Reviews
Box 1 | Tools for dissecting gut microbiota–metabolic health associations

Currently, there is major interest in exploring the potential role of the bacterial isolate. In addition, associations with health or disease status
intestinal microbiome and linked faecal, blood and urine metabolomes can be characterized for individual microbial genes, taxa or modules
as determinants for metabolic health and various metabolic diseases. of microbial functional potentials14,17,45. Classifiers such as supervised
Recent advances in shotgun sequencing technologies and a reduction in machine learning or random forest can be used to assign each sample to a
sequencing costs together with advances in bioinformatics have made it certain category. To validate associations identified by microbiome-wide
possible to capture a more comprehensive view of the entire community association studies, additional metagenomics data sets such as samples
of gut microorganisms and their functional potentials. Similarly, various from other independent studies are included.
sensitive and precise technologies for profiling the metabolome of body
Integrated analyses of microbiome and linked metabolomes
fluids are available. Here, we summarize some of the recent strategies
Gut bacteria can produce various bioactive metabolites, which can enter
that have been applied to uncover relationships between the gut
the bloodstream of the host through absorption into the enterohepatic
microbiome and various metabolomes.
circulation. Specific metabolites associated with a disease phenotype can
Microbiome-wide association studies be identified by mass spectrometry or nuclear magnetic resonance-based
The microbiome-wide association study approach largely mimics metabolomics of faeces, plasma, urine or other biofluids, making it
genome-wide association studies identifying genetic variants in case– possible to conduct joint analyses of the microbiome, metabolome and
control studies that are associated with a phenotype, often a disease host phenotypes to identify potential mechanistic links15. The analysis
state. In the case of microbiome-wide association studies, microbial DNA framework includes at least four steps: first, as described above, clustering
is purified from intestinal or faecal samples and subjected to deep of microbial genes, the functional potential of which can be constructed
shotgun-based sequencing. Individually, the identified and assembled by organizing the genes into Kyoto Encyclopedia of Genes and Genomes
genes are then integrated to construct a cross-sampled, non-redundant functional modules based on sequence similarity to proteins with
microbial gene catalogue. The abundance of each gene is determined by known functional characteristics15. The second step involves clustering of
counting the reads of matching sequences in the individual sample. metabolites showing co-abundance to reduce dimension of metabolomics
Applying various bioinformatic algorithms and correlation coefficients, features, given that metabolites are often shaped by common pathways
assembled genes in the microbiome data are clustered into specific that could be regulated by precursors14. By using weighted gene
groups, such as metagenomic linkage groups8, metagenomic species62, co-expression network analysis, which was originated from gene expression
co-abundance gene groups168 or metagenomic species pan-genomes169. analysis, co-abundant metabolites can be clustered174. Of note, one of the
Sequence reads from individual samples that map to the metagenomic advantages of this algorithm is the use of a bin cluster for unannotated
linkage groups, metagenomic species, co-abundance gene groups, metabolites instead of clustering all metabolites. The next step is to apply
metagenomic species pan-genomes and their contigs, respectively, are non-parametric statistical tests to correlated clustered genes and
then extracted and assembled into high-quality draft genomes, each metabolites. Finally, based on the hypothesis that microbial functional
of which is defined as a species or a group of highly related species. correlations could be driven by one or few species or by a group of species,
Recently, it has also become possible to define strain-level signatures it is recommended to perform leave-one-out cross-validation to evaluate
in gut microbiomes. These tools include StrainPhlAn that is based on the contribution of each gene cluster by removing it from the associations14.
applying single-nucleotide variant analysis to the core genome of Correlation analysis between these specific metabolites and prevalent
sequenced microbial isolates170, MetaMLST that is based on species- gene clusters may lead to the identification of specific species that
specific hypervariable loci171 and MetaPanPhlAn that is based on pan- harbour the genes needed to produce precursors of the disease-relevant
genome profiling172. However, these methods rely on the coverage and metabolites that are produced either by the gut microbiota or by both the
quality of reference genomes173. Results from strain-level profiling still host and the gut microbiota14,45. Eventually, identified microbial candidates
need to be validated by obtaining a pure culture containing the target are evaluated in rodent models (Box 2).
2004 when it was shown in mice that the gut microbiota future epidemiological and experimental studies (Box 2
regulates host capacity for harvesting energy from the and Box 3).
diet as well as energy storage12. Subsequently, and spear-
headed by application of next-generation microbiome Healthy gut microbiota
sequencing, generation of comprehensive microbial gene A fundamental assumption for claiming disruption of
Supervised machine learning catalogues, development of targeted bioinformatics and gut microbiota in states of metabolic disorders is knowl-
A form of applied artificial availability of high data storage and operational capac- edge on the composition and function of the gut micro-
intelligence where the algorithm
ity, the research field has delivered an enormous amount biota of metabolically healthy people. However, a healthy
learns by experience to classify
new data according to prior of new knowledge about the potential role of the gut human gut microbiota has not been defined at any pro-
labels. microbiota in metabolism13–15 (Box 1). found taxonomic resolution. The relative distribution of
In this Review, we discuss taxonomic and functional gut bacteria and archaea is unique to an individual, partly
Random forest characteristics of the intestinal microbiota in the meta- owing to strain-level diversities and differences in micro-
A machine learning approach
where a multitude of algorithms
bolically healthy human host and the implication of a bial growth rates16 and in structural variants within the
(decision trees) are used reduction in gut microbial diversity for host metabo- microbial genes17,18, and partly owing to influence from
to optimize, for instance, lism. This is followed by highlighting aspects of abnor- the considerable inter-individual variation in environ-
classification or regression mal gut microbiota composition and function, and the mental exposures and host genetics19. Overall, however,
of data sets.
relationship to dysmetabolism in obesity, T2D, CMD, a high taxa diversity, high microbial gene richness and
Kyoto Encyclopedia of metabolic liver disease and malnutrition. We evaluate stable microbiome functional cores characterize healthy
Genes and Genomes a series of recently discovered gut microbiota-derived gut microbial communities18. Yet it is worth noting that
A publicly available database compounds and their potential role as messengers high gut bacterial diversity and richness alone are not
in bioinformatics and system between the gut microbiota and host metabolism in unbiased indicators of a healthy microbiota, because the
medicine-driven analyses with
information on omics-generated
physiological and diseased states. We highlight exam- intestinal transit time affects microbial richness20. A pro-
data, biological pathways, ples of microbiota-based therapies to improve meta- longed transit time may result in an increased richness
chemicals and drugs. bolic health and, finally, we outline potential avenues for but not necessarily in a healthy gut microbiota.
56 | January 2021 | volume 19 www.nature.com/nrmicro

Reviews
Box 2 | A framework to study causality of gut microbial components

To progress from microbiome–disease
associations to functions of the microbiome a b
in a given disease, additional experiments
are suggested to explore potential causality
Control Germ- Control Antibiotic- Conven-
between microbial components and the disease Control Disease free treated tionalized
of interest. Biological samples from affected
and control individuals are analysed using
Serum Serum
shotgun sequencing and untargeted (that is, Faeces Faeces
Intensity
discovery-based and global but semi-quantitative) + Urine Urine
metabolomics profiling to identify microbial
features or molecules differently abundant in m/z Q1 Q2 Q3
various biofluids (see the figure, part a). In an Metagenomic Untargeted
profiling metabolomic
untargeted run, the analyte is characterized by profiling
Ion
its retention time and its mass to charge ratio source
(m/z value). The identification of a novel Precursor ion Fragmentation Product ion
selection selection
compound is confirmed by generation of a
Metabolome
–1 0 1
nuclear magnetic resonance spectrum or Spearman Targeted metabolomics
correlation
matching fragmental behaviour and retention coefficient
time to synthetic reference standards175.
Both microbiome and metabolome data sets are Microbiome Control Antibiotic- Conven-
subjected to integrative analysis to capture Integrative analysis treated tionalized
candidate biosynthetic gene clusters (BGCs),
bacteria or communities that are highly related Standard
Intensity
Intensity Control
to metabolites. To identify whether the Analyte Germ-
compound of interest is produced in a free
microbiota-dependent manner, triple
Time Time
quadrupole mass spectrometry can be used
Candidate BGCs or Metabolite
to perform targeted (that is, validation of bacteria or community identification Microbiota-dependent fashion
well-defined metabolites often in a quantitative
manner) profiling of samples from both c d
germ-free and control mice (see the figure,
part b). In addition, the conventionalization BGCs or bacteria or community of interest Gene knockout Control
of antibiotic-treated mice to rescue gut Colonization of
microbiota can also specifically demonstrate germ-free mice Potential targeted
the role of gut microbiota in generating gene deficient
specific compounds. When employing
reaction-selecting modes in targeted capturing Gene knockout
of the candidate analytes, both the first Serum
Intensity
Faeces Control
quadrupole (Q1) and the last quadrupole (Q3) Urine
are set at a specific mass, allowing only distinct
monitoring of the ion fragment from a certain BGC or bacteria or Time
Intensity
precursor, which results in an increased community

detecting sensitivity. Microbial-based
When the selected reaction monitoring is Germ-free drug design
applied to multiple product ions from one or
Time
more precursor ions, the pattern is called
Comparative targeted metabolomics
multiple reaction monitoring176.
Human intestinal microbiome data and +
microbial-derived compounds can be 15 Prevotella copri
Serum glucose (mol–1)
Sham control 50
associated to select candidate BGCs, microbial Disease Microbial-
species or a microbial community whose 40
10 ** targeted therapy
HOMA-IR
** 30
presence correlates with the production of *
specific compounds (see the figure, part c). 5 20
Dimensionality reduction is achieved by 10
clustering genes that encode the biosynthetic 0 0
0 50 100 0 20 40 60
enzymes for secondary metabolite pathways, Minutes post P. copri abundance
binning co-abundant genes and assembling insulin injection (a.u.)
co-occurring microbes that are present in a Phenotypic profiling Healthy metabolism
defined ecological habitat in space and
time14,177.
Phenotypic profiling of germ-free mice harbouring a reference BGC, microbial species or microbial community can be profiled in a comparative
targeted manner to examine the abundance changes of the compound of interest177.
To identify downstream signalling pathways that microbial-derived compounds may be involved in, a genetically altered mouse can be generated to see
the change of microorganism-generated compound because of the inactivated downstream effects (see the figure, part d). The evidence collected above
may stimulate to develop microbiota-directed interventions (Box 3).
HOMA-IR, homeostatic model assessment of insulin resistance. Part c of figure adapted from ref.14, Springer Nature Limited.

Reviews
The mode of birth and access to breastfeeding remains relatively stable in late childhood, adolescence
shape the infant gut microbiota, and it matures grad- and adulthood until a decline in diversity occurs at
ually during childhood in response to environmental advanced age, likely owing, in part, to a change in
exposures21,22. Thereafter, the intestinal microbiota immune functions22. The mature healthy intestinal
Box 3 | Microbiota-changing interventions

Owing to the exponentially growing Microbiome-directed interventions
knowledge gained from
epidemiological and experimental
studies about the impact of the
intestinal microbiome on metabolic Untargeted Targeted
health, there is an interest in testing
both targeted and untargeted
microbiome-directed interventions A
E B
in humans with disrupted gut D C
microbiomes (see the figure).

Untargeted interventions Exercises Individualized Faecal microbiota Bio-engineered Drugs targeting selected
Diet or exercise interventions nutrition transplantation commensals microbial metabolism
In rodents, feeding various
combinations of nutrients changes
the gut microbiome and metabolism.
In particular, high-saturated fat feeding
causes endotoxaemia and insulin
Prebiotics Probiotics Synbiotics Postbiotics Phage therapy CRISPR–Cas9-based
resistance178 (Fig. 1). A systematic therapy
review has summarized the results
of six interventions and nine cross-
sectional studies investigating the General improvement in Specific modification in
effects of dietary fat on intestinal microbial composition and functions metabolism-related gut microbiota
microbiota in humans179, and
demonstrated that diets high in saturated or monounsaturated fat
negatively influenced the microbiota whereas diets high in Heterologous and autologous faecal microbiota transplantation
polyunsaturated fat appeared to be neutral with respect to the microbiota. It is currently unsettled whether heterologous faecal microbiota
Similarly, high-polysaccharide diet interventions have resulted in altered transplantation will be an option in preventing or treating metabolic
gut microbiota linked with increased faecal, serum or urine concentrations dysfunctions with aetiologies far more complex than chronic infections
of short-chain fatty acids, weight loss and improvements of cytokine and caused by Clostridioides difficile196. Challenges include lack of knowledge
metabolome profiles 180–183
. Likewise, interventions with increased physical about the importance of dieting for stool graft colonization and survival,
activity have shown adaptive and, in some cases, transmissible changes of optimal anaerobic handling of donor faeces, immunological compatibility
the intestinal microbiome linked with an increased capacity for breakdown between donor and recipient, and the role of bacteriophages and fungi for
of lactate, branched-chain amino acids, an increased potential for a successful faecal microbiota transplantation. In parallel, there is a need
synthesis of short-chain fatty acids and improvements of cardiorespiratory for examining the feasibility and efficacy of the much less complicated
fitness and insulin sensitivity. autologous faecal microbiota transplantation stored prior to development
of a metabolic disease, that is, faecal samples biobanked in childhood or
Trials testing probiotics, prebiotics or postbiotics adolescence.
Bifidobacterium, Lactobacillus and Saccharomyces spp. have a long history
as safe probiotics . In addition, potential next-generation probiotics,
184 Targeted interventions
Bio-engineered commensals and drugs
which are not yet marketed, include Faecalibacterium prausnitzii185,
In mice, gavaging engineered Escherichia coli overexpressing the satiety
Akkermansia muciniphila186 and several Clostridia spp.187. Systematic
factor N-acylphosphatidylethanolamine alleviated high-fat diet-induced
reviews of short-term randomized controlled trials in healthy individuals
obesity, insulin resistance and hepatosteatosis197; and in rats with diabetes,
show no consistent effect of probiotics on gut microbial composition . 188
a genetically modified Lactobacillus gasseri strain with the ability to express
However, it is possible that some probiotics, for example A. muciniphila
and secrete glucagon-like peptide 1 (GLP-1) increased insulin release and
strains, do not even need to colonize the intestine to provide beneficial
reduced hyperglycaemia198. Whether delivering genetically modified
metabolic health effects189,190. Some non-digestible polysaccharides called
organisms carrying microbial genes to the human gut is acceptable for
prebiotics change the relative abundance of fermenting microbiota in the
health and drug authorities or consumers is at present unsettled. Targeting
colon of rodents and induce improved gut barrier functions and in some
specific microbial-synthesized metabolites by delivering tailored drugs is
cases improved metabolism . Of note, randomized controlled trials
191
another emerging potential frontier to optimize metabolism. A prime
testing various prebiotics have demonstrated that inulin-type fructans
example is the specific drug blockade of the microbial production of
changed the gut microbiota composition in adult women with obesity,
trimethylamine78.
leading to modest changes in host metabolism192. In children who are
overweight or have obesity, it is reported that oligofructose-enriched Bacteriophage therapy and CRISPR–Cas9 editing
inulin prebiotics alter the intestinal microbiota and modestly reduce body As most bacteriophages and archaeal viruses are specific to bacterial
weight, adiposity and inflammatory markers193. The same holds for the and archaeal strains, respectively, they may in future experiments
combination of various probiotic strains and prebiotics often named be applied to target dysbiotic parts of the microbiome in individuals
synbiotics. Postbiotics represent the pasteurized version of probiotics or with metabolic disorders199. Similarly, it is at present unknown whether
parts of microbial strains possessing health-promoting effects194. A pilot CRISPR–Cas can be applied to edit and reset dysbiotic parts of a
trial of pasteurized A. muciniphila and its membrane protein Amuc_1100 human gut microbiome without causing undesired effects for the
demonstrated positive effects on markers of human metabolism186,195. host200.

