Deciphering the microbiome–metabolome landscape of an inflammatory bowel disease inception cohort

Deciphering the microbiome–metabolome landscape of an inflammatory bowel disease inception cohort | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Deciphering the microbiome–metabolome landscape of an inflammatory bowel disease inception cohort Shiva T Radhakrishnan, Benjamin H Mullish, Marton L Olbei, Nathan P Danckert, and 16 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6232048/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 17 Jul, 2025 Read the published version in Gut Microbes → Version 1 posted You are reading this latest preprint version Abstract The gut microbiota contributes to the etiopathogenesis of inflammatory bowel disease (IBD), but limitations of prior studies include the use of sequencing alone (restricting exploration of the contribution of microbiota functionality) and the recruitment of patients with well-established disease (introducing potential confounders, such as immunomodulatory medication). Here, we analyze a true IBD inception cohort and matched healthy controls (HCs) via stool 16S rRNA gene sequencing and multi-system metabolomic phenotyping (using nuclear magnetic spectroscopy and mass spectroscopy), with subsequent integrative network analysis employed to delineate novel microbiota-metabolome interactions in IBD. Marked differences in β diversity and taxonomic profiles were observed both between IBD and HCs, as well as between Crohn’s disease (CD) and ulcerative colitis (UC) patients. Multiple between-group metabolomic differences were also observed, particularly related to tryptophan-/indole-related metabolites; for example, UC patients had higher levels of serum metabolites including xanthurenic acid ( q = 0.0092) and picolinic acid ( q = 0.018). Network analysis demonstrated multiple unique interactions in CD compared to HCs with minimal overlap, indicating a loss of ‘health-associated’ interactions in CD. Compared to HCs, UC patients demonstrated increased pathway activity related to nitrogen and butanoate metabolism, whilst CD patients displayed increased leucine and valine synthesis. Networks from IBD patients overall showed negative correlation with health-specific associations, including an increase in taurine metabolism. Collectively, this work characterizes multiple novel perturbed microbiota-metabolome interactions that are present even at the diagnosis of IBD, which may inform potential future targets to aid diagnosis and direct therapeutic options. General Microbiology Gastroenterology & Hepatology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Inflammatory bowel diseases (IBD) are chronic, relapsing-remitting conditions primarily affecting the gastrointestinal (GI) tract. The main two subtypes of IBD are Crohn’s disease (CD) and ulcerative colitis (UC). Although the etiopathogenesis of IBD development is yet to be fully understood, a combination of environmental, genetic and immunological factors appears to play important roles ( 1 ) . The gut microbiota plays a crucial role in the etiopathogenesis and clinical trajectory of IBD, with IBD patients exhibiting an over-representation of traditionally pro-inflammatory microbes in addition to an overall reduction in microbial diversity ( 2 ) . Pro-inflammatory bacteria are thought to adversely impact upon the mucus barrier layer of the bowel, a proposed mechanism of intestinal inflammation ( 3 , 4 ) . Prior data document a reduction in α diversity within the gut microbiome of patients suffering from terminal ileal Crohn's disease (CD) when compared to healthy controls (HCs) ( 5 ) . However, when comparing patients with ulcerative colitis (UC) to those with CD, such significant differences in α diversity have not been observed ( 5 ) . Previous studies have characterized the gut microbiota in the context of IBD, however, most of these studies have focused on patients who have been previously diagnosed with IBD and have been exposed to numerous previous treatments ( 5 , 6 ) . It is therefore difficult to ascertain if the “microbiota fingerprint” associated with clinical IBD phenotypes is a cause, a consequence (e.g., related to medication use), or an epiphenomenon. Studies leveraging multi-omics approaches, notably metabolomics, have revealed that changes in microbial composition are closely linked with shifts in the production of microbial metabolites, many of which have immunomodulatory properties ( 7 , 8 ) . Furthermore, recent data have shown that a multi-omics approach outperforms single-omic analyses in predicting outcomes in IBD using machine learning ( 9 ) . Analyses of ‘newly diagnosed’ and untreated IBD cohorts have yielded interesting results. Multi-omic approaches have revealed that patients with CD characterized by “dysbiosis” exhibit a marked reduction in butyrate and secondary bile acids (BAs) such as lithocholate and deoxycholate compared to those with milder inflammation ( 8 ) . However, this study primarily involved pediatric patients, and there were instances where some CD patients did not contribute baseline samples. Reduced concentrations of SCFAs have been shown in UC patients ( 10 ) ; an increased urinary concentration of xanthurenic acid has also been associated with severe UC ( 11 ) . In another study, thirty-one microbial species and thirteen metabolites were found to effectively distinguish IBD from HCs ( 12 ) . These studies have also highlighted the difficulty of obtaining unadulterated samples from a treatment-naïve population. One study included patients with previous surgical resection for IBD, therefore this was not a truly treatment-naïve cohort ( 10 ) ; another included sample collection after initial colonoscopy ( 12 ) which may have affected results, as bowel purgatives may impact the gut microbiota and metabolite composition ( 13 , 14 ) . Tryptophan metabolites (TRPm) have generated great interest in IBD: the gut microbiota are known to be important in the metabolism pathways of TRPm ( 15 ) . Previous data have shown an increased serum concentration of quinolinic acid in IBD patients compared to healthy controls ( 16 ) , with a recent publication showing mucosal inflammation was associated with reduced concentrations of xanthurenic and kynurenic acid ( 17 ) . Network biology analyses have previously been used by large clinical cohorts to integrate microbiome data with other data modalities - including metabolomics, metagenomics, transcriptomics and clinical data - to help build hypotheses about disease mechanisms ( 8 ) . Theoretically, interrogating personalized networks in individual patients could reveal novel biomarkers of interest in addition to therapeutic targets that would be individualized to each patient ( 18 ) . There have been very few studies documenting microbial composition and functionality in newly diagnosed IBD patients, where samples were obtained prior to any endoscopic intervention, formal diagnosis, or (most importantly) treatment. Here, we present the gut microbial composition and functionality of an inception cohort of IBD prior to formal diagnosis. We find distinctive microbiome and metabolomic signatures in a true inception IBD cohort that are different from established IBD cohorts and integrate these data in network biology analyses. The aims of this study were to generate novel potential insights into IBD pathogenesis by: Documenting differences in the gut microbiome and metabolome between treatment-naïve, newly diagnosed CD, UC patients and healthy controls, using samples obtained before diagnosis. Undertaking integrated microbiome-metabolome network-level analyses by focusing on those interactions that were lost or gained when comparing CD, UC and healthy controls. Materials and Methods Study participants and sample collection: Patients with symptoms suggestive of IBD, but no formal diagnosis, were consecutively recruited in a single Gastroenterology unit in London, United Kingdom. Patients were excluded if they had a prior diagnosis of IBD. Healthy controls (HCs) were also consecutively recruited independently - these were healthy individuals with no medical conditions and on no regular medication. Research ethics approval was obtained (REC reference 18/LO/1207, protocol number 18SM4548, IRAS ID 243310). Both HCs and patients were excluded if antibiotics were administered within the previous three months. Demographic and clinical data, and biofluid samples, were collected prospectively prior to any treatment, bowel preparation, endoscopy or formal diagnosis. Whole fecal samples were collected with a FECOTAINER® fecal collection device ( 19 ) . Serum samples were obtained by venesection with subsequent centrifugation and were stored in microcentrifuge tubes. Midstream urine samples were collected in universal containers; urinary sediment was separated using centrifugation and urine was stored in microcentrifuge tubes. Samples were also collected from HCs. All samples collected from patients and HCs were immediately stored at 4°C, for a maximum of 30 minutes, prior to processing and storage at -80°C. Participant phenotypic data were collected including age, weight, height, ethnicity, sex and dietary information. To document IBD disease severity, the Harvey Bradshaw index (HBI) (for CD), and simple clinical colitis activity index (SCCAI) (for UC) were recorded ( 20 , 21 ) ( 22 ) . Fecal samples were processed for fecal calprotectin (FC) values in recruited patients, prior to endoscopic evaluation. Microbial DNA extraction: Microbial DNA was extracted from crude fecal samples using the DNeasy PowerLyser PowerSoil Kit (Qiagen, Hilden, Germany) using the manufacturer’s established protocols. An additional step was undertaken to homogenize samples using Bullet Blender Storm bead beater (Chembio, St Alban’s, UK). Total microbial biomass within each sample was calculated using quantitative real-time polymerase chain reaction (qPCR) analyses. BactQuant assay was chosen given the comprehensive coverage of actual bacterial DNA ( 23 ) . qPCR analyses were undertaken to enable transformation of compositional metataxonomic data into ecosystem abundance ( 24 ) , removing the need for rarefaction ( 25 ) . Sample libraries were prepared following Illumina’s 16S Metagenomic Sequencing Library Preparation Protocol using specifically designed V1-V2 hypervariable region primers, as detailed in Appendix 1 ( 26 ) . Nuclear Magnetic Resonance Spectroscopy (H NMR) methods: To analyze fecal and urine samples, buffer (created using previously detailed protocols ( 27 ) ) and samples were combined in a 1:9 ratio (buffer: sample) for final analyses. Serum analyses were undertaken using previously described protocols ( 27 ) . All biofluid samples were prepared for analyses as per previous protocols ( 28 ) . Sample extracts were analyzed with a Bruker 600 MHz AVANCE II 1 H-NMR spectrometer, at 300 Kelvin (K). 1D 1 H NMR spectra were acquired using a standard one-dimensional pulse sequence, with saturation of the water resonance (noesygppr1d pulse program) during both the relaxation delay (RD = 4s) and mixing time (tm = 10 ms). In total, 4 dummy scans, 128 scans and 64 K data points were collected. Further information concerning 1 H NMR parameters is given in Appendix 2 . Liquid Chromatography Mass Spectroscopy (LC-MS) methods: Fecal water was prepared from lyophilized stool as per previously published protocols ( 29 ) . Urine and serum LC-MS experiments were undertaken at the National Phenome Centre (NPC) and followed previously described protocols ( 29 , 30 ) to obtain targeted and global profiles for the serum, urine and fecal samples. Targeted assays included tryptophan (TRP) metabolites and bile acids (BA). To identify and quantify fecal and serum short chain fatty acids (SCFAs), previously published protocols were followed 33 . Global analyses were completed with fecal reverse phase (RP) negative and positive phases were used in addition to urinary RP negative phase. Data processing: 16S rRNA gene amplicon sequencing data processing was carried out using the DADA2 pipeline (v.1.18) ( 31 ) . Data were analyzed using R studio environment to construct the phylogenetic tree using the Phangorn package with default settings ( 32 ) . A suite of R packages were used to analyze the microbiome sequencing data including Phyloseq ( 33 ) , Vegan ( 34 ) and ggplot2 ( 35 ) . 1 H-NMR data were analyzed in SIMCA® (version 17, Sartorius, Sweden) and MATLAB (2014a, MathWorks) to visualize the data. The individual spectra of these metabolites were identified, as previously described, after cross-referencing internal and external databases ( 36 – 38 ) . Untargeted LC-MS data were processed using “peakPantheR”, an established R package to identify metabolites as per previous publications ( 39 ) . LC-MS data were processed using the online platform MetaboAnalyst (version 5.0) ( 40 ) . Statistical and network analyses: Non-parametric ANOVAs with Kruskal-Wallis test were used to study the significance of differences in α diversity, taxonomic abundance and metabolite concentration between cohorts. Principal coordinate analysis (PCoA) was used to analyze β diversity, with permutational multivariate analysis of variance (PERMANOVA) to assess statistical significance. Benjamini-Hochberg false discovery rate (FDR) adjustment was used for P value correction where multiple comparisons were performed ( 41 ) . To ascertain which metabolites were responsible for the differences observed, loadings plots for each orthogonal projections to latent structures discriminant analysis (OPLS-DA) were created in MATLAB® which enabled identification of the metabolites at extreme ends of the loading plot, driving the differences noted. Semi-targeted 1 H-NMR metabolite differences were compared across cohorts using column analyses after log-transformation. Non-parametric ANOVAs were used to compare the mean rank of each cohort with Kruskal-Wallis tests with FDR P value correction. The processed fecal metabolite and bacterial abundance data were read into R (version 4.4.1). The individual microbial samples were aggregated to a genus level. Strains present in at least 20% of all patients were removed to filter out overly promiscuous strains. Microbial network analyses were formulated by calculating the Spearman correlation coefficient between the measured fecal metabolites and the bacterial amplicon sequencing variants (ASVs) for each disease state (UC, CD, Healthy), using the rcorr function from the Hmisc package (version 5.1) in R (version 4.4.1). Results were filtered for significance (FDR P = 0.5). Visualization was carried out with the ggplot ( 35 ) , ggraph ( 42 ) and tidygraph ( 43 ) packages. The corresponding bacterial ASVs were collapsed into their respective taxonomic genera, resulting in state specific, strong bacterial genus – metabolite associations. Functional enrichment analysis of the strongly correlating microbiota associated metabolites was carried out using the MetaboAnalyst web server ( 44 ) . The enrichment ratio was computed by dividing the number of expected hits with the observed hits, as returned by MetaboAnalyst. For the analysis of matching associations with opposite correlations across states, the median of correlation values was calculated