Functional interactions between the gut microbiota and host metabolism

Journal name:
Nature
Volume:
489,
Pages:
242–249
Date published:
DOI:
doi:10.1038/nature11552
Published online

Abstract

The link between the microbes in the human gut and the development of obesity, cardiovascular disease and metabolic syndromes, such as type 2 diabetes, is becoming clearer. However, because of the complexity of the microbial community, the functional connections are less well understood. Studies in both mice and humans are helping to show what effect the gut microbiota has on host metabolism by improving energy yield from food and modulating dietary or the host-derived compounds that alter host metabolic pathways. Through increased knowledge of the mechanisms involved in the interactions between the microbiota and its host, we will be in a better position to develop treatments for metabolic disease.

At a glance

Figures

  1. Effects of colonic fermentation of dietary fibres.
    Figure 1: Effects of colonic fermentation of dietary fibres.

    Complex carbohydrates, such as dietary fibre, are metabolized by the colonic microbiota to oligosaccharides and monosaccharides and then fermented to short-chain fatty acid end-products, mainly acetate, propionate and butyrate. Short-chain fatty acids are absorbed in the colon, where butyrate provides energy for colonic epithelial cells, and acetate and propionate reach the liver and peripheral organs, where they are substrates for gluconeogenesis and lipogenesis. In addition to being energy sources, short-chain fatty acids control colonic gene expression by inhibiting the enzyme histone deacetylase (HDAC) and metabolic regulation by signalling through G-protein-coupled receptors (GPCRs), such as GPR41 or GPR43.

  2. Features of the gut microbiota that promote obesity and insulin resistance.
    Figure 2: Features of the gut microbiota that promote obesity and insulin resistance.

    Alterations to the composition and metabolic capacity of gut microbiota in obesity promote adiposity and influence metabolic processes in peripheral organs, such as the control of satiety in the brain; the release of hormones from the gut (shown as PYY and GLP-1); and the synthesis, storage or metabolism of lipids in the adipose tissue, liver and muscle. Microbial molecules also increase intestinal permeability, leading to systemic inflammation and insulin resistance.

  3. Diet-independent and -dependent microbial effects on host metabolism.
    Figure 3: Diet-independent and -dependent microbial effects on host metabolism.

    The gut microbiota produces pro-inflammatory molecules, such as lipopolysaccharide and peptidoglycan, which may affect host metabolism through proteins produced by the host to mediate the immune response. Choline, cholesterol and polysaccharides obtained from the diet are metabolized by the gut microbiota and either directly or through further host–microbial co-metabolization generate bioactive compounds. In the case of choline, this can lead to cardiovascular disease; for cholesterol, activation of TGR5 can increase energy expenditure and GLP-1 secretion or protection against heart disease; and for polysaccharides, short-chain fatty acids can be used as an energy source or can bind to GPR41 or GPR43 to regulate hormones and modulate inflammation. FMO, flavin-containing monooxygenase; TLR4, Toll-like receptor 4; TMA, trimethylamine; TMAO, trimethylamine-N-oxide.

  4. Different microbial innate immune mechanisms affect host metabolism in the gut and liver.
    Figure 4: Different microbial innate immune mechanisms affect host metabolism in the gut and liver.

    Plasma lipopolysaccharide seems to rise with higher fat intake, and those with high visceral adiposity have higher levels of microbial DNA in their blood. Both lipopolysaccharide and microbial DNA seem to be connected with gut permeability. NLRP3 and 6 are both important regulators of microbial ecology through the effector protein interleukin-18 (IL-18). The altered gut microbiota can stimulate CCL5 secretion, which can result in increased permeability and influx of microbial components. In the liver, lipopolysaccharide and bacterial DNA activate the receptors, TLR4 and 9, leading to increased tumour-necrosis-factor-α (TNFα) secretion and development of non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH).

Changes to lifestyle and an increase in the availability of energy-rich foods are important contributors to the worldwide obesity epidemic. The microbial inhabitants of the gut can also have an influence on metabolic processes, such as energy extraction from food, and should be considered an environmental factor that contributes to obesity and its comorbidities (such as insulin resistance, diabetes and cardiovascular disease).

Culture-independent methods to study microbial communities have advanced our knowledge of this human gut microbiota (Box 1) (see page 250 of this issue). Profiling of the common proxy for this community, the faecal microbiota, by 16S ribosomal RNA surveys and by direct sequencing of genetic material have shown that the human gut microbiota is a complex community of 100 trillion archaeal and bacterial cells distributed over more than 1,000 species1 (Box 2). The community is dominated by bacteria, with more than 90% of the species belonging to Firmicutes and Bacteroidetes. Each person has a distinct and highly variable microbiota, but a conserved set of gut colonizers (the core gut microbiota) and genes (the core microbiome) are shared among individuals1, 2 and may be required for the correct functioning of the gut.

Box 1: Terminology

  • Enterotype is the grouping of the microbiota of a given person into discrete configurations. But recent data have conflicted with this definition and suggest that enterotypes may be less discrete.
  • Gnotobiotics is the study of animals living in a microbiologically defined environment, either germ-free or colonized with known bacteria.
  • Inflammasomes are protein complexes containing an intracellular sensor (such as a nucleotide-binding oligomerization domain (NOD)-like receptor), the procaspase-1 precursor and the ASC (apoptosis-associated speck-like protein containing a caspase activation and recruitment domain) adaptor protein. These complexes recognize microbe- and the host-derived inflammatory signals, microbial-associated molecular patterns and damage-associated molecular patterns, and its activation leads to the maturation of inflammatory cytokines (such as interleukin-1β and interleukin-18). Inflammasomes participate in antimicrobial innate immune responses, but may also have a role in metabolic diseases, such as obesity, type 2 diabetes and atherosclerosis.
  • Metagenome is the total DNA that can be extracted from an environment. The human metagenome is the aggregate of the DNA of the host and the microbiota. Metagenome and microbiome are often used interchangeably.
  • Metagenomics is the study of the metagenome (microbiome). Metagenomics can either be targeted (usually 16S ribosomal RNA) or untargeted (shotgun sequencing).
  • Microbiota is the collective microbial community inhabiting a specific environment. Cellular density increases along the length of the gut, and the colonic microbiota is the densest and most diverse community in the gut, and in the whole human body.
  • Microbiome is the collective genomic content of a microbiota. It also indicates the total genetic capacity of the community.
  • Probiotics are defined as live microorganisms that, when administered in adequate amounts, confer a health benefit for the host84. Many bacterial strains in the Lactobacillus and Bifidobacterium genera are considered to be probiotic.
  • Prebiotics are non-digestible food ingredients that, when consumed in sufficient amounts, selectively stimulate the growth, activity or both of one or a limited number of microbial genera or species in the gut microbiota that confer(s) health benefits to the host85. Inulin and trans-galacto-oligosaccharides are defined as prebiotics because they are resistant to gastric digestion and hydrolysis by human enzymes; are fermented by specific members of the gut microbiota; and induce selective growth, activity or both of beneficial intestinal bacteria85. Both inulin and trans-galacto-oligosaccharides stimulate the growth of Bifidobacterium, an effect defined as bifidogenic.

