Abstract/Summary

Autism is a neurodevelopment disorder characterised by problems in communication and social interaction skills. In addition to psychological comorbidities, autistic individuals might suffer of GI problems that have been previously associated with gut bacteria, altered dietary intake and barrier function problems. In order to improve these aspects, exclusion diets, such as gluten and casein free diets (GFCF), and/or use of food supplements have been suggested in ASD, but little is known about their impact on gut microbiota composition and metabolic activity, especially for prebiotics. The aim of this work was to investigate the effect of a prebiotic B-GOS on different aspects of autism, in in vitro and in in vivo. B-GOS (65%GOS content) was tested in dose response batch culture experiments inoculated with faecal samples from healthy donors, compared to the commercial B-GOS (50%GOS content). It showed an increase in bifidobacteria and modulation of SCFAs production. 65% B-GOS was then tested on faecal samples from autistic and non-autistic children using a three-stage continuous fermentation system that mimicked various anatomical regions of the colon. Results showed that B-GOS administration significantly increased bifidobacterial populations in the models inoculated with both autistic and non-autistic samples and influences changes in other bacterial groups, such as Clostridium, Roseburia, Bacteroides, Atopobium, Faecalibacterium prausnitzii, Sutterella spp. and Veillonellaceae. In addition, B-GOS modulated SCFA production in both groups, and increased ethanol and lactate inocula from autistic children. Next, a parallel, double blind, randomised study was designed in order to confirm these results in in vivo. Baseline samples from 30 autistic children aged 4-11 years old, were analysed to understand the effect of the exclusion diets on the gut microbiota and metabolome. In addition, parents were asked to fill in GI symptoms and food diaries. Results showed ASD children following GFCF diet had significant reduction in abdominal pain and bowel movement, compared to the un-restricted dietary group, and reduced abundance of Bifidobacterium spp. and Veillonellaceae family, with increased F. prausnitzii and Bacteroides spp. In both groups, no significant differences were found in urine metabolome, but significant correlations were found between bacterial populations and amino acids in faeces. After 6 weeks B-GOS intervention, no effect was reported on GI issues, but there were improvements in sleep habits and antisocial behaviour. B-GOS significantly modulated bifidobacteria and Lachnospiraceae family, as well as the metabolome in children whose diet was not restricted. Butyrate and valerate were the main SFCAs produced and reduced amino acid excretion was detected in faecal samples of the intervention group. Urine samples were dominated by citrate, creatine, creatinine, DMA (dimethylamine), DMG (dimethylglycine), malonate, carnitine, TMAO (trimethylamine-N-oxide), and α-hydroxybutyrate, comparing to the control group, where PAG (phenylacetiyglycine), phenylalanine and β-hydroxybutyrate were detected. Overall, the results presented in this study demonstrated, for the first time, that a prebiotic B-GOS is able to modulate different aspects of autism and be considered as potential dietary therapeutic approach for ASD individuals.