Microbiomarkers in Inflammatory Bowel Diseases

Leading article

our disease understanding. In contrast Microbiomarkers in inflammatory bowel to the static human genome, micro- biome composition is more dynamic and Gut: first published as 10.1136/gutjnl-2016-313678 on 21 July 2017. Downloaded from diseases: caveats come with caviar can thus be more easily modified by, for example, probiotics, prebiotics, antibi- Felix Sommer,1 Malte Christoph Rühlemann,1 otics or faecal microbiota transfer (FMT) 1 1 1 2 from healthy donors to patients by enema Corinna Bang, Marc Höppner, Ateequr Rehman, Christoph Kaleta, or via the nasoduodenal route. The latter 3 4 5 Phillippe Schmitt-Kopplin, Astrid Dempfle, Stephan Weidinger, was demonstrated to be perhaps the best 1 2,6 7 method to treat life-threatening recurrent Eva Ellinghaus, Susanne Krauss-Etschmann, Dirk Schmidt-Arras, 6–8 1 8 1 1,8 Clostridium difficile infections, but is Konrad Aden, Dominik Schulte, David Ellinghaus, Stefan Schreiber, a controversial method for IBD and is Andreas Tholey,2 Jan Rupp,9 Matthias Laudes,8 John F Baines,2,10 therefore still regarded as an experimental Philip Rosenstiel,1 Andre Franke1 therapy. Nevertheless, a recent multi- centre placebo-controlled trial in Australia (the Faecal Microbiota Transplantation in The largest numbers of commensal gut, but also between luminal and mucosa- Ulcerative Colitis (FOCUS) study) demon- bacteria reside within our intestinal tract, attached communities of the same intes- strated efficacy of FMT for treating UC with an increasing density from mouth to tinal segment, and even along the crypt- (p=0.021; enemas given at 5 days a week anus. Recently, a new estimate for the total villus axis in the epithelium. Thus, for overall 8 weeks).9 Based on this finding, number of bacteria (3.8×1013) in the host–microbiota interactions are likely it was suggested that future work should 70 kg ‘reference man’ was reported.1 For regionally specific and the local crosstalk focus on defining the optimum treatment human cells, the same authors revised past determines intestinal function and physi- intensity and the role of donor–recipient estimates to 3.0×1013 cells, out of which ology. Probably each human individual matching based on microbial profiles. The approximately 90% belong to the haema- carries its own ‘microbial fingerprint’ growing body of studies using FMT and topoietic lineage. Hence, the widely cited (especially when considering genomic gnotobiotics started to move the field of 10:1 ratio of bacteria versus human cells variation within the bacterial species’ microbiota research away from correla- received an update, showing that the populations), which is why forensic scien- tion to causality by proving a functional number of bacteria in the body is actually tists started to exploit the use of this involvement of the microbiota for health of the same order as the number of human non-human organ.3 and disease.10–12 cells, and that the cumulative bacterial Recent large-scale analyses of popula- Microbiome studies in IBD have been mass is about 200 g. Still, this large number tion-based cohorts with >1000 samples consistent in finding reduced diver- of bacteria highlights their importance in validated that body mass index, age sity in patients compared with healthy maintaining health and metabolism. at sampling and gender are important controls (HC).13 14 A lack of microbial

Different parts of the intestinal tract have covariates that need to be included in diversity seems to be a general theme in http://gut.bmj.com/ different functions, tissue structure varies microbiome association analyses.4 In several diseases (obesity, diabetes, asthma, accordingly and gradients exist for several sum, Falony et al identified 18 covari- atopic dermatitis, and so on), suggesting physicochemical parameters such as nutri- ates, including stool consistency, dietary that a species-rich ecosystem is more ents, pH or oxygen levels.2 Consequently, factors and blood traits, cumulatively robust against environmental influences, microbiota composition varies along the explaining 7.7% of the total variation in as functionally related microbes exist in the gut microbiota, leaving 92.3% of the an intact ecosystem that can compen- on October 2, 2021 by guest. Protected copyright. 