Molecular Characterization of Carbapenem Resistant Klebsiella pneumoniae and Klebsiella quasipneumoniae Isolated from Lebanon

Abstract

Klebsiella pneumoniae is a Gram-negative organism and a major public health threat. In this study, we used whole-genome sequences to characterize 32 carbapenem-resistant K. pneumoniae (CRKP) and two carbapenem-resistant K. quasipneumoniae (CRKQ). Antimicrobial resistance was assessed using disk diffusion and E-test, while virulence was assessed in silico. The capsule type was determined by sequencing the wzi gene. The plasmid diversity was assessed by PCR-based replicon typing to detect the plasmid incompatibility (Inc) groups. The genetic relatedness was determined by multilocus sequence typing, pan-genome, and recombination analysis. All of the isolates were resistant to ertapenem together with imipenem and/or meropenem. Phenotypic resistance was due to bla_OXA-48, bla_NDM-1, bla_NDM-7, or the coupling of ESBLs and outer membrane porin modifications. This is the first comprehensive study reporting on the WGS of CRKP and the first detection of CRKQ in the region. The presence and dissemination of CRKP and CRKQ, with some additionally having characteristics of hypervirulent clones such as the hypermucoviscous phenotype and the capsular type K2, are particularly concerning. Additionally, mining the completely sequenced K. pneumoniae genomes revealed the key roles of mobile genetic elements in the spread of antibiotic resistance and in understanding the epidemiology of these clinically significant pathogens.

Figure 1

Sequence types, antibiotic resistance genes, and Inc groups detected by PCR and in silico. ST, Sequence type. Classes of antibiotic resistance genes are marked as follows: A, aminoglycoside resistance genes; S, sulfonamide resistance genes; Q, quinolone resistance genes; B, β-lactam resistance genes; T, tetracycline resistance genes; F, fosfomycin resistance genes; R, trimethoprim resistance genes; C, chloramphenicol resistance genes; M, macrolide resistance genes; Inc Groups, plasmid incompatibility groups.

Figure 2

Comparison of the genetic environment of bla_NDM-7 allocated on an IncX3 plasmid in KP28. bla_NDM-7 was carried by an IS5-$∆$ISAba125-bla_NDM-7-ble_MBL-trpF-dsbC-IS26 genetic element.

Figure 3

Prevalence of different virulence factors detected and the total number of virulence factors in a relationship with the total number of resistance genes. (a) Frequency of important VFs detected among the isolates. (b) Blue dots representing one or more isolate with a specific number of VFs and ARGs. Encircled is an emerging type of K. pneumoniae with a high number of both VFs and ARGs.

Figure 4

PFGE profiles, pulsotypes, seven-gene MLST profiles, ST, and K-type of the sequenced isolates. Dendogram generated by BioNumerics software version 7.6.1 showing the relationship of the isolates based on their banding patterns generated by XbaI restriction digestion. a, isolate name; b, pulsotype number; c, seven MLST housekeeping genes profile; d, sequence type; e, K-type.

Figure 5

Circular genomic map and genome comparison of 32 K. pneumoniae genomes. The circle is divided into arcs representing the genomes as labeled. The black histogram represents the GC content and purple-green histogram represents the GC deviation. The isolates are colored according to their STs. The function of genes which were part of major deletion events are labelled at the edge of the rings. KP1 was used as a reference genome.

Figure 6

Pan-genome similarity matrix, patient and isolate data, and maximum likelihood tree based on core and accessory genomes. (a) Maximum likelihood phylogenetic tree based on the alignment of accessory genomes was generated using FastTree 2. *indicates reference genomes. (b) Patient metadata including specimen type, age, ST, K-type, sex (M: male; F: female), country (LEB: Lebanon; SPA: Spain; USA: United States of America; CAN: Canada; GER: Germany), and collection date. (c) Pan-genome was constructed using Roary based on the core and accessory genes showing phylogenetic relatedness of the isolates by blue (present) and white (absent) fragments.

Figure 7

Core genome variation in recombination and SNPs. (a) Maximum likelihood phylogenetic tree based on core genome SNPs. *indicates reference genomes. (b) Recombination densities were detected between the samples using Gubbins.