Box 2: Dominant microbes

The human gut microbiota is dominated by five bacterial phyla (Firmicutes, Bacteroidetes, Actinobacteria, Proteobacteria and Verrucomicrobia) and one Archaea (Euryarchaeota). The less prevalent bacterial groups are distributed among Cyanobacteria, Fusobacteria, Lentisphaerae, Spirochaetes and TM7.

The Firmicutes phylum contains relevant genera, including Ruminococcus, Clostridium, Lactobacillus (several strains of which are probiotics), and the butyrate producers Eubacterium, Faecalibacterium and Roseburia.

In Bacteroidetes, Bacteroides, Prevotella and Xylanibacter degrade a variety of complex glycans.

The Actinobacteria phylum includes Collinsella and Bifidobacterium (which contains probiotic strains). Common Proteobacteria are Escherichia (from the Enterobacteriaceae family) and Desulfovibrio (which contains sulphate-reducing bacteria). Verrucomicrobia was recently discovered and includes Akkermansia (which are specialized for mucus degradation). Euryarchaeota contains the prevalent Methanobrevibacter (which is involved in the continuation of intestinal methanogenesis).

Germ-free mice are those born and reared without exposure to any live microbes, and they provide a powerful tool for understanding the effects of the gut microbiota on host physiology. These mice can be colonized either with selected microbial species or whole communities from mice or humans to examine the transmissibility of physiological and pathological phenotypes, and to test what role the microbiota has in a particular phenotype. The gut microbiota in these mice modulates bone-mass density3 and promotes fat storage4, intestinal angiogenesis5, 6 and the development of an immune response7, 8 (see page 231 of this issue). In this Review, we discuss the metagenomic and gnotobiotic-based evidence for the role of the gut microbiota in energy metabolism and the possible links with obesity.

Obesity

Gut microbiota composition is altered in people who are obese, and it can respond to changes in body weight. Genetically obese ob/ob mice9 are hyperphagic as a result of a mutation in the gene that encodes the satiety-promoting hormone leptin. The caecal microbiota of these mice contains more Firmicutes and fewer Bacteroidetes than that of their lean wild-type littermates, even when the mice are fed the same low-fat, polysaccharide-rich diet9. Similar changes have also been seen in the faecal microbiota of humans who are obese10. Bacteroidetes levels increase when weight is reduced, either by fat- or carbohydrate-restricted diets10, suggesting that Bacteroidetes may be responsive to calorie intake. A similar effect has also been observed in people who lost weight after a Roux-en-Y gastric bypass procedure. In these patients, increased levels of Bacteroides and Prevotella were negatively correlated with energy intake and adiposity11. Other studies showed no such shift in the Firmicutes–Bacteroidetes ratio12, 13, 14, but this may be because they used different clinical criteria (such as level of obesity, age, degree of weight loss and duration of calorie restriction), geographical locations, population sizes and microbiota-profiling methodologies. Although obesity and energy intake can affect the microbial composition, whether the gut microbiota contributes to obesity in humans is unclear.

A gastric bypass promotes sustained weight reduction and diminishes the risk of diabetes and cardiovascular disease for people who are obese15, 16. This knowledge has allowed the relationship between microbiota and obesity to be explored further. After a gastric bypass, diabetes can resolve before patients begin to lose weight, suggesting that this type of surgery has a direct antidiabetic effect. Exactly how this happens is not clear, but a shift in the composition of the faecal microbiota of humans11, 14 suggests the gut microbiota contributes to the improved metabolic phenotype after a gastric bypass. The beneficial microbe Faecalibacterium prausnitzii, in particular, is less abundant in patients who are obese and diabetic, but increases after surgery11. Levels of F. prausnitzii are negatively correlated with inflammatory markers, indicating that the bacterium may modulate systemic inflammation (common to diabetes and obesity) and contribute to the amelioration of diabetes. In addition, germ-free mice do not develop diet-induced obesity, and treatment of obese mice with antibiotics reduces adiposity and adipose inflammation, and improves glucose metabolism17, 18, 19, further supporting the benefits of inducing a shift in microbiota composition.

Energy harvest

Carbohydrates are important sources of energy for human and microbial cells. Human enzymes cannot degrade most complex carbohydrates and plant polysaccharides. Instead, the non-digestible carbohydrates, including cellulose, xylans, resistant starch and inulin, are fermented in the colon by its microbiota to yield energy for microbial growth and end products such as short-chain fatty acids (SCFAs) (Fig. 1), mainly acetate, propionate and butyrate, which have profound effects on gut health as, for example, an energy source, an inflammation modulator, a vasodilator and part of gut motility and wound healing. In addition, SCFAs are energy substrates for the colonic epithelium (butyrate) and peripheral tissues (acetate and propionate)20. The patterns of intestinal fermentation, and consequently the types and amount of SCFAs produced, are determined by how much carbohydrate is consumed and the composition of the gut microbiota. For example, fermentation of dietary fructans increases when gnotobiotic mice that have been colonized with Bacteroides thetaiotaomicron, are co-colonized with Methanobrevibacter smithii21. B. thetaiotaomicron produces more acetate and formate, and M. smithii uses formate for methanogenesis. The interactions promote more efficient carbohydrate fermentation and increased energy absorption from the gut, resulting in increased adiposity in the co-colonized mice compared with mice colonized with only B. thetaiotaomicron. The composition of the gut microbiota and the metabolic interactions between its species may therefore affect food digestion and energy harvest.

Figure 1: Effects of colonic fermentation of dietary fibres.
Effects of colonic fermentation of dietary fibres.

