{"paper_id":"13a8e514-e461-4f7e-a8f8-ddcaa7137d41","body_text":"1 \nTitle: Fecal Microbial and Metabolic Signatures in VEO-IBD: Implications for Unique \nPathophysiology \n \nAuthors: \nKine Eide Kvitne1, Simone Zuffa1,2, Vincent Charron-Lamoureux1,2, Ipsita Mohanty1,2, Abubaker \nPatan1,2, Helena Mannochio-Russo1,2, Jasmine Zemlin1,2,3, Lindsey A. Burnett4, Lisa S. Zhang5, \nMia C. Cecala5, Ceylan Ersoz6, James A. Connelly7, Natasha Halasa8, Maribeth Nicholson5, \nPieter C. Dorrestein1,2,3, Shirley M. Tsunoda1, Janet Markle6,*  \n \nAffiliations:  \n1 Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California San Diego, La Jolla, \nCA, USA \n2 Collaborative Mass Spectrometry Innovation Center, Skaggs School of Pharmacy and Pharmaceutical \nSciences, University of California San Diego, La Jolla, CA, USA \n3 Center for Microbiome Innovation, University of California San Diego, La Jolla, CA, USA \n4 Department of Obstetrics Gynecology and Reproductive Sciences, University of California San Diego, \nLa Jolla, CA, USA \n5 Department of Pediatrics, Division of Gastroenterology, Hepatology, and Nutrition, Vanderbilt University \nMedical Center, Nashville, Tennessee, USA \n6 Department of Pathology, Microbiology, and Immunology, Vanderbilt University Medical Center, \nNashville, Tennessee, USA \n7 Department of Pediatrics, Division of Hematology and Oncology, Vanderbilt University Medical Center, \nNashville, Tennessee, USA \n8 Department of Pediatrics, Division of Infectious Diseases, Vanderbilt University Medical Center, \nNashville, Tennessee, USA \n \n*Correspondence: janet.markle@vumc.org (J.M.) \n \nABSTRACT \n \nBackground and Aims: Very early onset inflammatory bowel disease (VEO-IBD) is a clinically \ndistinct form of IBD manifesting in children before the age of six years. Disease in these children \nis especially severe and often refractory to treatment. While previous studies have investigated \nchanges in the fecal microbiome and metabolome in adult and pediatric IBD, insights in \nVEO-IBD remain limited. This multi-omics analysis reveals changes in the fecal microbiome and \nmetabolome in VEO-IBD compared with healthy controls. \n \nMethods: Fecal samples were collected from children diagnosed with VEO-IBD and age- and \nsex-matched healthy controls. Both the fecal metabolome and microbiome were profiled in each \nsample, using untargeted liquid chromatography coupled with tandem mass spectrometry \n(LC-MS/MS) and 16S rRNA gene amplicon sequencing.  \n \nResults: Fecal microbial and metabolic profiles in VEO-IBD were significantly different from \nhealthy controls. Untargeted metabolomics analysis identified a depletion of short-chain N-acyl \nlipids and an enrichment of dipeptides, tripeptides, and oxo bile acids in VEO-IBD patients. \nDifferential abundance analysis of the gut microbiome showed lower abundance of beneficial \n \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n2 \nbacteria such as Bifidobacterium and Blautia, and higher abundance of Lachnospira, Veillonella, \nand Bacteroides in VEO-IBD. The joint analysis suggested a clear association between the \naltered gut microbiome composition and metabolic dysregulation, specifically for the N-acyl \nlipids. \n \nConclusions: This study offers unique insight into fecal microbial and metabolic signatures in \nVEO-IBD, paving the way for a better understanding of disease patterns and thereby more \neffective treatment strategies.  \n \nINTRODUCTION  \n \nInflammatory bowel disease (IBD) comprises a group of chronic conditions characterized by \ninflammation of the gastrointestinal (GI) tract, with ulcerative colitis (UC) and Crohn's disease \n(CD) as the two main subtypes 1. Very early onset inflammatory bowel disease (VEO-IBD) is a \nform of IBD manifesting in children younger than six years of age 2,3. Compared to IBD \ndiagnosed at later ages, VEO-IBD is generally associated with a more severe disease course 4,5, \ndelayed time to accurate diagnosis 6,7, and lower initial response rate to conventional \ntreatments8. The prolonged exposure to chronic inflammation is of significant concern for key \ndevelopmental processes such as growth, bone health, and immune function in children with \nVEO-IBD9. Additionally, monogenic etiologiesー  mutations in a single geneー  play a larger role \nin the pathogenesis of VEO-IBD, especially in infantile onset cases (<2 years) 7,10. The incidence \nof VEO-IBD is rapidly increasing worldwide 11,12, with the majority of children with VEO-IBD \n(70-80%) not having an identified genetic etiology 13. This underscores the need to better \nunderstand its multifactorial etiology, driven by complex interactions between host, microbial, \nand environmental factors14,15. \n \nThe gut microbiomeー  a key modulator of many physiological processes, including host \ndigestion and vitamin synthesis 16, drug metabolism 17, immune function 18, and metabolic \nhomeostasis19ー is increasingly recognized for its important role in the pathophysiology of IBD. \nIts role may be particularly relevant in children with VEO-IBD, as the gut microbiome is \nimmature, less diverse, and undergoes rapid changes during the first years of life 20. It was \nrecently demonstrated that there is a lower abundance of beneficial bacteria belonging to \nBifidobacterium, Collinsella, and Akkermansia, and a higher abundance of potential pathogens \nsuch as Escherichia, Ruminococcus, Clostridium, and Veillonella in children with VEO-IBD \ncompared to age-matched healthy controls 21. Alterations in the gut microbial profiles can disrupt \nthe production and modulation of microbial metabolites that act as signaling molecules, thereby \ninfluencing key processes such as immune function and homeostasis. Indeed, the levels of \nseveral metabolites belonging to the classes of bile acids (BAs) 22,23, short-chain fatty acids \n(SCFA)24,25, sphingolipids26, and amino acids (AA)25,26 have been shown to be altered in IBD.  \n \nTo the best of our knowledge, no multi-omics study has been previously conducted in VEO-IBD, \nlimiting our understanding of how the gut microbiome and metabolome may contribute to this \ndisease. In this study, we therefore combined untargeted metabolomics and 16S rRNA gene \namplicon sequencing to investigate characteristics of the fecal metabolome and microbiome in \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n3 \nchildren with VEO-IBD compared with healthy age- and sex-matched controls. We demonstrate \nan altered gut microbiome composition in VEO-IBD, combined with altered levels of important \nsignaling molecules belonging to the classes of N-acyl lipids, small peptides, and BAs. We also \nshow, through a joint analysis, a link between the changes in microbiome composition and \nmetabolic alterations in VEO-IBD. These findings will help advance our understanding of the \ndistinct biology of VEO-IBD and support the development of targeted diagnostic and therapeutic \nstrategies. \n \nMATERIALS AND METHODS \n \nParticipants \nThe study recruited 28 children initially diagnosed with isolated VEO-IBD and followed at \nMonroe Carell Jr. Children’s Hospital at Vanderbilt in the Pediatric Gastroenterology, \nHepatology, and Nutrition clinic, and 28 healthy age- and sex-matched controls. However, two \nindividuals in the VEO-IBD group were subsequently excluded due to their clinical phenotype, \ni.e. disease evolved to systemic autoimmunity and/or autoinflammation. The remaining 26 \nVEO-IBD patient samples were included in subsequent analyses. The healthy controls were not \nimmunocompromised and were free of any parental-reported vomiting or diarrhea for at least 14 \ndays prior to providing a stool specimen. Exact age at the time of sample collection was not \navailable for the healthy controls, however each control was matched to a VEO-IBD patient of \nsimilar age (±1 year). All participants were recruited at Vanderbilt University Medical Center \n(VUMC). \n \nEthics approval and informed consent \nThe study was approved by the Institutional Review Board at VUMC (studies #170067, \n#202017, #200412, and #111296) and performed in accordance with the ethical standards of the \ninstitution and with the Declaration of Helsinki. Written informed consent was obtained from all \nparents/legal guardians and assent of pediatric participants prior to any study procedures. \n \nFecal specimen collection \nFecal material from each child with VEO-IBD was collected at home by the participants’ parents \nfollowing a single bowel movement. Parents received oral and written instructions on how to \ncollect fecal material into a sterile plastic container, which was then stored in a home freezer \nuntil transport to VUMC. For healthy controls, stool samples were collected either at VUMC \nduring the visit, or if not possible, a courier was sent to the home to collect the sample. Each \nsample was transported to VUMC within 24 hours of collection, aliquoted into 1.5 mL cryovials, \nand stored at -80 ℃ until processing.  \n \nUntargeted metabolomics sample preparation \n50% methanol (MeOH:H2O) was used to extract metabolites from the fecal samples. Briefly, 800 \nµL of cold 50% MeOH was added to each sample, before they were homogenized at 25 Hz for 5 \nmin on a tissue lyzer, incubated for 30 min at 4 °C, and centrifuged at max speed (15,000 x g) \nfor 10 min at 4 °C . Supernatant (440 µL) was then collected, dried overnight in a CentriVap, and \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n4 \nstored at -80 °C for later acquisition. Prior to instrumental analysis, the samples were \nreconstituted in 200 µL 50% methanol with 1 μM sulfadimethoxine as the internal standard.  \n \nUntargeted metabolomics data acquisition \nSamples (5 μL) were injected into a Vanquish ultra-high-performance liquid chromatography \n(UHPLC) system coupled to a QExactive quadrupole orbitrap (Thermo Scientific) mass \nspectrometer. The chromatographic separation was done on a polar C18 column (Kinetex Polar \nC18, 100 mm x 2.1 mm, 2.6 µm particle size, 100 A pore size; Phenomenex) with a matching \nGuard cartridge (2.1 mm) at 40 °C column temperature. The mobile phase consisted of solvent \nA H 2O + 0.1% formic acid (FA), and solvent B acetonitrile (ACN) + 0.1% FA. Samples were \neluted at a flow rate of 0.5 mL/min using the following gradient: 0–0.5 min, 5% B, 0.5–1.1 min, \n5–25% B, 1.1–7.5 min, 25-60% B, 7.5–8.5 min, 60-99% B, 8.5–9.5 min, 99% B, 9.5–10 min, \n99-5% B, 10.0–10.5 min 5% B, 10.5–10.75, 5-99% B, 10.75–11.25 min, 99% B, 11.25–11.5 min, \n99-5% B, 11.5–12 min, 5% B. All solvents used were LC-MS grade. Data dependent acquisition \n(DDA) of tandem mass spectrometry (MS/MS) spectra was performed in positive mode with the \nfollowing parameters: sheath gas flow 53 L/min, aux gas flow rate 14 L/min, sweep gas flow 3 \nL/min, spray voltage 3.5 kV, inlet capillary 269 °C, aux gas heater 438 °C, and 50 V S-lens level. \nThe MS scan range was set to 150–1000 m/z with a resolution of 35,000 at m/z 200; maximum \nion injection time, 100 ms; automatic gain control (AGC) target, 5.0E4. MS/MS spectra were \ncollected with a resolution of 17,500 and an AGC target of 5E5 with a maximum injection time of \n150 ms, and fragmented the top 5 most abundant ions per cycle with a 1 m/z isolation window, \nisolation offset set to 0 m/z, stepped collision energies of 25, 40, 60, and a dynamic exclusion \nwindow of 10 s. \n \nUntargeted metabolomics data processing \nThe acquired .raw files were converted into .mzML format using MSConvert 27. Feature detection \nand extraction were performed using MZmine 4.2 28. For mass detection, MS1 and MS2 noise \nlevels were set to 5.0E4 and 1.0E3, respectively. Parameters for the chromatogram builder were \nset to minimum 5 consecutive scans, 1.0E5 minimum absolute height, and m/z tolerance 10 \nppm. Parameters for the local minimum resolver function were set to 85% for the \nchromatographic threshold, 0.2 min for minimum search range RT, and 1.7 for minimum ratio of \npeak top/edge. 13C isotope filter and finder were applied. Join aligner was used to align \nfeatures with weight for m/z set to 80 and RT tolerance set to 0.2 min. Before using the peak \nfinder function, features not detected in at least 2 samples were removed. metaCorrelate and \nion identity networking were performed. The .mgf spectra file and associated feature table were \nthen exported and used to generate a feature based molecular network (FBMN) 29 on GNPS230. \nBriefly, fragment tolerances for both parent and fragment ions were set at 0.02, while networking \nand annotation parameters were set to minimum 5 matching peaks and cosine similarity > 0.7. \nThe GNPS2 FBMN job is available at \nhttps://gnps2.org/status?task=8f1989e83449460a9a5748cd8f32df30. The network.graphml file \nfrom the FBMN was imported and manipulated in Cytoscape 31 (version 3.10.3). MS/MS spectra \nof unknown features were imported into SIRIUS 32 (version 6.0.7) and CANOPUS 33 was used to \npredict their chemical pathways or classes. Pathways or classes were retained if probability \nscores were > 0.7. Known and unknown features of interest were also searched using two \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n5 \ndomain-specific MASSTs 34, microbeMASST 35 (https://masst.gnps2.org/microbemasst/) and \ntissueMASST (https://masst.gnps2.org/tissuemasst/). MicrobeMASST is a taxonomically \ninformed mass spectrometry search tool within the GNPS ecosystem where MS/MS spectra can \nbe linked to their respective microbial producers by querying against a curated reference \ndatabase of >60,000 microbial monocultures. TissueMASST, a similar MS/MS search tool, \nenables investigation of molecules of interest across diverse biological contexts by the search of \nMS/MS spectra across publicly available metabolomics datasets acquired from preclinical \nanimal models (mouse and rat) and humans and with associated metadata on tissue \nlocalization and disease status.  \n \nUntargeted metabolomics data analysis \nMetabolomics data was then imported in R 4.4.2 (R Foundation for Statistical Computing, \nVienna, Austria) for downstream data analysis. Quality control (QC) samples were used to \nevaluate data quality. Features with a retention time (RT) < 0.70 or > 9.5 min were excluded \nfrom the analysis. Because polymers were detected in the LC-MS/MS run, the package \n`homologueDiscoverer v 0.0.0.9000` 36 was used to remove them. Six samples (VEO-IBD; n = 5, \nhealthy; n = 1) were excluded from the metabolomics analysis because of low intensity in the \ntotal ion chromatogram (TIC; n = 3) or high contamination (n = 3). Blank subtraction was \nperformed by removing features detected in blank samples where the mean peak areas were \nless than five times that observed in the pooled QC samples. Features with near zero variance \nwere removed using the package `caret v 6.0`. A validation of features annotated as BA \ncandidates was performed using massQL 37–39. Multivariate analysis was conducted using the \npackage `mixOmics v 6.30`. Principal component analysis (PCA) and partial least square \ndiscriminant analysis (PLS-DA) were performed on peak areas after robust center log ratio \ntransformation (rclr) using the package `vegan v 2.6.10`. PERMANOVA was used to evaluate \ngroup centroid separation. The performance of the PLS-DA models were evaluated using a \n4-fold cross-validation. Variable importance in projection (VIP) scores were calculated per \nfeature and features with VIP > 1 were considered significant. Natural log ratios of features of \ninterest were generated by summing the peak areas. Wilcoxon rank sum test, followed by \nBenjamini-Hochberg method to adjust for multiple comparisons, was used to investigate \ndifferences between the VEO-IBD group and the control group. Chi-squared test was used to \ncompare categorical variables between the two groups. P < 0.05 was considered statistically \nsignificant. Pearson correlation was used to investigate correlation between age and alpha \ndiversity.  \n \n16S rRNA gene amplicon sequencing and processing \nThe UC San Diego Microbiome Core performed DNA extraction and sequencing following the \npreviously developed standard protocols from the Earth Microbiome Project \n(http://earthmicrobiome.org/protocols-and-standards/16s/)40. Briefly, extraction was conducted \nusing the MagMAX Microbiome Ultra Nucleic Acid Isolation Kit (Thermo Fisher Scientific, USA) \nand KingFisher Flex robots (Thermo Fisher Scientific, USA). Negative (blanks) and positive \n(Zymo mock communities) controls were included in the analysis and used for quality controls. \nSequencing was conducted on the V4 region of 16S rRNA gene using the 515F forward and the \n806R reverse primers on a MiSeq System (Illumina, USA). Generated demultiplexed fastq files \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n6 \nwere then imported in Qiita 41 (ID 15748) for processing. Briefly, the Qiita default workflow for \n16S rRNA data was applied, which included trimming to 150 base pairs and Deblur for the \ngeneration of the final OTU (operational taxonomic unit) table. Taxonomic classification was also \nperformed in Qiita using the Greengenes 13.8 phylogeny database 42. The obtained .biom file \nwas then converted into a .tsv table using QIIME 2 43 and imported in R 4.4.2 for downstream \nanalysis. \n \nMicrobiome data analysis \nThe sequencing depth of the microbiome data was inspected and five samples (VEO-IBD; n = 1, \nhealthy; n = 4) were excluded from the downstream analysis due to low read counts (< 9,000 \nreads). The `phyloseq v 1.50` 44 package was used to manipulate the microbiome data. Alpha \ndiversity analysis was conducted by rarefying the data to the lowest number of reads observed \nin the cohort and calculated using the Shannon Diversity Index. Before ordination and \ndifferential abundance analysis, OTUs were collapsed at genus level. Beta diversity analysis \nwas conducted via PCA of the rclr transformed data. Significant differences in community \nprofiles were tested via PERMANOVA 45. A PLS-DA model was also generated to extract \nfeatures driving group separation. The performance was evaluated using a 4-fold \ncross-validation, and features with VIP > 1 were considered significant. Finally, differential \nabundance analysis was performed using ALDEx2 46 and features with adjusted p value < 0.05 \nwere considered significant.  \n \nMulti-omics analysis \nThe joint analysis between the microbiome and the metabolomics data was performed using \nDIABLO47, a supervised method using multi-block PLS-DA to identify feature correlations \nbetween datasets in relation to a categorical outcome. The model was first tuned to retain top \nfeatures for each omics block maximizing discriminatory covariance between groups. Model \nperformance was evaluated via leave-one-out cross validation. A circos plot was finally \ngenerated to display intra feature correlation with a cut off set to 0.5. \n \nRESULTS \n \nA total of 26 children previously diagnosed with VEO-IBD and 28 healthy age- and sex-matched \ncontrols were included in this study. One stool sample per subject was collected and used for  \n16S rRNA gene amplicon sequencing, to profile the fecal microbial communities, and \nuntargeted liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) to \nprofile both known and unknown metabolites (Fig. 1a). Mean age ± SD of study participants \nwere 6.7 ± 3.4 and 6.4 ± 3.5 years in the VEO-IBD group and control group (Wilcoxon test: p = \n0.71), respectively, with 54% being male in both groups (Chi-squared test: p =1). Demographic \ncharacteristics of the study participants are listed in Table 1. Most patients with VEO-IBD (85%) \nreceived pharmacological treatment, including tumor necrosis factor alpha (TNF-α)-inhibitors \n(54%), aminosalicylates (31%), antibiotics (8%), or a combination of these therapies; and 4 \nsubjects were diagnosed with monogenic VEO-IBD. Mean time from VEO-IBD diagnosis to \nsample collection was 3.3 ± 3.2 years, with 46% of samples being collected within the first year.  \n \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n7 \nTable 1. Characteristics of the study population. \nCharacteristics VEO-IBD \n(n = 26) \nControls \n(n = 28) \nAge at enrollment (years) 6.7 (1-14) 6.4 (1-14) \nSex (female/male) 12/14 13/15 \nMonogenic IBD, n (%) 4 (15%) NA \nPharmacological treatment, n (%) \n-            Aminosalicylate (sulfasalazine, 5-ASA, balsalazide) \n-            TNF-α-inhibitor (infliximab, adalimumab) \n-            Antibiotics (amoxicillin, azithromycin) \n22 (85%) \n8 (31%) \n14 (54%) \n2 (8%) \nNA \nNumbers are given as mean (absolute range) or count (%). Abbreviations: 5-ASA, 5-aminosalicylate; NA, not \napplicable; TNF-α, tumor necrosis factor alpha \n \nChildren with VEO-IBD display distinct metabolic profiles \nUsing untargeted LC-MS/MS metabolomics, we observed a significant separation in the fecal \nmetabolome between the VEO-IBD group and the controls (Fig. 1b ) (PERMANOVA: R2 = 0.039, \nF-statistic = 1.9, p = 0.001). Age also had a significant impact on the fecal metabolomic profiles \n(PERMANOVA: R 2 = 0.031, F-statistic = 1.5, p = 0.016). To identify discriminant metabolites \nbetween the two groups, a supervised Partial Least Squares - Discriminant Analysis (PLS-DA) \nmodel was generated (SI Fig. 1a, classification error rate (CER) = 0.21) and 1,678 known and \nunknown metabolic features with variable importance in projection (VIP) > 1 were extracted (SI \nTable 1). The natural log ratio of the extracted features significantly separated the controls from \nthe VEO-IBD group (Wilcoxon test, p = 9.0e-08) (Fig. 1c). One group of metabolites that stood \nout as key discriminant features between VEO-IBD and controls were the N-acyl lipids, signaling \nmolecules containing an amine group (head) and a fatty acid (tail), linked by an amide bond 38 \n(Fig. 1d). Several putatively annotated short-chain (C2-C6) N-acyl lipids with methionine (Met), \nphenylalanine (Phe), leucine (Leu)/isoleucine (Ile), and tyrosine (Tyr) as head groups were \nsignificantly depleted in children with VEO-IBD (Fig. 1d, SI Fig. 1b ). In contrast, we observed a \nhigher relative abundance of dipeptides and tripeptides in VEO-IBD than in controls (Wilcoxon \ntest: p = 0.018 and p = 0.0046, respectively) (Fig. 1e). Annotated dipeptides that were enriched \nin VEO-IBD included asparagine (Asn)-Phe, glutamic acid (Glu)-Phe, Phe-Met, Ile-Leu, Trp-Phe, \nPhe-Leu, Tyr-Phe, valine (Val)-Ile, threonine (Thr)-Phe, Met-Leu, Ile-Met, aspartic acid \n(Asp)-Phe, and Ala-Phe (SI Table 1). Annotated tripeptides included Ile-Proline(Pro)-Ile and \nThr-Val-Leu. Given the limited number of annotated tripeptides in our dataset, we also included \nunknown features predicted to be tripeptides by SIRIUS (probability > 0.7) in the analysis. A \ntotal of 43 tripeptides with a VIP > 1 were identified, and a group comparison of the relative \nabundance supported higher levels of tripeptides in VEO-IBD (Wilcoxon test: p = 0.00011, SI \nFig. 1c).  \n \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n8 \n \nFigure 1. N-acyl lipids, small peptides, oxo bile acids, and drug metabolites of aminosalicylates \nare main drivers of separation in the fecal metabolome between VEO-IBD and controls.  \n(a) Overview of the study design. Twenty six patients with VEO-IBD and 28 age- and sex-matched \nhealthy controls provided one stool sample for untargeted LC-MS/MS analysis and 16S rRNA gene \namplicon sequencing. (b) PCA of fecal metabolic profiles showed separation based on cohort \n(PERMANOVA, R2 = 0.039, F-statistic = 1.9, p = 0.001) and age (PERMANOVA, R 2 = 0.031, F-statistic = \n1.5, p = 0.016). (c) A PLS-DA model was constructed to extract features separating VEO-IBD and control. \nBoxplot highlights the natural log ratio of differential features from the PLS-DA model (numerator = \ncontrol, denominator = VEO-IBD). (d) Volcano plot of N-acyl lipids, with the dotted lines showing the \nsignificance threshold (p adjusted (Benjamini-Hochberg) < 0.05 and absolute log2FC > 2). (e) Boxplot of \nthe relative abundance of dipeptides and tripeptides with a VIP score > 1 from the PLS-DA model, \nrespectively. Relative abundance was calculated by dividing the sum of peak areas for relevant dipeptides \nand tripeptides by the total ion current. (f) Boxplot of the natural log ratio of BAs with a VIP score > 1 from \nthe PLS-DA model (numerator = VEO-IBD, denominator = control). (g) Boxplot of the annotated oxo BAs \n- 3ɑ-7ɑ-12-oxo (Wilcoxon test, p = 0.005) and 3ɑ-7-oxo-12-oxo (Wilcoxon test, p = 0.012). P values were \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n9 \nadjusted for multiple comparisons using the Benjamini-Hochberg method. (h) Barplot of other