Fecal Short-Chain Fatty Acids in Young Children and Their Mothers Are Not Associated With Progression to Islet Autoimmunity
Bree J. Tillett, Min Yan, Dovile Anderson, Vivian Zhang, Darren Creek, Simon C. Barry, Peter G. Colman, Jennifer J. Couper, Maria E. Craig, Elizabeth A. Davis, Aveni Haynes, Tony Huynh, Ki Wook Kim, Grant Morahan, Megan A.S. Penno, William D. Rawlinson, Georgia Soldatos, Timothy Spelman, Rebecca L. Thomson, Jason Tye-Din, Peter J. Vuillermin, John M. Wentworth, Leonard C. Harrison, Helena Oakey, Emma E. Hamilton-Williams, , Simon C. Barry, Peter G. Colman, Jennifer J. Couper, Maria E. Craig, Elizabeth A. Davis, Emma E. Hamilton-Williams, Leonard C. Harrison, Aveni Haynes, Tony Huynh, Dao Huynh, Ki Wook Kim, Gwynne N. Martin, Kelly J. McGorm, Grant Morahan, Megan A.S. Penno, William D. Rawlinson, Georgia Soldatos, Rebecca L. Thomson, Jason Tye-Din, Peter J. Vuillermin, John M. WentworthOBJECTIVE
Reduced microbial short-chain fatty acids (SCFA) have been implicated in type 1 diabetes (T1D) risk. We investigated fecal SCFA concentrations in mother-infant dyads from pregnancy to the onset of islet autoimmunity (IA) to test whether decreased fecal SCFA concentrations increase IA risk.
RESEARCH DESIGN AND METHODS
Children with a first-degree relative with T1D and their mothers were followed in the Environmental Determinants of Islet Autoimmunity (ENDIA) pregnancy-birth cohort study. A nested case-control design included n = 54 infant case subjects that developed one or more persistent islet autoantibodies/clinical T1D and n = 161 age- and sex-matched control subjects without IA. Stool (n = 1,120 samples) from each pregnancy trimester in mothers (median n = 2 samples) and from birth in infants (median n = 6 samples) up until IA onset were used to quantify SCFA by liquid chromatography–mass spectrometry. Weighted conditional logistic regression and linear mixed-effects models investigated associations between SCFA and IA onset.
RESULTS
Fecal SCFA were not associated with IA onset, and SCFA profiles through pregnancy and early childhood were similar between case subjects and control subjects. Infant fecal SCFA concentrations increased until age 1 year. Introduction of solid foods significantly increased acetate (false discovery rate [FDR] = 0.000028) and summed primary SCFA (FDR = 0.00784) concentrations. Higher infant BMI z scores correlated with higher fecal acetate (FDR = 0.0767) and butyrate (FDR = 0.0590). Mothers with T1D had lower acetate (FDR = 0.0696), butyrate (FDR = 0.0959), and propionate (FDR = 0.0696) than mothers without T1D during pregnancy. Concentrations of these three SCFA were lower in infants of mothers with versus without T1D (FDR = 0.0073, 0.0603, 0.0479, respectively).
CONCLUSIONS
These data do not support that lower fecal SCFA concentrations across pregnancy and early life are directly associated with IA development.