Complex carbohydrates, such as dietary fibre, are metabolized by the colonic microbiota to oligosaccharides and monosaccharides and then fermented to short-chain fatty acid end-products, mainly acetate, propionate and butyrate. Short-chain fatty acids are absorbed in the colon, where butyrate provides energy for colonic epithelial cells, and acetate and propionate reach the liver and peripheral organs, where they are substrates for gluconeogenesis and lipogenesis. In addition to being energy sources, short-chain fatty acids control colonic gene expression by inhibiting the enzyme histone deacetylase (HDAC) and metabolic regulation by signalling through G-protein-coupled receptors (GPCRs), such as GPR41 or GPR43.

Direct evidence for the role of the microbiota in energy harvest and fat deposition comes from germ-free rats, which have reduced intestinal levels of SCFAs22, and twice as much urinary and faecal excretion of calories as that of conventional rats fed the same polysaccharide-rich diet23. The germ-free rodents compensate for the reduced energy harvest by increasing their food intake23. Germ-free mice also have reduced adiposity compared with their conventional counterparts, but adiposity is normalized when they are colonized with a healthy microbiota for 14 days4, 19. Microbial energy harvest in obesity has been investigated in conventional genetically obese ob/ob mice, which have increased amounts of SCFAs in their caecum and reduced energy content in their faeces compared with their lean littermates24. Metagenomic sequencing of the caecal microbiota showed an enrichment of gene functions that were related to the degradation of dietary polysaccharides in the microbiome of ob/ob mice24. This finding was also true of humans: the faecal microbiota of people who are obese has an increased capacity to harvest energy2. In mice, the obese phenotype was transmissible through microbiota transplants, and germ-free mice colonized with the microbiota from obese donors gained twice as much fat as those colonized with the microbiota from lean donors24.

The role of the gut microbiota in promoting energy harvest from diet and fat deposition has been clearly demonstrated in mice, but most of the evidence in humans has come from indirect studies. For instance, people who are obese have higher levels of ethanol in their breath than lean people25, indicating altered fermentation and a greater number of faecal SCFAs13, which may suggest increased microbial energy harvest.

Diet alters the gut microbiota

Diet is known to modulate the composition of the gut microbiota in humans and mice. Long-term dietary habits have a considerable effect on the human gut microbiota. For example, children in a rural African village, who consumed high amounts of plant polysaccharides, had low levels of Firmicutes and increased levels of Bacteroidetes — mainly Prevotella and Xylanibacter — in their faecal microbiota compared with Italian children, who had high levels of Enterobacteriaceae — mainly Shigella and Escherichia26. Prevotella and Xylanibacter are known to degrade cellulose and xylans, and are associated with increased faecal SCFAs, suggesting that the gut microbiota of the children living in rural Africa had adapted to maximize energy extraction from a diet rich in fibre. Human gut microbiota can be divided into three discrete compositions. However, this concept is currently being challenged as enterotypes maybe more of a gradient than discrete entities. Each enterotype is dominated by a different genus — Bacteroides, Prevotella or Ruminococcus27 — but not affected by gender, age or nationality27. Enterotypes dominated by Bacteroides or Prevotella are associated with the consumption of a diet rich in protein and animal fat, or carbohydrates, respectively28. The Ruminococcus enterotype is not well separated and is partly merged with the Bacteroides enterotype28. This division supports the association between Prevotella and a diet high in carbohydrates, which was seen in children from rural Africa26. A 10-day dietary intervention, however, was not sufficient to alter the enterotype of an individual28, suggesting that a long-term change may be required to provoke a major shift in gut microbiota composition.

Changes in daily carbohydrate intake may affect specific groups of colonic bacteria over a short period of time. Consumption of the prebiotic inulin increases the levels of F. prausnitzii and Bifidobacterium sp. in humans29. Similarly, prebiotics promote a selective increase in Bifidobacterium sp. in diet-induced obese mice, and this increase is correlated with reduced adiposity and levels of microbe-derived inflammatory molecules, such as lipopolysaccharide, compared with mice that are fed a high-fat diet without prebiotics30. Human diets that are supplemented with resistant starch have increased faecal levels of Ruminococcus bromii and Eubacterium rectale, which correlates with fibre fermentation31. Consumption of resistant starch also improves insulin sensitivity32, but the variation in the microbial response to changes in resistant starch between individuals suggests successful dietary interventions need to be personalized31.

The gut microbiota also reacts to dietary fat. Mice fed on high-fat diets have reduced numbers of Bacteroidetes, and increased numbers of Firmicutes and Proteobacteria33, 34. This change is rapid, occurring within 24 hours35. Transplantation of the caecal microbiota from obese mice fed on high-fat diets into germ-free recipients increases adiposity significantly more than transplantation of a lean microbiota34. The altered microbial community of obese mice seems to have some role in promoting diet-induced obesity, but the mechanisms that cause this are unknown. A change in diet clearly alters the gut microbiota, and these alterations may contribute to the host's metabolic phenotype. Further metatranscriptomic and proteomic studies should provide insight into the response of microbial function as a result of a dietary shift.

Microbial processing of food constituents

Products of microbial metabolism act as signalling molecules and influence the host's metabolism. Microbial products directly affect intestinal function but may also affect the liver and brain, as well as adipose and muscle tissue, which consequently may affect the level of obesity and the associated comorbidities (Fig. 2). Microbial enzymatic activities can act directly on the fermentation of polysaccharides and bile-acid metabolism, or act in conjunction with the host on the metabolism of choline (Fig. 3).

Figure 2: Features of the gut microbiota that promote obesity and insulin resistance.
Features of the gut microbiota that promote obesity and insulin resistance.

Alterations to the composition and metabolic capacity of gut microbiota in obesity promote adiposity and influence metabolic processes in peripheral organs, such as the control of satiety in the brain; the release of hormones from the gut (shown as PYY and GLP-1); and the synthesis, storage or metabolism of lipids in the adipose tissue, liver and muscle. Microbial molecules also increase intestinal permeability, leading to systemic inflammation and insulin resistance.

Figure 3: Diet-independent and -dependent microbial effects on host metabolism.
Diet-independent and -dependent microbial effects on host metabolism.

