Functional verification of computationally predicted qnr genes.

Area of Science:

• Microbiology
• Genomics
• Bioinformatics

Background:

• Quinolone resistance (qnr) genes are prevalent in bacteria.
• Novel qnr genes were computationally identified in public DNA sequence data, including metagenomes.

Purpose of the Study:

• To functionally validate novel qnr gene candidates.
• To explore the potential of computational models in discovering antibiotic resistance genes.

Main Methods:

• Probabilistic modeling was used to identify potential qnr genes.
• Inducible recombinant expression systems in Escherichia coli were employed for functional evaluation.

Main Results:

• Four qnr candidates were tested; two novel genes, Vfuqnr and assembled qnr 1, conferred fluoroquinolone resistance.
• Resistance levels correlated with inducer concentrations.
• Co-expression of qnr genes indicated non-synergistic action.

Conclusions:

• A combined approach of computational modeling and experimental validation is effective for identifying novel antibiotic resistance genes.
• This strategy is applicable to both genomic and metagenomic datasets.