1Institute of Clinical Molecular Biology, Christian- interindividual microbial variation unex- sate for the function of another species Albrechts-Universität zu Kiel, Kiel, Germany plained. In a recent host-microbiome that has disappeared. Consequently, 2Institute for Experimental Medicine, Christian Albrechts genome-wide association study, 42 diversity seems to be a good indicator University of Kiel, Kiel, Germany 15 3 genetic loci were identified that explain of a ‘healthy gut.’ A large study that Helmholtz Zentrum München, Research Unit Analytical 5 analysed the microbiota in 447 treat- BioGeoChemistry, Deutsches Forschungszentrum für another 10% of microbiome variability. Gesundheit und Umwelt (GmbH), Neuherberg, Germany Although these percentages need to be ment-naïve patients with new-onset CD 4Institute of Medical Informatics and Statistics, confirmed in other cohorts and for other (treatment effects on the microbiota can University Hospital Schleswig-Holstein, Kiel, Germany likely be excluded in this unique cohort) 5 ethnicities, both studies show that (A) a Department of Dermatology, University Hospital large portion of microbiome variability also identified different bacteria, which Schleswig-Holstein, Kiel, Germany 6Research Center Borstel, Leibniz-Center for Medicine remains to be explained and (B) any were more abundant in patients compared 16 and Biosciences, Airway Research Center North (ARCN), microbiome study needs to account for with controls and vice versa. Gevers German Center for Lung Research (DZL), Borstel, covariates to prevent false-positive and et al also developed a microbial dysbiosis Germany 7 false-negative results. This means for the index that was correlated with the paedi- Institute of Biochemistry, Christian-Albrechts- atric CD activity index. The dysbiotic Universität zu Kiel, Kiel, Germany field that the meta-data for a particular 8Clinic of Internal Medicine I, University Hospital sample is at least as important for the signal in their study was stronger in the Schleswig-Holstein, Kiel, Germany data analysis as the actual microbiome mucosa-attached microbial communities 9 Department of Infectious Diseases and Microbiology, data itself. compared with faecal bacteria, suggesting University of Lubeck, Lubeck, Germany The main subtypes of IBD are Crohn’s that disease-relevant and more adherent 10Evolutionary Genomics, Max Planck Institute for Evolutionary Biology, Plon, Germany disease (CD) and UC, affecting together bacteria are probably ‘diluted’ in the more than 2.5 million European indi- stool. The recently published longitudinal Correspondence to Mr Andre Franke, Institute IBD microbiome study of Halfvarson and of Clinical Molecular Biology, Christian-Albrechts- viduals. More than 200 genetic suscepti- University of Kiel, Rosalind-Franklin-Str. 12, 24105 Kiel, bility loci have been identified for IBD, a colleagues revealed that the microbiome Germany; a.​ ​franke@mucosa.​ ​de contribution that significantly improved of patients with IBD varies more over time

1734 Sommer F, et al. Gut October 2017 Vol 66 No 10 Leading article

dietary variance in microbiota-related data and to explore dietary modulation of the metabolic properties of the gut ecosystem Gut: first published as 10.1136/gutjnl-2016-313678 on 21 July 2017. Downloaded from (eg, enrichment of dietary fibre) as a possible therapeutic approach for IBD. Figure 1 provides a schematic overview on known gut microbiome changes in IBD. In the highlighted study, Pascal and colleagues29 report on an analysis of the faecal microbiome for a large IBD panel from four different countries (40 HC (unrelated HCs)/71 HR (healthy rela- tives from patients)/34 CD/74 UC from Spain, 54 CD from Belgium, 977 healthy twins from the UK and 158 patients with anorexia from Germany). In sum, they included about 1246 non-IBD and 162 IBD samples. The hypervariable