Prediction of the intestinal resistome by a three-dimensional structure-based method

Area of Science:

• Microbiology
• Genomics
• Bioinformatics

Background:

• The intestinal microbiota is a potential reservoir for antibiotic resistance determinants (ARDs).
• Existing evidence is limited due to homology gaps between known ARDs and those in gut microbes.
• An accurate census of the intestinal resistome is needed.

Purpose of the Study:

• To develop and validate a method for predicting ARDs in the human intestinal microbiota.
• To characterize the intestinal resistome and assess ARD transfer potential.
• To investigate the relationship between resistome composition and host factors.

Main Methods:

• Developed a pairwise comparative modeling method based on 3D structure homology.
• Applied the method to predict ARDs in 3.9 million proteins from the human intestinal microbiota.
• Clustered 663 subjects from the MetaHIT cohort into six resistotypes based on resistome composition.

Main Results:

• Predicted 6,095 ARDs (pdARDs), mostly distantly related (29.8% amino acid identity) to known ARDs.
• Found limited evidence for ARD transfer between species within the gut microbiota.
• Identified six resistotypes linked to enterotypes and found pdARD abundance correlated with gene richness, except under antibiotic exposure.

Conclusions:

• The majority of intestinal ARDs are intrinsic to commensal bacteria and not readily shared.
• The findings challenge the view of the gut microbiota as a primary source for ARD transfer to pathogens.
• The study provides a comprehensive catalog of the intestinal resistome and its ecological context.