Research On Gut Colonization By Bifidobacterium Longum

Gut colonization by Lactobacillus and Bifidobacterium,and the involved mechanisms remain largely unknown.The experimental settings of these previous studies are abstracted from any natural history of the strains.Based on population genomic analysis,we categorized Lactobacillus and Bifidobacterium species into different lifestyles,and revealed distinct genomic features according to lifestyle,with the aim of predicting gut colonization ability of different species/strains using knowledge of natural history/genomic features,respectively.By using mono-colonization germ-free(GF)mouse model,we validated colonization phenotypes of Lactobacillus and Bifidobacterium species/strains of different lifestyles predicted in the previous section,and compared possible distinct effects of strains with different colonization ability on host physiology.Taking Bifidobacterium longum(B.longum),a probiotic species with the most potent colonization ability,as a representative example,we eclucidated its gut fitness-related evolution mechanisms via using population genomic analysis.In addition,we built a pipeline to detect and qualify colonized biomass of ingested B.longum strains at the strain level.Based on specific pathogen free(SPF)mouse model,we explored gut colonization mechanisms of ingested B.longum strains,preliminarily indeitified and validated the gut colonization-related gene in B.longum,and analyzed probiotic effects brought by strain colonization.The main finsings are as follows:In a systematic analysis of the comparative genomics,we tried to elucidate the genomic contents that account for the distinct host adaptability patterns of Lactobacillus and Bifidobacterium species.The Bifidobacterium species,with species-level phylogenetic structures affected by recombination among strains,broad mucin-foraging activity,and dietary-fiber degrading ability,represented niche conservatism and tended to be host-adapted.The Lactobacillus species stretched across three lifestyles,including free-living,nomadic and host-adapted,as characterized by the variations of bacterial occurrence time,GC(guanine-cytosine)content and genome size,and evolution event frequency,and the presence of human-adapted bacterial genes.The numbers and activity of host-adapted factors,such as bile salt hydrolase and intestinal tissue-anchored elements,were distinctly distributed among the three lifestyles.The strains of the three lifestyles could be separated with such a collection of colonization-related genomic contents(genes,together with genome size and GC content).The conclusions obtained above on lifestyle and gut colonization-related genomic features of Lactobacillus and Bifidobacterium were verified at both species level and strain level:Base on mono-colonization GF mouse model,we observed that the two strains(L.delbrueckii D1 and D12)belonging to species of free-living lifestyle could not initiate colonization,and the population levels of the strain(L.delbrueckii D8)that could successfully colonize were lower than that of strains belonging to species of nomadic(L.plantarum LP1,LP8 and LP13)and host-adapted lifestyle(L.salivarius 5-1,T2 and T5);3 L.salivarius strains belonging to species of host-adapted lifestyle showed the highest colonized biomass(10⁹-10¹¹ CFU/g feces vs 10⁸-10¹⁰ CFU/g for L.plantarum strains vs 10⁸ CFU/g for L.delbrueckii D8).The above findings verified the effectiveness of predicting colonization ability of Lactobacillus species/strains according to lifestyle.For the 4 analyzed strains of B. longum,they could colonize in the gut for 30 days after a single gavage,and the colonized biomass was 10⁸-10⁹ CFU/g feces.For the strain level,three strains of the same Lactobacillus species(owned different distribution modes of gut colonization-related genomic features)showed significant differences in colonization ability.Through the comparative genomic analysis between these strains of the same species but with different colonization phenotypes,we preliminarily identified the colonization-related genes of these analyzed species:BSH and?-galactosidase for L.delbrueckii,pili and serine-rich glycoprotein adhesion for L.plantarum,and mucus-binding protein,?-galactosidase and beta-galactosidase for L.salivarius,validating the effectiveness of using colonization-related genomic features revealed in chapter II to predict the colonization phenotypes of bacterial strains.However,we observed that four B.longum strains showed comparable colonized biomasses.The analysis on the effects of strain colonization on host physiology indicated that influences of tested Lactobacillus and Bifidobacterium strains on ileal transcriptomic profiles were of strain-specific,but also demonstrated some similarity.Meanwhile,for strains of the same species,the strain with higher colonization ability caused more significant alterations on host ileal transcriptomic profiles in terms of numbers of changed genes.By sequencing 109 fecal samples from Chinese subjects based on both 16S r DNA-based bacterial profiling and gro EL gene-based bifidobacterial profiling,we observed that B.longum,was prevalent and dominant within the studied cohort and showed a significant association with host age.By population genomic analysis of 418 B.longum strains,including143 newly sequenced in this study,three geographically-distinct gene pools/populations,BLAsia1,BLAsia2,and BLothers,were identified.Genes involved in cell wall biosynthesis,particularly peptidoglycan biosynthesis,varied considerably among the core genomes of the different populations,but accessory genes that contributed to the carbohydrate metabolism were significantly distinct.Although active transmission was observed inter-host,inter-country,inter-city,intra-community,and intra-family,a single B.longum clone seemed to reside within each individual.A significant negative association was observed between host age and relative abundance of B.longum,while there was a strong positive association between host age and strain genotype[e.g.,single nucleotide polymorphisms in the arginine biosynthesis pathway].Based on bioinformatics tools,we constructed a pipeline to design strain-specific qPCR primers to detect and quantify ingested B.longum strains at the strain level:based on 206 B.longum genomes,gene presence/absence analysis was conducted by Roary software,and the strain-specific gene sequences were thus identified for each of the four B.longum strains that were going to be used in the following animal experiments.After validation,strain-specific qPCR primers were designed targeting to these unique sequences.The colonization experiment of ingested B.longum strains based on SPF mouse model under two types of diets(western diet and standard polysaccharide-rich mouse chow diet)was conducted.The results indicated that only B.longum 274 could colonize successfully in the gut under standard diet after one-week gavage,while the other 3 B.longum strains could not initiate colonization under both diets,and B.longum 274 could not colonize under western diet.Through comparative genomic analysis between good colonizer B.longum 274 and the other 3 B.longum strains that could not initiate gut colonization,the gene encoding?-1,5-arabinanases(ABN)was identified as the colonization-related gene for B.longum.The analysis on the distribution of this gene among Lactobacillus,Bifidobacterium and Bacteroides species indicated that it was a rare gene in B.longum strains,but was distributed widely in various species of Bacteroides.The in vitro assays indicated that in the culture medium with arabinan as sole carbon source,all the tested B.longum strains with this gene showed growth,but the strains without this gene could not grow.Further colonization experiments based on SPF mouse model by using additional B.longum strains with ABN gene demonstrated that all these ABN-harboring strains could colonize in the gut successfully.In addition,the colonization of B.longum caused the significant alterations on composition of gut microbiota,including enrichment of Akkermansia,and significant regulation on the fecal metabolome(modulating the pathways of phenylalanine metabolism,arginine and proline metabolism,and taurine and hypotaurine metabolism).