Reviews
Dysmetabolism
microbial core functions include genes encoding glycos- at restoring the richness and diversity of gut microbial
Metabolic dysfunctions often aminoglycan degradation, the production of short-chain communities in chronic disorders, it is not possible to
including abdominal obesity, fatty acids (SCFAs) via fermentation of complex poly- evaluate whether a receding intestinal microbiota is
dyslipidaemia and higher saccharides and synthesis of specific lipopolysaccharides part of disease causation or a secondary adaptation in
blood glucose and higher
blood pressure than normal.
(LPS), and the biosynthesis of some essential amino states of chronic non-communicable disorders. Yet there
acids and vitamins13,18,23,24 (Fig. 1). is some evidence that dietary interventions may improve
Endotoxaemia In summary, a ‘gold standard’ reference of a human low microbial gene richness and host metabolism in
The presence of endotoxin(s) gut microbiota with the capacity to promote host meta individuals with obesity38.
within the blood, for example,
bolic health does not exist. This is owing to a tremen-
bacterial lipopolysaccharides.
dous variation, especially at taxonomy levels, between Gut microbiota of metabolic pathologies
Dyslipidaemia individuals of different demography, ethnicity, sex, age Gut microbiota is aberrant in obesity. The incidence of
An abnormal amount and an and health status. The implications of this variation are obesity and its metabolic co-morbidities in the devel-
abnormal relative distribution that epidemiological studies must include controls for oped countries has increased dramatically since the mid-
of various lipids in the blood.
contextual factors or adjust for the contextual differences. twentieth century39. Sedentary lifestyles and increased
food consumption in combination with a widespread
Receding gut microbial diversity polygenetic susceptibility are considered to be the major
Compared with people living traditional lifestyles (for causes for the obesity epidemic40, which is potentially
instance, people in East Africa or in the Amazonas) further aggravated by the widespread use of antibiotics29.
who have not been exposed to antimicrobial factors of In addition, evidence is accumulating for a role of the
modern living, most populations in technically devel- gut microbiota in mediating some of the environmental
oped countries of the world show reduced gut micro- effects in obesity pathogenesis. Following the discovery
bial diversity25. In parallel with urbanization, sewerage, in 2006 that a transferrable obesity-associated micro
higher standards of housing and improved hygiene biota can induce weight gain in lean mice41, subse-
in general, the abundance of Bacteroides, Prevotella, quent epidemiological studies have shown differences
Desulfovibrio, Lactobacillus and Oxalobacter genera in the gut microbiota of individuals with obesity and
in the gut microbiota is declining25–27. This decline in lean individuals. At the species level, twin studies have
diversity relates to a rise in the prevalence of common shown that the abundance of SCFA producers such as
chronic metabolic disorders28. Similarly, in both lean Eubacterium ventriosum and Roseburia intestinalis is
individuals and individuals with obesity, low microbial associated with obesity42, whereas butyrate producers
gene richness links with a relative increase in adipos- such as Oscillospira spp. 43 and the methanogenic
ity, insulin resistance, inflammation and dyslipidaemia28. archaeon Methanobrevibacter smithii may associate
The precise and specific causes of the receding microbial with leanness44. Another metagenome-wide association
richness and diversity in developed countries are elusive. study of lean individuals and individuals with obesity
However, besides the general improvement in standards showed that the abundance of Bacteroides thetaiotaomi
of living and hygiene, the use of antibiotics to combat cron, a glutamate-fermenting commensal, was markedly
infectious diseases is suspected to be a contributor29. decreased in individuals with obesity and was inversely
For instance, the use of antibiotics before pregnancy, correlated with serum glutamate concentration 45.
during pregnancy or in early childhood may change Furthermore, gavage of mice with B. thetaiotaomicron
the composition of the gut microbiota of infants and protected against adiposity, pointing to possible future
children, practices that have occurred in parallel with modalities for obesity intervention targeting the gut
increases in the incidence of early-onset obesity30,31. microbiota with potential probiotic or microbial com-
In adults, short-term treatment of young healthy indi- pounds (Box 3). Analyses of gut microbial pathways and
viduals with broad-spectrum antibiotics causes the gene families suggest that obesity is associated with a
long-term depletion of some commensal and mutual- decreased capacity for unidirectional conjugation,
istic bacteria32. However, causal relationships between which transfers genetic material between bacteria, and a
antibiotic use, disruption of the gut microbiota and reduction in superoxide reductase, potentially leading to
dysmetabolism have not been shown33–35. The effects of intestinal oxidative stress46 (Fig. 2). Observational studies
antibiotics on the gut microbiota and metabolism may have been complemented by a faecal microbiota trans-
be dependent on host genetics because studies in mice plantation study demonstrating that transplantation of
fed a high-fat diet demonstrate that the effect of anti- stools from twins discordant for obesity into germ-free
biotics on numerous metabolic variables is dependent mice transfers, in a diet-dependent manner, the pheno-
on the genetic background of the animals36. Below, we type of the human donor to the recipient animal47. This
discuss aspects of abnormal gut microbiota and the landmark study provides the rationale for the ongoing
relationships to metabolism in obesity, T2D, metabolic intense search for gut microbial messengers linking the
liver disease and CMD, where the diversity of the gut gut microbiota with whole-body energy metabolism. In
microbiota of the affected is often reported to be reduced studies of the potential impact of altered gut microbiota
compared with healthy individuals, and even more in the development of chronic disorders including obe-
so compared with people living traditional lifestyles28,37. sity, however, it is important to bear in mind that various
In addition, we discuss how severe malnutrition dur- drugs may influence the gut microbiota. For instance,
ing infancy and early childhood relates to a relative intake of statins, which are cholesterol-lowering drugs
depletion of certain bacteria in the gut microbiota. In widely prescribed for the prevention of arterioscle-
the absence of clinically controlled interventions aimed rosis, has been linked with a health-promoting gut

Reviews
a Weekly defecation b
Monday
Tuesday
Diet low in animal fat and protein Diet high in animal fat and protein
but high in plant fibres but low in plant fibres
Wednesday
Host enzyme Acid Host
Proteolysis
digestion hydrolysis proteases
Indigestible Peptides
polysaccharides Thursday
Fermentation
Sulfur Other Aromatic
Indigestible but amino acids amino acids amino acids
fermentable Friday
polysaccharides
• Methionine Tyrosine Phenylalanine Tryptophan
• Cystine
• Cysteine
Saturday • Taurine
Microbiota related to metabolic health Sunday Aberrant microbiota related to metabolic diseases
Sulfate • Deamination Miscellaneous

reduction • Decarboxylation reactions
SCFAs
↑ Mucus Hydrogen • Ammonia Phenylacetate ↑ Luminal pH
secretion sulfide • Amines
• H2 • p-Cresol • Indole
↓ Mucus • CO2 • Phenyl- propionate
Propionate Acetate Butyrate ↑ Methane secretion • CH4 propionate • Indole
• Organic acetate
acids • Indole
↓ Luminal pH
↑ Acetaldehyde
GPCR-41/43 GPCR-41/43
PPARγ
β-Oxidation
L cells
Gut hormones ↑ BCFAs ↑ TMAO PAMPs (LPS)
PYY or GLP-1
↓ Blood ↑ Energy ↑ Glucose-stimulated ↑ Blood ↑ Metabolic ↓ Glucose-stimulated

glucose expenditure insulin secretion glucose endotoxaemia insulin secretion
microbiota48. Obviously, observations of any effects of have shown that hyperglycaemia may increase intesti-
microbiota-modifying drugs ideally need to be followed nal barrier permeability through a GLUT2-dependent
up in interventional studies to address any causality of transcriptional reprogramming of intestinal epithelial
the relationships. cells and alteration of tight junction integrity, subse-
quently causing a leaky mucosa50. Therefore, there is
Type 2 diabetes and prediabetes link with altered gut major interest in understanding whether an aberrant
microbiota. Indirect evidence that the intestinal micro- gut microbiota is involved in triggering or sustaining
biota together with the hormonal secretion capacities elevated blood glucose in T2D and its precursor states.
of the distal gut may be involved in glucose regulation T2D represents ~90% of all diabetes cases51 and, like
comes from large-scale epidemiological studies show- obesity, is growing in incidence and prevalence, affect-
ing that, compared with individuals without colectomy, ing between 5 and 15% of the adult population, making
patients with total colectomy have an increased risk it the most common endocrine disorder52. The aetiology
of T2D49. In addition, mechanistic studies in rodents of T2D involves a combination of hundreds of genetic

Reviews
◀ Fig. 1 | Impact of diet on gut microbiota and host metabolism. The balance between with prediabetes, the gut microbiota exhibits a loss
metabolically healthy microbiota and dysbiosis is crucial. Homeostasis is maintained by of butyrate-producing taxa, a decrease in abundance of
various factors, such as host genetics, diet, daily number of defecations, physical activity, Akkermansia muciniphila and an increase in abundance
smoking and drug usage. a | A metabolically healthy microbiota (mainly achieved by a of bacteria with pro-inflammatory potentials11,58 (Fig. 2).
high-fibre, low animal fat and low animal protein diet) is illustrated. The indigestible but
Gestational diabetes mellitus (GDM), which may occur
fermentable polysaccharides are metabolized by the microbiota of the large intestine
and are fermented to produce an array of compounds, and to stimulate a thick intestinal
in the third trimester of pregnancy and is estimated to
mucus layer and strong barrier functions. Microbial production of short-chain fatty acids affect 7–10% of all pregnancies, is another prediabetic
(SCFAs) provides an additional energy source for colonocytes and causes a decrease in state59. In GDM, diagnostic criteria for plasma glucose
luminal pH. The SCFAs acetate, butyrate and propionate can bind to G protein-coupled are set at lower thresholds than for other forms of dia-
receptor (GPCR)-41 and GPCR-43, which are expressed on enteroendocrine L cells, and betes owing to the toxic effects on fetus development
subsequently induce secretion of glucagon-like peptide 1 (GLP-1) and peptide YY (PYY) of even slightly elevated glycaemia59. Interestingly, com-
that contribute to increased energy expenditure162, reduced food intake163 and improved pared with pregnant women with normoglycaemia in
glucose metabolism and insulin secretion164. Butyrate is an activator of the peroxisome the same trimester, women with GDM have a disrupted
proliferator-activated receptor-γ (PPARγ) and a stimulator for β-oxidation and oxygen gut microbiota composition, and differences in the gut
consumption in the gut, which maintains an anaerobic environment in the gut lumen114.
microbiota are still detectable between the two groups
b | Microbial dysbiosis induced by a high animal fat and protein diet, sedentary life,
smoking, alcohol intake and relatively infrequent defecation may result in a leaky
of women 8 months after pregnancy. Both during and
mucosa, intestinal and systemic inflammation and reduced production of SCFAs, leading after pregnancy, the gut microbiota of GDM resem-
to less gut hormones being secreted by L cells. In the fermentation process, complex bles the aberrant microbiota composition reported in
proteins are first cleaved by various bacterial peptidases, proteases and endopeptidases non-pregnant individuals with T2D60, where a reduced
to release free amino acids and short peptides that then undergo fermentation165. abundance of butyrate-producing bacteria has been con-
Of note, a dysbiotic microbiota is often associated with a prolonged colonic transit time, sistently found in addition to an often reduced bacterial
resulting in a shift in colonic metabolism leading to increased microbial proteolysis, even richness and an increase in opportunistic pathogens8,61.
though the preferred substrate for bacterial fermentation is fermentable dietary fibres At the functional potential level, the gut microbiome of
and bacteria will not switch to protein metabolism until fermentable polysaccharides are individuals with T2D is enriched in multiple pathways.
depleted166.As a result of increased protein fermentation, branched-chain fatty acids
These include an enrichment of pathways involved in
(BCFAs; 2-methyl butyrate, isobutyrate and isovalerate), trimethylamine, organic acids,
gases (H2S (malodorous), H2 and CO2) and trace amounts of phenols, amines, indoles and
sugar-related membrane transport to increase cellular
ammonia are produced, causing an increase in luminal pH167. Altogether, such changes in glucose uptake, branched-chain amino acid (BCAA)
the microbial environment and metabolites cause a leakage of pathogen-associated outward transport to induce insulin resistance, methane
molecular patterns (PAMPs) including lipopolysaccharides (LPS) that become increased metabolism related to an anaerobic intestinal environ-
in the blood and trigger systemic low-grade inflammation and insulin resistance147. ment, xenobiotic degradation, metabolism to promote
It should be noted, however, that some indole derivatives such as 3-indolepropionic drug resistance and sulfate reduction to reduce insulin
acid produced by fermentation of dietary fibres have been shown to improve glucose sensitivity8 (Fig. 2). However, interpretation of gut micro-
metabolism137. TMAO, trimethylamine N-oxide. biota profiles from individuals with T2D who are being
treated with multiple drugs is problematic owing to
variants and a series of environmental factors, which are drug confounders. Among the drugs that are frequently
shared with obesity53. Most individuals are either over- prescribed for T2D, the anti-hyperglycaemic drug met-
weight or obese, and the pathophysiology in the early formin appears to change the gut microbiota most,
stages of these cases is characterized by insulin resist- with effects on the relative abundance of multiple gen-
ance of primarily skeletal muscles and liver and adipose era and species, such as Escherichia and Intestinibacter,
tissues, and a compensatory increased biosynthesis and and enhancement of several microbiome functional
secretion of insulin. In the presence of continued insu- potentials, such as propionate and butyrate production
lin resistance, insulin biosynthesis declines, which is fol- that induces intestinal gluconeogenesis62–64. In fact, the
lowed by an aggravation of hyperglycaemia54,55. Besides gut microbiota has been suggested to mediate some of
hyperglycaemia, the T2D phenotype often exhibits the anti-hyperglycaemic effects of metformin through
dyslipidaemia with elevated circulating levels of triglyc- enhanced production of SCFAs and unconjugated bile
erides and low-density lipoprotein cholesterol, hyperten- acids, which activate intestinal gluconeogenesis, and
sion and an increased tendency for platelet aggregation. an increase in intestinal gluconeogenesis resulting in
All of these abnormalities are risk factors for premature a lowering of glycaemia65,66. However, any effects that
arteriosclerosis55,56. Consequently, individuals with metformin-induced changes of gut microbiota have on
T2D are, in addition to health behaviour modifications, overall glucose regulation may be modest. Consistent
treated with multiple medications targeting both ele- with this notion, it has been reported that in germ-free
vated plasma glucose and risk factors for cardiovascular or antibiotic-treated mice, the effect of metformin on
co-morbidities. However, the multidrug regimens affect measures of dysmetabolism remains largely unaffected67.
the gut microbiota of these individuals in many differ- Metformin treatment is also known to cause transient or
ent ways57. Therefore, recent epidemiological studies persistent intestinal discomfort in around one-third of
attempting to elucidate links between the gut microbiota individuals, which in part may be attributed to increased
and T2D have focused on the drug-naive early stages gas production by some Escherichia spp. and an
of T2D, called prediabetes. Individuals with prediabetes enrichment of various virulence factors63,64,68.
have elevated plasma glucose values that are within the Altogether, these studies investigating the role of
non-diabetes range but raised to a level where they are at the gut microbiota in T2D highlight several important
an increased risk of developing overt T2D and ischaemic findings and shortcomings. The microbiome in indi-
cardiovascular disease11. In these drug-naive individuals viduals with overt T2D and prediabetes appears to be