The gut microbiota produces pro-inflammatory molecules, such as lipopolysaccharide and peptidoglycan, which may affect host metabolism through proteins produced by the host to mediate the immune response. Choline, cholesterol and polysaccharides obtained from the diet are metabolized by the gut microbiota and either directly or through further host–microbial co-metabolization generate bioactive compounds. In the case of choline, this can lead to cardiovascular disease; for cholesterol, activation of TGR5 can increase energy expenditure and GLP-1 secretion or protection against heart disease; and for polysaccharides, short-chain fatty acids can be used as an energy source or can bind to GPR41 or GPR43 to regulate hormones and modulate inflammation. FMO, flavin-containing monooxygenase; TLR4, Toll-like receptor 4; TMA, trimethylamine; TMAO, trimethylamine-N-oxide.

Fermentation of polysaccharides

Non-digestible carbohydrates are important sources of energy for several members of the colonic microbiota. Species such as B. thetaiotaomicron and Bacteroides ovatus contain more than twice the number of glycosidase and lyase genes than the human genome and are capable of using nearly all of the main plant and host glycans (such as mucus-associated glycoproteins)36, 37. Of the SCFAs produced from microbial fermentation, butyrate is particularly important as an energy substrate for cellular metabolism in the colonic epithelium. The colonic epithelial cells of germ-free mice are severely energy-deprived and are characterized by increased activation of AMP-activated protein kinase (AMPK), which senses cellular energy status38. This is also true of the liver of germ-free mice39. The liver metabolism of germ-free and colonized mice differs considerably, possibly because of the increased influx of SCFAs into the liver of colonized mice (Fig. 1). Acetate and propionate are taken up by the liver and used as substrates for lipogenesis and gluconeogenesis. Colonized mice have higher levels of stored triglycerides in the liver and an increase in the synthesis of very-low-density lipoproteins40, which transport triglycerides from the liver to other tissues. Increased triglyceride production in the liver of colonized mice is associated with reduced expression of fasting-induced adipose factor, or ANGPTL4, in the small intestine4, 39. ANGPTL4 is a potent inhibitor of the enzyme lipoprotein lipase, which mediates cellular uptake of triglycerides. Germ-free Angptl4-deficient mice gained as much fat mass and body weight during high-fat feeding as colonized mice, indicating that ANGPTL4 directly mediates microbial regulation of adiposity in mice4, 39.

SCFAs also affect proliferation, differentiation and modulation of gene expression in mammalian colonic epithelial cells41. However, these effects have been attributed to butyrate acting as a potent histone deacetylase inhibitor and, as such, it may regulate 2% of the mammalian transcriptome41. In addition, SCFAs can regulate gene expression by binding to the G-protein-coupled receptors (GPCRs) GPR41 (also known as FFAR3) and GPR43 (also known as FFAR2). Signalling through these receptors affects several different functions depending on the cellular type. For example, SCFAs suppress inflammation through GPR43 signalling in immune cells, such as neutrophils42, 43, and modulate secretion of the hormone GLP-1 — which improves insulin secretion and has antidiabetic effects — by enteroendocrine L-cells in the distal small intestine and colon44. In addition, the gut microbiota induces Pyy expression by L-cells through a GPR41-dependent mechanism. Conventional Gpr41-deficient mice have reduced adiposity compared with conventional wild-type mice, whereas germ-free wild-type and Gpr41-deficient mice had similar adiposity45, indicating that the effect of the microbiota on fat deposition is dependent on this SCFA receptor.

Microbial fermentation of polysaccharides may affect host adiposity through several complementary mechanisms, so modulation of the microbiota and its fermentation capacity may provide new avenues for managing obesity.

Microbial regulation of bile-acid metabolism

The primary bile acids cholic acid and chenodeoxycholic acid are synthesized in the human liver from cholesterol, and are important for ensuring that cholesterol, dietary fats and fat-soluble vitamins from the small intestine are soluble and absorbable. Primary bile acids are conjugated to taurine in mice and to glycine in humans, and are taken up in the distal ileum for transport to the liver. However, bacteria in this part of the ileum deconjugate these bile acids, which then escape intestinal uptake and can be further metabolized by the gut microbiota into secondary bile acids. Because the gut microbiota transforms bile acids, germ-free rodents have more bile acid and a less diverse profile than their conventionally raised counterparts46, 47, 48.

Bile acids also function as signalling molecules and bind to cellular receptors49, such as the bile-acid-synthesis controlling nuclear receptor farnesoid X receptor (FXR)50 and the GPCR TGR5. Both FXR and TGR5 have been implicated in the modulation of glucose metabolism in mice, but FXR impairs, whereas TGR5 promotes, glucose homeostasis51, 52, 53. In contrast to FXR, which is activated by primary bile acids, TGR5 binds secondary bile acids such as deoxycholic acid (formed from cholic acid) and lithocholic acid (formed from chenodeoxycholic acid). TGR5 signalling in enteroendocrine L-cells induces secretion of GLP-1, thereby improving liver and pancreatic function and enhancing glucose tolerance in obese mice53. Bile acids are taken up from the gut and circulated throughout the body, so activation of TGR5 and FXR in peripheral organs may contribute to overall host metabolism. Activation of TGR5 in brown adipose tissue and muscle increases energy expenditure and protects against diet-induced obesity54. The gut microbiota may therefore contribute to the level of obesity and type 2 diabetes by controlling lipid and glucose metabolism though the composition of bile-acid pools and the modulation of FXR and TGR5 signalling.

Microbial metabolism of choline

Choline is an important component of cell membranes and is mostly obtained from foods such as red meat and eggs, but may also be synthesized by the host55. Choline is also important for lipid metabolism and synthesis of very-low-density lipoprotein in the liver, and insufficient levels in the diet are associated with altered gut microbial ecology and liver steatosis in mice56 and humans57. In particular, low quantities of Gamma-proteobacteria and high levels of Erysipelotrichi in human faecal microbiota are associated with hepatic steatosis57. Microbial and host enzymatic activities interact in choline's transformation into toxic methylamines, so trimethylamine that is produced by intestinal microbes can be further metabolized to trimethylamine-N-oxide in the liver58, 59. These transformations may decrease the levels of bioavailable choline and are suggested to trigger non-alcoholic fatty liver disease (NAFLD) in mice58. An altered gut microbial composition and its capacity to metabolize choline may have an important role in modulating NAFLD as well as glucose homeostasis58.

Furthermore, plasma levels of trimethylamine-N-oxide and its metabolites are correlated with cardiovascular disease60(Fig. 3). The effect of microbial choline metabolism in cardiovascular disease is shown by the reduction of atherosclerosis in atherosclerosis-prone Apoe−/− mice treated with broad-spectrum antibiotics60. A gut microbiome with different capacities to process cholesterol and choline may contribute to the development of cardiovascular diseases.

