image_310

Temporal development of the gut microbiome in early childhood from the TEDDY study.

RSS de esta página

PubMed ID: 30356187

Imagen Publicación

Stewart CJ, Ajami NJ, O'Brien JL, Hutchinson DS, Smith DP, Wong MC, Ross MC, Lloyd RE, Doddapaneni H, Metcalf GA, Muzny D, Gibbs RA, Vatanen T, Huttenhower C, Xavier RJ, Rewers M, Hagopian W, Toppari J, Ziegler AG, She JX, Akolkar B, Lernmark A, Hyoty H, Vehik K, Krischer JP, Petrosino JF

Nature. Oct 2018. doi: 10.1038/s41586-018-0617-x

COMMENT: This work is oriented to the characterization of the microbiome in early life in a large, multi-centre population as as part of the TEDDY (The Environmental Determinants of Diabetes in the Young) study (https://teddy.epi.usf.edu/TEDDY/). 12,500 stool samples from 903 children from three European countries (Germany, Sweden and Finland) and three US states (Colorado, Georgia and Washington) were analyzed representing those who seroconverted to islet cell autoantibody positivity (IA) or developed type 1 diabetes (T1D) and matched controls. Nested case–control analysis was used to investigate if the microbiome is a good predictor for the development of Islet Autoimmunity or Type 1 Diabetes.

The 16S rRNA region sequenced was the V4 region, using the MiSeq platform (Illumina) with the 2 x 250 bp paired-end read protocol. Each sample was rarefied to 3,000 reads.

The metagenomics shotgun sequencing was done in HiSeq 2000 platform (Illumina) using the 2 x100 bp paired-end read protocol. Each sample was rarefied to 100,000 reads.

16S rRNA and metagenomic illumina sequencing data were analyzed applying Dirichlet multinomial mixtures (DMM) modelling. The entire dataset formed ten distinct clusters and the distribution of these clusters along time allowed to the authors to define 3 different phases of the microbiome in the period between 3 months and 31 months of age. Schematically the authors consider the first year as developmental, the second one as transitional and the third one as the year in which the microbiome is stabilized.

Using linear mixed-effects modelling of the top five phyla and Shannon’s diversity index, we determined three distinct phases of microbiome progression: a developmental phase (months 3–14), a transitional phase (months 15–30), and a stable phase (≥31 months), in which all five phyla and the Shannon diversity index changed significantly during the developmental phase, two phyla (Proteobacteria and Bacteroidetes) and the Shannon diversity index changed significantly during the transitional phase, and all phyla and the Shannon diversity index were unchanged during the stable phase. 

These were the compositional differences detected between Islet Autoimmunity (IA) cohort and Type 1 Diabetes (T1D) with the control cohort:

The relative abundance of the top 50 most abundant genera from 16S rRNA gene sequencing showed only subtle compositional differences, with higher relative abundance of an unclassified Erysipelotrichaceae (P = 0.019) in cases of IA. In the T1D and control cohort, five bacterial genera were associated with T1D onset, with Parabacteroides the most significant (P < 0.001). Eleven bacterial genera were lower in T1D cases, including four unclassified Ruminococcaceae, Lactococcus (P = 0.020), Streptococcus (P = 0.032), and Akkermansia (P = 0.045)

Searching for significant factors associated with the microbiome profiles the authors find that birth mode is significantly associated with microbiome development with higher levels of Bacteroides spp. in delivered vaginally infants during the first year of life. They also find that living with siblings or with furry pets influences the microbiome profiles in early life, accelerating the maturation of the microbiome. The authors higlight that the first year of life is is very important in the microbiome development being breast milk the main factor that influences microbiome development over this period.

Data availability: TEDDY microbiome 16S rRNA gene sequencing and metagenomic sequencing data that support the findings of this study have been deposited in the NCBI database of Genotypes and Phenotypes (dbGaP) with the primary accession code phs001443. v1.p1, in accordance with the dbGaP controlled-access authorization process. Clinical metadata analysed during the current study will be made available in the NIDDK Central Repository at https://www.niddkrepository.org/studies/teddy.

Contributor

Raquel Tobes