Reviews
Obesity Taxa features Function features

• ↓ Bacteroidetes to Firmicutes ratio • ↑ LGC: degradation of β-glucuronide
• ↓ Akkermansia muciniphila and aromatic amino acids
• ↓ Bacteroides thetaiotaomicron • ↑ HGC: production of organic acids
• ↓ Clostridium histolyticum and H2
• ↓ Clostridium coccoides • ↑ Phosphotransferase system
• ↑ Dorea longicatena • ↓ Glycosaminoglycan degradation
• ↑ Eubacterium ventriosum • ↑ Glutamine or glutamate transport
• ↓ Faecalibacterium prausnitzii system
• ↓ Methanobrevibacter smithii • ↑ Phenylalanine, tyrosine and
tryptophan biosynthesis
• ↑ Roseburia intestinalis
• ↓ Valine, leucine and isoleucine
• ↑ Ruminococcus gnavus degradation
• ↑ Ruminococcus torques • ↓ Superoxide reductase
Type 2 diabetes
Malnutrition
Taxa features
Taxa features • ↓ Akkermansia muciniphila
• ↓ Anaerobes to facultative • ↑ Bacteroides vulgatus
anaerobes ratio • ↑ Clostridium clostridioforme
• ↓ Bifidobacterium longum • ↑ Clostridium hathewayi
• ↓ Bifidobacterium pseudolongum • ↓ Clostridiales sp. SS3/4
Function features • ↑ Prevotella copri
• ↓ Lantibiotics production Function features
• ↓ Energy harvest • ↑ Membrane transport of sugars
• ↓ Immune protection • ↑ BCAA transport
• ↓ Vitamin biosynthesis Healthy microbiota Dysbiosis • ↓ SCFA production
• ↑ Pathogenic factors • ↓ Metabolism of cofactors
and vitamins
Metabolic liver Cardio-metabolic

disease disease
Taxa features
Taxa features • ↓ Enterobacter aerogenes
• ↓ Coprococcus comes • ↓ Escherichia coli
• ↓ Faecalibacterium prausnitzii • ↑ Faecalibacterium prausnitzii
• ↑ Klebsiella pneumoniae • ↑ Klebsiella spp.
• ↑ Streptococcus anginosus • ↑ Ruminococcus gnavus
• ↑ Veillonella atypica • ↑ Streptococcus spp.
Function features Function features
• ↑ Oxidative damage • ↓ Activities of SCFAs
• ↑ γ-Aminobutyric acid biosynthesis • ↑ TMA lysases
• ↓ Haem biosynthesis • ↑ Phosphotransferase system
• ↑ Dentrification • ↓ Vitamin metabolism
• ↑ Ethanol production • ↑ Lipopolysaccharide biosynthesis
• ↑ Lipopolysaccharide and • ↑ Tryptophan production
peptidoglycan biosynthesis • ↓ SCFA metabolism
• ↑ BCAAs and AAA biosynthesis • ↓ Butyrate production
Fig. 2 | Some of the reported intestinal microbial taxonomic and functional features linked to common metabolic
disorders. An overview of selected key gut microbiome features related to metabolic diseases. Multiple studies have
implicated bacterial species and functional features in metabolic diseases; however, results differ between studies. The
association results from metagenomic studies have been summarized. Selected microbial taxonomic features are listed
in addition to their directions in metabolic diseases. The list is not comprehensive of all altered taxonomic or functional
features but represents common patterns observed across studies. AAA, aromatic amino acid; BCAA, branched-chain
amino acid; HGC, high gene count; LGC, low gene count; SCFA, short-chain fatty acid; TMA, trimethylamine; ↓, lower
abundance in metabolic diseases when compared with control; ↑, higher levels in metabolic diseases when compared
with control.
relatively depleted in bacterial butyrate producers and to Gut microbial dysbiosis linked to cardio-metabolic dis-
exhibit an increase in species with a pro-inflammatory eases. Arteriosclerosis is one of the leading causes of
functional potential. However, none of these changes morbidity and mortality in the western world69. People
is specific to T2D but they occur in many chronic who suffer from arteriosclerosis frequently have prior
non-communicable disorders characterized by a clin- clinically silent metabolic dysfunctions for many years,
ically silent low-grade inflammation. So far, attempts including elevated circulating concentrations of glu-
to transfer gut microbiota from drug-naive individuals cose, insulin70 and lipids71 in addition to insulin resist-
with T2D or prediabetes to germ-free mice to reproduce ance and low-grade inflammation72. Therefore, it is a
the T2D phenotype have failed11. major challenge to disentangle the possible impact of an

Reviews
aberrant gut microbiota on early-stage metabolic per- increased plasma TMAO concentrations but unex-
turbations from an imbalanced gut microbiota that, at pectedly reduced aortic lesion sizes without changes in
a stage distal to the dysmetabolic stage, is escalating the plasma lipids, thereby indicating a potential beneficial
pathogenesis of ischaemic heart disease. On top of this, role for l-carnitine in the prevention of atherosclerosis
Atherogenesis individuals with CMD are, as in T2D, heavily medicated, independently of an increase in plasma TMAO, in this
The dynamic process of
forming atheromas (also called
making it challenging to tease apart authentic arterio- specific mouse model81. Moreover, there are numerous
plaques), that is, accumulated sclerosis signals in the gut microbiota from signatures foods that are rich in TMAO itself, such as some cold
inflammatory cells, lipids, induced by complex medication and pre-morbidities water-dwelling fish, the intake of which is generally con-
cell debris, minerals and and co-morbidities. So far, these challenges have not sidered beneficial for cardiovascular health82. Lately, the
connective tissue in and on
been systematically addressed. A metagenome-wide contradictory results may have been reconciled to some
the walls of an artery forming
a swelling that narrows the association study of faecal samples from individuals extent83. Recent animal and human studies suggest that
arterial lumen and restricts with CMD reported that the gut microbiome of these the influence of TMAO on vascular biology is dependent
the flow of blood. individuals, without adjustments for pre-morbidities on dietary–microbiome interactions and host genetics,
or co-morbidities, exhibits enriched abundances of which may explain some of the apparent controversies
Thrombosis
The formation of a blood clot,
Enterobacteriaceae, including Escherichia coli, Klebsiella of TMAO’s involvement in atherothrombosis. In rodent
known as a thrombus, within spp. and Enterobacter aerogenes, as well as an increased experiments, it was found that TMAO can be directly
a blood vessel. The blood clot abundance of oral cavity species when compared with produced in the gut in addition to oxidation of TMA in
that consists of platelets, red healthy control subjects73 (Fig. 2). By contrast, individuals the liver, and blood TMAO levels increased after feed-
blood cells and fibrin proteins
with CMD had a reduced abundance of Bacteroides spp. ing the animals with a high-fibre diet or a high-choline
obstructs the flow of blood
through the blood vessel. and anti-inflammatory Faecalibacterium prausnitzii74. diet. In two different genetically modified mouse mod-
At the functional level, a decrease in some fermen- els that are predisposed to arteriosclerosis, investigators
Stroke tation capacities was seen but an enrichment in the could not find evidence of any relationships between
An acute brain insult where phosphotransferase system, amino acid transporters TMAO generation and accelerated atherogenesis 83.
compromised blood flow in
atherosclerotic arteries or
and enzymes involved in the synthesis of LPS and tri- However, in the atherosclerosis-prone animals, circu-
bleeding from brain arteries methylamine (TMA)73 was also seen. Recently, it was lating TMAO concentrations linked directly with fea-
causes damage to brain tissues shown that ischaemic heart failure is associated with a tures of plaque instability, including biomarkers of platelet
often resulting in various dysbiotic gut microbiota with an increased abundance of aggregation, intra-plaque bleeding and inflammation.
paresis.
Ruminococcus, Acinetobacter and Veillonella spp. and a In a population-based observational study of adults, the
Myocardial infarction decreased abundance of Alistipes, Faecalibacterium and same authors could not find evidence of correlations
(Also known as an acute heart Oscillibacter spp.75. At the functional level, the micro- between plasma TMAO levels and markers of arterio-
attack or acute coronary biome of these patients was enriched in genes involved sclerosis such as intima thickness of carotid arteries or
syndrome). An event that in the biosynthesis of LPS and trimethylamine N-oxide calcium scores of coronary arteries83,84. Thus, it appears
occurs when blood flow is
acutely compromised or
(TMAO)75. Taken together, these studies of CMD point that a major role of the partially gut microbiota-derived
is completely stopped to a to a more inflammatory and a less fermentative micro- TMAO, in relation to ischaemic vascular disorders,
part of the heart muscle biome, even though it remains to be determined which may be an aggravation of atherogenesis in individu-
(myocardium), causing severe components of the gut microbiota are involved in early- als with already established ischaemic heart disease or
damage to the heart.
stage dysmetabolism and which are related to a more stroke through mechanisms primarily involving plaque
Atherosclerosis distal atherogenesis stage. instability.
The build-up of cholesterol, Of special interest is the discovery of the TMAO
other lipids, inflammatory cells pathway with the involvement of the gut microbiota Dysbiosis of the intestinal microbiota in metabolic liver
and calcium in artery walls, and relations to ischaemic vascular diseases76. TMA is diseases. Non-alcoholic fatty liver disease (NAFLD),
which can restrict blood flow.
synthesized by gut microbiota from dietary phosphati- frequently regarded as the hepatic manifestation of
Atherothrombosis dylcholine, lecithin and l-carnitine, which are found the metabolic syndrome, occurs in many countries
The formation of a blood clot not only in meat from four-legged animals but also in with a prevalence of 20–40% of the adult population3.
within an artery that is affected poultry, seafood and eggs. TMA enters the portal circu- NAFLD comprises a wide spectrum of diseases ranging
by arteriosclerosis.
lation and is oxidized to TMAO in the liver. Experiments from simple steatosis to non-alcoholic steatohepatitis
Plaque instability in rodents demonstrate that feeding dietary TMAO or (NASH), which is the inflammatory, aggressive form of
Vulnerable arterial wall plague its dietary precursors causes an acceleration of arterio- NAFLD85. There is growing recognition that a disrupted
that intermittently ruptures sclerosis, induces platelet aggregation and enhances the gut microbiota may be one of several crucial factors in
giving rise to circulating plaque thrombosis potential77. Moreover, inhibition of intestinal the pathophysiology of NAFLD and NASH. Individuals
fragments called emboli, which
may cause myocardial
bacterial TMA production with 3,3-dimethyl-1-butanol with NAFLD have an increased abundance of species
infarction or stroke. (a TMA-lyase inhibitor) attenuates arteriosclerosis and assigned to Clostridium, Anaerobacter, Streptococcus,
thrombosis78. Epidemiological studies showing posi- Escherichia and Lactobacillus, whereas Oscillibacter,
Intima tive relationships between the plasma concentrations Flavonifaractor, Odoribacter and Alistipes spp.
The innermost coating
of TMAO and an increased prevalence of stroke and are less abundant86. In comparison, the abundance of
of the vessel wall including
the endothelial surface myocardial infarction as well as an increase in prema- Proteobacteria, Enterobacteriaceae and Escherichia
at the lumen. ture mortality of people with stable coronary artery spp. is elevated in patients with NASH87. Concordantly,
ischaemia lend support to the outcome of the mech- children with steatosis or NASH are depleted in
Steatosis anistic experiments that TMAO is a biomarker of Oscillospira spp. accompanied by higher abundance
An abnormal retention of lipids
within an organ. The term is
arteriosclerosis79,80. of Dorea and Ruminococcus spp. when compared with
most often used about a fatty Still, there are inconsistencies in available results. For controls88. The microbiota alterations are associated
liver. instance, mice fed a diet enriched in l-carnitine have with higher faecal concentrations of 2-butanone and

