chapter 11. The gut microbiota and obesity

Hervé M. Blottière CHAPTER 11 CHAPTER The human microbiota is com- portance outside the gut, especially a definition [3]. In a healthy symbi- posed of about as many microorgan- after the pioneering work of Gordon otic state, the colonic microbiota isms as there are cells in the human and collaborators [1]. is an important organ, interacting body. It is a very diverse ecosystem Recently, the development of with food (in particular dietary fibre, comprising more than 100 trillion mi- molecular tools and subsequently of enabling energy harvest from oth- crobes living in the intestines, the next-generation sequencing enabled erwise indigestible dietary com- mouth, the skin, the vagina, and the richness of the intestinal ecosys- pounds), interacting with cells (in- elsewhere in the body. Although it tem to be revealed [2]. Each individ- cluding immune cells, but also the was previously called the gastroin- ual harbours hundreds of different metabolic and nervous systems), testinal flora or microflora, the more species, most of which have not and protecting against pathogens pragmatic term “microbiota” is now yet been cultured. Studies have re- by acting as a barrier to infection preferred. vealed that 70–80% of the dominant (Fig. 11.1). The microbiome, the “other ge- species have no representative in nome” or “second genome” of the culture collections. Only a few doz- Gene catalogues of gut human body, is composed of about en species are conserved between microbiota 10 million genes, compared with about individuals, representing a core that 23 000 genes in the human genome, seems to be a stable community un- The first draft of the human genome and thus provides a very rich function- der healthy conditions. Although this was published in 2000. In 2010, the al potential. The colonic microbiome view is controversial, some people Metagenomics of the Human Intes- is the most diverse and also the best consider the gut microbiota to be tinal Tract (MetaHIT) consortium re- characterized microbial community. a true organ; as such, it could be leased the first catalogue of human Although the human microbiome transplanted. The recent success gut microbial genes, obtained after has fantastic potential, it has only of faecal microbiota transplantation, sequencing whole faecal microbiota been about 10 years since the sci- especially in the context of Clostridi- metagenomes from 124 European indi- entific community first realized its im- um difficile infection, argues for such viduals [4]. Interestingly, the 3.3 million

Chapter 11. The gut microbiota and obesity 89 Fig. 11.1. The gut microbiota. The gut microbiota

• An average of 650 000 genes per microbiome Human physiology • About 25–30 times as many genes as the human genome About 500–1000 dominant species per • Intestinal Immun e individual Nutrition microbiota defence s • A true organ

Barrier agains t pathogen s

gut bacterial genes in the MetaHIT ation when attempting to extrapolate [10], although Proteobacteria, Ver- catalogue were also well represent- results obtained in mouse models to rucomicrobia, and Fusobacteria are ed in the other metagenomes that the situation in humans. present to a lesser extent. About were available at the time, from 50% of individuals harbour Archaea faecal samples of individuals in the Colonization in their microbiota, especially Meth- USA and Japan. In parallel, the Hu- anobrevibacter smithii, which is re- man Microbiome Project published a The colonization process starts at sponsible for methane excretion. A catalogue of 178 reference bacterial birth, and the delivery type is the first core of species has been identified genomes distributed among different factor that has an impact. For infants as being present in most individuals, body sites and including 151 repre- that are vaginally delivered, the initial but with different relative abundanc- sentative gastrointestinal species gut microbiota resembles the moth- es. The number of species identi- [5]. er’s vaginal microbiota, dominated fied in the core depends on the an- In 2014, the MetaHIT consortium by bacteria of the genera Lacto- alytical method used: 66 from 16S published an integrated catalogue bacillus, Prevotella, and Sneathia, rDNA sequencing [11] or 57 from of 10 million bacterial genes de- whereas for infants delivered by whole-metagenome sequencing [5]. rived from 1267 human gut metage- caesarean section, the initial gut mi- Under healthy conditions, the in- nomes obtained from individuals crobiota resembles the mother’s skin testinal microbiota is considered to on three continents, including 760 microbial community, composed of be a stable community, influenced samples from Europe. As expect- Staphylococcus, Corynebacterium, by dietary habits as well as by the ed, the number of frequent genes and Propionibacterium [8]. Coloniza- physiology of its host. stopped increasing, whereas the tion is also strongly affected by the number of rare genes, present in administration of antibiotics in early Enterotypes not more than 1% of the cohort, life [9]. During the first 3 years of life, continued to increase [6]. Analyses the infant’s gut microbiota is highly Further analysis of several metage- of this close-to-complete catalogue unstable and is largely influenced nomes led to the discovery of three revealed country-specific signatures by feeding habits. Key factors are balanced ecological arrangements, for xenobiotic metabolism and nutri- the type of feeding (breastfeeding termed enterotypes; the three en- ent consumption for samples from or formula feeding), the weaning terotypes are dominated by Bac- individuals in China and Denmark. time and process, and food compo- teroides, Prevotella, and Rumino- More recently, a catalogue of sition, as well as the hygiene of the coccus, respectively [12]. The third the mouse gut metagenome was environment. enterotype is also linked to the pres- established, emphasizing the host By the time an individual reaches ence of M. smithii. This description specificity of the microbiota [7]. Only adulthood, the intestinal microbiota of community types is not limited to about 4.0% of the mouse gut micro- is composed of several hundreds of the gut [13]. These enterotypes or bial genes were shared with those different species, belonging only to a community types emerged as being of the human gut microbiome. It is few phyla, predominantly Firmicutes, independent of sex and country of important to take this into consider- Bacteroidetes, and Actinobacteria origin but probably associated with