region V4 of the 16S rRNA gene was targeted for microbiome profiling. Pascal et al replicate the findings of Gevers et al16 that F. prausnitzii is reduced/absent in patients with CD and that patients have Figure 1 Microbial signatures of a healthy gut and IBD. Under healthy homeostasis the a higher abundance of Escherichia and microbiota is diverse. Goblet cells produce a thick colonic mucus layer, which creates a physical Fusobacterium. barrier against the microbiota, but also harbours a specific mucus-resident microbiota enriched in, The key finding of their study is a for example, short-chain fatty acid producing bacteria Roseburia and Faecalibacterium prausnitzii. ‘microbial signature’ that may be used to Immune sensory cells such as DC and MΦ sample microbial patterns and induce a T cell profile discriminate between CD and non-CD dominated by IL-10 producing Treg lymphocytes leading to homeostasis. However, the composition by eight prokaryotic groups: Faecali- of the microbiota is less diverse in patients with IBD with fewer Bacteroidetes mainly attributed bacterium (genus from family Clostrid- to loss of Prevotella species and expansion of Actinobacteria, Proteobacteria such as adherent iaceae), Peptostreptococcaceae (family invasive Escherichia coli and Fusobacteria. Mucosal function is also altered, for example, lipid from order Clostridiales), Anaerostipes metabolism, illustrating a cometabolism of the metaorganism. A reduction in Paneth and goblet (genus from family of Lachnospiraceae), cell number along with their impaired functions—secretion of antimicrobial substances and Methanobrevibacter (genus from family mucus—leads to decreased mucus thickness, reduced mucosal integrity and finally to an impaired Methanobacteriaceae), Christensenel- http://gut.bmj.com/ barrier function. This increases bacterial translocation and stimulates activation of DCs and MΦs, laceae (family from order Clostridi- which then induce an altered T cell profile with increased IFNγ/TNFα-producing Th1, IL-6/TNFα- ales) and Collinsella (genus from family producing Th2 and IL-17-producing Th17 lymphocytes resulting in a proinflammatory response Coriobacteriaceae) being decreased in and tissue damage, which in turn stabilises the dysbiotic microbiota and a chronic inflammatory patients with CD; Fusobacterium (genus tone. DC, dendritic cells; IFNγ, interferon gamma; IL, interleukin; MΦ, macrophages; TNFα, tumour from family Fusobacteriaceae) and necrosis factor alpha. Escherichia (genus from family Entero- bacteriaceae) having a higher relative on October 2, 2021 by guest. Protected copyright. compared with controls, corroborating layer, which increases competition for abundance in patients with CD. The that patients with IBD harbour a disturbed beneficial mucus-resident bacteria such overall sensitivity of this diagnostic microbial ecosystem.17 The highest vola- as Akkermansia muciniphila or Faecali- signature was 80%, with specificities of tility of the microbiota, that is, dynamic bacterium prausnitzii, which in turn shifts 89%–94% depending on the compar- deviation from a ‘healthy plane,’ was epithelial responses, in particular goblet ison. In comparison, assessing faecal observed for patients with CD with ileal cell function.23 24 Oral but even intrave- calprotectin (a heterodimer of S100A8 inflammation (lower volatility in colonic nous iron replacement therapy distinctly (calgranulin A) and S100A9 (calgranulin CD and UC). In an elegant visualisation alters the gut microbiota and metabolome B)), an established, non-invasive test that study, Swidsinski et al demonstrated in patients with IBD.25 Recent studies measures intestinal inflammation, has a that the number of mucosa-associated even suggest that changes in nutrition sensitivity of 83% and specificity of 84% bacteria increases with disease severity18 during the course of human history may for distinguishing organic from func- and together with reduced diversity, an have led to decreased microbiota diversity tional intestinal diseases in symptomatic increase in bacterial load seems to be a and the increased