Reviews
Portal endotoxaemia
4-methyl-2-pentanone that cause hepatocellular toxic- early gut microbiota in infants, partly through secretion
Endotoxins, primarily bacterial ity in individuals with metabolic liver diseases compared of broad-spectrum lantibiotics101. Following this, SAM
lipopolysaccharides, which are with healthy controls88. dysbiosis is mainly accompanied by an aberrant inver-
absorbed from the intestines Additionally, given that the microbiota associated sion of the ratio of anaerobes to facultative anaerobes.
into mensenteric and liver
veins (portal drainage).
with NALFD is enriched in ethanol-producing bac- This inversion might be corrected under appropriate
teria such as E. coli, it has been hypothesized that the dietary treatment, eventually leading to microbial com-
Liver cirrhosis aberrant gut microbiome of individuals with NAFLD munities enriched in anaerobes102. However, in cases
A chronic liver disease caused, produce more ethanol than microbiomes of healthy where malnutrition continues, the loss of the healthy
for instance, by alcohol abuse
individuals, as evidenced by increased concentrations mature anaerobic gut microbiota gradually leads to a
or virus infection with
impairment of multiple liver
of intrinsically generated ethanol in the circulation and deficiency in energy harvest41, immune responses103 and
functions owing to replacement breath87,89 (Fig. 2). Ethanol activates nuclear factor-κB vitamin synthesis104, and links with chronic malabsorp-
of normal liver tissues by scar (NF-κB) signalling pathways and causes tissue damage tion, diarrhoea and systemic invasion from pathogenic
tissue. by impairing gut barrier function and contributing to bacteria105. Current dietary therapeutic regimens for
portal endotoxaemia90,91. In the liver of individuals with SAM are not based on knowledge about the develop-
Hepatic encephalopathy
A spectrum of cognitive and NALFD, the detoxification pathway is weakened, pro- mental biology of the intestinal microbiota. However, a
neuro-psychiatric viding a constant source of reactive oxygen species that recent study designed diets using pig and mouse mod-
abnormalities such as potentially cause oxidative damage to the hepatocytes, els to nudge the microbiota into a mature post-weaning
personality changes, anxiety,
which in turn may induce hepatic inflammation and state, which could entrain maturation of the microbiota
confusion, fatigue, shaky hands
or seizures caused by severely
steatohepatitis92. of children with SAM and restore their metabolic and
impaired liver function. Similarly, individuals with liver cirrhosis have a sub- growth profiles to a healthier trajectory106. Such dietary
stantially altered microbiota, including an enrichment regimens urgently need to be tested in long-term ran-
Holobiont of Proteobacteria and Fusobacteria phyla9. The func- domized clinical trials to evaluate the effects of matu-
The unique and discrete
collective of a macro-organism
tional potentials of the gut microbiome in individuals rating the gut microbiota on weight gain and overall
— a host — and the complex with liver cirrhosis and liver failure are massively altered physiology of affected children.
microbial communities for compared with healthy individuals. The changes include
which the macro-organism is significant enrichment of both assimilation and dissim- Bacterial messengers affect metabolism
the habitat.
ilation of nitrate to or from ammonia as signatures of An overwhelming number of reports demonstrate that
Anorexigenic hormones
hepatic encephalopathy. An increase in denitrification disruptions in gut microbiome taxonomy and func-
These appetite-decreasing capacity may relate to the higher nitrogen production, tional potential relate to numerous pathological pheno-
hormones include whereas increased γ-aminobutyric acid (GABA) bio- types. The majority of studies in humans and animals
glucagon-like peptide 1 synthesis and GABA shunt functions relate to hydro- are observational and lack experimental mechanistic
(GLP-1) and peptide YY (PYY),
both produced by specialized
dynamic venous shunting in liver failure. An increase data. However, as revealed by the genetic repertoire of
intestinal cells and leptin of microbial haem biosynthesis may contribute to liver the human gut microbiome, the trillions of commen-
produced by adipocytes and damage because of protoporphyrin toxicity, whereas an sal or mutualistic bacteria and archaea are an immense
intestinal cells. increase in phosphotransferase systems may indicate chemical factory that can synthesize a multitude of
an increased demand for protein following liver tissue compounds needed for their own existence and survival
Leptin
A hormone predominantly
damage9,93. with their host, but also compounds that affect the entire
synthesized in adipose cells Apart from the above-mentioned observations, holobiont. In the following sections, we selectively review
and enterocytes in the small indicating links between gut microbial composition some reported intestinal microbial products that affect
intestine that helps to regulate and function and metabolic liver disorders, human and host energy homeostasis, body adiposity, glucose toler-
energy balance by inhibiting
hunger, which in turn
animal faecal transplantation studies show that a ance, insulin sensitivity, inflammation and endocrine
diminishes fat storage in high abundance of the alcohol-producing bacterial regulation (Fig. 3).
adipocytes. species Klebsiella pneumoniae in the gut accelerates
NAFLD pathogenesis94,95. Additionally, an integrative Microbiota affecting energy homeostasis and body adi-
multi-omics approach in women with obsesity without posity. For their own energy supply, the gut microbiota
diabetes identified phenylacetate as a microbial metabo- ferment energy-yielding nutrients, especially complex
lite contributing to the accumulation of lipids in the liver carbohydrates and, to lesser extent, proteins, mono
and, hence, to NASH96. saccharides, SCFAs and amino acids. However, the
microbiota produces large amounts of fermented nutri-
Disrupted gut microbiota in malnutrition. Globally, mal- ents that benefit their host, corresponding to 5–10%
nutrition affects ~160 million people and is the leading of the daily energy needs of a human107. Here, we first
cause of death in children under the age of 5 years97,98. focus on the SCFAs butyrate, propionate and acetate,
Breastfeeding, food and water security are major protec- which affect energy metabolism and body adiposity in
tive factors against malnutrition and are crucial factors various ways in addition to being simple substrates for
in the maturation of the healthy gut microbiota, char- overall host energy turnover and nutrients for colonic
acterized by a transient bifidobacterial bloom before cell metabolism. Both propionate and butyrate are pre-
a rise in anaerobes during late infancy99. Early loss of dominantly anti-obesogenic through their stimulation
Bifidobacterium longum and Bifidobacterium pseudo of anorexigenic hormones and leptin synthesis108,109. Mice
longum, two key members in mother milk, represents treated with a butyrate precursor drug (tributyrin) are
some of the first disruptions in the intestinal dysbiosis protected from diet-induced obesity, insulin resistance
of severe acute malnutrition (SAM)100 (Fig. 2). B. longum and hepatic steatosis110. Butyrate also suppresses inflam-
and B. pseudolongum appear to be crucial in shaping the mation in various tissues, enhances the differentiation of

Reviews
colonic anti-inflammatory regulatory T cells and induces and propionate113. Additionally, increased butyrate pro-
the NOD, LRR and pyrin domain-containing protein 3 duction activates the peroxisome proliferator-activated
(NLRP3) inflammasome in a G protein-coupled recep- receptor-γ (PPARγ), leading to β-oxidation and oxygen
tor manner111,112. Another G protein-coupled receptor, the consumption in the gut, a phenomenon that contributes
olfactory receptor 78, which may modulate blood pres- to the maintenance of anaerobic conditions in the gut
sure regulation, can also be directly activated by acetate lumen114 (Fig. 1).
Amino acids
Gut lumen
Dietary fibres Primary bile acids
Phosphatidylcholine
and L-carnitine
Specific
Specific bacteria
bacteria
Firmicutes/ Specific Gram-negative Increased
Bacteroidetes bacteria bacteria Specific goblet cell
Secondary bacteria O density
R1
bile acids R N
LPS BCAAs H ClpB
N-Acyl amides Amuc_1100
MECO-1
SCFAs Secondary bile acids LPS • Indole Entero-
Imidazole • Indole endocrine
Acetate Succinate TMA propionate derivatives cell MC1R TLR2
GPCRs TGR5 GPR-119 GLP-1/

GLP-1 AhR PYY
Improved gut
Ghrelin Neuro- barrier function
GLP-1 or PYY transmitters
UCP1 Cytokines
Gut peptide
FXR
TMAO
p38/mTORC1
FMO Vagus nerve • Body weight gain
• Fat mass gain
Satiety • Energy harvest Atherosclerosis • Fatty acid metabolism • Insulin resistance

• Fat storage • Inflammation • Glucose intolerance Inflammation
• Adipose
inflammation
• Satiety
• Neuronal effects
Fig. 3 | Microbial messengers regulate host metabolism. An overview of bacterial-derived branched-chain amino acids (BCAAs) correlate, in cases
some of the intestinal microbial compounds affecting host energy of high fat intake, with insulin resistance in both humans and rodents. Other
homeostasis, body adiposity, inflammation, glucose regulation, insulin bacterial metabolites such as indole and its derivatives bind to the aryl
sensitivity and hormone secretion. Fermentation of the dietary fibre by gut hydrocarbon receptor (AhR). Indolepropionic acid is linked with improved
Firmicutes and Bacteroidetes produces the short-chain fatty acids (SCFAs) insulin secretion and insulin sensitivity and a decreased risk of type 2
butyrate, propionate and acetate, which influence host metabolism in diabetes. Gut bacteria produce N-acyl amide, an endocannabinoid mimic
multiple ways by acting on G protein-coupled receptors (GPCRs) expressed that regulates host glucose metabolism by binding to the G protein-coupled
by enteroendocrine cells. Acetate and butyrate stimulates glucagon-like receptor 119 (GPR119). Proteins secreted by gut bacteria also modulate
peptide 1 (GLP-1) and peptide YY (PYY) release with effects on the pancreas paracrine or endocrine action. ClpB, a protein secreted by Escherichia coli,
(GLP-1-induced insulin biosynthesis) and on the brain (PYY-induced satiety), is involved in regulation of appetite. Melanocortin-like peptide of E. coli
and acetate may enhance fat storage by inducing secretion of ghrelin. (MECO-1), which is a structural and functional analogue to α-melanocyte-
Microbial-derived succinate drives expression of uncoupling protein 1 stimulating hormone and adrenocorticotropin, acts via the mammalian
(UCP1), thereby increasing thermogenesis in adipose tissue. However, melanocortin-1 receptor (MC1R), inhibiting cytokine release in response to
succinate has pro-inflammatory properties on lipopolysaccharide-activated pro-inflammatory stimuli. In the large intestine, pro-inflammatory changes
macrophages and may thus contribute to adipose tissue inflammation and involve increased production of cytokines such as IL-1β and IL-18, and
insulin resistance. Lipopolysaccharides (LPS) are pro-inflammatory reduced IL-10 and IL-22. Amuc_1100, expressed on the outer membrane of
compounds derived from Gram-negative bacterial membranes that Akkermansia muciniphila, improves gut barrier function with increased
promote inflammation. Primary bile acids are converted by gut microbiota goblet cell density through Toll-like receptor 2 (TLR2) and partially
to secondary bile acids that act through the TGR5 receptor to promote recapitulates the beneficial effect of the live A. muciniphila bacterium on
GLP-1 release enhancing thermogenesis in adipose tissue. Gut microbiota energy metabolism and insulin sensitivity. Host metabolism is also
produces trimethylamine (TMA) by metabolizing dietary phosphatidyl influenced by gut bacterially synthesized neurotransmitters (that is,
choline and l-c arnitine. TMA is N-o xidized in the liver by flavin- catecholamine, histamine, γ-aminobutyric acid and serotonin) or gaseous
containing monooxygenase (FMO) to trimethylamine N-oxide (TMAO), neurotransmitters (NO and H2S). Arrow represents stimulation, arrow with
which contributes to atherosclerosis. Imidazole propionate, a bacterial bar indicates inhibition. FXR, farnesoid X receptor; mTORC1, mechanistic
metabolite derived from histidine, contributes to insulin resistance. Gut target of rapamycin complex 1.

