Longitudinal study of stool-associated microbial taxa in sibling pairs with and without autism spectrum disorder

Abstract

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental disorder influenced by both genetic and environmental factors. Recently, gut dysbiosis has emerged as a powerful contributor to ASD symptoms. In this study, we recruited over 100 age-matched sibling pairs (between 2 and 8 years old) where one had an Autism ASD diagnosis and the other was developing typically (TD) (432 samples total). We collected stool samples over four weeks, tracked over 100 lifestyle and dietary variables, and surveyed behavior measures related to ASD symptoms. We identified 117 amplicon sequencing variants (ASVs) that were significantly different in abundance between sibling pairs across all three timepoints, 11 of which were supported by at least two contrast methods. We additionally identified dietary and lifestyle variables that differ significantly between cohorts, and further linked those variables to the ASVs they statistically relate to. Overall, dietary and lifestyle features were explanatory of ASD phenotype using logistic regression, however, global compositional microbiome features were not. Leveraging our longitudinal behavior questionnaires, we additionally identified 11 ASVs associated with changes in reported anxiety over time within and across all individuals. Lastly, we find that overall microbiome composition (beta-diversity) is associated with specific ASD-related behavioral characteristics.

Fig. 1. Overall study design.

Each sibling pair consisted of one ASD child and their respective TD sibling. Dietary, lifestyle, and other host variables collected can be viewed in Supplementary File 4. The DADA2 pipeline was used to process the 16S V4 amplicon sequences. Samples from sibling pairs with ASD phenotypes unverified by parent reports or home videos were removed, leaving 432 samples. ASVs that significantly varied between timepoints in a Friedman test or were not present in 3% or more of the samples were removed. 117 ASVs were found to be significantly enriched in either the TD or ASD cohort. 11 of those ASVs were identified by more than one of the contrast methods shown above.

Fig. 2. Relative abundance counts of ASVs significantly associated with the ASD cohort in two independent contrast methods.

ASVs taxonomic annotation of the 16S amplicon (at the families, genus, and species) and the corresponding relative abundance for the 11 taxa identified in at least two independent contrast methods (ANCOM and/or MetagenomeSeq and/or DESeq2) over the three timepoints.

Fig. 3. Overall effect size of diet/lifestyle vs. microbial compositional features.

A Logistic regression models were trained using age + sex (basic), basic + diet/lifestyle features, basic + microbiome features, and basic + diet/lifestyle + microbiome features. Null models included basic + random noise features that matched the range of the original variables. Compared to basic features (AUC = 0.69), diet/lifestyle variables improved cross validated performance significantly (AUC = 0.79) (p = 0.004 rank-sum test) while microbiome features did not (AUC = 0.67, p = 0.25). B Pearson correlation between diet and lifestyle variables significantly related to ASD phenotype within the logistic regression model (vertical) and all other lifestyle variables (horizontal). Columns are annotated by Z-score from a slope test within the combo (basic + diet/lifestyle + microbiome) logistic regression model. C Pearson correlation between axes of variation (Principal component analysis) that are related to ASD phenotype within the combo logistic regression model (vertical) and all lifestyle variables (horizontal). Columns are annotated by Z-score from a slope test within the combo model. D ASV abundances are ranked based on their scores across principal components. A set is either the eight biomarkers associated with ASD or the three associated with TD (Table 1). Axes where biomarkers appear significantly skewed to one end or the other (as compared to randomly distributed) as determined by gene set enrichment analysis are represented.

Fig. 4. Correlations between changes in anxiety and log2 fold changes in relative taxa abundances.

A Change of ASVs abundance correlated with changes in anxiety score across the entire cohort. Positive/negative values on the x axis signify increases/decreases in anxiety respectively between timepoints within an individual. Positive/negative values on the y axis represent an increased/decreased log2 fold change between the relative abundance of an ASV between timepoints within the same individual. R² and p values represent results from a spearman correlation. B ASVs correlated with changes in anxiety scores across both cohorts, and still significant when considering the ASD cohort only. C ASVs that correlate negatively with anxiety in the ASD cohort also correlate with alpha diversity (shannon) of samples.