differential \nfeatures between VEO-IBD and control (VIP score > 1). (i) Molecular network of 5-ASA, the major \nmetabolite of sulfasalazine and balsalazide, and its drug metabolites. Orange nodes represent features \nwith a VIP > 2, with larger nodes indicating higher VIP scores. All boxplots show the first (lower), median, \nand third (upper) quartiles, with whiskers 1.5 times the interquartile range. ‡Candidate dihydroxylated BA \n(∆ 135.033). †MS/MS spectral match to 2-acetamidophenol, a stereoisomer of acetaminophen, which is \nindistinguishable from acetaminophen via mass spectrometry. \n \nSeveral annotated and putatively annotated BAs were important discriminant features (Wilcoxon \ntest, p = 0.0035) (Fig. 1f). Notably, children with VEO-IBD displayed a higher abundance of oxo \nBAs compared with controls, specifically putative 3ɑ-7ɑ-12-oxo and 3ɑ-7-oxo-12-oxo (Fig. 1g ). \nAmong the other annotated oxo BAs, there was a trend towards a higher abundance of three \ndifferent 3ɑ-7ɑ-12-oxo isomers and 3ɑ-7-oxo-12-oxo in VEO-IBD, while 3ɑ-12-oxo did not differ \nbetween the groups (SI Fig. 1d). Microbial AA conjugated BAs such as Glu-cholic acid (CA) was \nalso enriched in VEO-IBD (Fig. 1h ). We did not observe any difference in host-derived CA, \nmicrobially derived deoxycholic acid (DCA), or taurine or glycine conjugated BAs (SI Fig. 1e).  \n \nAdditional discriminant metabolic features enriched in VEO-IBD included drug metabolites of the \naminosalicylates, acetaminophen/2-acetamidophenol (stereoisomers and thereby \nindistinguishable via mass spectrometry), a candidate dihydroxylated BA (refers to a BA with \nuncharacterized modification) 37, and 5'-methylthioadenosine. On the other hand, N-acetyl-Met, \nN-acetyl-Ile/Leu, Arg-Phe, Val-Trp, and stercobilin were enriched in control (Fig. 1h ). It should \nbe noted that we observed poor alignment (38%) between acetaminophen detection and \nreported use, possibly due to limited MS/MS fragmentation affecting annotation or inaccuracies \nin self-reported use of acetaminophen. Around 30% of the children with VEO-IBD reported use \nof an aminosalicylate. Using the GNPS drug library 48, a resource encompassing MS/MS data of \ndrugs and their metabolites/analogs, combined with a molecular network of the \naminosalicylates, we were able to annotate N-acetyl 5-ASA and N-propionyl 5-ASA as key \ndiscriminant features between VEO-IBD and controls (Fig. 1i). These gut microbially derived \nmetabolites49 were detected in all patients reported to receive treatment with an aminosalicylate. \n5-aminosalicylate (5-ASA) eluted in the dead volume. Sulfasalazine was not a discriminant \nfeature (Wilcoxon test: p = 0.25) and no spectral matches to balsalazide were observed, \nhighlighting the importance of also considering drug metabolites. Several other unannotated \nmetabolic features were also driving the separation between VEO-IBD and controls. To predict \nthese metabolites’ chemical classes, their MS/MS spectra were processed with SIRIUS. \nFeatures with a VIP > 2.5 were further investigated. Of the metabolites enriched in VEO-IBD, \nmany were predicted to be dipeptides and tripeptides, while depleted metabolites included \npiperidine, purine, and simple indole alkaloids (SI Fig. 2a). To investigate their possible \nmicrobial origin, we also searched the MS/MS spectra via microbeMASST (see methods). The \noutput showed that out of the 20 most discriminant features (top 10 in each group), 14 have \nbeen previously observed in microbial monocultures. \n \nGut microbiome composition is altered in VEO-IBD versus controls \nWe next looked at the microbial alpha diversity by measuring the Shannon Diversity Index, and \nfound no statistical significant difference between VEO-IBD and controls  (Wilcoxon test, p = \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n10 \n0.48) (Fig. 2a). However, when investigating age, we observed a moderate positive correlation \nbetween age and the Shannon Diversity Index in the VEO-IBD group (Pearson, R = 0.41, p = \n0.041), but not in the control group (p = 0.33) (Fig. 2b). There was also a significant dissimilarity \nin beta diversity based on PCA of the robust center log ratio (rclr) transformed data \n(PERMANOVA, R 2 = 0.087, F-statistic = 4.5, p < 0.001) (Fig. 2c). A PLS-DA model was \ngenerated to identify discriminant taxa at genus level (SI Fig. 3a, CER = 0.11). We identified 32 \ngenera with VIP > 1; 17 that were enriched in the control group and 15 that were enriched in the \nVEO-IBD group (SI Table 2). Natural log ratios of the selected taxa confirmed a significant \ndifference in the fecal microbiome composition between the two groups (Wilcoxon test, p = \n3.1e-09) (Fig. 2d ). Key discriminant genera, with a higher abundance in VEO-IBD, included \nLachnospira, Sutterella, Clostridium, Bacteroides, Alistipes, and unknown genera of the \nEnterobacteriaceae family (Fig. 2e, SI Fig. 3b ). On the other hand, genera including \nCoprococcus, Bifidobacterium, Collinsella, Blautia, and Dorea were depleted in VEO-IBD. To \nfurther investigate differences in fecal microbiome composition, differential abundance analysis \nwas also conducted using ALDEx2 46. The output showed similar findings as from the PLS-DA \nmodel, with higher abundance of Veillonella, Sutterella, Lachnospira, Clostridium, and \nBacteroides and lower abundance of Coprococcus, Collinsella, and Turibacter in VEO-IBD \ncompared with controls (Fig. 2f).  \n \n \nFigure 2. Fecal microbiome composition varies significantly between VEO-IBD and controls.  \n(a) Boxplot of alpha diversity measured as Shannon's Diversity Index (Wilcoxon test, p = 0.48). (b) \nScatter plot showing the correlation between age and Shannon's Diversity Index (Pearson’s correlation, \nVEO-IBD: R = 0.41, p = 0.041; control: R = 0.21, p = 0.33). (c) PCA of rclr transformed data showing a \nsignificant difference in beta diversity between the two cohorts (PERMANOVA, R 2 = 0.087, F-statistic = \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n11 \n4.5, p < 0.001) with no effect of age. (d) Boxplot of the natural log ratio of differential genera extracted \nfrom the PLS-DA model (numerator = control, denominator = VEO-IBD). (e) Boxplot of selected \ndifferential genera obtained from the PLS-DA model. P values were adjusted for multiple comparisons \nusing the Benjamini-Hochberg method. (f) Balloon plot of differential abundance taxa collapsed at genus \nlevel using ALDEx2. Only taxa with adjusted p value < 0.05 are reported. Taxa names reported as \nFamily_Genera. All boxplots show the first (lower), median, and third (upper) quartiles, with whiskers 1.5 \ntimes the interquartile range.  \n \nJoint analysis identifies link between N-acyl lipid dysregulation and the gut microbiome \nTo investigate if any of the metabolic changes we detected could be explained by changes in \nthe composition of the gut microbiome, a multi-omics analysis was performed using DIABLO 47. \nThe output of the first component (correlation cut-off > 0.5) showed a  correlation between the \nshort-chain N-acyl lipids Met-C3:0, Met-C4:0, Phe-C3:0, Phe-C4:0, and Ile/Leu-C4:0, and the \ngenera Coprococcus, Lachnospira, Collinsella, Bifidobacterium, Blautia, Dorea, Alistipes, and \nBacteroides (Fig. 3a). More specifically, N-acyl lipids were positively correlated with \nCoprococcus, Collinsella, Bifidobacterium, Blautia, and Dorea, genera more abundant in the \nhealthy controls, and negatively correlated with Lachnospira, Bacteroides, and Alistipes, \nenriched in VEO-IBD. Similarly, N-acetyl-Ile/Leu and N-acetyl-Met, both more abundant in \ncontrols, were positively correlated with Coprococcus, Collinsella, and Blautia and inversely \ncorrelated with Lachnospira and Alistipes. Additionally, we observed correlations between \nspecific genera and unannotated metabolic features (Fig. 3a). Constructed sub-molecular \nnetworks did not contain any annotated features. SIRIUS predictions indicated that these \nunknown metabolic features belonged to the pathways or classes of alkaloids (m/z 243.1128, \nm/z 197.1284, and m/z 188.128), dipeptides (m/z 395.1811), and tripeptides (m/z 346.1975) \n(Fig. 3b). Next, we searched them via microbeMASST and found that three of them have been \npreviously observed in microbial monocultures. Of particular interest was the unknown feature \nwith m/z 346.1975 enriched in VEO-IBD, which SIRIUS-predicted to be a tripeptide, matching to \ngenera such as Bacteroides and Akkermansia (SI Fig. 3c). We also searched this feature (m/z \nof 346.1975) via tissueMASST to investigate its presence in other publicly available \nmetabolomics IBD datasets. The output showed that this metabolic feature is detected with \nhigher frequency in feces of individuals with IBD compared with healthy (Chi-squared test: p = \n2.2e-16), and also in other chronic inflammatory diseases such as rheumatoid arthritis \ncompared with healthy (Chi-squared test: p = 5.106e-08) (Fig. 3b).  \n \nThis observation motivated us to explore the detection frequency of other discriminant di- and \ntripeptides from the metabolomics PLS-DA model in other publicly available metabolomics IBD \ndatasets. The MS/MS spectra of Glu-Phe, Asn-Phe, Thr-Val-Leu, Ile-Pro-Ile, and an unknown \nfeature SIRIUS-predicted to be a tripeptide, were therefore searched via tissueMASST. We \nfocused on the output from datasets comprising human stool samples. The output confirmed a \nhigher detection rate of all these features in human fecal samples from individuals with IBD \ncompared with healthy individuals (Chi-squared test: all p < 2.2e-16) (Fig. 3c). We also \nsearched the MS/MS spectra of these features via microbeMASST. The output confirmed that \nthey are microbial metabolites. The spectra of these metabolites were most frequently observed \nin cultures of Bacteroides, Veillonella, and Akkermansia, although they matched to several \nbacterial cultures, as illustrated for Thr-Val-Leu (Fig. 3d).  \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n12 \n \nFigure 3. Multi-omics analysis of the microbiome and metabolome data reveals significant \ncorrelation between specific microbial genera, N-acyl lipids, and unknown metabolic features.  \n(a) Integration of the metabolomics and microbiome data using DIABLO. The correlation cut-off was set to \n> 0.5. Red lines show positive correlation, blue lines show negative correlation. (b) Sub-molecular \nnetworks of unknown metabolic features highlighted by DIABLO. MS/MS spectra of the unknown \nmetabolites were processed using SIRIUS to predict their chemical pathways or classes. The MS/MS \nspectra of one of the unknown features enriched in VEO-IBD (m/z 346.1975), predicted to be a tripeptide, \nwas searched across publicly available metabolomics datasets using tissueMASST. The output showed a \nsignificantly higher detection rate of this feature in fecal samples from individuals with IBD compared to \nhealthy (Chi-squared test: p < 2.2e-16). Pie charts show the proportion of spectral matches found in \ndeposited datasets, with blue representing a match and yellow representing a non-match. (c) Other key \ndiscriminant features belonging to the classes of di- and tripeptides were therefore also searched via \ntissueMASST. Barplots showing a higher detection frequency in human fecal samples from individuals \nwith IBD than healthy individuals (Chi-squared test: all p < 2.2e-16). The Y-axis shows the percentage of \nsamples where the feature was detected vs not detected. (d) To investigate if they are of microbial origin, \nthese features were also searched against microbeMASST. The output for Thr-Val-Leu is illustrated, but \nthey all matched to microbial monocultures. Pie charts show the proportion of spectral matches found in \ndeposited datasets, with blue representing a match and yellow representing a non-match. Significance: \n**** p < 0.0001.  \n \n \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n13 \nDISCUSSION \n \nA more thorough understanding of gut microbiome composition and function in IBD may set the \nstage for more effective and personalized treatments. For most patients with IBD, the peak age \nat onset is between 20-40 years 50. However, a small but growing percentage of IBD patients are \ndiagnosed with VEO-IBD at age 6 or younger, and disease in these children is especially severe \nand often treatment-refractory3. The timely selection of an individualized treatment plan is critical \nfor these children, since VEO-IBD is associated with poor long-term outcomes, including \nsurgical resection, increased risk of colorectal cancer, and even death 7. The gut microbiome \nrepresents a promising therapeutic target in IBD. Significant research efforts have been \ndedicated to investigate changes in the gut microbiome and metabolome in pediatric 26,51,52 and \nadult IBD 53,54. However, studies specifically looking at VEO-IBD remain scarce 21, limiting our \nunderstanding of whether gut microbial and metabolic changes may contribute to the distinct \nclinical phenotype of VEO-IBD. Given that the gut microbiome and immune system evolve \nrapidly in early childhood 20, VEO-IBD may involve unique perturbations that differ, not only from \nhealthy peers, but also from those seen in IBD diagnosed at later ages. Untargeted LC-MS/MS \nmetabolomics allows the identification of thousands of known and unknown metabolites in a \ngiven sample. When combined with 16S rRNA gene amplicon sequencing, which characterizes \nthe gut microbiome composition, it can provide valuable insights into the intricate relationship \nbetween host, microbes, and metabolites. To our knowledge, this is the first multi-omics analysis \ninvestigating changes in the fecal microbiome and metabolome in children diagnosed with \nVEO-IBD compared with healthy age- and sex-matched controls.  \n \nN-acyl lipids, an understudied class of molecules, appear to play an important role in several \nbiological functions, such as immune regulation 38,55. In this study, short-chain N-acyl lipids were \namong the top discriminant features between VEO-IBD and healthy participants. Several \nputatively annotated N-acyl lipids had a significantly lower abundance in VEO-IBD compared \nwith controls. This depletion of short-chain N-acyl lipids in VEO-IBD compared to healthy \ncontrols has, to our knowledge, not been previously described in the literature. Whether this \nrepresents a unique characteristic of VEO-IBD or is a shared trait across IBD subtypes is not yet \nknown. Our study also demonstrated a significant correlation between the abundance of these \nshort-chain N-acyl lipids and the abundance of specific bacterial genera in a multi-omics \nanalysis of the fecal metabolome and the microbiome. Met-C3:0, Met-C4:0, Phe-C3:0, \nPhe-C4:0, and Ile/Leu-C4:0 were negatively correlated with bacterial genera enriched in \nVEO-IBD, and positively correlated with genera more abundant in healthy controls. This pattern \nsuggests that the dysregulation of N-acyl lipids in VEO-IBD may be linked to underlying gut \nmicrobiome dysbiosis. We recently demonstrated that many of these N-acyl lipids are \nmicrobially produced when both the amine group and lipid chain are present 38. We speculate \nthat a lower abundance of specific short-chain fatty acid (SCFA) producing bacteria in VEO-IBD, \nsuch as Coprococcus, Blautia, Dorea, and Bifidobacterium56, may alter SCFAs availability, \nthereby limiting their ability to form N-acyl lipids with an amine group. \n \nRecent studies have demonstrated a correlation between higher dipeptide abundance and \nincreased disease activity in UC 57. The elevated dipeptide levels have been linked to an \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n14 \noverabundance of microbial proteases – particularly serine proteases and dipeptidases \nproduced by Bacteroides, specifically B. vulgatus57. Consistent with this finding, we observed a \nhigher relative abundance of dipeptides and tripeptides combined with a higher abundance of \nBacteroides in VEO-IBD compared with healthy controls. Furthermore, a microbeMASST search \nconfirmed a higher detection frequency of several di- and tripeptides in cultures of Bacteroides \nand Veillonella, both of which were enriched in VEO-IBD. These findings suggest that the \nelevated levels of small peptides are likely secondary to increased proteolytic activity associated \nwith the altered gut microbiome in VEO-IBD. While it remains unknown whether these peptides \ndirectly contribute to VEO-IBD pathophysiology