incidence of chronic patients.30 However, diagnosis of IBD, key feature of several diseases.19 Notably, inflammatory diseases and obesity.24 26 27 and especially distinguishing between diet can rapidly alter the gut microbiome Another recent study has highlighted the the different subtypes of IBD, still relies with effects on disease susceptibility potential effect of a disturbed host–micro- on a combination of diagnostic tests, such as pathogenic infection or chronic biota crosstalk in IBD. This is evidenced including endoscopic and histological inflammation.20–22 For example, removal by a striking uncoupling of host transcrip- analyses, which showed high accuracy of carbohydrates from the diet promotes tional patterns and microbiota signatures in a 12-month long follow-up, with only the growth of mucus-utilising bacteria in IBD, with the affected pathways indi- 1%–2% of patients with CD and UC and alters microbial biogeography by cating a loss of cometabolic functions.28 It being identified as false positives, but attracting luminal bacteria to the mucus therefore seems imperative to correct for 9%–12% of patients with CD and UC

Sommer F, et al. Gut October 2017 Vol 66 No 10 1735 Leading article Gut: first published as 10.1136/gutjnl-2016-313678 on 21 July 2017. Downloaded from High cost Reconstruction of multiple complete genomes from complex communities More abundant taxa over- represented Emulsion or droplet-based single-cell amplification Tedious culturing and Tedious selection approaches oxygen); media, (eg, high cost Reconstruction of multiple complete genomes from complex communities Any NGS technology HiSeq/NextSeq Anaerobic and hard-to- culture bacteria are under- represented Multiple complete genomes Single bacterial cells Cultivation-based isolation of Cultivation-based single bacterial clones High cost; high computational burden; large amounts of unannotated data of 16S rRNA Removal required Information on actively expressed functional content HiSeq/NextSeq Transcriptionally more Transcriptionally active bacteria are over- represented All transcribed RNA molecules Sequencing of entire RNA extracted from samples High cost; high computational burden; large amounts of unannotated data Information on encoded functional repertoire; high-resolution taxonomic assignment; fungi and captures also viruses, archaea HiSeq/NextSeq All DNA molecules Sequencing of entire DNA extracted from samples http://gut.bmj.com/ No information on functional repertoire; low resolution on species level Low cost; high throughput amplification bias; gene amplification bias; copy-number bias eg, V1-V2; V3-V4; V4; V6 V4; V3-V4; V1-V2; eg, PCR-based amplification of target followed by sequencing Targeted 16S rRNA gene Targeted amplicon sequencing Metagenome sequencing Metatranscriptomics Single-cell analysis ta composition and function on October 2, 2021 by guest. Protected copyright. No information on functional repertoire; high error rates; higher cost per base than MiSeq species-level resolution PacBio or Oxford NanoporePacBio MiSeq Long-read-based amplification and sequencing of large 16S rRNA gene fragments No information on functional repertoire; low throughput; much hands-on time Species-level resolution High throughput; Gene copy-number bias Amplification bias Primer/region-specific Entire 16S rRNA gene Entire 16S rRNA gene regions, Variable Clade-specific amplification of 16S rRNA gene, cloning and vector-based sequencing Full-length 16S rRNA gene sequencing Overview of sequencing-based methods to characterise microbio

Disadvantages Advantages Sequencing technology Sanger Potential bias Potential Target Technique NGS, next generation sequencing. NGS, Table 1 Table