for each genus – metabolite pair (to capture the behavior of genera with multiple associating strains), and the results were visualized on a scatterplot ( Supplementary Fig. 1–3 ). KEGG pathway ( 45 ) enrichment analysis of the appropriate anticorrelation clusters found on the scatterplot (i.e. Healthy > = 0.4 & CD <= -0.4) was carried out using the MetaboAnalyst web server ( Supplementary Fig. 4 ). Corrected P values ( q values) of < 0.1 were deemed to be significant given the exploratory nature of these analyses, with previous comparable microbiota datasets using a q value of < 0.1 as significant ( 46 ) . Node rewiring capturing bacteria with the most variable interaction profiles was calculated using the DyNet app ( 47 ) (version: 1.0) in Cytoscape ( 48 ) (version: 3.10.) with default settings (undirected networks). The 16S rRNA gene DNA data have been deposited at the European Bioinformatics Institute’s (EBI) European Nucleotide Archive ENA database under the accession number PRJEB86986. Results Patient phenotypes In total 80 participants were recruited, comprising 60 IBD (23 CD and 37 UC) patients, newly diagnosed with IBD using established biochemical, endoscopic, histological, and radiological criteria ( 49 ) , and 20 HCs. Between HCs and patients with IBD, there were no significant differences in covariates such as age, BMI, diet, ethnicity, sex or smoking status. There was no significant difference in the median Fecal Calprotectin (FC) value between CD and UC patients. Additionally, the median disease activity index (DAI) score did not significantly differ between UC and CD patients - as measured by SCCAI score and HBI scores respectively. These data are displayed in Table 1 . Table 1 Phenotypic characteristics of inception cohort at baseline . General Characteristics CD UC HC P value Number of participants 23 37 20 Age range, (median) 18–66, (35) 18–75, (32) 26–54, (34) 0.713* Sex (M/F) (M = 13/F = 10) (M = 16/F = 21) (M = 12/F = 8) 0.403 φ BMI range group (25.0) (5/13/5) (6/22/9) (1/18/1) 0.131 φ Diet (Meat eater/pescetarian/vegetarian) (17/3/3) (27/6/4) (14/5/1) 0.790 φ Ethnicity (Caucasian/non-Caucasian) (12/11) (26/11) (12/8) 0.358 φ Smoking status (current/non-smoker) (1/22) (2/35) (1/19) 0.983 φ Calprotectin median, (range) 796, (259–2255) 1269, (258–7900) N/A 0.146 l DAI median, (range) 8, (4–14) 7, (4–15) N/A 0.713 l KEY : *=One way ANOVA; φ = Chi-square test; l = Mann-Whitney U IBD patients were classified as per the Montreal classification ( 50 ) . These data are displayed in Table 2 A and 2 B. Table 2 A. UC cohort characteristics by Montreal classification ( 50 ) . Extent of UC E1 E2 E3 Number of participants* 7 16 14 KEY : *=FC value between phenotypes was not statistically significant ( P = 0.65) Table 2 B. CD Cohort characteristics by Montreal classification ( 50 ) . Crohn’s cohort demographics Location of CD* L1 = 10 L2 = 4 L3 = 9 Behavior of CD B1 = 16 B2 = 3 B3 = 4 Evidence of perianal disease NO = 19 YES = 4 KEY : *=FC value between phenotypes was not statistically significant ( P = 0.40) A total of 238 samples were analyzed: 78 fecal samples, 80 serum and 80 urine samples. Gut microbiome diversity metrics and taxonomic features having a defined profile in an inception IBD cohort : β diversity between the cohorts, as plotted by PCoA, illustrated that HCs were significantly different to both CD and UC populations ( P = 0.0018) (Fig. 1A). No significant difference in β diversity was demonstrated between CD and UC. Although the α diversity for HCs was increased compared to both UC and CD patients across multiple indices, these trends were not found to be statistically significant in a linear mixed model (Fig. 1B-D). At phylum level, the median abundance of Bacteroidetes was significantly different, with an overall ANOVA of P < 0.0001. Across inter-cohort comparisons, Bacteroidetes abundance was significantly different, with enrichment noted in CD compared to both HC ( q < 0.0001) and to UC ( q < 0.0001). UC patients also demonstrated an increased median abundance of Bacteroidetes compared to HC ( q HC ( q = 0.0008) and UC > HC ( q = 0.011). Although the median abundance of Firmicutes was higher in CD compared to UC, this observation did not reach statistical significance ( q = 0.17). Similar trends were observed in Proteobacteria abundance with a significant overall ANOVA P = 0.0104. As expected, Proteobacteria median abundance was significantly higher in CD compared to HCs ( q = 0.0075). This was replicated in a UC population who were enriched with Proteobacteria compared to HCs ( q = 0.043). Key phyla differences are presented in Fig. 2 . Deeper analyses revealed that a particular Clostridia ASV from the Firmicutes phylum, identified as ASV455 (part of the Family XIII AD3011 genus group), was enriched in HCs compared to both CD and UC patients ( q < 0.001). Metabolomic profiles demonstrate distinct differences between controls and IBD in addition to between CD and UC Multivariate analyses were undertaken on 1 H NMR data with principal component analyses (PCAs) demonstrating no outlying samples. OPLS-DA analyses comparing CD, UC and HC cohorts to each other demonstrated significant differences between HCs and each IBD cohort. OPLS-DA analyses between serum samples of CD and UC demonstrated significant differences between the cohorts, but similar findings were not demonstrated in fecal and urine samples. These data are listed in Appendix 3 . Semi-targeted 1 H NMR analyses, with individual metabolite spectra identified using previously described databases ( 36 – 38 ) , demonstrated that serum N-acetylglucosamine/galactosamine (GlycA) and sialic acid (GlycB) were higher in both CD and UC compared to HCs. Serum GlycA was also at a higher concentration in CD patients compared to UC patients. A similar trend was observed with serum pyruvate at a higher concentration in both IBD cohorts compared to HCs (Fig. 3A-C). Fecal nicotinate was shown to be higher in CD patients (Fig. 3D). Urinary hippurate was noted to be higher in HCs compared to both CD and UC (Fig. 3E). Targeted LC-MS analyses demonstrated multiple differences between IBD cohorts and HCs in addition to between CD and UC. Serum tryptophan metabolites were markedly different between IBD and healthy controls, additionally notable differences were demonstrated between CD and UC. Serum xanthurenic acid ( q = 0.0092), picolinic acid ( q = 0.018), kynurenic acid ( q = 0.036) and 3-hydroxyanthranilic acid (3-HAA) ( q = 0.055) were lowest in CD patients compared to both HCs and UC patients. UC patients exhibited higher serum concentrations of neopterin ( q = 0.036) and quinolinic acid ( q = 0.055) compared to HCs. Additionally, serum 5-hydroxyindole acetic acid (5-HIAA) was consistently shown to be lowest in HCs compared to CD and UC patients ( q < 0.001). These findings are demonstrated in Fig. 4 . Multiple differences were noted in serum SCFAs with isobutyrate ( q = 0.039), 2-methylbutyrate ( q = 0.039) and lactate all increased in UC patients compared to HCs. Serum butyrate ( q < 0.001) concentration and isovalerate ( q = 0.053) concentration were lowest in HCs compared to both CD and UC patients. These serum SCFA findings are shown in Fig. 5 . Serum BAs analyses revealed multiple differences between each disease group. Serum glycochenodeoxycholic acid ( q = 0.014) and glycoursodeoxycholic acid 3-sulphate ( q = 0.077) concentration were shown to be lowest in HCs compared to both CD. A similar trend was observed with UC patients exhibiting a higher concentration of serum glycochenodeoxycholic acid ( q = 0.0081) and glycoursodeoxycholic acid 3-sulphate ( q = 0.072) compared to HCs. Serum glycocholic acid was lower in HC compared to UC patients ( q = 0.0062) although similar differences were not demonstrated between CD and HCs. Murocholic acid was shown to be lowest in UC patients compared to both CD ( q = 0.0099) and to HCs ( q = 0.070). The opposite trend was demonstrated with serum taurocholic acid concentration, as this was highest in UC patient compared to both CD ( q = 0.059) and HCs ( q = 0.029). 5-cholenic acid-3-beta-ol was noted to be highest in HC compared to both CD ( q = 0.011) and UC patients ( q = 0.010). These serum BA findings are shown in Fig. 6 . Global profiles using RP modes for fecal and urinary samples noted multiple metabolite concentration differences between HCs and IBD patients in addition to between CD and UC patients and these are listed in Appendix 4 . Microbial – metabolite correlation network analyses reveal key differences between IBD and healthy controls To further explore and explain the microbiota and metabolomic differences, these data were integrated with correlation network analyses. We observed strong (r > = 0.5) microbe-metabolite correlations (Fig. 7A), unique to each condition, with the healthy and CD patients containing the most distinct correlations. The overlaps among groups were small, suggesting a loss of correlations associated with health, demonstrated in Fig. 7B . Multiple bacterial genera have strong associations in each cohort, with the genus Bacteroides, Faecalibacterium and Blautia correlating with the most metabolites from genera present in all cohorts (Fig. 7C ) . Bacteroides, Faecalibacterium and Blautia are the genera correlating with the most different metabolites (Fig. 7D), indicating that the genera with the largest number of correlations also associated with the most different metabolites across CD and UC patients, and healthy controls. These strong, positive condition-specific microbe–metabolite correlations correspond to distinct functionally enriched metabolic KEGG pathways (Fig. 7E). Notably, the metabolites from the microbe–metabolite correlations of UC patients are overrepresented in nitrogen and butanoate metabolism pathways, while CD patients are enriched in valine, leucine and isoleucine biosynthesis, and both IBD subtypes demonstrate caffeine metabolism pathways. The healthy controls are characterized by the phenylalanine, tyrosine and tryptophan biosynthesis and metabolism and alanine, aspartate and glutamate metabolism pathways, all of which are lacking in the IBD subtypes. To gain further confidence in the disease relevance of these associations, we analyzed microbe–metabolite correlations, which were strongly positive in health but simultaneously negative in one or more IBD subtypes (Fig. 7F). KEGG pathway overrepresentation analysis of the metabolites from these matching microbe–metabolite correlations between health and IBD indicated that the taurine and hypotaurine metabolism pathway was lost in both the CD and UC comparisons. Discussion Here we present the analyses of biosamples from a true inception cohort of IBD patients with matched healthy controls, demonstrating multiple microbial and metabolomic differences between UC, CD and health. Additionally, we document microbe-host interaction with network analyses, demonstrating numerous novel microbiome–metabolome correlations that have not previously been well characterized in IBD. Previous studies have investigated the gut microbiome and metabolome in IBD, with IBD patients characterized by a reduction in microbial diversity ( 5 ) , an enrichment in Proteobacteria ( 6 ) and a perceived “dysbiotic environment”, as well as a low bile acid concentration compared to a healthy population ( 8 ) . However, previous data have not usually been derived from true inception cohorts, with patients having undergone prior surgery or having been exposed to medical treatment at the time of sample collection. Additionally, prior study sampling was often obtained after endoscopic diagnosis (with the known effects of bowel purgatives upon the gut microbiota and metabolome) ( 13 , 14 ) . Our findings demonstrate notable inter-disease differences between CD and UC patients, including abundance of Bacteroidetes and numerous differences in serum TRPm concentration. In some areas, our findings corroborate previous data ( 17 ) : TRPm concentration was increased in healthy controls compared to IBD patients; active IBD resulted in a reduced concentration of kynurenic and xanthurenic acid compared to quiescent disease. Of note, Michaudel et al. did not find a difference in TRPm concentration between CD and UC patients as seen in our cohort. Utilizing correlation network analysis, we observed strong disease-specific microbe-metabolite correlations (r ≥ 0.5), with unique interactions primarily in healthy and CD patients and with minimal overlap between cohorts, suggesting a loss of health-associated interactions. Bacterial genera such as Bacteroides, Faecalibacterium , and Blautia were linked to the most varied metabolites across conditions, and these strong positive correlations mapped to distinct metabolic KEGG pathways, with UC patients showing increased nitrogen and butanoate metabolism, CD increased valine, leucine, and isoleucine biosynthesis, and both IBD subtypes enhanced caffeine metabolism. Healthy controls showed enrichment in phenylalanine, tyrosine, tryptophan, and alanine-related pathways. The increase in butanoate metabolism pathways seen in UC patients may inform as to the nature of gut inflammation: previous data have identified butyrate as an inhibitor of neutrophils (isolated from IBD patients) secreting pro-inflammatory cytokines ( 51 ) . Further analysis of shared, but anticorrelating microbe-metabolite pairs revealed health-specific associations negatively correlated with IBD, identifying the taurine and hypotaurine metabolism pathway as activated in both CD and UC ( Supplementary Fig. 4 ). This finding correlates with recent data illustrating low serum taurine levels in both CD and UC compared to controls, implicating increased taurine metabolism ( 52 ) . Our novel finding of an increase in caffeine metabolism in IBD patients is of interest: previous murine models of colitis have illustrated the protective role of caffeine ( 53 ) , although to fully understand this interaction further prospective data with detailed caffeine intake linked to intestinal inflammation are required. The strengths of our data are numerous. Importantly, samples were obtained prior to endoscopic confirmation of disease (bowel purgatives having been shown to affect the gut microbial composition and metabolome) ( 13 , 14 ) . Additionally, as samples were obtained prior to any medical or surgical treatment (which have been shown to alter both microbes and metabolites) ( 54 ) , we have been able to present an unadulterated sample cohort. We were able to ensure almost complete sample donation from our entire cohort: 78 out of 80 participants were able to donate a full complement of fecal, serum and urine samples - this is far higher than in comparable clinical studies. Although our study recruited from one London center, there was heterogeneity in population age, ethnicity and sex allowing our findings to be applied to other diverse populations. Our findings have combined the microbiota and metabolite findings in the correlation network analyses, thereby linking the findings to truly understand the relationship between the microbe and the host. Moving forward, our work should be tested in a larger population of newly diagnosed patients across multiple centers to confirm that our findings can be