Regional Variation And Diagnosis Modeling Of Gut Microbiome For Inflammatory Bowel Diseases

Background:IBD(inflammatory bowel diseases)is chronic intestinal inflammation associated with genetic component and environmental factors,and can be broadly divided into Crohn’s disease(CD)and ulcerative colitis(UC)from their clinical phenotypes,with the development of industrial society,the incidence has been increasing in worldwide.The diagnosis of IBD can be missed or delayed due to the non-specific clinical features.Furthermore,standard clinical diagnose is complicated and harmful.Advances in microbiome discovery,many studies have been reported dysbiosis of the gut microbial composition in IBD,suggesting that gut microbiome has potential value in the diagnosis and RISK assessment of IBD.Furthermore,recently,we found that due to host location strongest associate with microbiota variations,microbiota-based metabolic disease models cannot be extrapolated,but dysbiosis of IBD patients gut microbial composition are stronger than metabolic disorders,whether we develop an diagnostic model across population with the potential to discriminate IBD and non-IBD?Objective:1.To observe gut microbiota in patients with inflammatory bowel disease(IBD)in different regions and identify microbiome biomarkers in IBD patients in different regions and to explore the characteristics of their gut microbiota functional changes.2.To develop an diagnostic model with the potential to discriminate IBD and non-IBD and test the extrapolated efficiency of models.3.To observe gut microbiota in patients with CD and UC patients in different regions and establish diagnostic models with the potential to discriminate CD and UC and test the extrapolated efficiency of models.Methods:Observational study,based on our research group IBD patients gut microbiota cohort study,all 16s rRNA gene sequences of 533 IBD patients and 276 healthy controls were collected from China,Spanish,Czech,Chicago,RISK and Boston cohort at corresponding database.QIIME(v1.9.1),microbiome analysis software,was used to analyze microbiota data.LEfSe,microbiome analysis software,was used to identify biomarkers for IBD.Random forest method was used to establish the algorithm to distinguish IBD from HC.PICRUSt was used to predict the functional changes of gut microbiota in IBD patients.Results:The gut microbiome diversity variations between HC and IBD patients in different regions showed different manifestations,influenced by location and other factors.The gut microbiome of IBD patients showed a region-specific dysbiosis pattern,but also shared elements between the different region cohorts.The significant decrease of Ruminococcus,typically prodUCing short chain fatty acids,was the most consistent biomarker in the intestinal tract of IBD patients across regions,and the increase in bacillus and enterobacteriaceae were also the consistent biomarker across regions.A loss of beneficial microorganisms was more associated with patients with IBD than a gain of more pathogenic ones.The glutathione metabolism fUCtion,an important antioxidant substance in human body,was significantly enhanced in the intestinal of IBD patients in Chinese cohort,Spanish cohort,Chicago cohort and the Boston cohort,suggesting that oxidative stress occurred in the intestinal of IBD patients in many regions,which may be a common functional change in the intestinal of IBD patients.Metabolic disease model extrapolation efficiency is limited by location factors,but the extent of IBD patients gut microbiome dysbiosis was greater than the location factor,the model established based on Spanish cohort samples was used to test others population and showed that China,Czech,Chicago and Boston cohorts having AUCs of 71.3%,75.9%,81.3%and 95.4%respectively,the average of 80.96%,which is best model for extrapolation efficiency in this study,moreover,comprehensive diagnostic model was used to test others population and showed that China,Spanish,Czech,Chicago and Boston cohorts having AUCs of 88.3%、77.5%、72.9%.、71.1%and 89.0%,suggesting that there was a good chance of establishing a diagnose model with the potential to discriminate IBD and non-IBD across population.Although a model with better extrapolation efficiency was obtained in this study,the location and other factors were not completely avoided,after all,the average AUC of the extrapolation model established by the Chinese cohort was 69.25%.The difference of gut microbiome between CD and UC patients wes greatly affected by location and other factors,and the extrapolation of trans-regional IBD subtype classification model is limited.Conclusion:Dysbiosis pattern of the gut microbial composition in IBD patients was affected by location and population,etc factors,but also shared elements between the different region cohorts.Therefore,it is feasible to establish a cross-population IBD diagnosis model.