New Delhi Metallo-β-Lactamase 5–Producing Klebsiella pneumoniae Sequence Type 258, Southwest China, 2017

We isolated a New Delhi metallo-β-lactamase 5 (NDM-5)–producing Klebsiella pneumoniae sequence type (ST) 258 strain in southwest China during 2017. The blaNDM-5 gene was acquired by horizontal plasmid transfer from NDM-5–producing Escherichia coli. We identified genomic characteristics in ST258 strains that differed from those of global K. pneumoniae carbapenemase–producing strains.

the province where he lived. However, he reported previous hospitalizations in numerous local healthcare centers.
One week after the surgery, the patient had diarrhea, nausea, and vomiting, and a subsequent stool sample positive for Clostridioides difficile toxin A/B, which was suggestive of a C. difficile infection. After he was given vancomycin and metronidazole for 9 days, the assay result was negative for C. difficile toxin A/B. However, bacteremia developed in the patient, and an extended-spectrum--lactamase phenotype Klebsiella pneumoniae (Kp2588) was isolated from a blood culture. The patient was then given piperacillin/tazobactam and moxifloxacin. Two days later, a carbapenem-resistant Escherichia coli (Ec2551) was recovered from a urine culture, although no major urinary tract symptoms were observed. One week later, the patient had symptoms of a severe urinary tract infection and a high fever (41°C). Subsequently, a carbapenemase-resistant K. pneumoniae strain (Kp2573) was recovered from a urine culture. His drug regimen was then changed to tigecycline and later with a combination of fosfomycin and amikacin.
Two weeks later, the carbapenemase-resistant K. pneumoniae appeared to have been cleared from the urinary tract. However, pneumonia developed, and extended-spectrum--Page 2 of 4 lactamase-positive K. pneumoniae, Acinetobacter baumannii, methicillin-resistant Staphylococcus aureus, and Candida albicans were isolated from multiple respiratory tract samples. He also had a high fever. One month later, the health of the patient condition continued to deteriorate. The patient then refused further treatment and requested to be discharged.

Whole-Genome Sequence Analysis
We sequenced 3 isolates by using the Hiseq platform (Illumina, https://www.illumina.com), which produced an average of 9.7 million (range 4.7 million-12.7 million) paired end reads of 150 bp in forward and reverse directions with a predicted minimum coverage of 200×. Illumina raw reads from whole-genome sequencing were trimmed by using Trimmomatic version 0.36 (1), followed by de novo assembly using SPAdes version 3.11.1 (2).
We de novo assembled sequencing reads by using SPAdes version 3.11.1 (2). We closed gaps between contigs by using a PCR, followed by standard Sanger sequencing.
The genomes sequenced in this study (Kp2573 and Kp2588) were compared with genomes deposited in the National Center for Biotechnology Information assembly database (https://www.ncbi.nlm.nih.gov/assembly). All genomes were mapped to the K. pneumoniae reference genome NJST258_2 (GenBank accession no. CP006918) by using Snippy (https://github.com/tseemann/snippy). Prophages were predicted by using PHASTER (6), repeated region were examined by using MUMmer (7), and putative regions of recombination were predicted by using Gubbins (8), followed by filter-using vcftools (9). A recombination-free single-nucleotide polymorphism (SNP) phylogenetic tree was generated by using RAxML version 8.2.4 (https://cme.h-its.org/exelixis/web/software/raxml/) and a general time reversible model of nucleotide substitution with a Gamma model of rate heterogeneity and 4 rate categories (10). SNPs were annotated by using SnpEff 4.3 (11). Phylogenetic clades were identified by using hierarchical Bayesian analysis of population structure. The phylogenetic tree was annotated by using iTOL (12).

Nucleotide Sequence Accession Numbers
Draft genome assemblies and raw reads from our study were deposited in GenBank Bioproject No. PRJNA354234. The draft genome sequences of the 3 strains isolated in this study were deposited in the GenBank whole-genome shotgun database under accession nos, RXHE00000000, RXHG00000000, RXHF00000000, respectively. The raw reads were deposited in sequence read archive database under biosample accession nos. SAMN10583475, SAMN10583476, and SAMN10583477, respectively.