Reviews
In two independent studies, structural variations in One of the primary bile acids, chenodeoxycholic acid, is
the Anaerostipes hadrus genome that encode metabolic a highly efficacious FXR ligand127, whereas the second-
modules for inositol catabolism and for metaboliz- ary bile acids lithocholic acid (LCA) and taurolithocholic
ing 3-hydroxybutanoyl-CoA to butyrate show inverse acid, which in rodents are transformed from primary
relationships with body weight, waist circumference bile acids by intestinal bacteria123, are the two most
and body mass index, potentially pointing to mecha- potent endogenous TGR5 ligands128. TGR5 signalling
nistic links between these metabolic pathways and host in enteroendocrine L cells induces secretion of GLP-1,
metabolic health17. thereby improving liver function and enhancing glucose
A mouse study demonstrated that acetate, butyrate tolerance in obese mice126. LCA, a high-affinity TGR5
and propionate may have anabolic effects on host agonist, stimulates thermogenesis through browning of
metabolism115, whereas another study demonstrated both white and brown adipose tissues129, whereas deoxy
that long-term administration of an inulin-propionate cholic acid (DCA) and LCA have been suggested as
ester significantly reduced weight gain, intra-abdominal partial antagonists of FXR130. Recently, a subgroup of sec-
adipose tissue distribution, intra-hepatocellular lipid ondary bile acids, C-6-hydroxylated secondary bile acids,
content and insulin resistance in overweight adults116. have attracted attention as these bile acids may be asso-
Nevertheless, other investigators have reported conflict- ciated with leanness in rodents131 and early weight loss
ing findings for effects of propionate on host metabo- in people with obesity following sleeve gastrectomy132.
lism. A host genetics-driven increase in gut microbial Studies of the intestinal microbiome in several
production of butyrate improved the insulin response chronic metabolic disorders show increased functional
after an oral glucose tolerance test, whereas propionate potentials for metabolism of amino acids133. Whether
causally increased the risk of T2D117. In line with the these observations reflect primary disease-linked mech-
latter finding, another study in mice suggested that pro- anisms or secondary phenomena is unclear. However, in
pionate impairs insulin signalling through stimulation the case of microbial break down of tryptophan, both
of the production of glucagon and fatty acid-binding in vitro and in vivo studies suggest that indole, a trypto-
protein 4 (FABP4)118. Intake of a propionate-containing phan microbial catabolite (Fig. 1), improves intestinal epi-
meal in humans resulted in a postprandial increase in thelial barrier functions134,135. Indole is also a signalling
plasma concentrations of glucagon, FABP4 and norep- molecule involved in enhancing the release of the gut
inephrine, leading to insulin resistance and compensa- hormone GLP-1 from enteroendocrine L cells136, thereby
tory hyperinsulinemia118. Finally, findings in a human pilot indirectly influencing insulin secretion and appetite
study suggested that glucose metabolism is more respon- regulation. Similarly, the circulating concentration of
sive to oral butyrate supplementation in lean individu- 3-indolepropionic acid (IPA), a catabolite involved in
als than in people with the metabolic syndrome, which the tryptophan metabolic pathway, is in epidemiolog-
argues against oral butyrate supplementation as a poten- ical studies reported to associate with improved insulin
tial therapeutic remedy to improve glucose regulation secretion and insulin sensitivity, and a lower likelihood
in individuals with dysmetabolism119. These conflicting of developing T2D137.
experimental results call for human studies in which the
physiological effects of each SCFA are tested separately Microbiota affecting insulin sensitivity and endocrine
as well as in combinations, relative to their proportional regulation. Elevated circulating concentrations of the
concentrations in circulation, in order to properly and BCAAs leucine, isoleucine and valine are a strong bio-
systematically investigate their net effects on glucose marker for insulin resistance and an increased risk of
homeostasis, insulin secretion and insulin sensitivity. T2D138. In individuals with normoglycaemia who are
Unlike butyrate and propionate, acetate may pre- insulin resistant, the gut microbiome has an increased
dominantly have obesogenic properties because it acts potential to synthesize BCAAs, which is largely driven
as a substrate for hepatic and adipocyte lipogenesis120. by an increased abundance of Prevotella copri and
In mice, increased acetate production induced by an Bacteroides vulgatus, but a reduced potential for micro-
altered gut microbiota promotes hyperphagia through bial BCAA uptake and BCAA catabolism primar-
increased ghrelin secretion and enhances fat storage by ily driven by Butyrivibrio crossotus and Eubacterium
increasing glucose-stimulated insulin secretion121. In siraeum14. These findings, supported by experiments in
humans, plasma concentrations of the intestinal hor- mice, suggest that the gut microbiome may contribute
Hyperinsulinemia mones, peptide YY (PYY) and glucagon-like peptide to increased plasma BCAA concentrations and insulin
A state where the
concentration of insulin in
1 (GLP-1) are reported to increase following rectal or resistance under conditions of unhealthy dieting138. In
blood is higher than what is intravenous administration of acetate122. heart failure, the catabolism of BCAAs is impaired and
considered normal. Host energy homeostasis and body adiposity are the accumulated BCAAs induce oxidative stress and
also influenced by bile acids, and the gut microbiota disrupt mitochondrial function139.
Hyperphagia
converts primary bile acids to various secondary bile Imidazole propionate is yet another microbial com-
An abnormally great desire for
food. acids123. Both categories of bile acid exert their meta- pound that contextually impairs insulin signalling140. In
bolic functions by binding to cellular receptors in various portal and peripheral blood of individuals with T2D,
Ghrelin organs of the body, including nuclear farnesoid X recep- imidazole propionate was present at higher concentra-
A circulating hormone that is tor (FXR) and TGR5, a G protein-coupled receptor124. tions than in healthy individuals. Gut bacterial species
produced mainly by stomach
cells and that stimulates
Signalling via these receptors is implicated in glucose such as Streptococcus mutans and Eggerthella lenta were
appetite and promotes fat metabolism, yet in opposite directions; FXR impairs125 verified as producers of imidazole propionate. In mice
storage. whereas TGR5 promotes126 carbohydrate metabolism. experiments, imidazole propionate worsened glucose

Reviews
tolerance and insulin signalling through activation of the is structurally similar to a human ligand for GPR119,
mechanistic target of rapamycin complex 1 (mTORC1) the mice had an improved glucose tolerance when
pathway. In line with these findings, overexpression of the compared with control mice that were gavaged with
mTORC1-mediated insulin signalling pathway was found E. coli harbouring an empty vector154. Finally, MECO-1,
in liver tissue isolated from individuals with T2D140. which is a bacterial melanocortin-like peptide that has
Both the host and the intestinal microbiome produce structural similarities to α-MSH and adrenocorticotro-
succinate, which may have multiple roles as a metabolic pin (ACTH)155, inhibits the release of cytokines when
intermediate. One role relates to insulin sensitivity and exposed to pro-inflammatory stimuli in vitro and res-
inflammation, with effects that are also apparently con- cues mice from death caused by endotoxins. Whether
textual. In wild-type mice, microbial-derived succinate the microbial peptide also has endocrine functions like
not only prevented obesity but also improved insulin its human homologous hormones is unknown.
sensitivity and glucose tolerance141,142. By contrast, suc- In summary, an overwhelming amount of data gen-
cinate might act as a metabolite in innate immune sig- erated both in humans and in animals suggest important
nalling. Thus, in LPS-activated macrophages, succinate involvements of the primary saccharolytically derived
stabilized the transcription factor hypoxia-inducible fac- microbial fermentation products SCFAs, succinate and
tor 1α, which increased IL-1β production and triggered ethanol in host metabolism. We know less about posi-
inflammation. Therefore, succinate may under some tive or undesired health implications of proteolytic or
conditions contribute to insulin resistance that is trig- lipolytic fermentation by the gut microbiota (Fig. 1). An
gered by inflammation143. Unexpectedly, pharmacolog- exception is the microbial fermentation of the dietary
ically raised circulating succinate concentrations in mice amino acids tryptophan and tyrosine that are metabo-
have been shown to be a driver for uncoupling protein lized to indoles and phenols, respectively156. In the liver,
1 (UCP1)-dependent thermogenesis in brown adipose the compounds are converted to indoxyl sulfate and
tissue, thereby boosting protection against diet-induced p-cresyl sulfate, the blood concentrations of which are
obesity and improving glucose tolerance142. Nevertheless, elevated in CMD. In chronic kidney disease, elevated cir-
outcomes of studies of the impact of the gut micro culating levels of the two metabolites are associated with
biome on regulation of UCP1 in various mouse models progression of chronic kidney disease, risk of CMD and
remain inconsistent144,145. Whether microbial-derived all-cause mortality157. In addition, experiments in mice
succinate is relevant to comparable processes in human suggest that microbiota-produced gluconeogenic amino
metabolism is unknown. acid metabolites, including indoxyl sulfate, contribute to
Bacterial LPS are the major outer surface membrane maintaining glucose homeostasis158.
components present in most Gram-negative bacteria146.
LPS are released at bacterial cell death and become Conclusions and perspectives
strong stimulators of host immunity. When in excess, The past two decades of research within the microbiome
for instance owing to habitual intake of diets high in field has made it evident that human biology means the
animal fat or in an extreme state such as sepsis-elicited biology of the holobiont Homo sapiens. In other words,
endotoxaemia, LPS trigger local and systemic inflamma- human biology is contextual on the coexisting microor-
tion and insulin resistance147. Generally, several common ganisms, with the majority living in the digestive tract
chronic disorders that appear to be characterized by a from where they produce or modify various chemicals,
relative disruption of the gut barrier, which elicits mul- or trigger host reactions that affect various physiolog-
tiple imbalances in both intestinal and systemic immune ical functions including immunity, neurobiology and,
competences, converge to cause insulin resistance, not least, metabolism. As appears from our discussion of
eventually leading to metabolic syndrome148,149. the pertinent literature on metabolism and the intestinal
Above, we have discussed how, for instance, SCFAs microbiota, the research field is still at a juvenile stage
and indole interfere with endocrine regulation, but both in its basic and translational dimensions. With
there are more examples of how gut bacterial com- the focus on the influence of the gut microbiome on the
pounds interact with hormone secretion or act as lig- overall functional metabolic read-outs, much still needs
ands for known hormone receptors. Several proteins to be learned. The knowledge gaps to be filled include
secreted by E. coli, including ClpB, which mimics the annotation of hundreds of as yet unknown chemical
host peptide α-m elanocyte-s timulating hormone compounds in the metabolomes and peptidomes of
(α-MSH), affect food intake150 and meal patterns in various body fluids as well as approaches to decipher
mice151. Mechanistically, the ClpB protein stimulates compounds of solely host, microbial and dietary origin,
release of intestinal GLP-1 and PYY and activates brain or of combined origin159. Most studies are snapshots of
anorexigenic pathways inducing satiety. In humans, microbiome landscapes and more information is needed
higher plasma ClpB levels have been reported in indi- on the short-term and long-term dynamics of the intes-
viduals with eating disorders such as anorexia ner- tinal microbiome. Importantly, very little of the novel
vosa compared with circulating ClpB levels in healthy knowledge is validated or has maturated to stages where
individuals152. Furthermore, some gut commensal bac- it can be translated to guide personal everyday health
teria harbour N-acyl amide synthase genes, and their behaviour (public health) or clinical practice. To achieve
encoded lipids mimic binding of classical eukaryotic such goals, several conceptual and analytical challenges
signalling molecules to G protein-coupled receptors153. need attention in future studies where data on the micro-
In experiments where mice were fed an engineered bac- biome are integrated with other omics read-out and
terium that produces the lipid N-acyl serinol, which bioclinical variables (Box 2).

