Infant gut microbiome composition correlated with type 1 diabetes acquisition in the general population: the ABIS study

Aims/hypothesis While autoantibodies are traditional markers for type 1 diabetes development, we identified gut microbial biomarkers in 1-year-old infants associated with future type 1 diabetes up to 20 years before diagnosis. Methods Infants enrolled in the longitudinal general population cohort All Babies In Southeast Sweden (ABIS) provided a stool sample at a mean age of 12.5 months. Samples (future type 1 diabetes, n =16; healthy controls, n =268) were subjected to 16S ribosomal RNA (rRNA) sequencing and quantitative PCR. Microbial differences at the taxonomic and core microbiome levels were assessed. PICRUSt was used to predict functional content from the 16S rRNA amplicons. Sixteen infants, with a future diagnosis of type 1 diabetes at a mean age of 13.3±5.4 years, and one hundred iterations of 32 matched control infants, who remained healthy up to 20 years of age, were analysed. Results Parasutterella and Eubacterium were more abundant in healthy control infants, while Porphyromonas was differentially more abundant in infants with future type 1 diabetes diagnosis. Ruminococcus was a strong determinant in differentiating both control infants and those with future type 1 diabetes using random forest analysis and had differing trends of abundance when comparing control infants and those with future type 1 diabetes. Flavonifractor and UBA1819 were the strongest factors for differentiating control infants, showing higher abundance in control infants compared with those with future type 1 diabetes. Alternatively, Alistipes (more abundant in control infants) and Fusicatenibacter (mixed abundance patterns when comparing case and control infants) were the strongest factors for differentiating future type 1 diabetes. Predicted gene content regarding butyrate production and pyruvate fermentation was differentially observed to be higher in healthy control infants. Conclusions/interpretation This investigation suggests that microbial biomarkers for type 1 diabetes may be present as early as 1 year of age, as reflected in the taxonomic and functional differences of the microbial communities. The possibility of preventing disease onset by altering or promoting a ‘healthy’ gut microbiome is appealing. Data availability The forward and reverse 16S raw sequencing data generated in this study are available through the NCBI Sequence Read Archive under BioProject PRJNA875929. Associated sample metadata used for statistical comparison are available in the source data file. R codes used for statistical comparisons and figure generation are available at: https://github.com/PMilletich/T1D_Pipeline . Graphical abstract