Whole genome analysis of Gram-negative bacteria using the EPISEQ CS application and other bioinformatic platforms

Abstract

Objectives: To determine genomic characteristics and molecular epidemiology of carbapenem non-susceptible Klebsiella pneumoniae, Escherichia coli, Acinetobacter baumannii, and Pseudomonas aeruginosa from medical centres of Mexico using whole genome sequencing data analysed with the EPISEQ^Ⓡ CS application and other bioinformatic platforms.

Methods: Clinical isolates collected from 28 centres in Mexico included carbapenem-non-susceptible K. pneumoniae (n = 22), E. coli (n = 24), A. baumannii (n = 16), and P. aeruginosa (n = 13). Isolates were subjected to whole genome sequencing using the Illumina (MiSeq) platform. FASTQ files were uploaded to the EPISEQ^Ⓡ CS application for analysis. Additionally, the tools Kleborate v2.0.4 and Pathogenwatch were used as comparators for Klebsiella genomes, and the bacterial whole genome sequence typing database was used for E. coli and A. baumannii.

Results: For K. pneumoniae, both bioinformatic approaches detected multiple genes encoding aminoglycoside, quinolone, and phenicol resistance, and the presence of bla_NDM-1 explained carbapenem non-susceptibility in 18 strains and bla_KPC-3 in four strains. Regarding E. coli, both EPISEQ^Ⓡ CS and bacterial whole genome sequence typing database analyses detected multiple virulence and resistance genes: 20 of 24 (83.3%) strains carried bla_NDM, 3 of 24 (12.4%) carried bla_OXA-232, and 1 carried bla_OXA-181. Genes that confer resistance to aminoglycosides, tetracyclines, sulfonamides, phenicols, trimethoprim, and macrolides were also detected by both platforms. Regarding A. baumannii, the most frequent carbapenemase-encoding gene detected by both platforms was bla_OXA-72, followed by bla_OXA-66. Both approaches detected similar genes for aminoglycosides, carbapenems, tetracyclines, phenicols, and sulfonamides. Regarding P. aeruginosa, bla_VIM, bla_IMP, and bla_GES were the more frequently detected. Multiple virulence genes were detected in all strains.

Conclusion: Compared to the other available platforms, EPISEQ^Ⓡ CS enabled a comprehensive resistance and virulence analysis, providing a reliable method for bacterial strain typing and characterization of the virulome and resistome.

Keywords: Carbapenem resistance; Drug resistance; EPISEQ; Pathogenwatch; Whole genome sequencing.