90 long-term dietary habits [14]. Wu et to be essential for the maintenance of 60% increase in body fat mass, ac- al. [14] were able to associate con- a healthy state, and several reports companied by increased leptin and sumption of protein and animal fat have shown that a state of dysbiosis insulin levels and linked to increased with the Bacteroides enterotype and is often associated with diseases, in- absorption of monosaccharides from consumption of carbohydrates with cluding inflammatory bowel disease, the gut lumen, with resulting induc- the Prevotella enterotype. Interest- allergies, colorectal cancer, and liver tion of hepatic de novo lipogenesis ingly, by analysing samples from diseases, as well as obesity, diabe- [16]. A comparison of the microbiota volunteers randomized to a high-fat, tes, and cardiovascular diseases [2]. of lean and obese mice revealed that low-fibre diet or a low-fat, high-fibre Dysbiosis may be defined as an im- in obese mice (ob/ob animals), the diet for 10 days, this study revealed balanced microbiota, including four relative abundance of Bacteroidetes rapid changes in microbiome com- types of imbalance: (i) loss of key- was lower and that of Firmicutes was position; however, the enterotype stone species, (ii) reduced richness higher [17]. Moreover, transplanting of an individual did not seem to be or diversity, (iii) increased pathogens microbiota from obese animal to affected by this relatively short-term or pathobionts, or (iv) modification germ-free mice resulted in a greater dietary intervention. Transit time of or shift in metabolic capacities [9] increase in total body fat compared food through the gut has also been (Fig. 11.2). with transplanting microbiota from correlated with enterotypes [15]. lean animals, highlighting the con- The link with obesity tributory role of microbiota to obesity Dysbiosis [18]. In a study comparing the mi- The first link between gut microbio- crobiota from a dozen obese people The human gut microbiota is very ta and obesity came from studies in with that of a few lean controls, the complex and diversified. The micro- germ-free rodents. These animals authors reported that the decreased biome of an individual has more than eat more, move less, develop less proportion of Bacteroidetes and the 25 times as many genes as there are fat content, and are resistant to di- increased proportion of Firmicutes in the human genome. The fitness of et-induced obesity. Conventionaliza- observed in obese mice were also this well-balanced symbiosis seems tion of germ-free mice resulted in a observed in obese people [19]. They

Fig. 11.2. Intestinal microbiota dysbiosis in obesity and physiological perturbation. AngPTL4, angiopoietin-like 4; 11 CHAPTER BA, bile acids; FA, fatty acids; GLP-1, glucagon-like peptide 1; LPL, lipoprotein lipase; LPS, lipopolysaccharide; PYY, peptide YY; SCFA, short-chainInt efattyst iacids;nal TG,mi ctriglycerides;robiot aTMA, dy trimethylamine;sbiosis TMAO, trimethylamine N-oxide.