replicated. Additionally, further investigation of murine models and human organoids should be undertaken to understand the biological mechanisms of our findings. In conclusion, this is the first analysis to comprehensively integrate microbiota composition and metabolic function in the study of a truly treatment-naïve inception cohort of patients newly diagnosed with IBD. Using state-of-the-art network biology techniques, we have elucidated novel pathways which may be critical to host-microbe interactions in disease pathogenesis. Our findings will enable mechanistic studies to probe the impact of these microbial metabolites on aberrant host immune function in IBD and will better inform translational research using interventions including diet and microbial therapeutics such as intestinal microbiota transplant to modulate IBD disease course. Declarations Acknowledgements: JPT is supported by the Chain Florey Clinical PhD Fellowship jointly funded by the National Institute for Health Research (NIHR) Imperial Biomedical Research Centre (BRC) and the UKRI Medical Research Council (MRC) Laboratory of Medical Sciences (LMS). DM acknowledges financial support from Imperial College London through an Imperial College Research Fellowship grant award. LG and TK were supported by the UKRI BBSRC Institute Strategic Programme Food Microbiome and Health BB/X011054/1 and its constituent project BBS/E/F/000PR13631.NP is supported by the Wellcome Trust (WT101159), Crohn’s and Colitis UK, and the NIHR Imperial Biomedical Research Centre (BRC). JVL is supported by Medical Research Council (MRC) New Investigator Research Grant (MR/P002536/1) and ERC Starting Grant (715662). STR, TK, JR, NP, JLA and HRTW are supported by the NIHR Imperial Biomedical Research Centre (BRC). The Division of Digestive Diseases at Imperial College London receives financial and infrastructure support from the NIHR Imperial BRC based at Imperial College Healthcare NHS Trust and Imperial College London. The views expressed are those of the authors and not necessarily those of the NIHR or the UK Department of Health and Social Care. References Zhang Y-Z, Li Y-Y (2014) Inflammatory bowel disease: pathogenesis. World J gastroenterology: WJG 20(1):91 Wlodarska M, Kostic AD, Xavier RJ (2015) An integrative view of microbiome-host interactions in inflammatory bowel diseases. Cell Host Microbe 17(5):577–591 Fyderek K, Strus M, Kowalska-Duplaga K, Gosiewski T, Wedrychowicz A, Jedynak-Wasowicz U et al (2009) Mucosal bacterial microflora and mucus layer thickness in adolescents with inflammatory bowel disease. World J Gastroenterol 15(42):5287–5294 Larsson JM, Karlsson H, Crespo JG, Johansson ME, Eklund L, Sjovall H et al (2011) Altered O-glycosylation profile of MUC2 mucin occurs in active ulcerative colitis and is associated with increased inflammation. Inflamm Bowel Dis 17(11):2299–2307 Imhann F, Vich Vila A, Bonder MJ, Fu J, Gevers D, Visschedijk MC et al (2018) Interplay of host genetics and gut microbiota underlying the onset and clinical presentation of inflammatory bowel disease. Gut 67(1):108–119 Morgan XC, Tickle TL, Sokol H, Gevers D, Devaney KL, Ward DV et al (2012) Dysfunction of the intestinal microbiome in inflammatory bowel disease and treatment. Genome Biol 13(9):R79 Franzosa EA, Sirota-Madi A, Avila-Pacheco J, Fornelos N, Haiser HJ, Reinker S et al (2019) Gut microbiome structure and metabolic activity in inflammatory bowel disease. Nat Microbiol 4(2):293–305 Lloyd-Price J, Arze C, Ananthakrishnan AN, Schirmer M, Avila-Pacheco J, Poon TW et al (2019) Multi-omics of the gut microbial ecosystem in inflammatory bowel diseases. Nature 569(7758):655–662 O'Sullivan J, Patel S, Leventhal GE, Fitzgerald RS, Laserna-Mendieta EJ, Huseyin CE et al (2025) Host-microbe multi-omics and succinotype profiling have prognostic value for future relapse in patients with inflammatory bowel disease. Gut Microbes 17(1):2450207 Vich Vila A, Hu S, Andreu-Sanchez S, Collij V, Jansen BH, Augustijn HE et al (2023) Faecal metabolome and its determinants in inflammatory bowel disease. Gut 72(8):1472–1485 Aldars-Garcia L, Gil-Redondo R, Embade N, Riestra S, Rivero M, Gutierrez A et al (2024) Serum and Urine Metabolomic Profiling of Newly Diagnosed Treatment-Naive Inflammatory Bowel Disease Patients. Inflamm Bowel Dis 30(2):167–182 Ning L, Zhou YL, Sun H, Zhang Y, Shen C, Wang Z et al (2023) Microbiome and metabolome features in inflammatory bowel disease via multi-omics integration analyses across cohorts. Nat Commun 14(1):7135 Jalanka J, Salonen A, Salojarvi J, Ritari J, Immonen O, Marciani L et al (2015) Effects of bowel cleansing on the intestinal microbiota. Gut 64(10):1562–1568 Powles STR, Gallagher KI, Chong LWL, Alexander JL, Mullish BH, Hicks LC et al (2022) Effects of bowel preparation on intestinal bacterial associated urine and faecal metabolites and the associated faecal microbiome. BMC Gastroenterol 22(1):240 Paeslack N, Mimmler M, Becker S, Gao Z, Khuu MP, Mann A et al (2022) Microbiota-derived tryptophan metabolites in vascular inflammation and cardiovascular disease. Amino Acids 54(10):1339–1356 Nikolaus S, Schulte B, Al-Massad N, Thieme F, Schulte DM, Bethge J et al (2017) Increased Tryptophan Metabolism Is Associated With Activity of Inflammatory Bowel Diseases. Gastroenterology. ;153(6):1504-16 e2 Michaudel C, Danne C, Agus A, Magniez A, Aucouturier A, Spatz M et al (2022) Rewiring the altered tryptophan metabolism as a novel therapeutic strategy in inflammatory bowel diseases. Gut Modos D, Bulusu KC, Fazekas D, Kubisch J, Brooks J, Marczell I et al (2017) Neighbours of cancer-related proteins have key influence on pathogenesis and could increase the drug target space for anticancer therapies. NPJ Syst Biol Appl 3:2 MEDGADGET To Fill a Void, AT Medical Introduces Fecotainer 2011 [cited 2023 18/02/2023]. Available from: https://www.medgadget.com/2011/09/to-fill-a-void-at-medical-introduces-fecotainer.html Harvey RF, Bradshaw JM, A SIMPLE INDEX OF, CROHN'S-DISEASE ACTIVITY (1980) Lancet 315(8167):514 Best WR (2006) Predicting the Crohn_s Disease Activity Index From the Harvey-Bradshaw Index. Inflamm Bowel Dis 12(4):304–310 Jowett SL, Seal CJ, Phillips E, Gregory W, Barton JR, Welfare MR (2003) Defining relapse of ulcerative colitis using a symptom-based activity index. Scand J Gastroenterol 38(2):164–171 Liu CM, Aziz M, Kachur S, Hsueh P-R, Huang Y-T, Keim P et al (2012) BactQuant- An enhanced broad-coverage bacterial quantitative real-time PCR assay. BMC Microbiol. ;12(56) Gloor GB, Macklaim JM, Pawlowsky-Glahn V, Egozcue JJ (2017) Microbiome Datasets Are Compositional: And This Is Not Optional. Front Microbiol 8:2224 McMurdie PJ, Holmes S (2014) Waste not, want not: why rarefying microbiome data is inadmissible. PLoS Comput Biol 10(4):e1003531 Mullish BH, Pechlivanis A, Barker GF, Thursz MR, Marchesi JR, McDonald JAK (2018) Functional microbiomics: Evaluation of gut microbiota-bile acid metabolism interactions in health and disease. Methods 149:49–58 Dona AC, Jimenez B, Schafer H, Humpfer E, Spraul M, Lewis MR et al (2014) Precision high-throughput proton NMR spectroscopy of human urine, serum, and plasma for large-scale metabolic phenotyping. Anal Chem 86(19):9887–9894 Gratton J, Phetcharaburanin J, Mullish BH, Williams HRT, Thursz MR, Nicholson JK et al (2016) An optimized sample handling strategy for metabolic profiling of human feces. Anal Chem 88(9):4661–4668 Lewis MR, Chekmeneva E, Camuzeaux S, Sands C, Yuen A, David M et al (2022) An Open Platform for Large Scale LC-MS-Based Metabolomics. ChemRxiv. https://chemrxiv.org/engage/chemrxiv/article-details/61ebd6fa0716a8529e3823dc Want EJ, Wilson ID, Gika H, Theodoridis G, Plumb RS, Shockcor J et al (2010) Global metabolic profiling procedures for urine using UPLC-MS. Nat Protoc 5(6):1005–1018 Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJ, Holmes SP (2016) DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods 13(7):581–583 Schliep KP (2011) phangorn: phylogenetic analysis in R. Bioinformatics 27(4):592–593 McMurdie PJ, Holmes S (2013) phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE 8(4):e61217 Oksanen J, Gavin L, Simpson GL, Guillaume Blanchet F, Kindt R, Legendre P, Minchin PR et al Package ‘vegan’ Community Ecology Package https:// cran.r-project.org/web/packages/vegan/vegan.pdf2022 [updated 17/04/. 2.6-2:[Available from: https://cran.r-project.org/web/packages/vegan/vegan.pdf Wickham H (2016) ggplot2 Elegant Graphics for Data Analysis. Second eds. Gentleman R, Hornik K, Parmigiani G (eds): Springer Nature Wishart DS, Feunang YD, Marcu A, Guo AC, Liang K, Vazquez-Fresno R et al (2018) HMDB 4.0: the human metabolome database for 2018. Nucleic Acids Res 46(D1):D608–D17 Hoch JC, Baskaran K, Burr H, Chin J, Eghbalnia HR, Fujiwara T et al (2023) Biological Magnetic Resonance Data Bank. Nucleic Acids Res 51(D1):D368–D76 Garcia-Perez I, Posma JM, Serrano-Contreras JI, Boulange CL, Chan Q, Frost G et al (2020) Identifying unknown metabolites using NMR-based metabolic profiling techniques. Nat Protoc 15(8):2538–2567 Wolfer AM, Correia GDS, Sands CJ, Camuzeaux S, Yuen AHY, Chekmeneva E et al (2021) peakPantheR, an R package for large-scale targeted extraction and integration of annotated metabolic features in LC-MS profiling datasets. Bioinformatics 37(24):4886–4888 Pang Z, Chong J, Zhou G, de Lima Morais DA, Chang L, Barrette M et al (2021) MetaboAnalyst 5.0: narrowing the gap between raw spectra and functional insights. Nucleic Acids Res 49(W1):W388–W96 Benjamini Y, Hochberg Y (1995) Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J Roy Stat Soc 57(1):289–300 Pedersen TL An Implementation of Grammar of Graphics for Graphs and Networks. R package version 2.2.1.9000 2024 [Available from: https://ggraph.data-imaginist.com Pedersen TL tidygraph: A Tidy API for Graph Manipulation. R package version 1.3.1.9000 2024 [Available from: https://tidygraph.data-imaginist.com Pang Z, Lu Y, Zhou G, Hui F, Xu L, Viau C et al (2024) MetaboAnalyst 6.0: towards a unified platform for metabolomics data processing, analysis and interpretation. Nucleic Acids Res 52(W1):W398–W406 Kanehisa M, Araki M, Goto S, Hattori M, Hirakawa M, Itoh M et al (2008) KEGG for linking genomes to life and the environment. Nucleic Acids Res 36(Database issue):D480–D484 Routy B, Lenehan JG, Miller WH Jr., Jamal R, Messaoudene M, Daisley BA et al (2023) Fecal microbiota transplantation plus anti-PD-1 immunotherapy in advanced melanoma: a phase I trial. Nat Med 29(8):2121–2132 Goenawan IH, Bryan K, Lynn DJ (2016) DyNet: visualization and analysis of dynamic molecular interaction networks. Bioinformatics 32(17):2713–2715 Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D et al (2003) Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 13(11):2498–2504 Maaser C, Sturm A, Vavricka SR, Kucharzik T, Fiorino G, Annese V et al (2019) ECCO-ESGAR Guideline for Diagnostic Assessment in IBD Part 1: Initial diagnosis, monitoring of known IBD, detection of complications. J Crohns Colitis 13(2):144–164 Silverberg MS, Satsangi J, Ahmad T, Arnott IDR, Bernstein CN, Brant SR et al (2005) Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: Report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol 19:5A–36A Li G, Lin J, Zhang C, Gao H, Lu H, Gao X et al (2021) Microbiota metabolite butyrate constrains neutrophil functions and ameliorates mucosal inflammation in inflammatory bowel disease. Gut Microbes 13(1):1968257 Frascatani R, Mattogno A, Iannucci A, Marafini I, Monteleone G (2024) Reduced Taurine Serum Levels in Inflammatory Bowel Disease. Nutrients. ;16(11) Mizoguchi E, Sadanaga T, Okada T, Minagawa T, Akiba J (2023) Does caffeine have a double-edged sword role in inflammation and carcinogenesis in the colon? Intest Res 21(3):306–317 Radhakrishnan ST, Alexander JL, Mullish BH, Gallagher KI, Powell N, Hicks LC et al (2022) Systematic review: the association between the gut microbiota and medical therapies in inflammatory bowel disease. Aliment Pharmacol Ther 55(1):26–48 Additional Declarations The authors declare potential competing interests as follows: STR has received conference fees from Pfizer and Vifor with advisory fees from Galapagos. BHM has received consultancy fees from Finch Therapeutics Group, Akebia Therapeutics Inc, and Ferring Pharmaceuticals, and speaker fees from Yakult. SB has received conference fees from Dr Falk and Ferring. RWP declares conference fee support from Ferring. NP declares advisory and/or speaker fees from Abbvie, Allergan, Celgene, Debiopharm, Ferring and Vitor Pharma, and lecture fees from Allergan, Dr Falk Pharma, Janssen, Takeda and Tillotts Pharma. JRM has been paid for consultancy by Cultech Ltd and Enterobiotix Ltd. JLA has received conference fees from Celltrion, Tillots Pharma and Takeda with speaker fees from Abbvie and Johnson & Johnson. HRTW has received financial support to attend conferences from Takeda and has been an advisory board member for Pfizer. Supplementary Files SupplementaryAppendices.docx Supplementary Appendices SupplementaryFigure1.jpg Supplementary Figure 1: Microbe-metabolite correlations for healthy controls. SupplementaryFigure2.jpg Supplementary Figure 2: Microbe-metabolite correlations for CD patients. SupplementaryFigure3.jpg Supplementary Figure 3: Microbe-metabolite correlations for UC patients. SupplementaryFigure4.jpg Supplementary Figure 4: Shared anti-correlating microbe-metabolite pairs shown on scatterplots, CD and UC, respectively. Enrichment analysis of the associations to the top left quadrant (positively associated with health, negative with UC/CD, highlighted in blue). Cite Share Download PDF Status: Published Journal Publication published 17 Jul, 2025 Read the published version in Gut Microbes → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6232048","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":429292616,"identity":"67a10b56-266f-4dd0-ae84-e6388aba37b9","order_by":0,"name":"Shiva T Radhakrishnan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAr0lEQVRIiWNgGAWjYJCCAx8qJGBMYtSzMTA+nHGGRC3MxpxtcAuJ0CE/v/eZNOM8C3kG9sMPmHnOEKHF4Bi7mXThNgnDBp40A2aeG8RoYWNjk565TSKBgSGHgZnnAzEOawNq4Z0D1ML/hkgtDMfYmI15G4BaJEC2EOWwY2nAQD4mYdgm8czg4BxivC/ffAwYlTV18vz8yQ8fvDlGjMNggI2ByIgcBaNgFIyCUUAEAAACNC3TZYxwnQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-8311-9714","institution":"Imperial College London","correspondingAuthor":true,"prefix":"","firstName":"Shiva","middleName":"T","lastName":"Radhakrishnan","suffix":""},{"id":429292617,"identity":"db967b36-0877-41e3-9aeb-0e029a51beab","order_by":1,"name":"Benjamin H Mullish","email":"","orcid":"https://orcid.org/0000-0001-6300-3100","institution":"Imperial College London","correspondingAuthor":false,"prefix":"","firstName":"Benjamin","middleName":"H","lastName":"Mullish","suffix":""},{"id":429292618,"identity":"526242b8-0fd9-4803-a804-690cf4a23388","order_by":2,"name":"Marton L Olbei","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Marton","middleName":"L","lastName":"Olbei","suffix":""},{"id":429292619,"identity":"2d158c09-0bf3-428b-a7f3-f1741295cebf","order_by":3,"name":"Nathan P Danckert","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Nathan","middleName":"P","lastName":"Danckert","suffix":""},{"id":429292818,"identity":"98357a6a-96e4-4ee8-94fa-c9e35f68a578","order_by":4,"name":"Maria