Regulation of permeability and inflammation

Obesity, insulin resistance and development of type 2 diabetes are associated with systemic and adipose tissue inflammation61. The gut microbiota is a rich source of molecules such as lipopolysaccharide and peptidoglycan that may cause inflammation in peripheral tissues of the body. Colonization of germ-free mice with Escherichia coli is sufficient to augment macrophage infiltration of adipose tissue and polarize macrophages towards the expression of pro-inflammatory cytokines19. Plasma lipopolysaccharide levels increase in patients with type 2 diabetes62, and feeding lipopolysaccharide to mice for 4 weeks increase adipose tissue inflammation and reduce insulin sensitivity63. These findings suggest that the gut microbiota may affect host metabolism by altering adipose tissue inflammation. Higher numbers of T cells64, 65 and mast cells66, and lower numbers of regulatory T cells67 are also involved, but if and how the gut microbiota affects these cells and whether such interactions contribute to metabolic abnormalities is unclear.

Plasma lipopolysaccharide levels seem to rise with higher fat intake in mice63 and humans68, 69. Two hypotheses have been made to explain the mechanism: lipopolysaccharide is taken up with dietary fats in chylomicrons70, or lipopolysaccharide reaches the circulation because the gut is more permeable in obese mice63. A connection between metabolism and the function of the epithelial barrier is thought to exist. Targeted deletion of fatty acid synthase — encoded by the Fas gene — in the gut epithelium of mice showed that epithelial de novo lipogenesis is required to maintain barrier function71. Fas-deficient epithelium has increased permeability and, as a result, increased colonic levels of proinflammatory cytokines and high serum lipopolysaccharide. These phenotypes were corrected by antibiotic treatment, suggesting a reciprocal interaction between microbiota, altered epithelial permeability and host metabolism.

A similar connection between gut permeability and type 2 diabetes in humans could also be present. Permeability is correlated with increased visceral adiposity and hepatic steatosis72, and those with high visceral adiposity and type 2 diabetes have increased levels of bacterial DNA in their blood73. However, inflammation may increase permeability in the gut, and further investigation into whether increased permeability causes adipose inflammation or increased inflammation contributes to increased permeability is needed. Either way, the gut microbiota modulates permeability that may contribute to adipose inflammation and cause insulin resistance.

Lipopolysaccharide molecules bind to Toll-like receptor 4 (TLR4), and peptidoglycan to nucleotide-binding oligomerization domain (NOD) receptors, both of which activates proinflammatory signalling cascades63, 74, 75. Deletion of TLR4 in haematopoietic cells by generating bone-marrow chimaeras shows that TLR4 activation in macrophages of mice fed a high-fat diet is required for the development of fasting hyperinsulinaemia, and insulin resistance in liver and adipose tissue but not for the development of obesity76. The innate immune system, however, also modulates microbial composition, which may have autonomous effects on host metabolism. Mice deficient in TLR5 have an altered microbial ecology and exhibit metabolic-syndrome signs, such as obesity, insulin resistance and dyslipidaemia, which are, in part, associated with increased food consumption77. Transplantation of the gut microbiota from Tlr5-deficient and wild-type mice into germ-free recipients shows that the phenotypes are transmissible, and suggests that the gut microbiota alone can mediate disease.

Microbe-associated molecular patterns, including lipopolysaccharide and peptidoglycan, can be recognized by nucleotide-binding domain and leucine-rich-repeat-containing proteins (NLRPs), which form the inflammasome complex together with the apoptosis-associated speck-like protein containing a caspase activation and recruitment domain (ASC)78. Obesity is associated with increased adipose expression of NLRP3 in mice and ablation of NLRP3 enhances insulin signalling79. However, inflammasomes may be linked to gut microbiota and host metabolism (Fig. 4). NLRP3, NLRP6 and ASC are important regulators of microbial ecology in mice, and deletion of their genes increases the number of Bacteroidetes (Prevotellaceae) and TM756, 80. In particular, deficiency in Nlrp6 results in altered gut microbial ecology that predisposes mice to colitis80, and inflammasome complexes that do not contain NLRP3 or NLRP6 are associated with an altered gut microbiota and promote NAFLD and non-alcoholic steatohepatitis (NASH)56. Importantly, wild-type mice housed with disease-prone Asc−/− mice develop NAFLD or NASH, providing direct evidence that an altered gut microbiota may cause these diseases56. Alterations to gut microbiota composition are associated with an increased influx of TLR4 and TLR9 ligands — presumably lipopolysaccharide and bacterial DNA — respectively, to the liver through the portal vein56. Mice deficient in TLR signalling in the liver are therefore protected from developing conditions related to metabolic syndrome such as obesity, NAFLD and NASH56 (Fig. 4). The interaction between diet, host and gut microbiota may modulate gut permeability that leads to an influx of proinflammatory molecules and subsequent activation of inflammatory signalling pathways in peripheral tissues that may cause obesity, steatosis and insulin resistance.

Figure 4: Different microbial innate immune mechanisms affect host metabolism in the gut and liver.
Different microbial innate immune mechanisms affect host metabolism in the gut and liver.

Plasma lipopolysaccharide seems to rise with higher fat intake, and those with high visceral adiposity have higher levels of microbial DNA in their blood. Both lipopolysaccharide and microbial DNA seem to be connected with gut permeability. NLRP3 and 6 are both important regulators of microbial ecology through the effector protein interleukin-18 (IL-18). The altered gut microbiota can stimulate CCL5 secretion, which can result in increased permeability and influx of microbial components. In the liver, lipopolysaccharide and bacterial DNA activate the receptors, TLR4 and 9, leading to increased tumour-necrosis-factor-α (TNFα) secretion and development of non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH).

Future research

The gut microbiota is increasingly being accepted as an environmental factor that affects host metabolism and contributes to associated pathological conditions, such as obesity, diabetes and cardiovascular disease. However, the contribution that the gut microbiota makes to causing obesity and diabetes in humans is unclear. This is probably because the heterogeneous aetiology of obesity and diabetes can be associated with different microbes; studies are underpowered and include participants with diverse ethnic origin and food habits; the composition of the gut microbiota has large interpersonal variation; and different methods, with specific biases, have been used to profile the microbiota. Cheaper sequencing and improved bioinformatics tools for the analysis of the gut microbiota will allow more researchers to use metagenomic sequencing and avoid primer and polymerase chain reaction biases linked to 16S rRNA gene surveys. Although useful, metagenomic approaches should be complemented by metatranscriptomics and metaproteomics to assess which microbial genes and proteins are expressed in specific conditions. One of the main challenges is to obtain robust predictive biomarkers for obesity and diabetes on the basis of the gut microbiota, for which improved study designs and analytical methods are essential. Much of the focus has been on the faecal microbiota, but many metabolic functions also occur in the small intestine. Sampling intestinal specimens might contribute to the identification of microbial biomarkers for health and disease, and although challenging, more emphasis should be placed on examining its microbiota and the effects on host metabolism.