or are simply byproducts of the altered gut \nmicrobiome composition, there is evidence pointing to a potential role in disease. For instance, \nprotease inhibitors have been shown to prevent colitis in B. vulgatus monocolonized, \nIL10-deficient mice 57, suggesting that microbial protease activity could be a viable therapeutic \ntarget in IBD. Additionally, a tissueMASST search found that key discriminant di- and tripeptides \nwere more frequently detected in fecal samples from individuals with IBD than from healthy \nsubjects. This suggests the possibility that certain di- or tripeptides may serve as biomarkers for \nIBD or disease activity and/or severity. \n \nPrevious studies have shown dysregulation of BA metabolism in IBD 22,23. BAs are \ncholesterol-derived signaling molecules synthesized in the liver that activate receptors such as \nfarnesoid X receptor (FXR) and pregnane X receptor (PXR) 58. After being conjugated with \ntaurine and glycine, BAs are secreted into the small intestine via the biliary system where they \nundergo extensive modifications via microbial metabolism 58. Consistent with previous IBD \nliterature23,59, we observed a shift in the bile acid pool in VEO-IBD, characterized by a higher \nabundance of microbial conjugated BAs such as putative Glu-CA. Bile salt hydrolase (BSH), an \nenzyme widely expressed across most bacterial phyla 60, plays a key role in the modification of \nBAs, including AA conjugation of BA 61,62. Notably, Glu-CA has been shown to be more resistant \nto BSH-mediated hydrolysis than other microbial conjugated BAs in an in vitro model with 17 gut \nbacterial isolates exposed to AA conjugated BAs 63. We also observed elevated levels of some, \nbut not all, 12-oxo BAs in VEO-IBD. These oxo BAs which contain a ketone in the carbon 12 \nposition in the steroid ring, have been implicated in activating the pyrin inflammasome and \npromoting secretion of pro-inflammatory cytokines, such as interleukin (IL)-18 and IL-1β 64. This \nsuggests a potential role for the oxo BAs in immune regulation and disease pathophysiology in \nVEO-IBD. Interestingly, unlike previous reports in pediatric IBD 51,65, we did not observe \ndifferences in levels of host-derived CA or microbially derived DCA. These findings indicate that \nBA metabolism in VEO-IBD may follow a distinct pattern compared to IBD diagnosed later in \nchildhood or in adulthood.  \n \nIn line with previous findings in pediatric IBD 21,66, we observed significant differences in gut \nmicrobiome composition between children with VEO-IBD compared with age- and sex-matched \ncontrols. This was evidenced by a significant difference in microbial community dissimilarity \nbetween the groups. Specifically, we found a significantly lower abundance of beneficial \ncommensal gut bacteria including Bifidobacterium, Blautia, and Coprococcus – microorganisms \nassociated with host health 67,68. Supporting this, Conrad et al. recently reported increased \nmicrobial dissimilarity in VEO-IBD compared with IBD diagnosed in older children (> 6 years) 21, \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n15 \nhighlighting that VEO-IBD likely involves unique perturbations that differ from IBD in older \nchildren. Consistent with their findings, we also observed a higher abundance of OTUs of the \ngenus Veillonella and Clostridium. Interestingly, alpha diversity did not differ between VEO-IBD \nand healthy controls, which contrasts earlier reports in VEO-IBD 21 and pediatric IBD69. However, \nwithin the VEO-IBD group, we found a positive correlation between age and microbial diversity– \na pattern not observed in healthy controls. This suggests that microbial diversification in \nVEO-IBD may be delayed, potentially due to early life dysbiosis combined with prolonged \nexposure to chronic intestinal inflammation and/or medications prescribed to control the \nsymptoms of IBD, disrupting normal microbiome development.  \n \nWe observed a lower abundance of N-acetyl-Met alongside a higher abundance of \n5'-methylthioadenosine (MTA) in VEO-IBD, indicating a potential disruption in the methionine \ncycle. Additionally, microbially derived metabolites of aminosalicylate drugs 49 were among the \nkey discriminant features of the VEO-IBD group. Our multi-omics analysis also identified several \nunknown metabolic features that strongly correlated with microbial genera either enriched or \ndepleted in VEO-IBD. Notably, using tissueMASST, we found that one of these metabolites – \nenriched in VEO-IBD – was also more frequently detected in the fecal samples from individuals \nwith IBD compared to healthy subjects when searching its spectra against publicly available \nmetabolomics datasets within the GNPS/MassIVE ecosystem. To explore the microbial origin of \nthis feature, we searched microbeMASST and identified matches across multiple bacterial \nphyla, including Veillonella and Bacteroides, both of which were more abundant in VEO-IBD. \nBased on SIRIUS predictions and manual inspection of the MS/MS spectra, this metabolite is \nlikely a modified tripeptide. However, further investigation is needed to confirm its structure and \nunderstand its putative role in the pathophysiology of VEO-IBD.  \n \nWe note limitations of this study. First, metabolite annotations are based on MS/MS spectral \nmatches with the GNPS spectral libraries, representing a level 2-3 annotation according to the \nMetabolomics Standard Initiative (MSI) 70. Generally, MS/MS data cannot differentiate between \nstereoisomers, such as acetaminophen and 2-acetamidophenol. The poor alignment between \nacetaminophen detection and reported use (38%) may be explained by the parameters used for \nthe FBMN job, which required at least 5 matching fragments for annotation, or inaccuracies in \nself-reported use of acetaminophen. Factors that can influence the gut microbiome composition \nand functionality, such as antibiotic exposure, probiotics intake, and dietary modifications were \nnot reported in detail. Finally, the study has a relatively small sample size, and findings should \nbe validated in a larger, independent cohort.  \n \nIn conclusion, this is the first study investigating changes in the fecal microbiome and \nmetabolome in children diagnosed with VEO-IBD compared to age- and sex-matched healthy \ncontrols. We identified distinct fecal microbial and metabolic signatures in VEO-IBD. Notably, we \nobserved a marked depletion of short-chain N-acyl lipids, along with elevated levels of several \ndi- and tripeptides and oxo BAs. Our integrative multi-omic analysis highlights a strong \nconnection between gut microbial dysbiosis and metabolic alterations in VEO-IBD. Importantly, \nseveral of these features appear to be unique to VEO-IBD, distinguishing it from IBD diagnosed \nin older children or adults. Manipulation of gut microbiota composition and function via \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n16 \npre/probiotics, fecal microbiota transplantation, or antibiotics may provide an opportunity to \ntarget processes that contribute to disease development and/or severity in children with \nVEO-IBD71,72. In general, these microbiome-directed therapies have been shown to be \nwell-tolerated with few side effects, making them attractive treatment options in the \nmanagement of VEO-IBD 71. Overall, the findings enhance our understanding of the distinct \ndisease biology of VEO-IBD and provide a foundation for the development of more targeted \ndiagnostic and therapeutic strategies.  \n \nFunding \nThis research was supported by an Opportunity Pool Grant (sub-award 9800-VU) to J.G.M., \nM.N. and J.A.C. from the Maternal and Pediatric Precision In Therapeutics (MPRINT) program, \nfunded by an NICHD/NIH award (5P30HD106451-03, PI: S. Quinney) and NICHD P50 award \n(P50HD106463, MPI: SM Tsunoda) \n \nAcknowledgement \nThe authors would like to acknowledge support from Laura Stewart, Rendie McHenry, Dan \nPayne and Jim Chappell (VUMC) with healthy donor sample collection. \n \nAuthor contributions \nJ.M. conceptualized the study. M.N., J.A.C., and N.H. recruited patients and oversaw sample \ncollection. Samples were processed by L.S.Z., M.C.C., and C.E. K.E.K., J.Z., V.C. performed \nsample extraction and LC-MS/MS analysis. K.E.K. conducted untargeted metabolomics \nanalysis. K.E.K. and S.Z. conducted microbiome analysis. K.E.K. drafted the manuscript. P.C.D., \nS.M.T., and J.M. acquired funding and supervised this project. All authors provided feedback, \nreviewed, and approved the manuscript.  \n \nDisclosures \nS.M.T. receives research funding from Veloxis Pharmaceuticals. P.C.D. is an advisor and holds \nequity in Cybele, Sirenas, and BileOmix, and he is a scientific co-founder, advisor, holds equity \nand/or receives income from Ometa, Enveda, and Arome with prior approval by UC San Diego. \nP.C.D. consulted for DSM Animal Health in 2023. J.A.C. is an advisor for Pharming \nPharmaceuticals. All other authors declare no conflicts of interest. \n \nData availability \nUntargeted LC-MS/MS data generated in this study are publicly available at GNPS/MassIVE \n(https://massive.ucsd.edu/) under the accession code MSV000097610. The associated FBMN \njob is publicly available at GNPS2: \nhttps://gnps2.org/status?task=8f1989e83449460a9a5748cd8f32df30. Additional information and \nprocessing pipelines for the microbiome data are available in Qiita: VEOIBD - ID 15748. \n \nCode availability \nThe code used for data analysis and to generate the figures can be found in the GitHub \nrepository: https://github.com/kinekvitne/manuscript_VEO-IBD  \n \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n17 \nREFERENCES \n1. Burisch J., Jess T., Martinato M., Lakatos PL., ECCO -EpiCom. The burden of inflammatory \nbowel disease in Europe. J Crohns Colitis 2013;7(4):322–37. \n2. Muise AM., Snapper SB., Kugathasan S. The age of gene discovery in very early onset \ninflammatory bowel disease. Gastroenterology 2012;143(2):285–8. \n3. Uhlig HH., Schwerd T., Koletzko S., Shah N., Kammermeier J., Elkadri A., et al. The \ndiagnostic approach to monogenic very early onset inflammatory bowel disease. \nGastroenterology 2014;147(5):990–1007.e3. \n4. Cucinotta U., Arrigo S., Dipasquale V., Gramaglia SMC., Laganà F., Romano C., et al. \nClinical course of very early-onset inflammatory bowel disease. J Pediatr Gastroenterol \nNutr 2023;76(5):590–5. \n5. Kelsen JR., Sullivan KE., Rabizadeh S., Singh N., Snapper S., Elkadri A., et al. \nNASPGHAN position paper on the evaluation and management for patients with very \nearly-onset inflammatory bowel disease (VEO-IBD). J Pediatr Gastroenterol Nutr \n2019;70(3):389–403. \n6. Kammermeier J., Dziubak R., Pescarin M., Drury S., Godwin H., Reeve K., et al. \nPhenotypic and genotypic characterisation of inflammatory bowel disease presenting before \nthe age of 2 years. J Crohns Colitis 2017;11(1):60–9. \n7. Ouahed J., Spencer E., Kotlarz D., Shouval DS., Kowalik M., Peng K., et al. Very early \nonset inflammatory bowel disease: A clinical approach with a focus on the role of genetics \nand underlying immune deficiencies. Inflamm Bowel Dis 2020;26(6):820–42. \n8. Oliva-Hemker M., Hutfless S., Al Kazzi ES., Lerer T., Mack D., LeLeiko N., et al. Clinical \npresentation and five-year therapeutic management of very early-onset inflammatory bowel \ndisease in a large north American cohort. J Pediatr 2015;167(3):527–32.e1–3. \n9. Rosen MJ., Dhawan A., Saeed SA. Inflammatory bowel disease in children and \nadolescents. JAMA Pediatr 2015;169(11):1053–60. \n10. Nambu R., Warner N., Mulder DJ., Kotlarz D., McGovern DPB., Cho J., et al. A systematic \nreview of monogenic inflammatory bowel disease. Clin Gastroenterol Hepatol \n2022;20(4):e653–63. \n11. Sýkora J., Pomahačová R., Kreslová M., Cvalínová D., Štych P., Schwarz J. Current global \ntrends in the incidence of pediatric-onset inflammatory bowel disease. World J \nGastroenterol 2018;24(25):2741–63. \n12. Benchimol EI., Mack DR., Nguyen GC., Snapper SB., Li W., Mojaverian N., et al. Incidence, \noutcomes, and health services burden of very early onset inflammatory bowel disease. \nGastroenterology 2014;147(4):803–13.e7; quiz e14–5. \n13. Charbit-Henrion F., Parlato M., Hanein S., Duclaux-Loras R., Nowak J., Begue B., et al. \nDiagnostic yield of next-generation sequencing in very early-onset inflammatory bowel \ndiseases: A multicentre study. J Crohns Colitis 2018;12(9):1104–12. \n14. Lloyd-Price J., Arze C., Ananthakrishnan AN., Schirmer M., Avila-Pacheco J., Poon TW., et \nal. Multi-omics of the gut microbial ecosystem in inflammatory bowel diseases. Nature \n2019;569(7758):655–62. \n15. Khor B., Gardet A., Xavier RJ. Genetics and pathogenesis of inflammatory bowel disease. \nNature 2011;474(7351):307–17. \n16. Cummings JH., Macfarlane GT. Role of intestinal bacteria in nutrient metabolism. JPEN J \nParenter Enteral Nutr 1997;21(6):357–65. \n17. Wilson ID., Nicholson JK. Gut microbiome interactions with drug metabolism, efficacy, and \ntoxicity. Transl Res 2017;179:204–22. \n18. Zheng D., Liwinski T., Elinav E. Interaction between microbiota and immunity in health and \ndisease. Cell Res 2020;30(6):492–506. \n19. Zarrinpar A., Chaix A., Xu ZZ., Chang MW., Marotz CA., Saghatelian A., et al. \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n18 \nAntibiotic-induced microbiome depletion alters metabolic homeostasis by affecting gut \nsignaling and colonic metabolism. Nat Commun 2018;9(1):2872. \n20. Tamburini S., Shen N., Wu HC., Clemente JC. The microbiome in early life: implications for \nhealth outcomes. Nat Med 2016;22(7):713–22. \n21. Conrad MA., Bittinger K., Ren Y., Kachelries K., Vales J., Li H., et al. The intestinal \nmicrobiome of inflammatory bowel disease across the pediatric age range. Gut Microbes \n2024;16(1):2317932. \n22. Duboc H., Rajca S., Rainteau D., Benarous D., Maubert M-A., Quervain E., et al. \nConnecting dysbiosis, bile-acid dysmetabolism and gut inflammation in inflammatory bowel \ndiseases. Gut 2013;62(4):531–9. \n23. Gentry EC., Collins SL., Panitchpakdi M., Belda-Ferre P., Stewart AK., Carrillo Terrazas M., \net al. Reverse metabolomics for the discovery of chemical structures from humans. Nature \n2024;626(7998):419–26. \n24. Bjerrum JT., Wang Y., Hao F., Coskun M., Ludwig C., Günther U., et al. Metabonomics of \nhuman fecal extracts characterize ulcerative colitis, Crohn’s disease and healthy \nindividuals. Metabolomics 2015;11(1):122–33. \n25. Marchesi JR., Holmes E., Khan F., Kochhar S., Scanlan P., Shanahan F., et al. Rapid and \nnoninvasive metabonomic characterization of inflammatory bowel disease. J Proteome Res \n2007;6(2):546–51. \n26. Kolho K-L., Pessia A., Jaakkola T., de Vos WM., Velagapudi V. Faecal and serum \nmetabolomics in paediatric inflammatory bowel disease. J Crohns Colitis \n2017;11(3):321–34. \n27. Chambers MC., Maclean B., Burke R., Amodei D., Ruderman DL., Neumann S., et al. A \ncross-platform toolkit for mass spectrometry and proteomics. Nat Biotechnol \n2012;30(10):918–20. \n28. Schmid R., Heuckeroth S., Korf A., Smirnov A., Myers O., Dyrlund TS., et al. Integrative \nanalysis of multimodal mass spectrometry data in MZmine 3. Nat Biotechnol \n2023;41(4):447–9. \n29. Nothias L-F., Petras D., Schmid R., Dührkop K., Rainer J., Sarvepalli A., et al. \nFeature-based molecular networking in the GNPS analysis environment. Nat Methods \n2020;17(9):905–8. \n30. Wang M., Carver JJ., Phelan VV., Sanchez LM., Garg N., Peng Y., et al. Sharing and \ncommunity curation of mass spectrometry data with Global Natural Products Social \nMolecular Networking. Nat Biotechnol 2016;34(8):828–37. \n31. Shannon P., Markiel A., Ozier O., Baliga NS., Wang JT., Ramage D., et al. Cytoscape: a \nsoftware environment for integrated models of biomolecular interaction networks. Genome \nRes 2003;13(11):2498–504. \n32. Dührkop K., Fleischauer M., Ludwig M., Aksenov AA., Melnik AV., Meusel M., et al. SIRIUS \n4: a rapid tool for turning tandem mass spectra into metabolite structure information. Nat \nMethods 2019;16(4):299–302. \n33. Dührkop K., Nothias L-F., Fleischauer M., Reher R., Ludwig M., Hoffmann MA., et al. \nSystematic classification of unknown metabolites using high-resolution fragmentation mass \nspectra. Nat Biotechnol 2021;39(4):462–71. \n34. Wang M., Jarmusch AK., Vargas F., Aksenov AA., Gauglitz JM., Weldon K., et al. Mass \nspectrometry searches using MASST. Nat Biotechnol 2020;38(1):23–6. \n35. Zuffa S., Schmid R., Bauermeister A., P Gomes PW., Caraballo-Rodriguez AM., El Abiead \nY., et al. microbeMASST: a taxonomically informed mass spectrometry search tool for \nmicrobial metabolomics data. Nat Microbiol 2024;9(2):336–45. \n36. Mildau K., van der Hooft JJJ., Flasch M., Warth B., El Abiead Y., Koellensperger G., et al. \nHomologue series detection and management in LC-MS data with homologueDiscoverer. \nBioinformatics 2022;38(22):5139–40. \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n19 \n37. Mohanty I., Mannochio-Russo H., Schweer JV., El Abiead Y., Bittremieux W., Xing S., et al. \nThe underappreciated diversity of bile acid modifications. Cell 2024;187(7):1801–18.e20. \n38. Mannochio-Russo H., Charron-Lamoureux V., van Faassen M., Lamichhane S., Nunes \nWDG., Deleray V., et al. The microbiome diversifies N-acyl lipid pools - including \nshort-chain fatty acid-derived compounds. bioRxivorg 2024:2024.10.31.621412. \n39. Jarmusch AK., Aron AT., Petras D., Phelan VV., Bittremieux W., Acharya DD., et al. A \nuniversal language for finding mass spectrometry data patterns. bioRxiv \n2022:2022.08.06.503000. Doi: 10.1101/2022.08.06.503000. \n40. Caporaso JG., Lauber CL., Walters WA., Berg-Lyons D., Huntley J., Fierer N., et al. \nUltra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq \nplatforms. ISME J 2012;6(8):1621–4. \n41. Gonzalez A., Navas-Molina JA., Kosciolek T., McDonald D., Vázquez-Baeza Y., Ackermann \nG., et al. Qiita: rapid, web-enabled microbiome meta-analysis. Nat Methods \n2018;15(10):796–8. \n42. McDonald D., Price MN., Goodrich J., Nawrocki EP., DeSantis TZ., Probst A., et al. An \nimproved Greengenes taxonomy with explicit ranks for ecological and evolutionary \nanalyses of bacteria and archaea. ISME J 2012;6(3):610–8. \n43. Bolyen E., Rideout JR., Dillon MR., Bokulich NA., Abnet CC., Al-Ghalith GA., et al. \nReproducible, interactive, scalable and extensible microbiome data science using QIIME 2. \nNat Biotechnol 2019;37(8):852–7. \n44. McMurdie PJ., Holmes S. phyloseq: an R package for reproducible interactive analysis and \ngraphics of microbiome census data. PLoS One 2013;8(4):e61217. \n45. Anderson MJ. Permutational multivariate analysis of variance (PERMANOVA). Wiley \nStatsRef: Statistics Reference Online 2017:1–15. Doi: 10.1002/9781118445112.stat07841. \n46. Fernandes AD., Reid JN., Macklaim JM., McMurrough TA., Edgell DR., Gloor GB. Unifying \nthe analysis of high-throughput sequencing datasets: characterizing RNA-seq, 16S rRNA \ngene sequencing and selective growth experiments by compositional data analysis. \nMicrobiome 2014;2(1):15. \n47. Singh A., Shannon CP., Gautier B., Rohart F., Vacher M., Tebbutt SJ., et al. DIABLO: an \nintegrative approach for identifying key molecular drivers from multi-omics assays. \nBioinformatics 2019;35(17):3055–62. \n48. Zhao HN., Kvitne KE., Brungs C., Mohan S., Charron-Lamoureux V., Bittremieux W., et al. \nEmpirically establishing drug exposure records directly from untargeted metabolomics data. \nbioRxiv 2024:2024.10.07.617109. Doi: 10.1101/2024.10.07.617109. \n49. Mehta RS., Mayers JR., Zhang Y., Bhosle A., Glasser NR., Nguyen LH., et al. Gut microbial \nmetabolism of 5-ASA diminishes its clinical efficacy in inflammatory bowel disease. Nat \nMed 2023;29(3):700–9. \n50. Uhlig HH. Monogenic diseases associated with intestinal inflammation: implications for the \nunderstanding of inflammatory bowel disease. Gut 2013;62(12):1795–805. \n51. Bushman FD., Conrad M., Ren Y., Zhao C., Gu C., Petucci C., et al. Multi-omic analysis of \nthe interaction between Clostridioides difficile infection and pediatric inflammatory bowel \ndisease. Cell Host Microbe 2020;28(3):422–33.e7. \n52. Bosch S., Struys EA., van Gaal N., Bakkali A., Jansen EW., Diederen K., et al. Fecal amino \nacid analysis can discriminate DE Novo treatment-naïve pediatric inflammatory bowel \ndisease from controls. J Pediatr Gastroenterol Nutr 2018;66(5):773–8. \n53. Ning L., Zhou Y-L., Sun H., Zhang Y., Shen C., Wang Z., et al. Microbiome and metabolome \nfeatures in inflammatory bowel disease via multi-omics integration analyses across cohorts. \nNat Commun 2023;14(1):7135. \n54. Franzosa EA., Sirota-Madi A., Avila-Pacheco J., Fornelos N., Haiser HJ., Reinker S., et al. \nGut microbiome structure and metabolic activity in inflammatory bowel disease. Nat \nMicrobiol 2019;4(2):293–305. \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint \n\n20 \n55. Chang F-Y., Siuti P., Laurent S., Williams T., Glassey E., Sailer AW., et al. Gut-inhabiting \nClostridia build human GPCR ligands by conjugating neurotransmitters with diet- and \nhuman-derived fatty acids. Nat Microbiol 2021;6(6):792–805. \n56. Louis P., Flint HJ. Formation of propionate and butyrate by the human colonic microbiota. \nEnviron Microbiol 2017;19(1):29–41. \n57. Mills RH., Dulai PS., Vázquez-Baeza Y., Sauceda C., Daniel N., Gerner RR., et al. \nMulti-omics analyses of the ulcerative colitis gut microbiome link Bacteroides vulgatus \nproteases with disease severity. Nat Microbiol 2022;7(2):262–76. \n58. Mohanty I., Allaband C., Mannochio-Russo H., El Abiead Y., Hagey LR., Knight R., et al. \nThe changing metabolic landscape of bile acids - keys to metabolism and immune \nregulation. Nat Rev Gastroenterol Hepatol 2024;21(7):493–516. \n59. Quinn RA., Melnik AV., Vrbanac A., Fu T., Patras KA., Christy MP., et al. Global chemical \neffects of the microbiome include new bile-acid conjugations. Nature \n2020;579(7797):123–9. \n60. Song Z., Cai Y., Lao X., Wang X., Lin X., Cui Y., et al. Taxonomic profiling and populational \npatterns of bacterial bile salt hydrolase (BSH) genes based on worldwide human gut \nmicrobiome. Microbiome 2019;7(1):9. \n61. Guzior DV., Okros M., Shivel M., Armwald B., Bridges C., Fu Y., et al. Bile salt hydrolase \nacyltransferase activity expands bile acid diversity. Nature 2024;626(8000):852–8. \n62. Rimal B., Collins SL., Tanes CE., Rocha ER., Granda MA., Solanki S., et al. Bile salt \nhydrolase catalyses formation of amine-conjugated bile acids. Nature \n2024;626(8000):859–63. \n63. Quinn R., Fu Y., Guzior D., Okros M., Bridges C., Rosset S., et al. Balance between bile \nacid conjugation and hydrolysis activity can alter outcomes of gut inflammation. Research \nSquare 2024. Doi: 10.21203/rs.3.rs-5005563/v1. \n64. Alimov I., Menon S., Cochran N., Maher R., Wang Q., Alford J., et al. Bile acid analogues \nare activators of pyrin inflammasome. J Biol Chem 2019;294(10):3359–66. \n65. Ejderhamn J., Rafter JJ., Strandvik B. Faecal bile acid excretion in children with \ninflammatory bowel disease. Gut 1991;32(11):1346–51. \n66. Fitzgerald RS., Sanderson IR., Claesson MJ. Paediatric inflammatory bowel disease and its \nrelationship with the microbiome. Microb Ecol 2021;82(4):833–44. \n67. Holmberg SM., Feeney RH., Prasoodanan P K V., Puértolas-Balint F., Singh DK., \nWongkuna S., et al. The gut commensal Blautia maintains colonic mucus function under \nlow-fiber consumption through secretion of short-chain fatty acids. Nat Commun \n2024;15(1):3502. \n68. Hidalgo-Cantabrana C., Delgado S., Ruiz L., Ruas-Madiedo P., Sánchez B., Margolles A. \nBifidobacteria and their health-promoting effects. Microbiol Spectr 2017;5(3). Doi: \n10.1128/microbiolspec.BAD-0010-2016. \n69. Kowalska-Duplaga K., Gosiewski T., Kapusta P., Sroka-Oleksiak A., Wędrychowicz A., \nPieczarkowski S., et al. Differences in the intestinal microbiome of healthy children and \npatients with newly diagnosed Crohn’s disease. Sci Rep 2019;9(1):18880. \n70. Sumner LW., Amberg A., Barrett D., Beale MH., Beger R., Daykin CA., et al. Proposed \nminimum reporting standards for chemical analysis Chemical Analysis Working Group \n(CAWG) Metabolomics Standards Initiative (MSI): Chemical Analysis Working Group \n(CAWG) Metabolomics Standards Initiative (MSI). Metabolomics 2007;3(3):211–21. \n71. Uhlig HH., Powrie F. Translating immunology into therapeutic concepts for inflammatory \nbowel disease. Annu Rev Immunol 2018;36(1):755–81. \n72. Sullivan KE., Conrad M., Kelsen JR. Very early-onset inflammatory bowel disease: an \nintegrated approach. Curr Opin Allergy Clin Immunol 2018;18(6):459–69. \n \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted April 26, 2025. ; https://doi.org/10.1101/2025.04.26.650779doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}