A Valdivia-Garcia","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"A","lastName":"Valdivia-Garcia","suffix":""},{"id":429292819,"identity":"63b39be2-8b8b-4924-a76b-0eec160c7e26","order_by":5,"name":"Jose I Serrano-Contreras","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Jose","middleName":"I","lastName":"Serrano-Contreras","suffix":""},{"id":429292820,"identity":"2427005d-5893-4261-b7be-d8fde49b3c88","order_by":6,"name":"Despoina Chrysostomou","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Despoina","middleName":"","lastName":"Chrysostomou","suffix":""},{"id":429292821,"identity":"efec05d2-c48e-4610-bf6d-d0d0cdc616e4","order_by":7,"name":"Sharmili Balarajah","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Sharmili","middleName":"","lastName":"Balarajah","suffix":""},{"id":429292822,"identity":"7b883552-7b48-4ff4-93fa-e0ff82089c43","order_by":8,"name":"Robert W Perry","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Robert","middleName":"W","lastName":"Perry","suffix":""},{"id":429292823,"identity":"33c308b5-a98e-40ba-8b1a-2d1c8e8d8bea","order_by":9,"name":"John P Thomas","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"John","middleName":"P","lastName":"Thomas","suffix":""},{"id":429292824,"identity":"ab26a279-73d2-4c12-b559-81bcc1b3b75b","order_by":10,"name":"Lejla Potari-Gul","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Lejla","middleName":"","lastName":"Potari-Gul","suffix":""},{"id":429292825,"identity":"35a53983-ef30-4572-95e0-917af66d41e8","order_by":11,"name":"Dezso Modos","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Dezso","middleName":"","lastName":"Modos","suffix":""},{"id":429292826,"identity":"57a0ad22-e773-4409-bb3b-ae9bdac3b3d1","order_by":12,"name":"Lucy C Hicks","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Lucy","middleName":"C","lastName":"Hicks","suffix":""},{"id":429292827,"identity":"5506b3d7-ab8a-4dfb-8d92-df02d17ff3d3","order_by":13,"name":"Nick Powell","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Nick","middleName":"","lastName":"Powell","suffix":""},{"id":429292828,"identity":"7fc8154c-7e22-4a1d-bba7-e5345566829a","order_by":14,"name":"Timothy R Orchard","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Timothy","middleName":"R","lastName":"Orchard","suffix":""},{"id":429292829,"identity":"c4b31e6e-135b-429b-96f5-07ff077a2d14","order_by":15,"name":"Jia V Li","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Jia","middleName":"V","lastName":"Li","suffix":""},{"id":429292830,"identity":"31fd1f9d-252a-472d-8554-7a9b87c55232","order_by":16,"name":"Julian R Marchesi","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Julian","middleName":"R","lastName":"Marchesi","suffix":""},{"id":429292831,"identity":"c9e511f7-e9d6-4377-8518-fadc8eb3716b","order_by":17,"name":"Tamas Korcsmaros","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Tamas","middleName":"","lastName":"Korcsmaros","suffix":""},{"id":429292832,"identity":"8c5c7cd6-26a1-45a6-b01f-5bbdd92af2e8","order_by":18,"name":"James L Alexander","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"James","middleName":"L","lastName":"Alexander","suffix":""},{"id":429292833,"identity":"c963b724-8784-4e97-a6bf-720f36392f0b","order_by":19,"name":"Horace R T Williams","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Horace","middleName":"R T","lastName":"Williams","suffix":""}],"badges":[],"createdAt":"2025-03-15 10:03:32","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":true,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":true,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6232048/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6232048/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1080/19490976.2025.2527863","type":"published","date":"2025-07-18T00:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":78734560,"identity":"c4d631a0-178c-4afc-be6e-3105fc6f6bd1","added_by":"auto","created_at":"2025-03-18 08:01:20","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":391856,"visible":true,"origin":"","legend":"\u003cp\u003eA. PCoA of Beta diversity measurements HC vs UC vs CD (Ellipses representing 95% CI). Alpha diversity measurements HC vs UC vs CD utilising B. Shannon Index, C. Inverse Simpson Index, D. Chao 1 Index, E. Faith’s PD. (Median signified by middle horizontal line, 25% and 75% confidence intervals (CI) by box ends and range signified by vertical line extents)\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/fd8016b1a094ab46145c1c8b.png"},{"id":78735606,"identity":"e3bb44d6-0d5b-4e61-8edf-99c815eeb716","added_by":"auto","created_at":"2025-03-18 08:09:20","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":430527,"visible":true,"origin":"","legend":"\u003cp\u003eMedian bacterial phyla count (qPCR data), HC vs CD vs UC for A. Bacteroidetes (ANOVA P\u0026lt;0.001, CD vs HC q\u0026lt;0.001, UC vs HC q\u0026lt;0.001, CD vs UC q\u0026lt;0.001). B. Proteobacteria (ANOVA P=0.01, CD vs HC q=0.0075, UC vs HC q=0.043, CD vs UC q=0.16). C. Firmicutes (ANOVA P=0.001, CD vs HC q\u0026lt;0.001, UC vs HC q=0.011, CD vs UC q=0.17).\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/69ddc69648b5458dc0c63ac5.png"},{"id":78735611,"identity":"42a00758-5985-418d-be6c-2d8df5db9915","added_by":"auto","created_at":"2025-03-18 08:09:21","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":337995,"visible":true,"origin":"","legend":"\u003cp\u003eUnivariate analyses of\u0026nbsp;1H-NMR data from inception cohort – A:\u0026nbsp;\u003cstrong\u003eSerum GlycA\u003c/strong\u003e\u0026nbsp;(ANOVA P\u0026lt;0.001, CD vs HC q\u0026lt;0.001, UC vs HC q=0.019, CD vs UC q=0.048). B:\u0026nbsp;\u003cstrong\u003eSerum GlycB\u0026nbsp;\u003c/strong\u003e(ANOVA P\u0026lt;0.001, CD vs HC q\u0026lt;0.001, UC vs HC q=0.0017, CD vs UC q=0.10). C:\u0026nbsp;\u003cstrong\u003eSerum\u003c/strong\u003e \u003cstrong\u003ePyruvate\u003c/strong\u003e\u0026nbsp;(ANOVA P=0.0079, CD vs HC q=0.028, UC vs HC q=0.0072, CD vs UC q=0.64). D:\u0026nbsp;\u003cstrong\u003eFecal\u003c/strong\u003e \u003cstrong\u003eNicotinate\u0026nbsp;\u003c/strong\u003e(ANOVA P=0.0047, CD vs HC q=0.0032, UC vs HC q=0.071, CD vs UC q=0.083). E:\u0026nbsp;\u003cstrong\u003eUrine\u003c/strong\u003e \u003cstrong\u003eHippurate\u003c/strong\u003e\u0026nbsp;(ANOVA P\u0026lt;0.001, CD vs HC q=0.0027, UC vs HC q\u0026lt;0.001, CD vs UC q=0.74).\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/0e5c9463069f1f1b0f7f043d.png"},{"id":78734570,"identity":"c144845b-f5ed-4d13-a180-c41aa3787592","added_by":"auto","created_at":"2025-03-18 08:01:21","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":386739,"visible":true,"origin":"","legend":"\u003cp\u003eUnivariate analyses of serum tryptophan metabolites, inception cohort \u003cstrong\u003eA. 5-HIAA \u003c/strong\u003e(ANOVA P\u0026lt;0.001, CD vs HC q=0.0057, UC vs HC q\u0026lt;0.001, CD vs UC q=0.35) \u003cstrong\u003eB. Xanthurenic acid \u003c/strong\u003e(ANOVA P=0.0018, CD vs HC q=0.0018, UC vs HC q=0.20, CD vs UC q=0.013). \u003cstrong\u003eC. Picolinic acid \u003c/strong\u003e(ANOVA P=0.0055, CD vs HC q=0.0092, UC vs HC q=0.56, CD vs UC q=0.0092). \u003cstrong\u003eD. Neopterin \u003c/strong\u003e(ANOVA P=0.018, CD vs HC q=0.19, UC vs HC q=0.015, CD vs UC q=0.19). \u003cstrong\u003eE. Kynurenic acid \u003c/strong\u003e(ANOVA P=0.018, CD vs HC q=0.015, UC vs HC q=0.17, CD vs UC q=0.10).\u003cstrong\u003e F. 3-HAA \u003c/strong\u003e(ANOVA P=0.034, CD vs HC q=0.031, UC vs HC q=0.24, CD vs UC q=0.12). \u003cstrong\u003eG. Quinolinic acid \u003c/strong\u003e(ANOVA P=0.039, CD vs HC q=0.23, UC vs HC q=0.034, CD vs UC q=0.23).\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/6ec0b34f69011ac626129873.png"},{"id":78736927,"identity":"039414f7-67e1-4dac-b21b-30c136aa7acb","added_by":"auto","created_at":"2025-03-18 08:25:21","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":281047,"visible":true,"origin":"","legend":"\u003cp\u003eUnivariate analyses of serum SCFAs, inception cohort \u003cstrong\u003eA. Butyrate \u003c/strong\u003e(ANOVA P\u0026lt;0.001, CD vs HC q\u0026lt;0.001, UC vs HC q\u0026lt;0.001, CD vs UC q=0.58). \u0026nbsp;\u003cstrong\u003eB. Lactate \u003c/strong\u003e(ANOVA P=0.0049, CD vs HC q=0.11, UC vs HC q=0.0034, CD vs UC q=0.11). \u003cstrong\u003eC. Isobutyrate \u003c/strong\u003e(ANOVA P=0.029, CD vs HC q=0.054, UC vs HC q=0.032, CD vs UC q=0.80). \u003cstrong\u003eD. 2-Methylbutyrate \u003c/strong\u003e(ANOVA P=0.012, CD vs HC q=0.24, UC vs HC q=0.011, CD vs UC q=0.13). \u003cstrong\u003eE. Isovalerate \u003c/strong\u003e(ANOVA P=0.029, CD vs HC q=0.054, UC vs HC q=0.032, CD vs UC q=0.80).\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/2662c95485bd8aee77865bd9.png"},{"id":78734573,"identity":"4c4f2be7-388e-4bf9-82e2-9771f131b3f9","added_by":"auto","created_at":"2025-03-18 08:01:21","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":303535,"visible":true,"origin":"","legend":"\u003cp\u003eUnivariate analyses of serum BAs inception cohort. \u003cstrong\u003eA. Glycochendeoxycholic acid \u003c/strong\u003e(ANOVA P=0.0065, CD vs HC q=0.014, UC vs HC q=0.0081, CD vs UC q=0.89). \u003cstrong\u003eB. 5-Cholenic Acid-3-beta-ol \u003c/strong\u003e(ANOVA P=0.0071, CD vs HC q=0.011, UC vs HC q=0.010, CD vs UC q=0.98). \u003cstrong\u003eC. Glycochenodeoxycholic Acid 3-Sulfate \u003c/strong\u003e(ANOVA P=0.059, CD vs HC q=0.078, UC vs HC q=0.072, CD vs UC q=0.90). \u003cstrong\u003eD. Glycocholic Acid\u003c/strong\u003e (ANOVA P=0.0072, CD vs HC q=0.19, UC vs HC q=0.0062, CD vs UC q=0.13). \u003cstrong\u003eE. Murocholic Acid \u003c/strong\u003e(ANOVA P=0.0082, CD vs HC q=0.45, UC vs HC q=0.070, CD vs UC q=0.0099). \u003cstrong\u003eF. Taurocholic Acid \u003c/strong\u003e(ANOVA P=0.017, CD vs HC q=0.58, UC vs HC q=0.029, CD vs UC q=0.059).\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/c0debe76da7aa08c8c4d7c3a.png"},{"id":78735609,"identity":"b78f1998-701f-46e9-9bb1-347def31ee69","added_by":"auto","created_at":"2025-03-18 08:09:21","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":392742,"visible":true,"origin":"","legend":"\u003cp\u003eA: Distribution of correlation values across disease and health states B: Venn-diagram of strong positive (r \u0026gt;= 0.5) associations across cohorts. C: Number of strong positive associations per genera and state D: Genera with the most variable strong positive associations across states E: KEGG Pathway enrichment of strong positive associations across states F: Associations strongly positively correlated with health and negatively correlated with ulcerative colitis and/or Crohn's disease.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/cfdf367402250cfbad1c2a6c.png"},{"id":87848634,"identity":"41ed223d-97d6-453b-8133-7feb94c21870","added_by":"auto","created_at":"2025-07-29 15:37:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3471753,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/69d288e7-96d1-4979-8e40-77966a4d8fca.pdf"},{"id":78734561,"identity":"826a3510-c4c4-492a-9f57-0a9778208341","added_by":"auto","created_at":"2025-03-18 08:01:20","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":24087,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Appendices\u003c/p\u003e","description":"","filename":"SupplementaryAppendices.docx","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/bb9c88b8a3eddffdfaf4aa53.docx"},{"id":78734564,"identity":"c5b4d9d7-4db5-4b21-9dd6-465a84b0bcab","added_by":"auto","created_at":"2025-03-18 08:01:21","extension":"jpg","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":271367,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cu\u003eSupplementary Figure 1:\u003c/u\u003e \u003cem\u003eMicrobe-metabolite correlations for healthy controls.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"SupplementaryFigure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/a3584558fafde4eaaabab1e1.jpg"},{"id":78734574,"identity":"a93e14f0-9d89-48dd-8aed-1c593b364c1b","added_by":"auto","created_at":"2025-03-18 08:01:21","extension":"jpg","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":179656,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cu\u003eSupplementary Figure 2:\u003c/u\u003e \u003cem\u003eMicrobe-metabolite correlations for CD patients.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"SupplementaryFigure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/946d093e3996981ee6600835.jpg"},{"id":78735607,"identity":"34e6570a-a425-4fed-841a-f2d7e1dffa49","added_by":"auto","created_at":"2025-03-18 08:09:21","extension":"jpg","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":132545,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cu\u003eSupplementary Figure 3:\u003c/u\u003e \u003cem\u003eMicrobe-metabolite correlations for UC patients.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"SupplementaryFigure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/c506e0af14b5031427a8f8e0.jpg"},{"id":78734566,"identity":"e0750b83-d2ae-42e2-9b3e-bdef494ef409","added_by":"auto","created_at":"2025-03-18 08:01:21","extension":"jpg","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":218999,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cu\u003eSupplementary Figure 4:\u003c/u\u003e\u003cem\u003e Shared anti-correlating microbe-metabolite pairs shown on scatterplots, CD and UC, respectively. Enrichment analysis of the associations to the top left quadrant (positively associated with health, negative with UC/CD, highlighted in blue).\u003c/em\u003e\u003c/p\u003e","description":"","filename":"SupplementaryFigure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6232048/v1/90666ff15b274e679974d40b.jpg"}],"financialInterests":"The authors declare potential competing interests as follows: STR has received conference fees from Pfizer and Vifor with advisory fees from Galapagos. BHM has received consultancy fees from Finch Therapeutics Group, Akebia Therapeutics Inc, and Ferring Pharmaceuticals, and speaker fees from Yakult. SB has received conference fees from Dr Falk and Ferring. RWP declares conference fee support from Ferring. NP declares advisory and/or speaker fees from Abbvie, Allergan, Celgene, Debiopharm, Ferring and Vitor Pharma, and lecture fees from Allergan, Dr Falk Pharma, Janssen, Takeda and Tillotts Pharma. JRM has been paid for consultancy by Cultech Ltd and Enterobiotix Ltd. JLA has received conference fees from Celltrion, Tillots Pharma and Takeda with speaker fees from Abbvie and Johnson \u0026 Johnson. HRTW has received financial support to attend conferences from Takeda and has been an advisory board member for Pfizer.