Dysbiosis: • Loss of keystone species • Loss of richness • Increased pathobionts • Metabolic shift in the ecosystem Brain Reduced satiety Liver SCFA, Increased gluconeogenesis TMA, Increase TMAO production Increased inflammation BA, Reduced PYY, GLP-1 LPS, Increased permeability etc. Increased endotoxaemia Adipose tissue Increased metabolite Increased LPL activity absorption Increase TG incorporation Reduced AngPTL4 Increased inflammation

Muscle Reduced FA oxidation Immune system Recruitment of cells Cell activation Cytokine secretion

Chapter 11. The gut microbiota and obesity 91 also reported that obese people los- count and those with high bacterial insulin resistance are diverse. They ing weight on a low-calorie diet had a richness [21]. Among the species that are often derived from mouse mod- more balanced microbiota, with an in- are more prevalent in individuals with els and still require complete valida- creased proportion of Bacteroidetes high bacterial richness, the analysis tion in humans. Dysbiosis is linked and a decreased proportion of Firmi- highlighted two species: Faecalibac- to increased energy harvest from cutes, more similar to the microbiota terium prausnitzii, a bacterium that food, altered fermentation of fibres, of lean controls. was previously described as lacking and increased endotoxaemia. These After this pioneering work, other in patients with inflammatory bowel changes in microbiota functions researchers developed approaches disease and that has anti-inflamma- have an impact on different tissues, to better understand the mechanisms tory properties [23], and Akkermansia including the intestine, muscles, adi- by which the microbiota can contrib- muciniphila, a bacterium that was pose tissues, the liver, and the brain ute to metabolic syndrome and obesi- found to be associated with body fat [26]. ty [20]. Large cohorts of patients were mass and glucose intolerance in mice In the intestine, the changes re- studied. and that was further confirmed to be sult in increased permeability of the The MetaHIT consortium investi- linked with a healthier metabolic phe- epithelium, allowing translocation of gated the composition of the human notype and better clinical outcomes bacteria as well as bacterial prod- gut microbiota in a population sample after a hyper-low-calorie diet in over- ucts, such as lipopolysaccharides. of 123 non-obese and 169 obese in- weight or obese adults [24]. Among Moreover, secretion by enteroen- dividuals from a Danish cohort study the species that are more prevalent in docrine cells of hormones, including called Inter99 [21]. A quantitative individuals with low bacterial richness peptide YY (PYY), glucagon-like pep- metagenomic pipeline was applied, are Bacteroides strains and Rumino- tide 1 (GLP-1), and neurotensin, is and the study found two groups of coccus gnavus, which are considered impaired, with effects on the brain, individuals that differed by the num- to be pro-inflammatory and are often resulting in reduced satiety, as well ber of genes in their metagenome, found in patients with inflammatory as on the liver and on gut motility. and thus the gut bacterial richness. bowel disease. The short-chain fatty acids ace- About a quarter of the population Such a phylogenetic shift has also tate and propionate are taken up had low bacterial richness. Individu- been confirmed at the functional lev- by hepatocytes and serve as sub- als with a low gene count had higher el. Low bacterial richness is associ- strates for lipogenesis and gluco- adiposity, reduced insulin sensitivi- ated with a reduction in butyrate-pro- neogenesis. Thus, increased tri- ty, higher dyslipidaemia, and higher ducing bacteria, reduced production glyceride production by the liver, inflammatory status compared with of hydrogen and methane, increased associated with reduced expression those with a high gene count. The sulfate reduction and mucin degrada- of angiopoietin-like 4 (AngPTL4), an obese individuals in the group with a tion, increased endotoxaemia, and a inhibitor of lipoprotein lipase, by the low gene count gained more weight higher capacity to manage exposure small intestine, leads to increased during the 10 years of follow-up to oxygen/oxidative stress [21]. triglyceride incorporation in adipose before stool sampling [21]. Dietary habits seem to be associ- tissues [26]. Increased inflammation Similar observations were made ated with microbiota richness [25]. A is also observed in different tissues, in a cohort of obese individuals in dietary pattern with high consumption including gut, liver, and adipose tis- France who were recruited to follow a of potatoes, confectionery, and sug- sues. A reduction of fatty acid oxi- hyper-low-calorie diet with increased ary drinks and low intake of fruits and dation by muscles is also observed. intake of protein and fibre [22]. Al- yogurt was correlated with low mi- Finally, the metabolism of bile though the microbial gene richness of crobiota richness, whereas a dietary acids and choline is affected. Pertur- the participants increased by 25% af- pattern with low consumption of con- bation of choline metabolism results ter the 6-week diet, the obese individ- fectionery and sugary drinks and high in increased production by intestinal uals with low bacterial richness bene- intake of fruits, vegetables, soups, microbes of trimethylamine, which fited the least from the diet, whereas and yogurt was correlated with higher is further metabolized by hepato- those with higher bacterial richness at microbiota richness. cytes to trimethylamine N-oxide, a the start of the diet lost more weight compound that is associated with and had a larger improvement in met- Mechanisms liver and cardiovascular diseases abolic status. [27]. Primary bile acids are trans- Interestingly, only a few bacterial The proposed mechanisms by which formed by the intestinal microbiota species are sufficient to distinguish gut microbiota dysbiosis and loss of to secondary bile acids, which are between individuals with a low gene richness can promote obesity and potent signalling molecules through