Studies in humans tend to be correlative, so the role of the microbiota in obesity and its comorbidities in humans remains to be proven. However, this role can be examined in animal studies. Germ-free mice can be 'humanized' by colonizing them with human intestinal communities, providing tools for examining the function of a specific human microbiota and testing how it interacts with specific diets. Genetically engineered germ-free mice could help to identify the molecular mechanisms by which the gut microbiota affects host metabolism. Pigs have similar gastrointestinal tracts and diets to humans, so they could be useful animal models in which to test dietary interventions and to manipulate the gut microbiota to improve health and prevent disease.

Accumulating evidence indicates that the gut microbiota may be a target for treating metabolic diseases81. Supplementing the diet with non-digestible food ingredients, or prebiotics, that stimulate the expansion of specific microbes to improve metabolic regulation can be a therapy. Probiotics may be an interesting approach for prevention of obesity and related diseases. But to determine the effects of both of these therapies, double-blind, placebo-controlled studies are required. Therapies that replace unhealthy with a healthy microbiota through transplantation have been used successfully since 1958 for the treatment of antibiotic-related diarrhoeal colitis82. Recently, transplantation of healthy lean microbiota improved insulin signalling in participants with metabolic syndrome83. Although a promising technique, transmission of unknown and potentially pathogenic bacteria and viruses from an unfractionated gut microbiota may have risks for the recipient. Using microbiota-based interventions to treat obesity will require probiotics that are selected for specific clinical manifestations of metabolic syndrome.