Reviews
The fact that the human intestinal microbiota is rodents (Box 2). However, many experimental studies
unique at deep resolution levels, with substantial struc- fall short owing to lack of appropriate test animals with
tural variations in species and subspecies that influ- a biology and behaviour close to those of humans and,
ence microbiome functionality, is a major conceptual especially when unravelling the role of the gut micro-
and statistical challenge when comparing individual biome in disease causation, there is often a lack of ani-
microbiota irrespective of the health status of study mal models with genetic susceptibilities comparable
subjects17. Moreover, there are several more technical to those of the human disorders under examination.
and database-related shortcomings, including incom- Nevertheless, experimental studies are not exclusively
plete microbial genome databases and a lack of func- done in cellular systems or animal models. In fact, clin-
tional annotation of the majority of microbial genes. ically controlled trials are by their nature mechanis-
Similarly, knowledge about the regulation of transcrip- tic experiments. Because of the massive and complex
tional and translational activities of the gut microbiome inter-individual variation in the gut microbiome, the
is sparse, and when it comes to the interpretation of the ideal human trial design is to perform various con-
outcome of untargeted metabolome profiling, which is trolled crossover interventions in the same individuals
often compared with microbiome features, the majority for years. Intra-individual trials have been done but are
of read-outs are neither annotated nor quantified70. very resource demanding and have no potential for gen-
When considering the translational implications of eralization to the population level161. Therefore, the only
microbiome research, it recently became evident that realistic way forward to gain population-relevant mecha-
most drugs influence microbiome, metatranscriptome nistic insights into how the gut microbiome mediates or
and metabolome profiles in addition to their well-known modifies the effects of diet, exercise, drugs and so forth
effects on bioclinical variables57,62–64. Blindness to the in the human setting appears to be through carefully
impact of medication on ‘omics’ analyses may there- prepared randomized controlled trials of long duration.
fore have inflated the reported findings in human stud- That means the inclusion of homogeneous groups of
ies of any relationships between the gut microbiome people who have undergone deep and extensive pheno-
and host metabolism, which we have discussed in this typing in a sufficient number to have statistical power to
Review. Ideally, any such omics-based studies should be address the hypothesis in focus.
performed in drug-naive individuals. This demand is Finally, the complexity of the gut microbiome
difficult to accommodate in studies of individuals with is daunting, and the global intestinal microbiome is
chronic metabolic diseases who often are prescribed far more than bacteria and archaea. It also includes
multiple medications. Instead, more studies are cur- fungi, bacteriophages and eukaryotic virus. Future,
rently undertaken in individuals at high risk of devel- sequencing-based and culture-based gut microorganism
oping metabolic diseases and prior to medication, for surveys combined with mechanistic exploitations of the
example, in treatment-naive pre-morbid individuals. Yet gut bacteriome, archaeome, phageome, virome and myco
another shortcoming of microbiome studies is the fact biome will exponentially expand our knowledge about
that microbiome analysis has so far largely been limited the interactions within the global intestinal microbial
to relative abundances, failing therefore to account for community. Moreover, not least, exciting new knowl-
biologically relevant differences between microbiomes edge about the multitude of chemicals that the global
owing to considerable sample variation in the bacterial gut microbiome produces affecting host physiology and
cell density of stools or intestinal contents160. numerous pathologies may foster novel efficacious paths
Experimental studies of the targets and mechanisms to stabilize metabolic health of the human holobiont and
of action of a given microbial compound are, for obvi- prevent or combat common human metabolic disorders.
ous reasons, done in vitro in assumed target tissues
or cells and are complemented with in vivo studies in Published online 4 September 2020
1. Lynch, S. V. & Pedersen, O. The human intestinal 9. Qin, N. et al. Alterations of the human gut microbiome 16. Korem, T. et al. Growth dynamics of gut microbiota in
microbiome in health and disease. N. Engl. J. Med. in liver cirrhosis. Nature 513, 59–64 (2014). health and disease inferred from single metagenomic
375, 2369–2379 (2016). 10. Koeth, R. A. et al. Intestinal microbiota metabolism samples. Science 349, 1101–1106 (2015).
2. Jaacks, L. M. et al. The obesity transition: stages of of l-carnitine, a nutrient in red meat, promotes 17. Zeevi, D. et al. Structural variation in the gut
the global epidemic. Lancet Diabetes Endocrinol. 7, atherosclerosis. Nat. Med. 19, 576–585 (2013). microbiome associates with host health. Nature 568,
231–240 (2019). 11. Allin, K. H. et al. Aberrant intestinal microbiota in 43–48 (2019).
3. Younossi, Z. M. et al. Global epidemiology of individuals with prediabetes. Diabetologia 61, 18. Huttenhower, C. et al. Structure, function and diversity
nonalcoholic fatty liver disease — meta-analytic 810–820 (2018). of the healthy human microbiome. Nature 486,
assessment of prevalence, incidence, and outcomes. 12. Bäckhed, F. et al. The gut microbiota as an 207–214 (2012).
Hepatology 64, 73–84 (2016). environmental factor that regulates fat storage. 19. Rothschild, D. et al. Environment dominates over host
4. Zheng, Y., Ley, S. H. & Hu, F. B. Global aetiology Proc. Natl Acad. Sci. USA 101, 15718–15723 genetics in shaping human gut microbiota. Nature
and epidemiology of type 2 diabetes mellitus and its (2004). 555, 210–215 (2018).
complications. Nat. Rev. Endocrinol. 14, 88 (2018). 13. Qin, J. et al. A human gut microbial gene catalogue 20. Falony, G., Vieira-Silva, S. & Raes, J. Richness and
5. Reddy, K. S. & Yusuf, S. Emerging epidemic of established by metagenomic sequencing. Nature 464, ecosystem development across faecal snapshots
cardiovascular disease in developing countries. 59–65 (2010). of the gut microbiota. Nat. Microbiol. 3, 526–528
Circulation 97, 596–601 (1998). 14. Pedersen, H. K. et al. Human gut microbes impact (2018).
6. Lakka, H.-M. et al. The metabolic syndrome and total host serum metabolome and insulin sensitivity. Nature 21. Jakobsson, H. E. et al. Decreased gut microbiota
and cardiovascular disease mortality in middle-aged 535, 376–381 (2016). diversity, delayed Bacteroidetes colonisation and
men. JAMA 288, 2709–2716 (2002). This study identifies P. copri and B. vulgatus as reduced Th1 responses in infants delivered by
7. Müller, O. & Krawinkel, M. Malnutrition and health the main species driving the association between caesarean section. Gut 63, 559–566 (2014).
in developing countries. CMAJ 173, 279–286 (2005). biosynthesis of BCAAs and insulin resistance under 22. Rodríguez, J. M. et al. The composition of the gut
8. Qin, J. et al. A metagenome-wide association study conditions of high fat intake. microbiota throughout life, with an emphasis on
of gut microbiota in type 2 diabetes. Nature 490, 15. Pedersen, H. K. et al. A computational framework early life. Microb. Ecol. Health Dis. 26, 26050
55–60 (2012). to integrate high-throughput ‘-omics’ datasets for (2015).
This is the first metagenome-wide association study to the identification of potential mechanistic links. 23. Flint, H. J., Scott, K. P., Louis, P. & Duncan, S. H.
identify links between the gut microbiome and T2D. Nat. Protoc. 13, 2781 (2018). The role of the gut microbiota in nutrition and health.

Reviews
Nat. Rev. Gastroenterol. Hepatol. 9, 577–589 48. Vieira-Silva, S. et al. Statin therapy is associated with 72. Hansson, G. K. Inflammation, atherosclerosis, and
(2012). lower prevalence of gut microbiota dysbiosis. Nature coronary artery disease. N. Engl. J. Med. 352,
24. Shafquat, A., Joice, R., Simmons, S. L. & 581, 310–315 (2020). 1685–1695 (2005).
Huttenhower, C. Functional and phylogenetic assembly This study observes that obesity-associated 73. Jie, Z. et al. The gut microbiome in atherosclerotic
of microbial communities in the human microbiome. microbiota dysbiosis is negatively associated with cardiovascular disease. Nat. Commun. 8, 845 (2017).
Trends Microbiol. 22, 261–266 (2014). statin treatment, indicating statins as a possible 74. Sokol, H. et al. Faecalibacterium prausnitzii is an anti-
25. Clemente, J. C. et al. The microbiome of uncontacted target for the development of future drug-based inflammatory commensal bacterium identified by gut
Amerindians. Sci. Adv. 1, e1500183 (2015). strategies for the modulation of the intestinal microbiota analysis of Crohn disease patients. Proc.
26. Smits, S. A. et al. Seasonal cycling in the gut microbiome microbiota. Natl Acad. Sci. USA 105, 16731–16736 (2008).
of the Hadza hunter-gatherers of Tanzania. Science 357, 49. Jensen, A. B. et al. Increase in clinically recorded type 2 75. Cui, X. et al. Metagenomic and metabolomic analyses
802–806 (2017). diabetes after colectomy. eLife 7, e37420 (2018). unveil dysbiosis of gut microbiota in chronic heart
27. Tyakht, A. V. et al. Human gut microbiota community 50. Thaiss, C. A. et al. Hyperglycemia drives intestinal failure patients. Sci. Rep. 8, 1–15 (2018).
structures in urban and rural populations in Russia. barrier dysfunction and risk for enteric infection. 76. Wang, Z. et al. Gut flora metabolism of
Nat. Commun. 4, 1–9 (2013). Science 359, 1376–1383 (2018). phosphatidylcholine promotes cardiovascular disease.
28. Le Chatelier, E. et al. Richness of human gut microbiome 51. Association, A. D. Diagnosis and classification of Nature 472, 57–63 (2011).
correlates with metabolic markers. Nature 500, diabetes mellitus. Diabetes Care 37, S81–S90 (2014). This work discovers a relationship between the
541–546 (2013). 52. Deshpande, A. D., Harris-Hayes, M. & Schootman, M. gut microbiota-dependent metabolism of dietary
This work shows that adults with low bacterial gene Epidemiology of diabetes and diabetes-related phosphatidylcholine and pathogenesis of
richness in their gut microbiome feature insulin complications. Phys. Ther. 88, 1254–1264 (2008). arteriosclerosis.
resistance, pro-inflammation dyslipidaemia and 53. Grarup, N., Sandholt, C. H., Hansen, T. & Pedersen, O. 77. Zhu, W. et al. Gut microbial metabolite TMAO
increased body adiposity compared with individuals Genetic susceptibility to type 2 diabetes and obesity: enhances platelet hyperreactivity and thrombosis risk.
with high bacterial gene richness. from genome-wide association studies to rare variants Cell 165, 111–124 (2016).
29. Cox, L. M. & Blaser, M. J. Antibiotics in early life and and beyond. Diabetologia 57, 1528–1541 (2014). 78. Wang, Z. et al. Non-lethal inhibition of gut microbial
obesity. Nat. Rev. Endocrinol. 11, 182–190 (2015). 54. Kasuga, M. Insulin resistance and pancreatic β cell trimethylamine production for the treatment of
30. Ajslev, T., Andersen, C., Gamborg, M., Sørensen, T. failure. J. Clin. Invest. 116, 1756–1760 (2006). atherosclerosis. Cell 163, 1585–1595 (2015).
& Jess, T. Childhood overweight after establishment 55. Bornfeldt, K. E. & Tabas, I. Insulin resistance, This study discovers a structural analogue of
of the gut microbiota: the role of delivery mode, hyperglycemia, and atherosclerosis. Cell Metab. 14, choline that inhibits microbial TMA lyases and the
pre-pregnancy weight and early administration of 575–585 (2011). production of TMA by the gut microbiota.
antibiotics. Int. J. Obes. 35, 522 (2011). 56. Laakso, M. & Kuusisto, J. Insulin resistance and 79. Haghikia, A. et al. Gut microbiota-dependent
31. Mor, A. et al. Prenatal exposure to systemic hyperglycaemia in cardiovascular disease development. trimethylamine N-oxide predicts risk of cardiovascular
antibacterials and overweight and obesity in Danish Nat. Rev. Endocrinol. 10, 293–302 (2014). events in patients with stroke and is related to
schoolchildren: a prevalence study. Int. J. Obes. 39, 57. Vila, A. V. et al. Impact of commonly used drugs on proinflammatory monocytes. Arterioscl. Throm. Vas.
1450 (2015). the composition and metabolic function of the gut Biol. 38, 2225–2235 (2018).
32. Palleja, A. et al. Recovery of gut microbiota of healthy microbiota. Nat. Commun. 11, 1–11 (2020). 80. Senthong, V. et al. Intestinal microbiota-generated
adults following antibiotic exposure. Nat. Microbiol. 3, 58. Zhong, H. et al. Distinct gut metagenomics metabolite trimethylamine-N-oxide and 5-year
1255–1265 (2018). and metaproteomics signatures in prediabetics and mortality risk in stable coronary artery disease: the
33. Thuny, F. et al. Vancomycin treatment of infective treatment-naïve type 2 diabetics. EBioMedicine 47, contributory role of intestinal microbiota in a
endocarditis is linked with recently acquired obesity. 373–383 (2019). COURAGE-like patient cohort. J. Am. Heart Assoc. 5,
PLoS ONE 5, e9074 (2010). 59. Whiting, D. R., Guariguata, L., Weil, C. & Shaw, J. e002816 (2016).
34. Mikkelsen, K. H. et al. Effect of antibiotics on gut IDF diabetes atlas: global estimates of the prevalence 81. Collins, H. L. et al. l-Carnitine intake and high
microbiota, gut hormones and glucose metabolism. of diabetes for 2011 and 2030. Diabetes Res. Clin. trimethylamine N-oxide plasma levels correlate
PLoS ONE 10, e0142352 (2015). Pract. 94, 311–321 (2011). with low aortic lesions in ApoE−/− transgenic mice
35. Reijnders, D. et al. Effects of gut microbiota 60. Crusell, M. K. W. et al. Gestational diabetes is expressing CETP. Atherosclerosis 244, 29–37 (2016).
manipulation by antibiotics on host metabolism associated with change in the gut microbiota 82. He, K. et al. Accumulated evidence on fish consumption
in obese humans: a randomized double-blind composition in third trimester of pregnancy and and coronary heart disease mortality: a meta-analysis
placebo-controlled trial. Cell Metab. 24, 63–74 postpartum. Microbiome 6, 89 (2018). of cohort studies. Circulation 109, 2705–2711 (2004).
(2016). 61. Karlsson, F. H. et al. Gut metagenome in European 83. Koay, Y. C. et al. Plasma levels of TMAO can be
36. Fujisaka, S. et al. Antibiotic effects on gut microbiota women with normal, impaired and diabetic glucose increased with ‘healthy’ and ‘unhealthy’ diets and do
and metabolism are host dependent. J. Clin. Invest. control. Nature 498, 99–103 (2013). not correlate with the extent of atherosclerosis but
126, 4430–4443 (2016). 62. Forslund, K. et al. Disentangling type 2 diabetes and with plaque instability. Cardiovasc. Res. 8, cvaa094
37. Yatsunenko, T. et al. Human gut microbiome viewed metformin treatment signatures in the human gut (2020).
across age and geography. Nature 486, 222–227 microbiota. Nature 528, 262–266 (2015). 84. Tang, W. W. et al. Intestinal microbial metabolism of
(2012). This work outlines a paradigm to disentangle phosphatidylcholine and cardiovascular risk. N. Engl.
38. Cotillard, A. et al. Dietary intervention impact on gut disease microbiome features from secondary J. Med. 368, 1575–1584 (2013).
microbial gene richness. Nature 500, 585–588 (2013). changes in the microbiome induced by medication. 85. Brunt, E. M. et al. Nonalcoholic fatty liver disease
39. World Health Organization. Obesity: Preventing 63. Wu, H. et al. Metformin alters the gut microbiome of (NAFLD) activity score and the histopathologic
and Managing the Global Epidemic (World Health individuals with treatment-naive type 2 diabetes, diagnosis in NAFLD: distinct clinicopathologic
Organization, 2000). contributing to the therapeutic effects of the drug. meanings. Hepatology 53, 810–820 (2011).
40. McAllister, E. J. et al. Ten putative contributors to Nat. Med. 23, 850–858 (2017). 86. Jiang, W. et al. Dysbiosis gut microbiota associated
the obesity epidemic. Crit. Rev. Food Sci. Nutr. 49, 64. Bryrup, T. et al. Metformin-induced changes of the gut with inflammation and impaired mucosal immune
868–913 (2009). microbiota in healthy young men: results of a non- function in intestine of humans with non-alcoholic fatty
41. Turnbaugh, P. J. et al. An obesity-associated gut blinded, one-armed intervention study. Diabetologia liver disease. Sci. Rep. 5, 8096 (2015).
microbiome with increased capacity for energy 62, 1024–1035 (2019). 87. Zhu, L. et al. Characterization of gut microbiomes
harvest. Nature 444, 1027–1031 (2006). 65. Sun, L. et al. Gut microbiota and intestinal FXR in nonalcoholic steatohepatitis (NASH) patients:
42. Tims, S. et al. Microbiota conservation and BMI mediate the clinical benefits of metformin. Nat. Med. a connection between endogenous alcohol and NASH.
signatures in adult monozygotic twins. ISME J. 7, 24, 1919 (2018). Hepatology 57, 601–609 (2013).
707–717 (2013). 66. De La Cuesta-Zuluaga, J. et al. Metformin is 88. Del Chierico, F. et al. Gut microbiota profiling of
43. Gophna, U., Konikoff, T. & Nielsen, H. B. Oscillospira associated with higher relative abundance of mucin- pediatric nonalcoholic fatty liver disease and obese
and related bacteria—from metagenomic species to degrading Akkermansia muciniphila and several patients unveiled by an integrated meta-omics-based
metabolic features. Environ. Microbiol. 19, 835–841 short-chain fatty acid-producing microbiota in the gut. approach. Hepatology 65, 451–464 (2017).
(2017). Diabetes Care 40, 54–62 (2017). 89. Nair, S., Cope, K., Terence, R. H. & Diehl, A. M.
44. Miller, T. L., Wolin, M., de Macario, E. C. & Macario, A. 67. Adeshirlarijaney, A., Zou, J., Tran, H. Q., Chassaing, B. Obesity and female gender increase breath ethanol
Isolation of Methanobrevibacter smithii from human & Gewirtz, A. T. Amelioration of metabolic syndrome concentration: potential implications for the
feces. Appl. Environ. Microbiol. 43, 227–232 (1982). by metformin associates with reduced indices of pathogenesis of nonalcoholic steatohepatitis. Am. J.
45. Liu, R. et al. Gut microbiome and serum metabolome low-grade inflammation independently of the gut Gastroenterol. 96, 1200–1204 (2001).
alterations in obesity and after weight-loss intervention. microbiota. Am. J. Physiol. Endocrinol. Metab. 317, 90. Rao, R., Seth, A. & Sheth, P. Recent advances in
Nat. Med. 23, 859–868 (2017). E1121–E1130 (2019). alcoholic liver disease I. Role of intestinal permeability
This study links intestinal microbiota alterations, 68. Pryor, R. et al. Host–microbe–drug–nutrient screen and endotoxemia in alcoholic liver disease. Am. J.
circulating amino acids and obesity, and suggests identifies bacterial effectors of metformin therapy. Cell Physiol. Gastrointest. Liver Physiol. 286, G881–G884
that a possible way to intervene in obesity is by 178, 1299–1312.e29 (2019). (2004).
targeting the gut microbiota. 69. Herrington, W., Lacey, B., Sherliker, P., Armitage, J. 91. Xu, J. et al. Synergistic steatohepatitis by moderate
46. Thingholm, L. B. et al. Obese individuals with and & Lewington, S. Epidemiology of atherosclerosis obesity and alcohol in mice despite increased
without type 2 diabetes show different gut microbial and the potential to reduce the global burden of adiponectin and p-AMPK. J. Hepatol. 55, 673–682
functional capacity and composition. Cell Host atherothrombotic disease. Circ. Res. 118, 535–546 (2011).
Microbe 26, 252–264.e10 (2019). (2016). 92. de Medeiros, I. C. & de Lima, J. G. Is nonalcoholic
47. Ridaura, V. K. et al. Gut microbiota from twins 70. Fan, Y. et al. Comprehensive metabolomic fatty liver disease an endogenous alcoholic fatty
discordant for obesity modulate metabolism in mice. characterization of coronary artery diseases. J. Am. liver disease? — A mechanistic hypothesis.
Science 341, 1241214 (2013). Coll. Cardiol. 68, 1281–1293 (2016). Med. Hypotheses 85, 148–152 (2015).
This study shows that adiposity is transmissible in 71. Michos, E. D., McEvoy, J. W. & Blumenthal, R. S. Lipid 93. Schwenger, K. J., Clermont-Dejean, N. & Allard, J. P.
a diet-dependent manner from human to mouse management for the prevention of atherosclerotic The role of the gut microbiome in chronic liver
and is associated with alterations in serum levels cardiovascular disease. N. Engl. J. Med. 381, disease: the clinical evidence revised. JHEP Rep. 1,
of BCAAs. 1557–1567 (2019). 214–226 (2019).

