Markers Mining For Common Intestinal Diseases And Intervention Of Colitis By Synbiotics Through Metagenomics Analysis

In recent years,the incidences of common intestinal diseases,including inflammatory bowel disease(IBD)and colorectal cancer(CRC),have been increasing in the world,imposing health and economic burdens on communities.Early detection and timely intervention could reduce the difficulty of treatment and mortality effectively.Owing to the development of cultivation methods and sequencing technology,many researches have suggest that the compromised intestinal barrier(CIB)and gut microbiota play a pivotal role in the pathogenesis and progression of the IBD and CRC,which are manifested by the accelerated shedding of intestinal epithelial cells,decreased commensals and increased pathogens.These findings have driven the discovery of novel biomarkers that indicate the intestinal barrier function and gut microbiota dysbiosis,as well as the exploration of therapeutic approaches which target intestinal microorganisms,such as the use of probiotics,prebiotics and synbiotics.However,the heterogeneity among individuals and different analytic methods have led to the inconsistent microbial biomarkers across studies for the same disease.Besides,compared with single probiotics or prebiotics,the synbiotics composed of traditional probiotics and prebiotics have better benefits in health,but have limited therapeutic effects on intestinal diseases.It is urgent to discover next-generation probiotics and prebiotics to produce novel synbiotics,for better and more efficient treatment and intervention.To address the above issues,this study collected 13 fecal metagenomics datasets for CRC and two IBD subtypes(Crohn’s disease(CD)and ulcerative colitis(UC)),and performed meta-analysis.This study identified a novel biomarker for CIB and consistent biomarkers across studies,and explored the possibile mechanism that the gut microbial dysbiosis contributed to these common intestinal diseases.This study also constructed multi-class machine-learning models for predicting multiple diseases,supporting the feasibility of metagenome-based non-invasive test for common intestinal diseases.In addition,this study also explored the possibility of disease-intervention using novel synbiotics.Novel synbiotics with different combinations of human gut commensals,probiotics and prebiotics were generated,and tested in dextran sulfate sodium(DSS)-induced colitis mouse models.Those synbiotics with anti-inflammatory effects were screened for future validation and optimization.These results provide experimental basis for clinical intervention of IBD.The main results are as follows:1)This study identified stool-metagenome-derived host DNA contents(HDC)as novel marker of CIB in CRC and CD.Previous researches suggest that CIB could lead to increased HDC in feces of patients with intestinal diseases.According to the characteristic that shotgun sequencing for metagenomics could amplify all DNA in samples,this study calculated HDC as a quantified index of CIB.This study found that HDC were significantly higher in CRC and CD,and associated with gut microbiota dysbiosis.In CD patients with treatment,HDC-related microbial features significantly changed along with reduced HDC.Including HDC as an additional feature to microbiome-based classifiers could improve their performances in disease stratification and treatment response.These results revealed the relationship between intestinal barrier and gut microbiota dysbiosis,and supported the fecal metagenomics as a means for non-invasive diagnosis and assessment of therapeutic response of intestinal diseases.2)This study constructed metagenomics-based multi-class machine learning models for CD,UC and CRC.IBD and CRC manifest as microbial dysbiosis and share similar clinical symptoms.Besides,IBD are considered at high risk of developing CRC.However,it lacks a systemic analysis to investigate the exclusive microbial shifts of these common intestinal diseases.This study collected available fecal metagenomics data,and performed meta-analysis and cross-disease comparisons.This study identified the consistent taxonomic and functional alterations in each disease,revealed phylogenic relationship and interaction pattern among microbes,and mined the potential targets for disease treatment.For clinical applications,this study used the marker species and pathways to construct four machine learning models,of which could predict three diseases and controls with an accuracy of 0.75,and predict three diseases with an accuracy of 0.9.This study improved the understanding of the gut microbiota associated with these common intestinal diseases,and provided powerful support for the application of gut metagenome-based multidisease classifications.3)This study screened out the synbiotic combinations with anti-inflammatory effect by in vivo mouse models and gut microbiome sequencing.The second results in this study demonstrated that there is a close association between gut microbial dysbiosis and IBD.Previous studies showed that synbiotics could modulate intestinal microbes and consequently promote host health.But existing synbiotics fail to meet the requirements of treating IBD.According to the phenomenon that commensals reduce in IBD patients,and the potential benefits of gut commensals,this study collected and mixed commensals isolated from human,traditional probiotics and prebiotics,finally got 28 synbiotic combinations.Through feeding synbiotics to mice and inducing colitis by dextran sulfate sodium(DSS),this study found that fourteen synbiotic combinations could prevent intestinal inflammation significantly,and named them as “beneficial group(BG)”.The other fourteen synbiotics were named as “worthless group(WG)”.This study revealed the significant differences in taxonomic and functional composition between BG and WG,regardless DSS induction.This study also revealed the effects of prebiotics on the colonization of exogenous microbes,indicating the diversity in modulating gut microbiota among prebiotics.The metabonomics and microbiome analysis suggest that different prebiotics have various approaches in reducing inflammation.These results deepened the understanding of the interaction pattern between prebiotics and microorganisms in vivo,and provided experimental basis for the further optimization of synbiotic formula.In summary,based on the gut microbiome,this study identified a novel marker for CIB and a series of microbial markers in IBD and CRC,and explored the intervention effects of novel synbiotics on colitis.This study confirmed the feasibility of fecal microbiome as a non-invasive detection test,constructed the first multi-class models to distinguish IBD and CRC,as well as provided the experimental basis for biotherapey targeting intestinal microbes.