References

  1. Qin, J. et al. A human gut microbial gene catalogue established by metagenomic sequencing. Nature 464, 5965 (2010).
    This article is the first catalogue of the human metagenome.
  2. Turnbaugh, P. J. et al. A core gut microbiome in obese and lean twins. Nature 457, 480484 (2009).
    This paper reports that the diversity of the gut microbiota differs between lean and obese individuals.
  3. Sjogren, K. et al. The gut microbiota regulates bone mass in mice. J. Bone Miner. Res. 27, 13571367 (2012).
  4. Bäckhed, F. et al. The gut microbiota as an environmental factor that regulates fat storage. Proc. Natl Acad. Sci. USA 101, 1571815723 (2004).
    This article reports a link between the gut microbiota and adiposity.
  5. Reinhardt, C. et al. Tissue factor and PAR1 promote microbiota-induced intestinal vascular remodelling. Nature 483, 627631 (2012).
  6. Stappenbeck, T. S., Hooper, L. V. & Gordon, J. I. Developmental regulation of intestinal angiogenesis by indigenous microbes via Paneth cells. Proc. Natl Acad. Sci. USA 99, 1545115455 (2002).
  7. Larsson, E. et al. Analysis of gut microbial regulation of host gene expression along the length of the gut and regulation of gut microbial ecology through MyD88. Gut 61, 11241131 (2011).
  8. Gaboriau-Routhiau, V. et al. The key role of segmented filamentous bacteria in the coordinated maturation of gut helper T cell responses. Immunity 31, 677689 (2009).
  9. Ley, R. E. et al. Obesity alters gut microbial ecology. Proc. Natl Acad. Sci. USA 102, 1107011075 (2005).
  10. Ley, R. E., Turnbaugh, P. J., Klein, S. & Gordon, J. I. Microbial ecology: human gut microbes associated with obesity. Nature 444, 10221023 (2006).
  11. Furet, J. P. et al. Differential adaptation of human gut microbiota to bariatric surgery-induced weight loss: links with metabolic and low-grade inflammation markers. Diabetes 59, 30493057 (2010).
  12. Duncan, S. H. et al. Human colonic microbiota associated with diet, obesity and weight loss. Int. J. Obes. (Lond.) 32, 17201724 (2008).
  13. Schwiertz, A. et al. Microbiota and SCFA in lean and overweight healthy subjects. Obesity 18, 190195 (2010).
  14. Zhang, H. et al. Human gut microbiota in obesity and after gastric bypass. Proc. Natl Acad. Sci. USA 106, 23652370 (2009).
  15. Sjostrom, L. et al. Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N. Engl. J. Med. 351, 26832693 (2004).
  16. Sjostrom, L. et al. Effects of bariatric surgery on mortality in Swedish obese subjects. N. Engl. J. Med. 357, 741752 (2007).
  17. Cani, P. D. et al. Changes in gut microbiota control metabolic endotoxemia-induced inflammation in high-fat diet-induced obesity and diabetes in mice. Diabetes 57, 14701481 (2008).
  18. Membrez, M. et al. Gut microbiota modulation with norfloxacin and ampicillin enhances glucose tolerance in mice. FASEB J. 22, 24162426 (2008).
  19. Caesar, R. et al. Gut-derived lipopolysaccharide augments adipose macrophage accumulation but is not essential for impaired glucose or insulin tolerance in mice. Gut http://dx.doi.org/10.1136/gutjnl-2011-301689 (25 April, 2012).
  20. Bergman, E. N. Energy contributions of volatile fatty acids from the gastrointestinal tract in various species. Physiol. Rev. 70, 567590 (1990).
  21. Samuel, B. S. & Gordon, J. I. A humanized gnotobiotic mouse model of host–archaeal–bacterial mutualism. Proc. Natl Acad. Sci. USA 103, 1001110016 (2006).
  22. Hoverstad, T. & Midtvedt, T. Short-chain fatty acids in germfree mice and rats. J. Nutr. 116, 17721776 (1986).
  23. Wostmann, B. S., Larkin, C., Moriarty, A. & Bruckner-Kardoss, E. Dietary intake, energy metabolism, and excretory losses of adult male germfree Wistar rats. Lab. Anim. Sci. 33, 4650 (1983).
  24. Turnbaugh, P. J. et al. An obesity-associated gut microbiome with increased capacity for energy harvest. Nature 444, 10271031 (2006).
    This article demonstrates that the microbiota of those who are obese is enriched in genes for energy harvest and that obesity can be transmitted by transplantation of microbiota.
  25. Nair, S., Cope, K., Risby, T. H. & Diehl, A. M. Obesity and female gender increase breath ethanol concentration: potential implications for the pathogenesis of nonalcoholic steatohepatitis. Am. J. Gastroenterol. 96, 12001204 (2001).
  26. De Filippo, C. et al. Impact of diet in shaping gut microbiota revealed by a comparative study in children from Europe and rural Africa. Proc. Natl Acad. Sci. USA 107, 1469114696 (2010).
    This article demonstrates the affect of diet in shaping the gut microbiota.
  27. Arumugam, M. et al. Enterotypes of the human gut microbiome. Nature 473, 174180 (2011).
  28. Wu, G. D. et al. Linking long-term dietary patterns with gut microbial enterotypes. Science 334, 105108 (2011).
  29. Ramirez-Farias, C. et al. Effect of inulin on the human gut microbiota: stimulation of Bifidobacterium adolescentis and Faecalibacterium prausnitzii. Br. J. Nutr. 101, 541550 (2009).
  30. Cani, P. D. et al. Selective increases of bifidobacteria in gut microflora improve high-fat-diet-induced diabetes in mice through a mechanism associated with endotoxaemia. Diabetologia 50, 23742383 (2007).
  31. Walker, A. W. et al. Dominant and diet-responsive groups of bacteria within the human colonic microbiota. ISME J. 5, 220230 (2010).
  32. Robertson, M. D., Bickerton, A. S., Dennis, A. L., Vidal, H. & Frayn, K. N. Insulin-sensitizing effects of dietary resistant starch and effects on skeletal muscle and adipose tissue metabolism. Am. J. Clin. Nutr. 82, 559567 (2005).
  33. Hildebrandt, M. A. et al. High-fat diet determines the composition of the murine gut microbiome independently of obesity. Gastroenterology 137, 17161724 (2009).
  34. Turnbaugh, P. J., Bäckhed, F., Fulton, L. & Gordon, J. I. Diet-induced obesity is linked to marked but reversible alterations in the mouse distal gut microbiome. Cell Host Microbe 3, 213223 (2008).
  35. Turnbaugh, P. J. et al. The effect of diet on the human gut microbiome: a metagenomic analysis in humanized gnotobiotic mice. Sci. Transl. Med. 1, 6ra14 (2009).
  36. Xu, J. et al. A genomic view of the human–Bacteroides thetaiotaomicron symbiosis. Science 299, 20742076 (2003).
  37. Martens, E. C. et al. Recognition and degradation of plant cell wall polysaccharides by two human gut symbionts. PLoS Biol. 9, e1001221 (2011).
  38. Donohoe, D. R. et al. The microbiome and butyrate regulate energy metabolism and autophagy in the mammalian colon. Cell Metab. 13, 517526 (2011).
  39. Bäckhed, F., Manchester, J. K., Semenkovich, C. F. & Gordon, J. I. Mechanisms underlying the resistance to diet-induced obesity in germ-free mice. Proc. Natl Acad. Sci. USA 104, 979984 (2007).
  40. Velagapudi, V. R. et al. The gut microbiota modulates host energy and lipid metabolism in mice. J. Lipid Res. 51, 11011112 (2010).
  41. Davie, J. R. Inhibition of histone deacetylase activity by butyrate. J. Nutr. 133, 2485S2493S (2003).
  42. Maslowski, K. M. et al. Regulation of inflammatory responses by gut microbiota and chemoattractant receptor GPR43. Nature 461, 12821286 (2009).
  43. Sina, C. et al. G protein-coupled receptor 43 is essential for neutrophil recruitment during intestinal inflammation. J. Immunol. 183, 75147522 (2009).
  44. Tolhurst, G. et al. Short-chain fatty acids stimulate glucagon-like peptide-1 secretion via the G-protein-coupled receptor FFAR2. Diabetes 61, 364371 (2012).
  45. Samuel, B. S. et al. Effects of the gut microbiota on host adiposity are modulated by the short-chain fatty-acid binding G protein-coupled receptor, Gpr41. Proc. Natl Acad. Sci. USA 105, 1676716772 (2008).
  46. Wostmann, B. S. Intestinal bile acids and cholesterol absorption in the germfree rat. J. Nutr. 103, 982990 (1973).