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eDeciphering the microbiome–metabolome landscape of an inflammatory bowel disease inception cohort\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eInflammatory bowel diseases (IBD) are chronic, relapsing-remitting conditions primarily affecting the gastrointestinal (GI) tract. The main two subtypes of IBD are Crohn\u0026rsquo;s disease (CD) and ulcerative colitis (UC). Although the etiopathogenesis of IBD development is yet to be fully understood, a combination of environmental, genetic and immunological factors appears to play important roles\u003csup\u003e(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e)\u003c/sup\u003e. The gut microbiota plays a crucial role in the etiopathogenesis and clinical trajectory of IBD, with IBD patients exhibiting an over-representation of traditionally pro-inflammatory microbes in addition to an overall reduction in microbial diversity\u003csup\u003e(\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e)\u003c/sup\u003e. Pro-inflammatory bacteria are thought to adversely impact upon the mucus barrier layer of the bowel, a proposed mechanism of intestinal inflammation\u003csup\u003e(\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e)\u003c/sup\u003e. Prior data document a reduction in α diversity within the gut microbiome of patients suffering from terminal ileal Crohn's disease (CD) when compared to healthy controls (HCs)\u003csup\u003e(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e)\u003c/sup\u003e. However, when comparing patients with ulcerative colitis (UC) to those with CD, such significant differences in α diversity have not been observed\u003csup\u003e(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003ePrevious studies have characterized the gut microbiota in the context of IBD, however, most of these studies have focused on patients who have been previously diagnosed with IBD and have been exposed to numerous previous treatments\u003csup\u003e(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e)\u003c/sup\u003e. It is therefore difficult to ascertain if the \u0026ldquo;microbiota fingerprint\u0026rdquo; associated with clinical IBD phenotypes is a cause, a consequence (e.g., related to medication use), or an epiphenomenon.\u003c/p\u003e \u003cp\u003eStudies leveraging multi-omics approaches, notably metabolomics, have revealed that changes in microbial composition are closely linked with shifts in the production of microbial metabolites, many of which have immunomodulatory properties\u003csup\u003e(\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e)\u003c/sup\u003e. Furthermore, recent data have shown that a multi-omics approach outperforms single-omic analyses in predicting outcomes in IBD using machine learning\u003csup\u003e(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAnalyses of \u0026lsquo;newly diagnosed\u0026rsquo; and untreated IBD cohorts have yielded interesting results. Multi-omic approaches have revealed that patients with CD characterized by \u0026ldquo;dysbiosis\u0026rdquo; exhibit a marked reduction in butyrate and secondary bile acids (BAs) such as lithocholate and deoxycholate compared to those with milder inflammation\u003csup\u003e(\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e)\u003c/sup\u003e. However, this study primarily involved pediatric patients, and there were instances where some CD patients did not contribute baseline samples. Reduced concentrations of SCFAs have been shown in UC patients\u003csup\u003e(\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e)\u003c/sup\u003e; an increased urinary concentration of xanthurenic acid has also been associated with severe UC\u003csup\u003e(\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e)\u003c/sup\u003e. In another study, thirty-one microbial species and thirteen metabolites were found to effectively distinguish IBD from HCs\u003csup\u003e(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e)\u003c/sup\u003e. These studies have also highlighted the difficulty of obtaining unadulterated samples from a treatment-na\u0026iuml;ve population. One study included patients with previous surgical resection for IBD, therefore this was not a truly treatment-na\u0026iuml;ve cohort\u003csup\u003e(\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e)\u003c/sup\u003e; another included sample collection after initial colonoscopy\u003csup\u003e(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e)\u003c/sup\u003e which may have affected results, as bowel purgatives may impact the gut microbiota and metabolite composition\u003csup\u003e(\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTryptophan metabolites (TRPm) have generated great interest in IBD: the gut microbiota are known to be important in the metabolism pathways of TRPm\u003csup\u003e(\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e)\u003c/sup\u003e. Previous data have shown an increased serum concentration of quinolinic acid in IBD patients compared to healthy controls\u003csup\u003e(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e)\u003c/sup\u003e, with a recent publication showing mucosal inflammation was associated with reduced concentrations of xanthurenic and kynurenic acid\u003csup\u003e(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eNetwork biology analyses have previously been used by large clinical cohorts to integrate microbiome data with other data modalities - including metabolomics, metagenomics, transcriptomics and clinical data - to help build hypotheses about disease mechanisms\u003csup\u003e(\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e)\u003c/sup\u003e. Theoretically, interrogating personalized networks in individual patients could reveal novel biomarkers of interest in addition to therapeutic targets that would be individualized to each patient\u003csup\u003e(\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThere have been very few studies documenting microbial composition and functionality in newly diagnosed IBD patients, where samples were obtained prior to any endoscopic intervention, formal diagnosis, or (most importantly) treatment. Here, we present the gut microbial composition and functionality of an inception cohort of IBD prior to formal diagnosis. We find distinctive microbiome and metabolomic signatures in a true inception IBD cohort that are different from established IBD cohorts and integrate these data in network biology analyses.\u003c/p\u003e \u003cp\u003eThe aims of this study were to generate novel potential insights into IBD pathogenesis by:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eDocumenting differences in the gut microbiome and metabolome between treatment-na\u0026iuml;ve, newly diagnosed CD, UC patients and healthy controls, using samples obtained before diagnosis.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eUndertaking integrated microbiome-metabolome network-level analyses by focusing on those interactions that were lost or gained when comparing CD, UC and healthy controls.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy participants and sample collection:\u003c/h2\u003e \u003cp\u003ePatients with symptoms suggestive of IBD, but no formal diagnosis, were consecutively recruited in a single Gastroenterology unit in London, United Kingdom. Patients were excluded if they had a prior diagnosis of IBD. Healthy controls (HCs) were also consecutively recruited independently - these were healthy individuals with no medical conditions and on no regular medication.\u003c/p\u003e \u003cp\u003e Research ethics approval was obtained (REC reference 18/LO/1207, protocol number 18SM4548, IRAS ID 243310). Both HCs and patients were excluded if antibiotics were administered within the previous three months.\u003c/p\u003e \u003cp\u003eDemographic and clinical data, and biofluid samples, were collected prospectively prior to any treatment, bowel preparation, endoscopy or formal diagnosis. Whole fecal samples were collected with a FECOTAINER\u0026reg; fecal collection device\u003csup\u003e(\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e)\u003c/sup\u003e. Serum samples were obtained by venesection with subsequent centrifugation and were stored in microcentrifuge tubes. Midstream urine samples were collected in universal containers; urinary sediment was separated using centrifugation and urine was stored in microcentrifuge tubes. Samples were also collected from HCs. All samples collected from patients and HCs were immediately stored at 4\u0026deg;C, for a maximum of 30 minutes, prior to processing and storage at -80\u0026deg;C. Participant phenotypic data were collected including age, weight, height, ethnicity, sex and dietary information. To document IBD disease severity, the Harvey Bradshaw index (HBI) (for CD), and simple clinical colitis activity index (SCCAI) (for UC) were recorded\u003csup\u003e(\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e) (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e)\u003c/sup\u003e. Fecal samples were processed for fecal calprotectin (FC) values in recruited patients, prior to endoscopic evaluation.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMicrobial DNA extraction:\u003c/h3\u003e\n\u003cp\u003eMicrobial DNA was extracted from crude fecal samples using the DNeasy PowerLyser PowerSoil Kit (Qiagen, Hilden, Germany) using the manufacturer\u0026rsquo;s established protocols. An additional step was undertaken to homogenize samples using Bullet Blender Storm bead beater (Chembio, St Alban\u0026rsquo;s, UK). Total microbial biomass within each sample was calculated using quantitative real-time polymerase chain reaction (qPCR) analyses. BactQuant assay was chosen given the comprehensive coverage of actual bacterial DNA\u003csup\u003e(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e)\u003c/sup\u003e. qPCR analyses were undertaken to enable transformation of compositional metataxonomic data into ecosystem abundance\u003csup\u003e(\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e)\u003c/sup\u003e, removing the need for rarefaction\u003csup\u003e(\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e)\u003c/sup\u003e. Sample libraries were prepared following Illumina\u0026rsquo;s 16S Metagenomic Sequencing Library Preparation Protocol using specifically designed V1-V2 hypervariable region primers, as detailed in \u003cb\u003eAppendix 1\u003c/b\u003e\u003csup\u003e(\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e\n\u003ch3\u003eNuclear Magnetic Resonance Spectroscopy (H NMR) methods:\u003c/h3\u003e\n\u003cp\u003eTo analyze fecal and urine samples, buffer (created using previously detailed protocols\u003csup\u003e(\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e)\u003c/sup\u003e) and samples were combined in a 1:9 ratio (buffer: sample) for final analyses. Serum analyses were undertaken using previously described protocols\u003csup\u003e(\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e)\u003c/sup\u003e. All biofluid samples were prepared for analyses as per previous protocols\u003csup\u003e(\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e)\u003c/sup\u003e. Sample extracts were analyzed with a Bruker 600 MHz AVANCE II \u003csup\u003e1\u003c/sup\u003eH-NMR spectrometer, at 300 Kelvin (K). 1D \u003csup\u003e1\u003c/sup\u003eH NMR spectra were acquired using a standard one-dimensional pulse sequence, with saturation of the water resonance (noesygppr1d pulse program) during both the relaxation delay (RD\u0026thinsp;=\u0026thinsp;4s) and mixing time (tm\u0026thinsp;=\u0026thinsp;10 ms). In total, 4 dummy scans, 128 scans and 64 K data points were collected. Further information concerning \u003csup\u003e1\u003c/sup\u003eH NMR parameters is given in \u003cb\u003eAppendix 2\u003c/b\u003e.\u003c/p\u003e\n\u003ch3\u003eLiquid Chromatography Mass Spectroscopy (LC-MS) methods:\u003c/h3\u003e\n\u003cp\u003eFecal water was prepared from lyophilized stool as per previously published protocols\u003csup\u003e(\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e)\u003c/sup\u003e. Urine and serum LC-MS experiments were undertaken at the National Phenome Centre (NPC) and followed previously described protocols\u003csup\u003e(\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e)\u003c/sup\u003e to obtain targeted and global profiles for the serum, urine and fecal samples. Targeted assays included tryptophan (TRP) metabolites and bile acids (BA). To identify and quantify fecal and serum short chain fatty acids (SCFAs), previously published protocols were followed\u003csup\u003e33\u003c/sup\u003e. Global analyses were completed with fecal reverse phase (RP) negative and positive phases were used in addition to urinary RP negative phase.\u003c/p\u003e\n\u003ch3\u003eData processing:\u003c/h3\u003e\n\u003cp\u003e16S rRNA gene amplicon sequencing data processing was carried out using the DADA2 pipeline (v.1.18)\u003csup\u003e(\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e)\u003c/sup\u003e. Data were analyzed using R studio environment to construct the phylogenetic tree using the Phangorn package with default settings\u003csup\u003e(\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e)\u003c/sup\u003e. A suite of R packages were used to analyze the microbiome sequencing data including Phyloseq\u003csup\u003e(\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e)\u003c/sup\u003e, Vegan\u003csup\u003e(\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e)\u003c/sup\u003e and ggplot2\u003csup\u003e(\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003csup\u003e1\u003c/sup\u003eH-NMR data were analyzed in SIMCA\u0026reg; (version 17, Sartorius, Sweden) and MATLAB (2014a, MathWorks) to visualize the data. The individual spectra of these metabolites were identified, as previously described, after cross-referencing internal and external databases\u003csup\u003e(\u003cspan additionalcitationids=\"CR37\" citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eUntargeted LC-MS data were processed using \u0026ldquo;peakPantheR\u0026rdquo;, an established R package to identify metabolites as per previous publications\u003csup\u003e(\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e)\u003c/sup\u003e. LC-MS data were processed using the online platform MetaboAnalyst (version 5.0)\u003csup\u003e(\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eStatistical and network analyses:\u003c/h2\u003e \u003cp\u003eNon-parametric ANOVAs with Kruskal-Wallis test were used to study the significance of differences in α diversity, taxonomic abundance and metabolite concentration between cohorts. Principal coordinate analysis (PCoA) was used to analyze β diversity, with permutational multivariate analysis of variance (PERMANOVA) to assess statistical significance. Benjamini-Hochberg false discovery rate (FDR) adjustment was used for P value correction where multiple comparisons were performed\u003csup\u003e(\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTo ascertain which metabolites were responsible for the differences observed, loadings plots for each orthogonal projections to latent structures discriminant analysis (OPLS-DA) were created in MATLAB\u0026reg; which enabled identification of the metabolites at extreme ends of the loading plot, driving the differences noted. Semi-targeted \u003csup\u003e1\u003c/sup\u003eH-NMR metabolite differences were compared across cohorts using column analyses after log-transformation. Non-parametric ANOVAs were used to compare the mean rank of each cohort with Kruskal-Wallis tests with FDR P value correction.\u003c/p\u003e \u003cp\u003eThe processed fecal metabolite and bacterial abundance data were read into R (version 4.4.1). The individual microbial samples were aggregated to a genus level. Strains present in at least 20% of all patients were removed to filter out overly promiscuous strains.