92 the activation of FXR, a nuclear re- is linked to more severe metabol- cific nutrition, prebiotics, and probi- ceptor, and TGR5, a G protein-cou- ic syndrome and lower sensitivity otics may be efficient avenues for pled receptor; these receptors are to weight loss after caloric restric- the prevention of obesity. The recent expressed in intestinal enteroendo- tion. The role of the gut microbiota success of a diet rich in non-digest- crine cells, resulting in the modifica- in the development and chronicity ible carbohydrates in children with tion of glucose homeostasis [26]. of obesity still needs to be clarified, Prader–Willi syndrome, resulting in and the mechanisms of action in weight loss and reduction of inflam- Conclusions humans remain to be deciphered. mation as well as structural changes Strategies to transiently modulate of the intestinal microbiota, highlights Dysbiosis in intestinal microbiota the human intestinal microbiota and the feasibility of dietary modulation has been associated with obesity. to potentially increase its richness of the gut microbiome to manage A loss of bacterial gene richness need to be explored [22, 25]. Spe- metabolic diseases [28].

Key points • The human microbiota is a dense and diverse microbiome. • It includes 100 trillion microorganisms, as many as the number of cells in the human body. • Each individual harbours hundreds of different species, most of which (70–80% of the dominant species) have not yet been cultured. • A few dozen species are conserved between individuals (a core), representing a stable community. • The gut microbiota is a true organ, protecting health and well-being throughout all life stages. • The colonic microbiota is a key organ, interacting with food (fermentation), interacting with cells (the immune and nervous systems), and protecting against pathogens (barrier function). • Dysbiosis has been observed in several chronic diseases. • Dysbiosis is observed in obesity, and a loss of microbiota richness and diversity is associated with inflammatory status. 11 CHAPTER

Research needs • Standardization of analysis tools and processes is required. • Longitudinal studies are needed. • The impact of medication/drugs should be considered. • Mechanisms of action remain to be deciphered. • Holistic studies should be designed, associating excellent phenotyping of patients and deep characterization using metabolomics, immunomics, transcriptomics, and metagenomics. • An ecological understanding of the intestinal ecosystem is needed.