1736 Sommer F, et al. Gut October 2017 Vol 66 No 10 Leading article being reclassified as another subform of properties and frequency or abdominal call for more large-scale efforts by the IBD.31 pain have been developed. These scores, community. While the study of Pascal et al employed however, do not correlate well with 34 Contributors All authors edited the commentary and Gut: first published as 10.1136/gutjnl-2016-313678 on 21 July 2017. Downloaded from a large sample panel in order to identify mucosal inflammation and therefore designed the figure. potential microbiomarkers of interest, a inflammatory markers such as C-reactive Competing interests None declared. few words of caution are necessary. First protein or faecal calprotectin are addi- of all, the findings still need to be vali- tionally used to evaluate disease activity. Patient consent Parental/guardian consent obtained. dated by other centres using well-pow- Still, levels of C-reactive protein are not Provenance and peer review Commissioned; ered sample collections that include HC, specific for the intestine, but rather a internally peer reviewed. CD, UC and IBD-related diseases, that general marker of systemic inflamma- is, differential diagnoses such as IBS. tion and faecal calprotectin correlates In a first attempt, Pascal et al applied with UC but not CD and thus seems to their test to another cohort sequenced be a marker of inflammation only for in another centre in France with another the colon.35 36 Thus, other non-invasive method (V3-V5 instead of V4). This biomarkers are needed as surrogates for Open Access This is an Open Access article led to a significantly lower sensitivity a healthy or inflamed gut ecosystem. distributed in accordance with the Creative Commons and specificity for the prediction of Explorative metabolomics of faecal Attribution Non Commercial (CC BY-NC 4.0) license, CD versus UC (60% and 68% respec- water showed already good differenti- which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their tively) and a lower sensitivity (60%) for ation of colonic versus ileal CD with a derivative works on different terms, provided the the prediction of CD versus HC. While clear association of metabolites to the original work is properly cited and the use is non- these results still suggest a non-random core metabolome or to specific micro- commercial. See: http://creativecommons.​ ​org/licenses/​ ​ signal in the data, they also imply that biota,37 motivating thus also the use of by-nc/​ ​4.0/​ larger and multicentre trials are clearly high-resolution analytical technologies © Article author(s) (or their employer(s) unless needed to produce a more robust and for the description of novel molecule otherwise stated in the text of the article) 2017. All rights reserved. No commercial use is permitted unless clinically useful diagnostic signature. It biomarkers. The concept of determining otherwise expressly granted. is noteworthy that Pascal et al included non-invasive microbiomarkers in stool, FS and MCR contributed equally. a very large UK control panel, making which can aid gastroenterologists in the up almost half of their sample size, while diagnosis of IBD and in distinguishing they did not include a matched patient between different intestinal diseases, panel from the same geographical is a very attractive one and warrants region. Anthropometric features as well future investigations.29 38 However, To cite Sommer F, Rühlemann MC, Bang C, et al. Gut as nutritional patterns can have a large specific bacterial species that are clearly 2017;66:1734–1738. impact on the gut’s microbial community associated with disease activity are not Received 29 April 2017 and are a valuable resource in the inves- yet identified. With IBD on the rise Revised 7 June 2017 tigation of connections between disease worldwide and a strong geographical Accepted 8 June 2017 http://gut.bmj.com/ states and the microbiota. While these influence on intestinal microbiota signa- Published Online First 21 July 2017 data will certainly lead to more complex tures,39 analysis of a core taxa signature Gut 2017;66:1734–1738. models for the determination and distin- (combination of bacterial taxa) in large doi:10.1136/gutjnl-2016-313678 guishing of dysbiotic states, they will transethnic cohorts or determining most certainly lead to higher sensitivity functional deficits of the microbial and specificity of microbiome-based community seems to be required.16 29 40 References classifiers