\u003c/p\u003e \u003cp\u003eMicrobial network analyses were formulated by calculating the Spearman correlation coefficient between the measured fecal metabolites and the bacterial amplicon sequencing variants (ASVs) for each disease state (UC, CD, Healthy), using the rcorr function from the Hmisc package (version 5.1) in R (version 4.4.1). Results were filtered for significance (FDR \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;=\u0026thinsp;0.05) and magnitude (\u003cem\u003er\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;=\u0026thinsp;0.5). Visualization was carried out with the ggplot\u003csup\u003e(\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e)\u003c/sup\u003e, ggraph\u003csup\u003e(\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e)\u003c/sup\u003e and tidygraph\u003csup\u003e(\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e)\u003c/sup\u003e packages.\u003c/p\u003e \u003cp\u003eThe corresponding bacterial ASVs were collapsed into their respective taxonomic genera, resulting in state specific, strong bacterial genus \u0026ndash; metabolite associations. Functional enrichment analysis of the strongly correlating microbiota associated metabolites was carried out using the MetaboAnalyst web server\u003csup\u003e(\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e)\u003c/sup\u003e. The enrichment ratio was computed by dividing the number of expected hits with the observed hits, as returned by MetaboAnalyst. For the analysis of matching associations with opposite correlations across states, the median of correlation values was calculated for each genus \u0026ndash; metabolite pair (to capture the behavior of genera with multiple associating strains), and the results were visualized on a scatterplot (\u003cb\u003eSupplementary Fig.\u0026nbsp;1\u0026ndash;3\u003c/b\u003e). KEGG pathway\u003csup\u003e(\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e)\u003c/sup\u003e enrichment analysis of the appropriate anticorrelation clusters found on the scatterplot (i.e. Healthy\u0026thinsp;\u0026gt;\u0026thinsp;=\u0026thinsp;0.4 \u0026amp; CD \u0026lt;= -0.4) was carried out using the MetaboAnalyst web server (\u003cb\u003eSupplementary Fig.\u0026nbsp;4\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eCorrected \u003cem\u003eP\u003c/em\u003e values (\u003cem\u003eq\u003c/em\u003e values) of \u0026lt;\u0026thinsp;0.1 were deemed to be significant given the exploratory nature of these analyses, with previous comparable microbiota datasets using a \u003cem\u003eq\u003c/em\u003e value of \u0026lt;\u0026thinsp;0.1 as significant\u003csup\u003e(\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eNode rewiring capturing bacteria with the most variable interaction profiles was calculated using the DyNet app\u003csup\u003e(\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e)\u003c/sup\u003e (version: 1.0) in Cytoscape\u003csup\u003e(\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e)\u003c/sup\u003e (version: 3.10.) with default settings (undirected networks).\u003c/p\u003e \u003cp\u003eThe 16S rRNA gene DNA data have been deposited at the European Bioinformatics Institute\u0026rsquo;s (EBI) European Nucleotide Archive ENA database under the accession number PRJEB86986.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003ePatient phenotypes\u003c/h2\u003e \u003cp\u003eIn total 80 participants were recruited, comprising 60 IBD (23 CD and 37 UC) patients, newly diagnosed with IBD using established biochemical, endoscopic, histological, and radiological criteria\u003csup\u003e(\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e)\u003c/sup\u003e, and 20 HCs. Between HCs and patients with IBD, there were no significant differences in covariates such as age, BMI, diet, ethnicity, sex or smoking status. There was no significant difference in the median Fecal Calprotectin (FC) value between CD and UC patients. Additionally, the median disease activity index (DAI) score did not significantly differ between UC and CD patients - as measured by SCCAI score and HBI scores respectively. These data are displayed in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003ePhenotypic characteristics of inception cohort at baseline\u003c/em\u003e.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGeneral Characteristics\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eUC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eP value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of participants\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge range, (median)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18\u0026ndash;66, (35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18\u0026ndash;75, (32)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e26\u0026ndash;54, (34)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.713*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex (M/F)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(M\u0026thinsp;=\u0026thinsp;13/F\u0026thinsp;=\u0026thinsp;10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(M\u0026thinsp;=\u0026thinsp;16/F\u0026thinsp;=\u0026thinsp;21)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(M\u0026thinsp;=\u0026thinsp;12/F\u0026thinsp;=\u0026thinsp;8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.403\u003csup\u003eφ\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI range group (\u0026lt;\u0026thinsp;18.5/18.6\u0026ndash;24.9/\u0026gt;25.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(5/13/5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(6/22/9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(1/18/1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.131\u003csup\u003eφ\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiet (Meat eater/pescetarian/vegetarian)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(17/3/3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(27/6/4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(14/5/1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.790\u003csup\u003eφ\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEthnicity (Caucasian/non-Caucasian)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(12/11)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(26/11)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(12/8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.358\u003csup\u003eφ\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmoking status (current/non-smoker)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1/22)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2/35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(1/19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.983\u003csup\u003eφ\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCalprotectin median, (range)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e796, (259\u0026ndash;2255)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1269, (258\u0026ndash;7900)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.146\u003csup\u003el\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDAI median, (range)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8, (4\u0026ndash;14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7, (4\u0026ndash;15)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.713\u003csup\u003el\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eKEY\u003c/em\u003e: *=One way ANOVA; φ\u0026thinsp;=\u0026thinsp;Chi-square test; l\u0026thinsp;=\u0026thinsp;Mann-Whitney U\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIBD patients were classified as per the Montreal classification\u003csup\u003e(\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e)\u003c/sup\u003e. These data are displayed in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e2\u003c/span\u003eB.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cb\u003eA.\u003c/b\u003e \u003cem\u003eUC cohort characteristics by Montreal classification\u003c/em\u003e\u003csup\u003e\u003cem\u003e(\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e)\u003c/em\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExtent of UC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eE1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eE2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eE3\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of participants*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003e\u003cem\u003eKEY\u003c/em\u003e: *=FC value between phenotypes was not statistically significant (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.65)\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cb\u003eB.\u003c/b\u003e \u003cem\u003eCD Cohort characteristics by Montreal classification\u003c/em\u003e\u003csup\u003e\u003cem\u003e(\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e)\u003c/em\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCrohn\u0026rsquo;s cohort demographics\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLocation of CD*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eL1\u0026thinsp;=\u0026thinsp;10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eL2\u0026thinsp;=\u0026thinsp;4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eL3\u0026thinsp;=\u0026thinsp;9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBehavior of CD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eB1\u0026thinsp;=\u0026thinsp;16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eB2\u0026thinsp;=\u0026thinsp;3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eB3\u0026thinsp;=\u0026thinsp;4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEvidence of perianal disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNO\u0026thinsp;=\u0026thinsp;19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYES\u0026thinsp;=\u0026thinsp;4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003e\u003cem\u003eKEY\u003c/em\u003e: *=FC value between phenotypes was not statistically significant (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.40)\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eA total of 238 samples were analyzed: 78 fecal samples, 80 serum and 80 urine samples.\u003c/p\u003e \u003cp\u003e \u003cb\u003eGut microbiome diversity metrics and taxonomic features having a defined profile in an inception IBD cohort\u003c/b\u003e:\u003c/p\u003e \u003cp\u003eβ diversity between the cohorts, as plotted by PCoA, illustrated that HCs were significantly different to both CD and UC populations (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0018) (Fig.\u0026nbsp;1A). No significant difference in β diversity was demonstrated between CD and UC. Although the α diversity for HCs was increased compared to both UC and CD patients across multiple indices, these trends were not found to be statistically significant in a linear mixed model (Fig.\u0026nbsp;1B-D). At phylum level, the median abundance of Bacteroidetes was significantly different, with an overall ANOVA of \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0001. Across inter-cohort comparisons, Bacteroidetes abundance was significantly different, with enrichment noted in CD compared to both HC (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0001) and to UC (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). UC patients also demonstrated an increased median abundance of Bacteroidetes compared to HC (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). The overall ANOVA for Firmicutes median abundance also reached significance with a \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0011. Firmicutes were increased in both IBD cohorts compared to HCs, with CD\u0026thinsp;\u0026gt;\u0026thinsp;HC (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0008) and UC\u0026thinsp;\u0026gt;\u0026thinsp;HC (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.011). Although the median abundance of Firmicutes was higher in CD compared to UC, this observation did not reach statistical significance (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.17). Similar trends were observed in Proteobacteria abundance with a significant overall ANOVA \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0104. As expected, Proteobacteria median abundance was significantly higher in CD compared to HCs (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0075). This was replicated in a UC population who were enriched with Proteobacteria compared to HCs (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.043). Key phyla differences are presented in \u003cb\u003eFig.\u0026nbsp;2\u003c/b\u003e. Deeper analyses revealed that a particular Clostridia ASV from the Firmicutes phylum, identified as ASV455 (part of the Family XIII AD3011 genus group), was enriched in HCs compared to both CD and UC patients (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e \u003cp\u003e \u003cb\u003eMetabolomic profiles demonstrate distinct differences between controls and IBD in addition to between CD and UC\u003c/b\u003e \u003c/p\u003e \u003cp\u003eMultivariate analyses were undertaken on \u003csup\u003e1\u003c/sup\u003eH NMR data with principal component analyses (PCAs) demonstrating no outlying samples. OPLS-DA analyses comparing CD, UC and HC cohorts to each other demonstrated significant differences between HCs and each IBD cohort. OPLS-DA analyses between serum samples of CD and UC demonstrated significant differences between the cohorts, but similar findings were not demonstrated in fecal and urine samples. These data are listed in \u003cb\u003eAppendix 3\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eSemi-targeted \u003csup\u003e1\u003c/sup\u003eH NMR analyses, with individual metabolite spectra identified using previously described databases\u003csup\u003e(\u003cspan additionalcitationids=\"CR37\" citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e)\u003c/sup\u003e, demonstrated that serum N-acetylglucosamine/galactosamine (GlycA) and sialic acid (GlycB) were higher in both CD and UC compared to HCs. Serum GlycA was also at a higher concentration in CD patients compared to UC patients. A similar trend was observed with serum pyruvate at a higher concentration in both IBD cohorts compared to HCs (Fig.\u0026nbsp;3A-C). Fecal nicotinate was shown to be higher in CD patients (Fig.\u0026nbsp;3D). Urinary hippurate was noted to be higher in HCs compared to both CD and UC (Fig.\u0026nbsp;3E).\u003c/p\u003e \u003cp\u003eTargeted LC-MS analyses demonstrated multiple differences between IBD cohorts and HCs in addition to between CD and UC. Serum tryptophan metabolites were markedly different between IBD and healthy controls, additionally notable differences were demonstrated between CD and UC. Serum xanthurenic acid (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0092), picolinic acid (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.018), kynurenic acid (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.036) and 3-hydroxyanthranilic acid (3-HAA) (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.055) were lowest in CD patients compared to both HCs and UC patients. UC patients exhibited higher serum concentrations of neopterin (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.036) and quinolinic acid (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.055) compared to HCs. Additionally, serum 5-hydroxyindole acetic acid (5-HIAA) was consistently shown to be lowest in HCs compared to CD and UC patients (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). These findings are demonstrated in \u003cb\u003eFig.