Chapter 11. The gut microbiota and obesity 93 References

1. Bäckhed F, Ley RE, Sonnenburg JL, Pe- 11. Tap J, Mondot S, Levenez F, Pelletier E, 21. Le Chatelier E, Nielsen T, Qin J, Prifti E, terson DA, Gordon JI (2005). Host-bacterial Caron C, Furet J-P, et al. (2009). Towards Hildebrand F, Falony G, et al.; MetaHIT con- mutualism in the human intestine. Science. the human intestinal microbiota phylogenetic sortium (2013). Richness of human gut microbi- 307(5717):1915–20. http://dx.doi.org/10.1126/ core. Environ Microbiol. 11(10):2574–84. http:// ome correlates with metabolic markers. Nature. science.1104816 PMID:15790844 dx.doi.org/10.1111/j.1462-2920.2009.01982.x 500(7464):541–6. http://dx.doi.org/10.1038/ PMID:19601958 nature12506 PMID:23985870 2. Blottière HM, de Vos WM, Ehrlich SD, Doré J (2013). Human intestinal metagenomics: state of 12. Arumugam M, Raes J, Pelletier E, Le Pasli- 22. Cotillard A, Kennedy SP, Kong LC, Prifti E, the art and future. Curr Opin Microbiol. 16(3):232–9. er D, Yamada T, Mende DR, et al.; MetaHIT Pons N, Le Chatelier E, et al.; ANR MicroObes http://dx.doi.org/10.1016/j.mib.2013.06.006 Consortium (2011). Enterotypes of the human consortium (2013). Dietary intervention im- PMID:23870802 gut microbiome. Nature. 473(7346):174– pact on gut microbial gene richness. Nature. 80. http://dx.doi.org/10.1038/nature09944 500(7464):585–8. http://dx.doi.org/10.1038/ 3. van Nood E, Vrieze A, Nieuwdorp M, Fuentes PMID:21508958 nature12480 PMID:23985875 S, Zoetendal EG, de Vos WM, et al. (2013). Duodenal infusion of donor feces for recurrent 13. Ding T, Schloss PD (2014). Dynam- 23. Sokol H, Pigneur B, Watterlot L, Lakhdari Clostridium difficile. N Engl J Med. 368(5):407– ics and associations of microbial commu- O, Bermúdez-Humarán LG, Gratadoux J-J, 15. http://dx.doi.org/10.1056/NEJMoa1205037 nity types across the human body. Nature. et al. (2008). Faecalibacterium prausnitzii is PMID:23323867 509(7500):357–60. http://dx.doi.org/10.1038/ an anti-inflammatory commensal bacterium nature13178 PMID:24739969 identified by gut microbiota analysis of Crohn 4. Qin J, Li R, Raes J, Arumugam M, Burgdorf disease patients. Proc Natl Acad Sci U S A. KS, Manichanh C, et al.; MetaHIT Consortium 14. Wu GD, Chen J, Hoffmann C, Bittinger K, 105(43):16731–6. http://dx.doi.org/10.1073/ (2010). A human gut microbial gene catalogue Chen Y-Y, Keilbaugh SA, et al. (2011). Linking pnas.0804812105 PMID:18936492 established by metagenomic sequencing. Na- long-term dietary patterns with gut microbi- ture. 464(7285):59–65. http://dx.doi.org/10.1038/ al enterotypes. Science. 334(6052):105–8. 24. Dao MC, Everard A, Aron-Wisnewsky J, nature08821 PMID:20203603 http://dx.doi.org/10.1126/science.1208344 Sokolovska N, Prifti E, Verger EO, et al.; PMID:21885731 MICRO-Obes Consortium (2016). Akkermansia 5. Nelson KE, Weinstock GM, Highlander SK, muciniphila and improved metabolic health Worley KC, Creasy HH, Wortman JR, et al.; Hu- 15. Vandeputte D, Falony G, Vieira-Silva S, during a dietary intervention in obesity: man Microbiome Jumpstart Reference Strains Tito RY, Joossens M, Raes J (2016). Stool relationship with gut microbiome richness and Consortium (2010). A catalog of reference ge- consistency is strongly associated with gut mi- ecology. Gut. 65(3):426–36. PMID:26100928 nomes from the human microbiome. Science. crobiota