and should be considered for More sophisticated sequencing-based 1 Sender R, Fuchs S, Milo R. Revised Estimates for the on October 2, 2021 by guest. Protected copyright. future approaches. approaches (see table 1) such as metag- Number of Human and Bacteria Cells in the Body. PLoS Biol 2016;14:e1002533. The regional specificity of the intestinal enomic/metatranscriptomic shotgun 41 2 Sommer F, Bäckhed F. Know your neighbor: Microbiota microbiota to different anatomical sites sequencing or metabolic modelling, and host epithelial cells interact locally to control might partially explain why the authors which capture the functional capacity of intestinal function and physiology. Bioessays failed for CD but were able to describe the microbiota, overcome many limita- 2016;38:455–64. UC-specific microbiomarkers.2 16 Faeces tions of 16S sequencing, yet should be 3 Franzosa EA, Huang K, Meadow JF, et al. Identifying personal microbiomes using metagenomic codes. Proc may be a suitable surrogate for the colonic complemented with phenotypic char- Natl Acad Sci U S A 2015;112:E2930–E2938. microbiota but not for that of the small acterisation of individual bacteria using 4 Falony G, Joossens M, Vieira-Silva S, et al. Population- intestine. Thus, the diagnosis of CD culture-based techniques, proteomics level analysis of gut microbiome variation. Science using microbiomarkers seems to require and metabolomics. With more complete 2016;352:560–4. 5 Wang J, Thingholm LB, Skiecevičienė J, et al. Genome- sampling of the local microbiota using microbiome reference data sets becoming wide association analysis identifies variation in vitamin biopsies. available, and with larger international D receptor and other host factors influencing the gut To date, endoscopy is the gold stan- sample panels being collated—similar microbiota. Nat Genet 2016;48:1396–406. dard to diagnose IBD and assess mucosal to the genetics community where 6 Hamilton MJ, Weingarden AR, Unno T, et al. inflammation. Since endoscopy is an several 10 000 DNAs and genome-wide High-throughput DNA sequence analysis reveals stable engraftment of gut Microbiota following expensive and invasive procedure with genetic data sets were analysed in a joint transplantation of previously frozen fecal Bacteria. Gut the risk for complications, alternative effort (see www.​ibdgenetics.​org)—we Microbes 2013;4:125–35. scores, for example, the Harvey-Brad- are certain that microbial signatures 7 Li SS, Zhu A, Benes V, et al. Durable coexistence of shaw index for CD32 or the simple of ‘caviar value’ may be derived in the donor and recipient strains after fecal Microbiota 33 transplantation. Science 2016;352:586–9. clinical colitis activity index for UC, future. Until then, we alert the scien- 8 Manichanh C, Reeder J, Gibert P, et al. Reshaping the which include non-invasive measures tists to the ‘caveats’ that come along gut microbiome with bacterial transplantation and such as general state of health, faecal with the field of biomarker research and antibiotic intake. Genome Res 2010;20:1411–9.

Sommer F, et al. Gut October 2017 Vol 66 No 10 1737 Leading article

9 Paramsothy S, Kamm MA, Kaakoush NO, et al. 20 Desai MS, Seekatz AM, Koropatkin NM, et al. A Dietary 31 Moum B, Ekbom A, Vatn MH, et al. Inflammatory Multidonor intensive faecal Microbiota transplantation Fiber-Deprived Gut Microbiota Degrades the Colonic bowel disease: re-evaluation of the diagnosis in a for active ulcerative colitis: a randomised placebo- Mucus Barrier and Enhances Pathogen Susceptibility. prospective population based study in south eastern