\u0026nbsp;4\u003c/b\u003e. Multiple differences were noted in serum SCFAs with isobutyrate (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.039), 2-methylbutyrate (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.039) and lactate all increased in UC patients compared to HCs. Serum butyrate (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) concentration and isovalerate (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.053) concentration were lowest in HCs compared to both CD and UC patients. These serum SCFA findings are shown in \u003cb\u003eFig.\u0026nbsp;5\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eSerum BAs analyses revealed multiple differences between each disease group. Serum glycochenodeoxycholic acid (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.014) and glycoursodeoxycholic acid 3-sulphate (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.077) concentration were shown to be lowest in HCs compared to both CD. A similar trend was observed with UC patients exhibiting a higher concentration of serum glycochenodeoxycholic acid (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0081) and glycoursodeoxycholic acid 3-sulphate (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.072) compared to HCs. Serum glycocholic acid was lower in HC compared to UC patients (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0062) although similar differences were not demonstrated between CD and HCs. Murocholic acid was shown to be lowest in UC patients compared to both CD (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0099) and to HCs (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.070). The opposite trend was demonstrated with serum taurocholic acid concentration, as this was highest in UC patient compared to both CD (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.059) and HCs (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.029). 5-cholenic acid-3-beta-ol was noted to be highest in HC compared to both CD (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.011) and UC patients (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.010). These serum BA findings are shown in \u003cb\u003eFig.\u0026nbsp;6\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eGlobal profiles using RP modes for fecal and urinary samples noted multiple metabolite concentration differences between HCs and IBD patients in addition to between CD and UC patients and these are listed in \u003cb\u003eAppendix 4\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003cb\u003eMicrobial\u003c/b\u003e\u0026ndash;\u003cb\u003emetabolite correlation network analyses reveal key differences between IBD and healthy controls\u003c/b\u003e\u003c/p\u003e \u003cp\u003eTo further explore and explain the microbiota and metabolomic differences, these data were integrated with correlation network analyses. We observed strong (r\u0026thinsp;\u0026gt;\u0026thinsp;=\u0026thinsp;0.5) microbe-metabolite correlations (Fig.\u0026nbsp;7A), unique to each condition, with the healthy and CD patients containing the most distinct correlations. The overlaps among groups were small, suggesting a loss of correlations associated with health, demonstrated in \u003cb\u003eFig.\u0026nbsp;7B\u003c/b\u003e. Multiple bacterial genera have strong associations in each cohort, with the genus \u003cem\u003eBacteroides, Faecalibacterium\u003c/em\u003e and \u003cem\u003eBlautia\u003c/em\u003e correlating with the most metabolites from genera present in all cohorts (Fig.\u0026nbsp;7C\u003cb\u003e)\u003c/b\u003e. \u003cem\u003eBacteroides, Faecalibacterium\u003c/em\u003e and \u003cem\u003eBlautia\u003c/em\u003e are the genera correlating with the most different metabolites (Fig.\u0026nbsp;7D), indicating that the genera with the largest number of correlations also associated with the most different metabolites across CD and UC patients, and healthy controls. These strong, positive condition-specific microbe\u0026ndash;metabolite correlations correspond to distinct functionally enriched metabolic KEGG pathways (Fig.\u0026nbsp;7E). Notably, the metabolites from the microbe\u0026ndash;metabolite correlations of UC patients are overrepresented in nitrogen and butanoate metabolism pathways, while CD patients are enriched in valine, leucine and isoleucine biosynthesis, and both IBD subtypes demonstrate caffeine metabolism pathways. The healthy controls are characterized by the phenylalanine, tyrosine and tryptophan biosynthesis and metabolism and alanine, aspartate and glutamate metabolism pathways, all of which are lacking in the IBD subtypes.\u003c/p\u003e \u003cp\u003eTo gain further confidence in the disease relevance of these associations, we analyzed microbe\u0026ndash;metabolite correlations, which were strongly positive in health but simultaneously negative in one or more IBD subtypes (Fig.\u0026nbsp;7F). KEGG pathway overrepresentation analysis of the metabolites from these matching microbe\u0026ndash;metabolite correlations between health and IBD indicated that the taurine and hypotaurine metabolism pathway was lost in both the CD and UC comparisons.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eHere we present the analyses of biosamples from a true inception cohort of IBD patients with matched healthy controls, demonstrating multiple microbial and metabolomic differences between UC, CD and health. Additionally, we document microbe-host interaction with network analyses, demonstrating numerous novel microbiome\u0026ndash;metabolome correlations that have not previously been well characterized in IBD.\u003c/p\u003e \u003cp\u003ePrevious studies have investigated the gut microbiome and metabolome in IBD, with IBD patients characterized by a reduction in microbial diversity\u003csup\u003e(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e)\u003c/sup\u003e, an enrichment in Proteobacteria\u003csup\u003e(\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e)\u003c/sup\u003e and a perceived \u0026ldquo;dysbiotic environment\u0026rdquo;, as well as a low bile acid concentration compared to a healthy population\u003csup\u003e(\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e)\u003c/sup\u003e. However, previous data have not usually been derived from true inception cohorts, with patients having undergone prior surgery or having been exposed to medical treatment at the time of sample collection. Additionally, prior study sampling was often obtained after endoscopic diagnosis (with the known effects of bowel purgatives upon the gut microbiota and metabolome)\u003csup\u003e(\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e)\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOur findings demonstrate notable inter-disease differences between CD and UC patients, including abundance of Bacteroidetes and numerous differences in serum TRPm concentration. In some areas, our findings corroborate previous data\u003csup\u003e(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e)\u003c/sup\u003e: TRPm concentration was increased in healthy controls compared to IBD patients; active IBD resulted in a reduced concentration of kynurenic and xanthurenic acid compared to quiescent disease. Of note, Michaudel \u003cem\u003eet al.\u003c/em\u003e did not find a difference in TRPm concentration between CD and UC patients as seen in our cohort. Utilizing correlation network analysis, we observed strong disease-specific microbe-metabolite correlations (r\u0026thinsp;\u0026ge;\u0026thinsp;0.5), with unique interactions primarily in healthy and CD patients and with minimal overlap between cohorts, suggesting a loss of health-associated interactions. Bacterial genera such as \u003cem\u003eBacteroides, Faecalibacterium\u003c/em\u003e, and \u003cem\u003eBlautia\u003c/em\u003e were linked to the most varied metabolites across conditions, and these strong positive correlations mapped to distinct metabolic KEGG pathways, with UC patients showing increased nitrogen and butanoate metabolism, CD increased valine, leucine, and isoleucine biosynthesis, and both IBD subtypes enhanced caffeine metabolism. Healthy controls showed enrichment in phenylalanine, tyrosine, tryptophan, and alanine-related pathways. The increase in butanoate metabolism pathways seen in UC patients may inform as to the nature of gut inflammation: previous data have identified butyrate as an inhibitor of neutrophils (isolated from IBD patients) secreting pro-inflammatory cytokines\u003csup\u003e(\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e)\u003c/sup\u003e. Further analysis of shared, but anticorrelating microbe-metabolite pairs revealed health-specific associations negatively correlated with IBD, identifying the taurine and hypotaurine metabolism pathway as activated in both CD and UC (\u003cb\u003eSupplementary Fig.\u0026nbsp;4\u003c/b\u003e). This finding correlates with recent data illustrating low serum taurine levels in both CD and UC compared to controls, implicating increased taurine metabolism\u003csup\u003e(\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e)\u003c/sup\u003e. Our novel finding of an increase in caffeine metabolism in IBD patients is of interest: previous murine models of colitis have illustrated the protective role of caffeine\u003csup\u003e(\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e)\u003c/sup\u003e, although to fully understand this interaction further prospective data with detailed caffeine intake linked to intestinal inflammation are required.\u003c/p\u003e \u003cp\u003eThe strengths of our data are numerous. Importantly, samples were obtained prior to endoscopic confirmation of disease (bowel purgatives having been shown to affect the gut microbial composition and metabolome)\u003csup\u003e(\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e)\u003c/sup\u003e. Additionally, as samples were obtained prior to any medical or surgical treatment (which have been shown to alter both microbes and metabolites)\u003csup\u003e(\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e)\u003c/sup\u003e, we have been able to present an unadulterated sample cohort. We were able to ensure almost complete sample donation from our entire cohort: 78 out of 80 participants were able to donate a full complement of fecal, serum and urine samples - this is far higher than in comparable clinical studies. Although our study recruited from one London center, there was heterogeneity in population age, ethnicity and sex allowing our findings to be applied to other diverse populations.\u003c/p\u003e \u003cp\u003eOur findings have combined the microbiota and metabolite findings in the correlation network analyses, thereby linking the findings to truly understand the relationship between the microbe and the host. Moving forward, our work should be tested in a larger population of newly diagnosed patients across multiple centers to confirm that our findings can be replicated. Additionally, further investigation of murine models and human organoids should be undertaken to understand the biological mechanisms of our findings.\u003c/p\u003e \u003cp\u003eIn conclusion, this is the first analysis to comprehensively integrate microbiota composition and metabolic function in the study of a truly treatment-na\u0026iuml;ve inception cohort of patients newly diagnosed with IBD. Using state-of-the-art network biology techniques, we have elucidated novel pathways which may be critical to host-microbe interactions in disease pathogenesis. Our findings will enable mechanistic studies to probe the impact of these microbial metabolites on aberrant host immune function in IBD and will better inform translational research using interventions including diet and microbial therapeutics such as intestinal microbiota transplant to modulate IBD disease course.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAcknowledgements:\u003c/h2\u003e \u003cp\u003eJPT is supported by the Chain Florey Clinical PhD Fellowship jointly funded by the National Institute for Health Research (NIHR) Imperial Biomedical Research Centre (BRC) and the UKRI Medical Research Council (MRC) Laboratory of Medical Sciences (LMS). DM acknowledges financial support from Imperial College London through an Imperial College Research Fellowship grant award. LG and TK were supported by the UKRI BBSRC Institute Strategic Programme Food Microbiome and Health BB/X011054/1 and its constituent project BBS/E/F/000PR13631.NP is supported by the Wellcome Trust (WT101159), Crohn\u0026rsquo;s and Colitis UK, and the NIHR Imperial Biomedical Research Centre (BRC). JVL is supported by Medical Research Council (MRC) New Investigator Research Grant (MR/P002536/1) and ERC Starting Grant (715662). STR, TK, JR, NP, JLA and HRTW are supported by the NIHR Imperial Biomedical Research Centre (BRC). The Division of Digestive Diseases at Imperial College London receives financial and infrastructure support from the NIHR Imperial BRC based at Imperial College Healthcare NHS Trust and Imperial College London. The views expressed are those of the authors and not necessarily those of the NIHR or the UK Department of Health and Social Care.