Reviews
94. Le Roy, T. et al. Intestinal microbiota determines insulin action in mice and humans. Sci. Transl Med. 142. Mills, E. L. et al. Accumulation of succinate controls
development of non-alcoholic fatty liver disease in 11, eaav0120 (2019). activation of adipose tissue thermogenesis. Nature
mice. Gut 62, 1787–1794 (2013). 119. Bouter, K. et al. Differential metabolic effects of oral 560, 102–106 (2018).
95. Yuan, J. et al. Fatty liver disease caused by high- butyrate treatment in lean versus metabolic syndrome 143. Tannahill, G. et al. Succinate is an inflammatory signal
alcohol-producing Klebsiella pneumoniae. Cell Metab. subjects. Clin. Trans. Gastroenterol 9, 155 (2018). that induces IL-1β through HIF-1α. Nature 496,
30, 675–688.e7 (2019). 120. Gao, X. et al. Acetate functions as an epigenetic 238–242 (2013).
96. Hoyles, L. et al. Molecular phenomics and metabolite to promote lipid synthesis under hypoxia. 144. Suárez-Zamorano, N. et al. Microbiota depletion
metagenomics of hepatic steatosis in non-diabetic Nat. Commun. 7, 11960 (2016). promotes browning of white adipose tissue and
obese women. Nat. Med. 24, 1070–1080 (2018). 121. Perry, R. J. et al. Acetate mediates a microbiome– reduces obesity. Nat. Med. 21, 1497 (2015).
This study demonstrates that individuals with liver brain–β-cell axis to promote metabolic syndrome. 145. Li, B. et al. Microbiota depletion impairs thermogenesis
steatosis have low microbial gene richness and Nature 534, 213–217 (2016). of brown adipose tissue and browning of white adipose
increased genetic potential for the processing of 122. Freeland, K. R. & Wolever, T. M. Acute effects of tissue. Cell Rep. 26, 2720–2737.e5 (2019).
dietary lipids and endotoxin biosynthesis, hepatic intravenous and rectal acetate on glucagon-like 146. Kamio, Y. & Nikaido, H. Outer membrane of
inflammation and dysregulation of aromatic and peptide-1, peptide YY, ghrelin, adiponectin and Salmonella typhimurium: accessibility of phospholipid
BCAA metabolism. tumour necrosis factor-α. Br. J. Nutr. 103, 460–466 head groups to phospholipase C and cyanogen
97. UNICEF–WHO–The World Bank Group: Joint Child (2010). bromide activated dextran in the external medium.
Malnutrition Estimates — levels and trends in 123. Ridlon, J. M., Kang, D.-J. & Hylemon, P. B. Bile salt Biochemistry 15, 2561–2570 (1976).
child malnutrition: key findings of the 2015 edition. biotransformations by human intestinal bacteria. 147. Cani, P. D. et al. Metabolic endotoxemia initiates
Global Database on Child Growth and Malnutrition J. Lipid Res. 47, 241–259 (2006). obesity and insulin resistance. Diabetes 56,
(WHO, 2015). 124. Fiorucci, S., Mencarelli, A., Palladino, G. & Cipriani, S. 1761–1772 (2007).
98. Black, R. E. et al. Maternal and child undernutrition: Bile-acid-activated receptors: targeting TGR5 and 148. Luck, H. et al. Gut-associated IgA+ immune cells regulate
global and regional exposures and health farnesoid-X-receptor in lipid and glucose disorders. obesity-related insulin resistance. Nat. Commun. 10,
consequences. Lancet 371, 243–260 (2008). Trends Pharmacol. Sci. 30, 570–580 (2009). 1–17 (2019).
99. Million, M., Diallo, A. & Raoult, D. Gut microbiota and 125. Prawitt, J. et al. Farnesoid X receptor deficiency 149. Winer, D. A., Luck, H., Tsai, S. & Winer, S. The
malnutrition. Microb. Pathog. 106, 127–138 (2017). improves glucose homeostasis in mouse models intestinal immune system in obesity and insulin
100. Smith, M. I. et al. Gut microbiomes of Malawian of obesity. Diabetes 60, 1861–1871 (2011). resistance. Cell Metab. 23, 413–426 (2016).
twin pairs discordant for kwashiorkor. Science 339, 126. Thomas, C. et al. TGR5-mediated bile acid sensing 150. Breton, J. et al. Gut commensal E. coli proteins
548–554 (2013). controls glucose homeostasis. Cell Metab. 10, activate host satiety pathways following nutrient-
101. Lee, J.-H., Li, X. & O’Sullivan, D. J. Transcription 167–177 (2009). induced bacterial growth. Cell Metab. 23, 324–334
analysis of a lantibiotic gene cluster from 127. Makishima, M. et al. Identification of a nuclear (2016).
Bifidobacterium longum DJO10A. Appl. Environ. receptor for bile acids. Science 284, 1362–1365 151. Tennoune, N. et al. Bacterial ClpB heat-shock protein,
Microbiol. 77, 5879–5887 (2011). (1999). an antigen-mimetic of the anorexigenic peptide α-MSH,
102. Mata, L. J. et al. Gastrointestinal flora of children with 128. Kawamata, Y. et al. A G protein-coupled receptor at the origin of eating disorders. Transl Psychiat. 4,
protein — calorie malnutrition. Am. J. Clin. Nutr. 25, responsive to bile acids. J. Biol. Chem. 278, e458–e458 (2014).
1118–1126 (1972). 9435–9440 (2003). 152. Breton, J. et al. Elevated plasma concentrations of
103. Kelly, D. et al. Commensal anaerobic gut bacteria 129. Pathak, P. et al. Intestine farnesoid X receptor agonist bacterial ClpB protein in patients with eating
attenuate inflammation by regulating nuclear- and the gut microbiota activate G-protein bile acid disorders. Int. J. Eat. Disord. 49, 805–808 (2016).
cytoplasmic shuttling of PPAR-γ and RelA. receptor-1 signaling to improve metabolism. 153. Cohen, L. J. et al. Functional metagenomic discovery of
Nat. Immunol. 5, 104–112 (2004). Hepatology 68, 1574–1588 (2018). bacterial effectors in the human microbiome and
104. Martens, J.-H., Barg, H., Warren, M. A. & Jahn, D. 130. Modica, S., Gadaleta, R. M. & Moschetta, A. isolation of commendamide, a GPCR G2A/132 agonist.
Microbial production of vitamin B 12. Appl. Microbiol. Deciphering the nuclear bile acid receptor FXR Proc. Natl Acad. Sci. USA 112, E4825–E4834 (2015).
Biotechnol. 58, 275–285 (2002). paradigm. Nucl. Recept. Signal. https://doi.org/ 154. Cohen, L. J. et al. Commensal bacteria make GPCR
105. Stecher, B. & Hardt, W.-D. The role of microbiota in 10.1621/nrs.08005 (2010). ligands that mimic human signalling molecules. Nature
infectious disease. Trends Microbiol. 16, 107–114 131. Spinelli, V. et al. Influence of Roux-en-Y gastric bypass 549, 48 (2017).
(2008). on plasma bile acid profiles: a comparative study 155. Qiang, X. et al. New melanocortin-like peptide of
106. Gehrig, J. L. et al. Effects of microbiota-directed foods between rats, pigs and humans. Int. J. Obes. 40, 1260 E. coli can suppress inflammation via the mammalian
in gnotobiotic animals and undernourished children. (2016). melanocortin-1 receptor (MC1R): possible endocrine-
Science 365, eaau4732 (2019). 132. Kindel, T. L. et al. Increased glycine-amidated like function for microbes of the gut. NPJ Biofilms
107. Jumpertz, R. et al. Energy-balance studies reveal hyocholic acid correlates to improved early weight loss Microbiomes 3, 31 (2017).
associations between gut microbes, caloric load, and after sleeve gastrectomy. Surg. Endosc. 32, 805–812 156. Elsden, S. R., Hilton, M. G. & Waller, J. M.
nutrient absorption in humans. Am. J. Clin. Nutr. 94, (2018). The end products of the metabolism of aromatic
58–65 (2011). 133. Canfora, E. E., Meex, R. C., Venema, K. & Blaak, E. E. amino acids by Clostridia. Arch. Microbiol. 107,
108. Lin, H. V. et al. Butyrate and propionate protect Gut microbial metabolites in obesity, NAFLD and 283–288 (1976).
against diet-induced obesity and regulate gut T2DM. Nat. Rev. Endocrinol. 15, 261–273 (2019). 157. Lin, C.-J., Wu, V., Wu, P.-C. & Wu, C.-J. Meta-analysis
hormones via free fatty acid receptor 3-independent 134. Bansal, T., Alaniz, R. C., Wood, T. K. & Jayaraman, A. of the associations of p-cresyl sulfate (PCS) and indoxyl
mechanisms. PLoS ONE 7, e35240 (2012). The bacterial signal indole increases epithelial-cell sulfate (IS) with cardiovascular events and all-cause
109. Xiong, Y. et al. Short-chain fatty acids stimulate leptin tight-junction resistance and attenuates indicators mortality in patients with chronic renal failure.
production in adipocytes through the G protein- of inflammation. Proc. Natl Acad. Sci. USA 107, PLoS ONE 10, e0132589 (2015).
coupled receptor GPR41. Proc. Natl Acad. Sci. USA 228–233 (2010). 158. Krisko, T. I. et al. Dissociation of adaptive thermogenesis
101, 1045–1050 (2004). 135. Shimada, Y. et al. Commensal bacteria-dependent from glucose homeostasis in microbiome-deficient
110. Vinolo, M. A. R. et al. Tributyrin attenuates obesity- indole production enhances epithelial barrier function mice. Cell Metab. 31, 592–604.e9 (2020).
associated inflammation and insulin resistance in high- in the colon. PLoS ONE 8, e80604 (2013). 159. Vojinovic, D. et al. Relationship between gut
fat-fed mice. Am. J. Physiol. Endocrinol. Metab. 303, 136. Chimerel, C. et al. Bacterial metabolite indole microbiota and circulating metabolites in
E272–E282 (2012). modulates incretin secretion from intestinal population-based cohorts. Nat. Commun. 10,
111. Maslowski, K. M. et al. Regulation of inflammatory enteroendocrine L cells. Cell Rep. 9, 1202–1208 1–7 (2019).
responses by gut microbiota and chemoattractant (2014). 160. Vandeputte, D. et al. Quantitative microbiome
receptor GPR43. Nature 461, 1282–1286 (2009). 137. De Mello, V. D. et al. Indolepropionic acid and novel profiling links gut community variation to microbial
112. Macia, L. et al. Metabolite-sensing receptors GPR43 lipid metabolites are associated with a lower risk of load. Nature 551, 507 (2017).
and GPR109A facilitate dietary fibre-induced gut type 2 diabetes in the Finnish Diabetes Prevention This study builds a workflow for the quantitative
homeostasis through regulation of the inflammasome. Study. Sci. Rep. 7, 46337 (2017). microbiome profiling of faecal material, showing
Nat. Commun. 6, 6734 (2015). This study shows that microbial indole propionate that quantitative microbiome profiling has a
113. Pluznick, J. L. Renal and cardiovascular sensory and additional metabolites associate with lower substantial effect on co-occurrence analyses and
receptors and blood pressure regulation. Am. J. risk of incident T2D. the characterization of disease-associated
Physiol. Ren. Physiol. 305, F439–F444 (2013). 138. Wang, T. J. et al. Metabolite profiles and the risk of microbiota perturbations.
114. Byndloss, M. X. et al. Microbiota-activated PPAR-γ developing diabetes. Nat. Med. 17, 448 (2011). 161. Zhou, W. et al. Longitudinal multi-omics of host–microbe
signaling inhibits dysbiotic Enterobacteriaceae 139. Sun, H. et al. Catabolic defect of branched-chain dynamics in prediabetes. Nature 569, 663–671
expansion. Science 357, 570–575 (2017). amino acids promotes heart failure. Circulation 133, (2019).
115. Tang, T. W. et al. Loss of gut microbiota alters immune 2038–2049 (2016). 162. Flint, A., Raben, A., Rehfeld, J., Holst, J. & Astrup, A.
system composition and cripples postinfarction 140. Koh, A. et al. Microbially produced imidazole The effect of glucagon-like peptide-1 on energy
cardiac repair. Circulation 139, 647–659 (2019). propionate impairs insulin signaling through mTORC1. expenditure and substrate metabolism in humans.
116. Chambers, E. S. et al. Effects of targeted delivery of Cell 175, 947–961.e17 (2018). Int. J. Obes. 24, 288 (2000).
propionate to the human colon on appetite regulation, This study demonstrates that imidazole propionate 163. Batterham, R. L. et al. Gut hormone PYY 3-36
body weight maintenance and adiposity in overweight is produced from histidine in a gut simulator at physiologically inhibits food intake. Nature 418, 650
adults. Gut 64, 1744–1754 (2015). higher concentrations when using faecal microbiota (2002).
117. Sanna, S. et al. Causal relationships among the gut from subjects with T2D than from individuals 164. Holz, I. V. IV, G. G., Kiihtreiber, W. M. & Habener, J. F.
microbiome, short-chain fatty acids and metabolic without T2D, and that it impairs glucose tolerance Pancreatic β-cells are rendered glucose-competent by
diseases. Nat. Genet. 51, 600 (2019). when administered to mice. the insulinotropic hormone glucagon-like peptide-1
This study presents data providing evidence of a 141. MacDonald, M. J., Fahien, L. A., Mertz, R. J. & (7–37). Nature 361, 362 (1993).
causal effect of the gut microbiome on metabolic Rana, R. S. Effect of esters of succinic acid and other 165. Macfarlane, S. & Macfarlane, G. T. Composition and
traits. citric acid cycle intermediates on insulin release and metabolic activities of bacterial biofilms colonizing
118. Tirosh, A. et al. The short-chain fatty acid propionate inositol phosphate formation by pancreatic islets. food residues in the human gut. Appl. Environ.
increases glucagon and FABP4 production, impairing Arch. Biochem. Biophys. 269, 400–406 (1989). Microbiol. 72, 6204–6211 (2006).

