Department of Genetics, University of Groningen, University Medical Center Groningen, Groningen 9713GZ, the Netherlands; Department of Pediatrics, University of Groningen, University Medical Center Groningen, Groningen 9713GZ, the Netherlands.
Department of Genetics, University of Groningen, University Medical Center Groningen, Groningen 9713GZ, the Netherlands.
Cell. 2021 Apr 29;184(9):2302-2315.e12. doi: 10.1016/j.cell.2021.03.024. Epub 2021 Apr 9.
By following up the gut microbiome, 51 human phenotypes and plasma levels of 1,183 metabolites in 338 individuals after 4 years, we characterize microbial stability and variation in relation to host physiology. Using these individual-specific and temporally stable microbial profiles, including bacterial SNPs and structural variations, we develop a microbial fingerprinting method that shows up to 85% accuracy in classifying metagenomic samples taken 4 years apart. Application of our fingerprinting method to the independent HMP cohort results in 95% accuracy for samples taken 1 year apart. We further observe temporal changes in the abundance of multiple bacterial species, metabolic pathways, and structural variation, as well as strain replacement. We report 190 longitudinal microbial associations with host phenotypes and 519 associations with plasma metabolites. These associations are enriched for cardiometabolic traits, vitamin B, and uremic toxins. Finally, mediation analysis suggests that the gut microbiome may influence cardiometabolic health through its metabolites.
通过对 338 名个体 4 年后的肠道微生物组、51 个人类表型和 1183 种代谢物的血浆水平进行随访,我们描述了微生物的稳定性和与宿主生理学的关系。利用这些个体特异性和时间稳定的微生物特征,包括细菌 SNP 和结构变异,我们开发了一种微生物指纹识别方法,在对相隔 4 年的宏基因组样本进行分类时,准确率高达 85%。将我们的指纹识别方法应用于独立的 HMP 队列,可使相隔 1 年的样本的准确率达到 95%。我们进一步观察到多种细菌物种、代谢途径和结构变异以及菌株替换的丰度的时间变化。我们报告了 190 个与宿主表型相关的纵向微生物关联和 519 个与血浆代谢物相关的关联。这些关联与心血管代谢特征、维生素 B 和尿毒症毒素富集。最后,中介分析表明,肠道微生物组可能通过其代谢物影响心血管代谢健康。