Bacterial Genes of Metabolite Biomarkers of Autism Spectrum Disorders in Gut Microbiota of Young Children: Detection by Real-Time PCR and In Silico Analysis

Abstract

Two different approaches, real-time PCR and in silico analysis, were used to study the abundance of bacterial genes encoding the enzymes producing biomarker metabolites in metagenomes of gut microbiota of children with autism spectrum disorders (ASD) and healthy children. As a result of the analysis of the published data, the following biomarker metabolites of ASD were chosen for the research: p-cresol, indole, propionic acid, and D-lactic acid. For the real-time PCR and the transcriptomic analysis, primer sequences were used specifically developed for conservative regions of the genes involved in production of these bacterial metabolites. Nucleotide sequences of the conservative regions were also united into a catalog and were searched in metagenomic assemblies as part of in silico analysis. Comparison of the results revealed a decrease in abundance of the genes in the metagenomes of children with ASD during both PCR and bioinformatics analyses. In addition, real-time PCR made it possible to detect a significant increase in abundance of the genes encoding p-hydroxyphenylacetate decarboxylase and D-lactate dehydrogenase in Bacteroides fragilis and Alistipes finegoldii species and methylmalonyl-CoA decarboxylase in B. fragilis, Alistipes shahii, and Eubacterium rectali species. A low level of the transcripts in total RNA was found in microbiota of a child with severe form of ASD.