controlled trial. Lancet 2017;389:1218–28. Cell 2016;167:e21:1339–53. Norway. Gut 1997;40:328–32. Gut: first published as 10.1136/gutjnl-2016-313678 on 21 July 2017. Downloaded from 10 Smirnov KS, Maier TV, Walker A, et al. Challenges of 21 Sonnenburg ED, Smits SA, Tikhonov M, et al. Diet- 32 Harvey RF, Bradshaw JM. A simple index of crohn’s- metabolomics in human gut Microbiota research. Int J induced extinctions in the gut Microbiota compound disease activity. Lancet 1980;1:514. Med Microbiol 2016;306:266–79. over generations. Nature 2016;529:212–5. 33 Walmsley RS, Ayres RC, Pounder RE, et al. A simple 11 Zhao L. The gut Microbiota and obesity: from 22 David LA, Maurice CF, Carmody RN, et al. Diet rapidly clinical colitis activity index. Gut 1998;43:29–32. correlation to causality. Nat Rev Microbiol and reproducibly alters the human gut microbiome. 34 Palmer KR. Complications of gastrointestinal 2013;11:639–47. Nature 2014;505:559–63. endoscopy. Gut 2007;56:456–7. 12 Saraswati S, Sitaraman R. Aging and the human 23 Earle KA, Billings G, Sigal M, et al. Quantitative 35 Vermeire S, Van Assche G, Rutgeerts P. Laboratory gut microbiota-from correlation to causality. Front Imaging of Gut Microbiota spatial Organization. Cell markers in IBD: useful, magic, or unnecessary toys? Microbiol 2014;5:764. Host Microbe 2015;18:478–88. 13 Manichanh C, Rigottier-Gois L, Bonnaud E, et al. 24 Sonnenburg ED, Sonnenburg JL. Starving our microbial Gut 2006;55:426–31. Reduced diversity of faecal Microbiota in Crohn’s self: the deleterious consequences of a diet deficient 36 af Björkesten CG, Nieminen U, Turunen U, et al. disease revealed by a metagenomic approach. Gut in microbiota-accessible carbohydrates. Cell Metab Surrogate markers and clinical indices, alone or 2006;55:205–11. 2014;20:779–86. combined, as indicators for endoscopic remission in 14 Ott SJ, Musfeldt M, Wenderoth DF, et al. Reduction in 25 Lee T, Clavel T, Smirnov K, et al. Oral versus intravenous anti-TNF-treated luminal Crohn’s disease. Scand J diversity of the colonic mucosa associated bacterial iron replacement therapy distinctly alters the gut Gastroenterol 2012;47:528–37. microflora in patients with active inflammatory bowel Microbiota and metabolome in patients with IBD. Gut 37 Jansson J, Willing B, Lucio M, et al. Metabolomics disease. Gut 2004;53:685–93. 2017;66:863–71. reveals metabolic biomarkers of crohn’s disease. PLoS 15 Sommer F, Anderson JM, Bharti R, et al. The resilience 26 Rampelli S, Schnorr SL, Consolandi C, et al. One 2009;4:e6386. of the intestinal Microbiota influences health and Metagenome sequencing of the Hadza Hunter- 38 Tedjo DI, Smolinska A, Savelkoul PH, et al. The fecal disease. Nat Rev Microbiol 2017. Gatherer Gut Microbiota. Curr Biol 2015;25:1682–93. Microbiota as a biomarker for disease activity in 16 Gevers D, Kugathasan S, Denson LA, et al. The 27 Schnorr SL, Candela M, Rampelli S, et al. Gut Crohn’s disease. Sci Rep 2016;6:35216. treatment-naive microbiome in new-onset crohn’s microbiome of the Hadza hunter-gatherers. Nat 39 Rehman A, Rausch P, Wang J, et al. Geographical disease. Cell Host Microbe 2014;15:382–92. Commun 2014;5:3654. patterns of the standing and active human gut 17 Halfvarson J, Brislawn CJ, Lamendella R, et al. 28 Häsler R, Sheibani-Tezerji R, Sinha A, et al. Uncoupling microbiome in health and IBD. Gut 2016;65:238–48. Dynamics of the human gut microbiome in of mucosal gene regulation, mRNA splicing and 40 Papa E, Docktor M, Smillie C, et al. Non-invasive inflammatory bowel disease. Nat Microbiol adherent microbiota signatures in inflammatory bowel mapping of the gastrointestinal Microbiota identifies 2017;2:17004. disease. Gut 2016:gutjnl-2016-311651. children with inflammatory bowel disease. PLoS One 18 Swidsinski A, Ladhoff A, Pernthaler A, et al. Mucosal 29 Pascal V, Pozuelo M, Borruel N, et al. A microbial 2012;7:e39242. flora in inflammatory bowel disease. Gastroenterology signature for crohn’s disease. Gut 2017;66:813–22. 2002;122:44–54. 30 Gisbert JP, McNicholl AG. Questions and answers 41 Bauer E, Zimmermann J, Baldini F, Thiele I, et al. 19 Sommer F, Bäckhed F. The gut microbiota – masters of on the role of faecal calprotectin as a biological BacArena: individual-based metabolic modeling of host development and physiology. Nat Rev Microbiol marker in inflammatory bowel disease. Dig Liver Dis heterogeneous microbes in complex communities. 2013;11:227–38. 2009;41:56–66. PLoS Comput Biol 2017;13:e1005544. http://gut.bmj.com/ on October 2, 2021 by guest. Protected copyright.

1738 Sommer F, et al. Gut October 2017 Vol 66 No 10