Reviews
166. Roager, H. M. et al. Colonic transit time is related to signs of metabolic syndrome: a randomized cross-over 193. Nicolucci, A. C. et al. Prebiotics reduce body fat
bacterial metabolism and mucosal turnover in the gut. trial. Clin. Nutr. 39, 67–79 (2019). and alter intestinal microbiota in children who are
Nat. Microbiol. 1, 16093 (2016). 181. Dieterich, W. et al. Influence of low FODMAP and overweight or with obesity. Gastroenterology 153,
This study indicates that the colonic transit time is gluten-free diets on disease activity and intestinal 711–722 (2017).
an important factor to consider in microbiome and microbiota in patients with non-celiac gluten 194. Tsilingiri, K. et al. Probiotic and postbiotic activity in
metabolomics studies. sensitivity. Clin. Nutr. 38, 697–707 (2019). health and disease: comparison on a novel polarised ex-
167. Russell, W. R. et al. Major phenylpropanoid-derived 182. Hansen, L. B. et al. A low-gluten diet induces changes vivo organ culture model. Gut 61, 1007–1015 (2012).
metabolites in the human gut can arise from microbial in the intestinal microbiome of healthy Danish adults. 195. Plovier, H. et al. A purified membrane protein
fermentation of protein. Mol. Nutr. Food Res. 57, Nat. Commun. 9, 4630 (2018). from Akkermansia muciniphila or the pasteurized
523–535 (2013). 183. Roager, H. M. et al. Whole grain-rich diet reduces bacterium improves metabolism in obese and diabetic
168. Flemer, B. et al. The oral microbiota in colorectal cancer body weight and systemic low-grade inflammation mice. Nat. Med. 23, 107 (2017).
is distinctive and predictive. Gut 67, 1454–1463 without inducing major changes of the gut microbiome: 196. Van Nood, E. et al. Duodenal infusion of donor feces
(2018). a randomised cross-over trial. Gut 68, 83–93 (2019). for recurrent Clostridium difficile. N. Engl. J. Med.
169. Plaza Oñate, F. et al. MSPminer: abundance-based 184. Sanders, M. E., Merenstein, D. J., Reid, G., Gibson, G. R. 368, 407–415 (2013).
reconstitution of microbial pan-genomes from shotgun & Rastall, R. A. Probiotics and prebiotics in intestinal 197. Kootte, R. S. et al. Improvement of insulin sensitivity
metagenomic data. Bioinformatics 35, 1544–1552 health and disease: from biology to the clinic. after lean donor feces in metabolic syndrome is
(2018). Nat. Rev. Gastroenterol. Hepatol. 16, 605–616 driven by baseline intestinal microbiota composition.
170. Truong, D. T., Tett, A., Pasolli, E., Huttenhower, C. & (2019). Cell Metab. 26, 611–619.e6 (2017).
Segata, N. Microbial strain-level population structure 185. Martín, R. et al. Functional characterization of 198. Duan, F. F., Liu, J. H. & March, J. C. Engineered
and genetic diversity from metagenomes. Genome novel Faecalibacterium prausnitzii strains isolated commensal bacteria reprogram intestinal cells into
Res. 27, 626–638 (2017). from healthy volunteers: a step forward in the use glucose-responsive insulin-secreting cells for the
171. Zolfo, M., Tett, A., Jousson, O., Donati, C. & Segata, N. of F. prausnitzii as a next-generation probiotic. treatment of diabetes. Diabetes 64, 1794–1803
MetaMLST: multi-locus strain-level bacterial typing Front. Microbiol. 8, 1226 (2017). (2015).
from metagenomic samples. Nucleic Acids Res. 45, 186. Depommier, C. et al. Supplementation with 199. Górski, A. et al. Perspectives of phage therapy in non-
e7–e7 (2016). Akkermansia muciniphila in overweight and obese bacterial infections. Front. Microbiol. https://doi.org/
172. Truong, D. T. et al. MetaPhlAn2 for enhanced human volunteers: a proof-of-concept exploratory 10.3389/fmicb.2018.03306 (2018).
metagenomic taxonomic profiling. Nat. Methods 12, study. Nat. Med. 25, 1096–1103 (2019). 200. Ramachandran, G. & Bikard, D. Editing the microbiome
902 (2015). This proof-of-concept study shows that the CRISPR way. Philos. Trans. R. Soc. Lond. B Biol. Sci.
173. Greenblum, S., Carr, R. & Borenstein, E. Extensive supplementation with A. muciniphila improves 374, 20180103 (2019).
strain-level copy-number variation across human gut several dysmetabolic features.
microbiome species. Cell 160, 583–594 (2015). 187. Cani, P. D. & Van Hul, M. Novel opportunities for Acknowledgements
174. Langfelder, P. & Horvath, S. WGCNA: an R package next-generation probiotics targeting metabolic This study was supported by Marie Skłodowska-C urie
for weighted correlation network analysis. BMC syndrome. Curr. Opin. Biotechnol. 32, 21–27 Individual Fellowship 797267 (granted to Y.F.). The Novo
Bioinforma. 9, 559 (2008). (2015). Nordisk Foundation Center for Basic Metabolic Research is
175. Wikoff, W. R. et al. Diacetylspermine is a novel 188. Kristensen, N. B. et al. Alterations in fecal microbiota an independent research centre at the University of
prediagnostic serum biomarker for non-small-cell composition by probiotic supplementation in healthy Copenhagen partially funded by an unrestricted donation
lung cancer and has additive performance with adults: a systematic review of randomized controlled from the Novo Nordisk Foundation.
pro-surfactant protein B. J. Clin. Oncol. 33, 3880 trials. Genome Med. 8, 52 (2016).
(2015). 189. Zmora, N. et al. Personalized gut mucosal colonization Author contributions
176. de Hoffmann, E. Tandem mass spectrometry: resistance to empiric probiotics is associated with The authors contributed equally to all aspects of the article.
a primer. J. Mass. Spectrom. 31, 129–137 (1996). unique host and microbiome features. Cell 174,
177. Donia, M. S. & Fischbach, M. A. Small molecules from 1388–1405.e21 (2018). Competing interests
the human microbiota. Science 349, 1254766 (2015). 190. Suez, J. et al. Post-antibiotic gut mucosal microbiome The authors declare no competing interests.
178. Cani, P. D. et al. Changes in gut microbiota control reconstitution is impaired by probiotics and improved
metabolic endotoxemia-induced inflammation in by autologous FMT. Cell 174, 1406–1423.e16 Peer review information
high-fat diet-induced obesity and diabetes in mice. (2018). Nature Reviews Microbiology thanks M. Nieuwdorp, C. Thaiss
Diabetes 57, 1470–1481 (2008). 191. Everard, A. et al. Responses of gut microbiota and and the other, anonymous, reviewer(s) for their contribution
179. Wolters, M. et al. Dietary fat, the gut microbiota, and glucose and lipid metabolism to prebiotics in genetic to the peer review of this work.
metabolic health — a systematic review conducted obese and diet-induced leptin-resistant mice. Diabetes
within the MyNewGut project. Clin. Nutr. 38, 60, 2775–2786 (2011). Publisher’s note
2504–2520 (2018). 192. Dewulf, E. M. et al. Insight into the prebiotic concept: Springer Nature remains neutral with regard to jurisdictional
180. Kjølbæk, L. et al. Arabinoxylan oligosaccharides and lessons from an exploratory, double blind intervention claims in published maps and institutional affiliations.
polyunsaturated fatty acid effects on gut microbiota study with inulin-type fructans in obese women. Gut
and metabolic markers in overweight individuals with 62, 1112–1121 (2013). © Springer Nature Limited 2020

Gut Microbiota in Human Metabolic Health and Disease: Yong Fan and Oluf Pedersen

Uploaded by

Copyright:

Available Formats

Gut Microbiota in Human Metabolic Health and Disease: Yong Fan and Oluf Pedersen

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Gut Microbiota in Human Metabolic Health and Disease: Yong Fan and Oluf Pedersen

Uploaded by

Copyright:

Available Formats

Reviews

Gut microbiota in human metabolic

naTure RevIeWS | MIcrobIology volume 19 | January 2021 | 55

Box 1 | Tools for dissecting gut microbiota–metabolic health associations

56 | January 2021 | volume 19 www.nature.com/nrmicro

Box 2 | A framework to study causality of gut microbial components

precursor, which results in an increased community

naTure RevIeWS | MIcrobIology volume 19 | January 2021 | 57

Box 3 | Microbiota-​changing interventions

microbiomes (see the figure).

58 | January 2021 | volume 19 www.nature.com/nrmicro

naTure RevIeWS | MIcrobIology volume 19 | January 2021 | 59

Sulfate • Deamination Miscellaneous

↓ Blood ↑ Energy ↑ Glucose-stimulated ↑ Blood ↑ Metabolic ↓ Glucose-stimulated

60 | January 2021 | volume 19 www.nature.com/nrmicro

naTure RevIeWS | MIcrobIology volume 19 | January 2021 | 61

Obesity Taxa features Function features

Metabolic liver Cardio-metabolic

62 | January 2021 | volume 19 www.nature.com/nrmicro

naTure RevIeWS | MIcrobIology volume 19 | January 2021 | 63

64 | January 2021 | volume 19 www.nature.com/nrmicro

GPCRs TGR5 GPR-119 GLP-1/

Satiety • Energy harvest Atherosclerosis • Fatty acid metabolism • Insulin resistance

naTure RevIeWS | MIcrobIology volume 19 | January 2021 | 65

66 | January 2021 | volume 19 www.nature.com/nrmicro

naTure RevIeWS | MIcrobIology volume 19 | January 2021 | 67

68 | January 2021 | volume 19 www.nature.com/nrmicro

naTure RevIeWS | MIcrobIology volume 19 | January 2021 | 69

70 | January 2021 | volume 19 www.nature.com/nrmicro

naTure RevIeWS | MIcrobIology volume 19 | January 2021 | 71

You might also like

Box 3 | Microbiota-changing interventions