  47. Swann, J. R. et al. Systemic gut microbial modulation of bile acid metabolism in host tissue compartments. Proc. Natl Acad. Sci. USA 108, 45234530 (2010).
  48. Gustafsson, B. E., Bergstrom, S., Lindstedt, S. & Norman, A. Turnover and nature of fecal bile acids in germfree and infected rats fed cholic acid-24–14C; bile acids and steroids 41. Proc. Soc. Exp. Biol. Med. 94, 467471 (1957).
  49. Hylemon, P. B. et al. Bile acids as regulatory molecules. J. Lipid Res. 50, 15091520 (2009).
  50. Sinal, C. J. et al. Targeted disruption of the nuclear receptor FXR/BAR impairs bile acid and lipid homeostasis. Cell 102, 731744 (2000).
  51. van Dijk, T. H. et al. An increased flux through the glucose 6-phosphate pool in enterocytes delays glucose absorption in Fxr−/− mice. J. Biol. Chem. 284, 1031510323 (2009).
  52. Prawitt, J. et al. Farnesoid X receptor deficiency improves glucose homeostasis in mouse models of obesity. Diabetes 60, 18611871 (2011).
  53. Thomas, C. et al. TGR5-mediated bile acid sensing controls glucose homeostasis. Cell Metab. 10, 167177 (2009).
  54. Watanabe, M. et al. Bile acids induce energy expenditure by promoting intracellular thyroid hormone activation. Nature 439, 484489 (2006).
  55. Vance, D. E. Role of phosphatidylcholine biosynthesis in the regulation of lipoprotein homeostasis. Curr. Opin. Lipidol. 19, 229234 (2008).
  56. Henao-Mejia, J. et al. Inflammasome-mediated dysbiosis regulates progression of NAFLD and obesity. Nature 482, 179185 (2012).
    This paper elegantly connects microbial recognition by inflammasomes in the gut with altered permeability and liver steatosis.
  57. Spencer, M. D. et al. Association between composition of the human gastrointestinal microbiome and development of fatty liver with choline deficiency. Gastroenterology 140, 976986 (2011).
  58. Dumas, M. E. et al. Metabolic profiling reveals a contribution of gut microbiota to fatty liver phenotype in insulin-resistant mice. Proc. Natl Acad. Sci. USA 103, 1251112516 (2006).
  59. Prentiss, P. G. et al. The metabolism of choline by the germfree rat. Arch. Biochem. Biophys. 94, 424429 (1961).
  60. Wang, Z. et al. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature 472, 5763 (2011).
  61. Osborn, O. & Olefsky, J. M. The cellular and signaling networks linking the immune system and metabolism in disease. Nature Med. 18, 363374 (2012).
  62. Creely, S. J. et al. Lipopolysaccharide activates an innate immune system response in human adipose tissue in obesity and type 2 diabetes. Am. J. Physiol. Endocrinol. Metab. 292, E740E747 (2007).
  63. Cani, P. D. et al. Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes 56, 17611772 (2007).
    This paper reports that endotoxin is sufficient to alter metabolic inflammation and insulin sensitivity.
  64. Nishimura, S. et al. CD8+ effector T cells contribute to macrophage recruitment and adipose tissue inflammation in obesity. Nature Med. 15, 914920 (2009).
  65. Winer, S. et al. Normalization of obesity-associated insulin resistance through immunotherapy. Nature Med. 15, 921929 (2009).
  66. Liu, J. et al. Genetic deficiency and pharmacological stabilization of mast cells reduce diet-induced obesity and diabetes in mice. Nature Med. 15, 940945 (2009).
  67. Feuerer, M. et al. Lean, but not obese, fat is enriched for a unique population of regulatory T cells that affect metabolic parameters. Nature Med. 15, 930939 (2009).
  68. Amar, J. et al. Energy intake is associated with endotoxemia in apparently healthy men. Am. J. Clin. Nutr. 87, 12191223 (2008).
  69. Erridge, C., Attina, T., Spickett, C. M. & Webb, D. J. A high-fat meal induces low-grade endotoxemia: evidence of a novel mechanism of postprandial inflammation. Am. J. Clin. Nutr. 86, 12861292 (2007).
  70. Ghoshal, S., Witta, J., Zhong, J., de Villiers, W. & Eckhardt, E. Chylomicrons promote intestinal absorption of lipopolysaccharides. J. Lipid Res. 50, 9097 (2009).
  71. Wei, X. et al. Fatty acid synthase modulates intestinal barrier function through palmitoylation of mucin 2. Cell Host Microbe 11, 140152 (2012).
  72. Gummesson, A. et al. Intestinal permeability is associated with visceral adiposity in healthy women. Obesity 19, 22802282 (2011).
  73. Amar, J. et al. Involvement of tissue bacteria in the onset of diabetes in humans: evidence for a concept. Diabetologia 54, 30553061 (2011).
  74. Amar, J. et al. Intestinal mucosal adherence and translocation of commensal bacteria at the early onset of type 2 diabetes: molecular mechanisms and probiotic treatment. EMBO Mol. Med. 3, 559572 (2011).
  75. Schertzer, J. D. et al. NOD1 activators link innate immunity to insulin resistance. Diabetes 60, 22062215 (2011).
  76. Saberi, M. et al. Hematopoietic cell-specific deletion of Toll-like receptor 4 ameliorates hepatic and adipose tissue insulin resistance in high-fat-fed mice. Cell Metab. 10, 419429 (2009).
  77. Vijay-Kumar, M. et al. Metabolic syndrome and altered gut microbiota in mice lacking Toll-like receptor 5. Science 328, 228231 (2010).
  78. Strowig, T., Henao-Mejia, J., Elinav, E. & Flavell, R. Inflammasomes in health and disease. Nature 481, 278286 (2012).
  79. Vandanmagsar, B. et al. The NLRP3 inflammasome instigates obesity-induced inflammation and insulin resistance. Nature Med. 17, 179188 (2011).
  80. Elinav, E. et al. NLRP6 inflammasome regulates colonic microbial ecology and risk for colitis. Cell 145, 745757 (2011).
  81. Murphy, E. F. et al. Divergent metabolic outcomes arising from targeted manipulation of the gut microbiota in diet-induced obesity. Gut http://dx.doi.org/10.1136/gutjnl-2011-300705 (16 February, 2012).
  82. Eiseman, B., Silen, W., Bascom, G. S. & Kauvar, A. J. Fecal enema as an adjunct in the treatment of pseudomembranous enterocolitis. Surgery 44, 854859 (1958).
  83. Vrieze, A. et al. Transfer of intestinal microbiota from lean donors increases insulin sensitivity in subjects with metabolic syndrome. Gastroenterology http://dx.doi.org/10.1053/j.gastro.2012.06.031 (20 June, 2012).
    This article reports that insulin resistance in obese humans can be improved by transplantation of a lean microbiota.
  84. Food and Agriculture Organization of the United Nations and World Health Organization. Working Group Report on Drafting Guidelines for the evaluation of Probiotics in Food. ftp://ftp.fao.org/es/esn/food/wgreport2.pdf (FAO and WHO, 2002).
  85. Roberfroid, M. et al. Prebiotic effects: metabolic and health benefits. Br. J. Nutr. 104, S1S63 (2010).

Download references

Acknowledgements

The authors are grateful to A. Hallén for contribution to the artwork and to R. Perkins for reading the manuscript. Work in the authors' laboratory is funded by the Swedish Research Council, the Swedish Foundation for Strategic Research, Torsten Söderberg's Foundation, Ragnar Söderberg's foundation, AFA Insurances, the Knut and Alice Wallenberg foundation, the Swedish heart lung foundation, the NovoNordisk foundation and the Swedish diabetes foundation.

Author information

Affiliations

  1. Wallenberg Laboratory for Cardiovascular and Metabolic Research, Sahlgrenska University Hospital, 413 45 Gothenburg, Sweden.

    • Valentina Tremaroli &
    • Fredrik Bäckhed
  2. Department of Molecular and Clinical Medicine, University of Gothenburg, 413 45 Gothenburg, Sweden.

    • Valentina Tremaroli &
    • Fredrik Bäckhed
  3. Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health Sciences, University of Copenhagen, Copenhagen DK-2200, Denmark.

    • Fredrik Bäckhed

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Reprints and permissions information is available at www.nature.com/reprints.

Author details

Additional data