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eZhang Y-Z, Li Y-Y (2014) Inflammatory bowel disease: pathogenesis. World J gastroenterology: WJG 20(1):91\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWlodarska M, Kostic AD, Xavier RJ (2015) An integrative view of microbiome-host interactions in inflammatory bowel diseases. Cell Host Microbe 17(5):577\u0026ndash;591\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFyderek K, Strus M, Kowalska-Duplaga K, Gosiewski T, Wedrychowicz A, Jedynak-Wasowicz U et al (2009) Mucosal bacterial microflora and mucus layer thickness in adolescents with inflammatory bowel disease. World J Gastroenterol 15(42):5287\u0026ndash;5294\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLarsson JM, Karlsson H, Crespo JG, Johansson ME, Eklund L, Sjovall H et al (2011) Altered O-glycosylation profile of MUC2 mucin occurs in active ulcerative colitis and is associated with increased inflammation. Inflamm Bowel Dis 17(11):2299\u0026ndash;2307\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eImhann F, Vich Vila A, Bonder MJ, Fu J, Gevers D, Visschedijk MC et al (2018) Interplay of host genetics and gut microbiota underlying the onset and clinical presentation of inflammatory bowel disease. Gut 67(1):108\u0026ndash;119\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorgan XC, Tickle TL, Sokol H, Gevers D, Devaney KL, Ward DV et al (2012) Dysfunction of the intestinal microbiome in inflammatory bowel disease and treatment. Genome Biol 13(9):R79\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFranzosa EA, Sirota-Madi A, Avila-Pacheco J, Fornelos N, Haiser HJ, Reinker S et al (2019) Gut microbiome structure and metabolic activity in inflammatory bowel disease. Nat Microbiol 4(2):293\u0026ndash;305\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLloyd-Price J, Arze C, Ananthakrishnan AN, Schirmer M, Avila-Pacheco J, Poon TW et al (2019) Multi-omics of the gut microbial ecosystem in inflammatory bowel diseases. Nature 569(7758):655\u0026ndash;662\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO'Sullivan J, Patel S, Leventhal GE, Fitzgerald RS, Laserna-Mendieta EJ, Huseyin CE et al (2025) Host-microbe multi-omics and succinotype profiling have prognostic value for future relapse in patients with inflammatory bowel disease. Gut Microbes 17(1):2450207\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVich Vila A, Hu S, Andreu-Sanchez S, Collij V, Jansen BH, Augustijn HE et al (2023) Faecal metabolome and its determinants in inflammatory bowel disease. Gut 72(8):1472\u0026ndash;1485\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAldars-Garcia L, Gil-Redondo R, Embade N, Riestra S, Rivero M, Gutierrez A et al (2024) Serum and Urine Metabolomic Profiling of Newly Diagnosed Treatment-Naive Inflammatory Bowel Disease Patients. Inflamm Bowel Dis 30(2):167\u0026ndash;182\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNing L, Zhou YL, Sun H, Zhang Y, Shen C, Wang Z et al (2023) Microbiome and metabolome features in inflammatory bowel disease via multi-omics integration analyses across cohorts. Nat Commun 14(1):7135\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJalanka J, Salonen A, Salojarvi J, Ritari J, Immonen O, Marciani L et al (2015) Effects of bowel cleansing on the intestinal microbiota. Gut 64(10):1562\u0026ndash;1568\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePowles STR, Gallagher KI, Chong LWL, Alexander JL, Mullish BH, Hicks LC et al (2022) Effects of bowel preparation on intestinal bacterial associated urine and faecal metabolites and the associated faecal microbiome. BMC Gastroenterol 22(1):240\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaeslack N, Mimmler M, Becker S, Gao Z, Khuu MP, Mann A et al (2022) Microbiota-derived tryptophan metabolites in vascular inflammation and cardiovascular disease. Amino Acids 54(10):1339\u0026ndash;1356\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNikolaus S, Schulte B, Al-Massad N, Thieme F, Schulte DM, Bethge J et al (2017) Increased Tryptophan Metabolism Is Associated With Activity of Inflammatory Bowel Diseases. Gastroenterology. ;153(6):1504-16 e2\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMichaudel C, Danne C, Agus A, Magniez A, Aucouturier A, Spatz M et al (2022) Rewiring the altered tryptophan metabolism as a novel therapeutic strategy in inflammatory bowel diseases. Gut\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eModos D, Bulusu KC, Fazekas D, Kubisch J, Brooks J, Marczell I et al (2017) Neighbours of cancer-related proteins have key influence on pathogenesis and could increase the drug target space for anticancer therapies. NPJ Syst Biol Appl 3:2\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMEDGADGET To Fill a Void, AT Medical Introduces Fecotainer 2011 [cited 2023 18/02/2023]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.medgadget.com/2011/09/to-fill-a-void-at-medical-introduces-fecotainer.html\u003c/span\u003e\u003cspan address=\"https://www.medgadget.com/2011/09/to-fill-a-void-at-medical-introduces-fecotainer.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarvey RF, Bradshaw JM, A SIMPLE INDEX OF, CROHN'S-DISEASE ACTIVITY (1980) Lancet 315(8167):514\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBest WR (2006) Predicting the Crohn_s Disease Activity Index From the Harvey-Bradshaw Index. Inflamm Bowel Dis 12(4):304\u0026ndash;310\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJowett SL, Seal CJ, Phillips E, Gregory W, Barton JR, Welfare MR (2003) Defining relapse of ulcerative colitis using a symptom-based activity index. Scand J Gastroenterol 38(2):164\u0026ndash;171\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu CM, Aziz M, Kachur S, Hsueh P-R, Huang Y-T, Keim P et al (2012) BactQuant- An enhanced broad-coverage bacterial quantitative real-time PCR assay. BMC Microbiol. ;12(56)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGloor GB, Macklaim JM, Pawlowsky-Glahn V, Egozcue JJ (2017) Microbiome Datasets Are Compositional: And This Is Not Optional. Front Microbiol 8:2224\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcMurdie PJ, Holmes S (2014) Waste not, want not: why rarefying microbiome data is inadmissible. PLoS Comput Biol 10(4):e1003531\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMullish BH, Pechlivanis A, Barker GF, Thursz MR, Marchesi JR, McDonald JAK (2018) Functional microbiomics: Evaluation of gut microbiota-bile acid metabolism interactions in health and disease. Methods 149:49\u0026ndash;58\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDona AC, Jimenez B, Schafer H, Humpfer E, Spraul M, Lewis MR et al (2014) Precision high-throughput proton NMR spectroscopy of human urine, serum, and plasma for large-scale metabolic phenotyping. Anal Chem 86(19):9887\u0026ndash;9894\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGratton J, Phetcharaburanin J, Mullish BH, Williams HRT, Thursz MR, Nicholson JK et al (2016) An optimized sample handling strategy for metabolic profiling of human feces. Anal Chem 88(9):4661\u0026ndash;4668\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLewis MR, Chekmeneva E, Camuzeaux S, Sands C, Yuen A, David M et al (2022) An Open Platform for Large Scale LC-MS-Based Metabolomics. ChemRxiv. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://chemrxiv.org/engage/chemrxiv/article-details/61ebd6fa0716a8529e3823dc\u003c/span\u003e\u003cspan address=\"https://chemrxiv.org/engage/chemrxiv/article-details/61ebd6fa0716a8529e3823dc\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWant EJ, Wilson ID, Gika H, Theodoridis G, Plumb RS, Shockcor J et al (2010) Global metabolic profiling procedures for urine using UPLC-MS. Nat Protoc 5(6):1005\u0026ndash;1018\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCallahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJ, Holmes SP (2016) DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods 13(7):581\u0026ndash;583\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchliep KP (2011) phangorn: phylogenetic analysis in R. Bioinformatics 27(4):592\u0026ndash;593\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcMurdie PJ, Holmes S (2013) phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE 8(4):e61217\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOksanen J, Gavin L, Simpson GL, Guillaume Blanchet F, Kindt R, Legendre P, Minchin PR et al Package \u0026lsquo;vegan\u0026rsquo; Community Ecology Package https://\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ecran.r-project.org/web/packages/vegan/vegan.pdf2022\u003c/span\u003e\u003cspan address=\"http://cran.r-project.org/web/packages/vegan/vegan.pdf2022\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [updated 17/04/. 2.6-2:[Available from: https://cran.r-project.org/web/packages/vegan/vegan.pdf\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickham H (2016) ggplot2 Elegant Graphics for Data Analysis. Second eds. Gentleman R, Hornik K, Parmigiani G (eds): Springer Nature\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWishart DS, Feunang YD, Marcu A, Guo AC, Liang K, Vazquez-Fresno R et al (2018) HMDB 4.0: the human metabolome database for 2018. Nucleic Acids Res 46(D1):D608\u0026ndash;D17\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoch JC, Baskaran K, Burr H, Chin J, Eghbalnia HR, Fujiwara T et al (2023) Biological Magnetic Resonance Data Bank. Nucleic Acids Res 51(D1):D368\u0026ndash;D76\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarcia-Perez I, Posma JM, Serrano-Contreras JI, Boulange CL, Chan Q, Frost G et al (2020) Identifying unknown metabolites using NMR-based metabolic profiling techniques. Nat Protoc 15(8):2538\u0026ndash;2567\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWolfer AM, Correia GDS, Sands CJ, Camuzeaux S, Yuen AHY, Chekmeneva E et al (2021) peakPantheR, an R package for large-scale targeted extraction and integration of annotated metabolic features in LC-MS profiling datasets. Bioinformatics 37(24):4886\u0026ndash;4888\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePang Z, Chong J, Zhou G, de Lima Morais DA, Chang L, Barrette M et al (2021) MetaboAnalyst 5.0: narrowing the gap between raw spectra and functional insights. Nucleic Acids Res 49(W1):W388\u0026ndash;W96\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBenjamini Y, Hochberg Y (1995) Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J Roy Stat Soc 57(1):289\u0026ndash;300\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePedersen TL An Implementation of Grammar of Graphics for Graphs and Networks. R package version 2.2.1.9000 2024 [Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://ggraph.data-imaginist.com\u003c/span\u003e\u003cspan address=\"https://ggraph.data-imaginist.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePedersen TL tidygraph: A Tidy API for Graph Manipulation. R package version 1.3.1.9000 2024 [Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://tidygraph.data-imaginist.com\u003c/span\u003e\u003cspan address=\"https://tidygraph.data-imaginist.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePang Z, Lu Y, Zhou G, Hui F, Xu L, Viau C et al (2024) MetaboAnalyst 6.0: towards a unified platform for metabolomics data processing, analysis and interpretation. Nucleic Acids Res 52(W1):W398\u0026ndash;W406\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKanehisa M, Araki M, Goto S, Hattori M, Hirakawa M, Itoh M et al (2008) KEGG for linking genomes to life and the environment. Nucleic Acids Res 36(Database issue):D480\u0026ndash;D484\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRouty B, Lenehan JG, Miller WH Jr., Jamal R, Messaoudene M, Daisley BA et al (2023) Fecal microbiota transplantation plus anti-PD-1 immunotherapy in advanced melanoma: a phase I trial. Nat Med 29(8):2121\u0026ndash;2132\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoenawan IH, Bryan K, Lynn DJ (2016) DyNet: visualization and analysis of dynamic molecular interaction networks. Bioinformatics 32(17):2713\u0026ndash;2715\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D et al (2003) Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 13(11):2498\u0026ndash;2504\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaaser C, Sturm A, Vavricka SR, Kucharzik T, Fiorino G, Annese V et al (2019) ECCO-ESGAR Guideline for Diagnostic Assessment in IBD Part 1: Initial diagnosis, monitoring of known IBD, detection of complications. J Crohns Colitis 13(2):144\u0026ndash;164\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilverberg MS, Satsangi J, Ahmad T, Arnott IDR, Bernstein CN, Brant SR et al (2005) Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: Report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol 19:5A\u0026ndash;36A\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi G, Lin J, Zhang C, Gao H, Lu H, Gao X et al (2021) Microbiota metabolite butyrate constrains neutrophil functions and ameliorates mucosal inflammation in inflammatory bowel disease. Gut Microbes 13(1):1968257\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrascatani R, Mattogno A, Iannucci A, Marafini I, Monteleone G (2024) Reduced Taurine Serum Levels in Inflammatory Bowel Disease. Nutrients. ;16(11)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMizoguchi E, Sadanaga T, Okada T, Minagawa T, Akiba J (2023) Does caffeine have a double-edged sword role in inflammation and carcinogenesis in the colon? Intest Res 21(3):306\u0026ndash;317\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRadhakrishnan ST, Alexander JL, Mullish BH, Gallagher KI, Powell N, Hicks LC et al (2022) Systematic review: the association between the gut microbiota and medical therapies in inflammatory bowel disease. Aliment Pharmacol Ther 55(1):26\u0026ndash;48\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"Imperial College London","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6232048/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6232048/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe gut microbiota contributes to the etiopathogenesis of inflammatory bowel disease (IBD), but limitations of prior studies include the use of sequencing alone (restricting exploration of the contribution of microbiota functionality) and the recruitment of patients with well-established disease (introducing potential confounders, such as immunomodulatory medication). Here, we analyze a true IBD inception cohort and matched healthy controls (HCs) via stool 16S rRNA gene sequencing and multi-system metabolomic phenotyping (using nuclear magnetic spectroscopy and mass spectroscopy), with subsequent integrative network analysis employed to delineate novel microbiota-metabolome interactions in IBD. Marked differences in β diversity and taxonomic profiles were observed both between IBD and HCs, as well as between Crohn\u0026rsquo;s disease (CD) and ulcerative colitis (UC) patients. Multiple between-group metabolomic differences were also observed, particularly related to tryptophan-/indole-related metabolites; for example, UC patients had higher levels of serum metabolites including xanthurenic acid (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0092) and picolinic acid (\u003cem\u003eq\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.018). Network analysis demonstrated multiple unique interactions in CD compared to HCs with minimal overlap, indicating a loss of \u0026lsquo;health-associated\u0026rsquo; interactions in CD. Compared to HCs, UC patients demonstrated increased pathway activity related to nitrogen and butanoate metabolism, whilst CD patients displayed increased leucine and valine synthesis. Networks from IBD patients overall showed negative correlation with health-specific associations, including an increase in taurine metabolism. Collectively, this work characterizes multiple novel perturbed microbiota-metabolome interactions that are present even at the diagnosis of IBD, which may inform potential future targets to aid diagnosis and direct therapeutic options.\u003c/p\u003e","manuscriptTitle":"Deciphering the microbiome–metabolome landscape of an inflammatory bowel disease inception cohort","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-18 08:01:15","doi":"10.21203/rs.3.rs-6232048/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9f3ccb65-369a-47c5-af5d-20ce5f50556c","owner":[],"postedDate":"March 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":45729988,"name":"General Microbiology"},{"id":45729989,"name":"Gastroenterology \u0026 Hepatology"}],"tags":[],"updatedAt":"2025-07-29T15:36:55+00:00","versionOfRecord":{"articleIdentity":"rs-6232048","link":"https://doi.org/10.1080/19490976.2025.2527863","journal":{"identity":"gut-microbes","isVorOnly":true,"title":"Gut Microbes"},"publishedOn":"2025-07-18 00:00:00","publishedOnDateReadable":"July 18th, 2025"},"versionCreatedAt":"2025-03-18 08:01:15","video":"","vorDoi":"10.1080/19490976.2025.2527863","vorDoiUrl":"https://doi.org/10.1080/19490976.2025.2527863","workflowStages":[]},"version":"v1","identity":"rs-6232048","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6232048","identity":"rs-6232048","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}