richness and composition, enterotypes 328(5981):994–9. http://dx.doi.org/10.1126/ and bacterial growth rates. Gut. 65(1):57–62 25. Kong LC, Holmes BA, Cotillard A, Habi- science.1183605 PMID:20489017 PMID:26069274 Rachedi F, Brazeilles R, Gougis S, et al. (2014). Dietary patterns differently associate 6. Li J, Jia H, Cai X, Zhong H, Feng Q, Sunagawa 16. Bäckhed F, Ding H, Wang T, Hooper LV, Koh with inflammation and gut microbiota in S, et al.; MetaHIT Consortium; MetaHIT Consor- GY, Nagy A, et al. (2004). The gut microbiota as overweight and obese subjects. PLoS One. tium (2014). An integrated catalog of reference an environmental factor that regulates fat stor- 9(10):e109434. http://dx.doi.org/10.1371/ genes in the human gut microbiome. Nat Biotech- age. Proc Natl Acad Sci U S A. 101(44):15718– journal.pone.0109434 PMID:25330000 nol. 32(8):834–41. http://dx.doi.org/10.1038/ 23. http://dx.doi.org/10.1073/pnas.0407076101 nbt.2942 PMID:24997786 PMID:15505215 26. Tremaroli V, Bäckhed F (2012). Func- tional interactions between the gut microbi- 7. Xiao L, Feng Q, Liang S, Sonne SB, Xia Z, 17. Ley RE, Bäckhed F, Turnbaugh P, Lozupone ota and host metabolism. Nature. 489(7415): Qiu X, et al. (2015). A catalog of the mouse CA, Knight RD, Gordon JI (2005). Obesity alters 242–9. http://dx.doi.org/10.1038/nature11552 gut metagenome. Nat Biotechnol. 33(10): gut microbial ecology. Proc Natl Acad Sci U S PMID:22972297 1103–8. http://dx.doi.org/10.1038/nbt.3353 A. 102(31):11070–5. http://dx.doi.org/10.1073/ PMID:26414350 pnas.0504978102 PMID:16033867 27. Dumas ME, Kinross J, Nicholson JK (2014). Metabolic phenotyping and sys- 8. Dominguez-Bello MG, Blaser MJ, Ley RE, 18. Turnbaugh PJ, Ley RE, Mahowald MA, tems biology approaches to understanding Knight R (2011). Development of the human Magrini V, Mardis ER, Gordon JI (2006). An metabolic syndrome and fatty liver disease. gastrointestinal microbiota and insights from obesity-associated gut microbiome with in- Gastroenterology. 146(1):46–62. http:// high-throughput sequencing. Gastroenterology. creased capacity for energy harvest. Nature. dx.doi.org/10.1053/j.gastro.2013.11.001 140(6):1713–9. http://dx.doi.org/10.1053/j. 444(7122):1027–31. http://dx.doi.org/10.1038/ PMID:24211299 gastro.2011.02.011 PMID:21530737 nature05414 PMID:17183312 28. Zhang C, Yin A, Li H, Wang R, Wu G, 9. Vangay P, Ward T, Gerber JS, Knights D 19. Ley RE, Turnbaugh PJ, Klein S, Gordon JI Shen J, et al. (2015). Dietary modulation (2015). Antibiotics, pediatric dysbiosis, and (2006). Microbial ecology: human gut microbes of gut microbiota contributes to alleviation disease. Cell Host Microbe. 17(5):553–64. associated with obesity. Nature. 444(7122): of both genetic and simple obesity in http://dx.doi.org/10.1016/j.chom.2015.04.006 1022–3. http://dx.doi.org/10.1038/4441022a children. EBioMedicine. 2(8):966–82. http:// PMID:25974298 PMID:17183309 dx.doi.org/10.1016/j.ebiom.2015.07.007 10. Lepage P, Leclerc MC, Joossens M, 20. Cani PD, Delzenne NM (2009). The role of the PMID:26425705 Mondot S, Blottière HM, Raes J, et al. (2013). A gut microbiota in energy metabolism and meta- metagenomic insight into our gut’s microbiome. bolic disease. Curr Pharm Des. 15(13):1546–58. Gut. 62(1):146–58. http://dx.doi.org/10.1136/ http://dx.doi.org/10.2174/138161209788168164 gutjnl-2011-301805 PMID:22525886 PMID:19442172

94