Characterization of resistant ST15-KL112 K. pneumoniae | IDR

Introduction

Based on the data of the CHINET antimicrobial resistance surveillance program from 2005 to 2021, the resistance rates of Klebsiella pneumoniae to imipenem and meropenem increased markedly.¹ The main mechanisms are carbapenemase production and efflux pumps diversification. Carbapenem-resistant Klebsiella pneumoniae (CRKP) often carry multiple drug resistance genes, the resistance rates to clinical commonly used antimicrobial agents are often more than 60%, only sensitive to tigecycline, polymyxin, and some aminoglycosides. For this class of multi-resistant or pan-resistant Enterobacteriaceae, only the more toxic polymyxin can be selected clinically, or the combination of drugs, such as the combination of tigecycline or aminoglycosides and β-lactamases.² Tigecycline as a derivative of minocycline derivatives not only overcomes the common resistance mechanism of tetracyclines but also was found to have good antimicrobial effects on multiple resistant bacteria.³ Aminoglycoside drugs interfere with protein synthesis by binding to the highly conserved A site of 16S rRNA of the 30S ribosomal subunit of prokaryotes, leading to bacterial death.⁴ However, some multi-resistance, pan-resistance, and “super-bacteria” K. pneumoniae still appear, so we need to understand the resistance mechanism and transmission characteristics of these strains to provide relevant strategies for clinical treatment.

According to data collected by worldwide surveillance programs, OXA-48-like carbapenemases are the predominant type of carbapenemases found in Enterobacterales strains across different nations. There have been over twelve various forms of OXA-48-like carbapenemase identified, with OXA-232 being the third most prevalent OXA-48-like mutation worldwide at present.⁵ After the first identification of OXA-232-producing CRKP in five newborn patients in China in 2017,⁶ there have been multiple small-scale outbreaks of this strain in hospitals affecting children and elderly individuals.⁷ Since tigecycline was first used in clinical treatment in 2005,⁸ cases of drug resistance have emerged, and the most common mechanism for conferring resistance is either overexpression of non-specific active efflux pumps or mutations within ribosomal drug-binding sites.⁹ The mechanism of resistance to aminoglycosides is mainly for bacteria to obtain modified enzymes of aminoglycosides, such as acetyltransferase, phosphorylase and adenosine transferase, and also to obtain plasmid-mediated 16S rRNA methylase resulting in high resistance to all aminoglycoside drugs.¹⁰ RmtF is a 16S-RMTase, first found in France in 2012, and has since only occurred alone in the United States, India, and some European countries.¹¹ RmtF was first reported in 2020 in Shanghai, China.¹² Although CRKP with high levels of tigecycline or aminoglycosides resistance appears. However, CRKP with concurrent resistance to both tigecycline and aminoglycosides was rarely reported.

In our study, we found ten OXA-232-producing K. pneumoniae strains appearance in a tertiary hospital, which occurred from July 2020 to January 2021. These strains were simultaneously resistant to carbapenems, tigecycline, as well as aminoglycosides. We characterized the resistance characteristics and analyzed the genetic evolutionary relationships and related resistance transmission mechanisms. The aim of this study is to describe the resistance mechanisms and transmission patterns of these high-risk bacteria.

Materials and Methods

Bacterial Isolates and Identification

A group of ten CRKP strains were collected from the nine patients in the Affiliated Hospital of Jiaxing University (Zhejiang, China) during July 2020 to January 2021. The strains were K. pneumoniae 7.1/ 7.2/ 11/ 12/ 13/ 15/ 16/ 17/ 20/ 30, and Kpn 7.1 / 7.2 strains morphology were different but come from the same patient The bacteria obtained from the culture were preserved in glycerol broth at a temperature of −80°C. Specimens were placed on Colombian blood Agar plates and then analyzed using matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS) following the guidelines provided by the manufacturer for identification. Clinical data were acquired from the Electronic Medical Records of the patients and the study were approved by the Ethical Committee of the Affiliated Hospital of the Jiaxing University. Escherichia coli American type culture collection (ATCC) 25,922 and Pseudomonas aeruginosa ATCC27853 were used as control strains for the identification and antimicrobial susceptibility test (AST). E. coli ATCC13883 was used as control strain for the quantitative real-time PCR (RT-qPCR) test.

Antimicrobial Susceptibility Testing

A total of 19 antimicrobial agents were tested, including imipenem (IPM), meropenem (MEM), ertapenem (ETP), amoxicillin/clavulanic acid (AMC), piperacillin-tazobactam (TZP), cefuroxime (CXM), cefuroxime axetil, cefoperazone/sulbactam (SCF), ceftriaxone (CRO), ceftazidime (CAZ), cefepime (FEP), cefoxitin (FOX), gentamicin (GEN), amikacin (AMK), Tobramycin (TOB), ciprofloxacin (CIP), levofloxacin (LEV), sulfamethoxazole (SXT) and tigecycline (TGC). Minimum inhibitory concentrations (MICs) for the tested bacteria were determined by using either the agar dilution method or broth microdilution method to assess the minimum inhibitory concentrations of antimicrobial agents. We interpreted the susceptibility breakpoints following the guidelines of the Clinical and Laboratory Standards Institute (CLSI 2024), except for tigecycline, for which we utilized the European Committee on Antimicrobial Susceptibility Testing (EUCAST, 2024) breakpoints.

Tigecycline Resistance and Phenotypic Inhibition Test

The MIC of tigecycline was determined by broth microdilution method. MIC results were interpreted according to the (EUCAST, 2024) clinical breakpoints (for tigecycline, ≤0.5 mg/L is susceptible, and >0.5 mg/L is resistant). E. coli ATCC 25922 was used for quality control in the susceptibility assays. Isolates that showed resistance to tigecycline also tested by adding efflux pump inhibitor Carbonyl cyanide m-chlo rophenylhy’drazone (CCCP) to the medium.

Real-Time Quantitative PCR (RT-qPCR)

RT-qPCR was used to analyze the mRNA levels of efflux pump genes (acrA/acrB and oqxA/oqxB) and their regulators (ramA). Overnight bacterial cultures were diluted 1/100 in Luria-Bertani (LB) broth (Sigma-Aldrich, USA) and grown to log phase at 37°C with vigorous shaking (220 rpm). Total RNA was extracted from the cells, and cDNA was synthesized using an RNA PCR kit (Takara Bio, Otsu, Japan). The yield and quality of RNA and cDNA were determined using a Nanodrop 2000C (Thermo, USA). Q-PCR was performed in a 20 µL reaction volume using Fast Start DNA Master SYBR Green II Mixture (Takara, Tokyo, Japan). A housekeeping gene (rpoB) was the internal reference in the RT-qPCR. The expression levels of each target gene were compared to the expression level of a tigecycline-susceptible E. coli ATCC13883 isolate (expression = 1), which was used as the reference sample. Analysis was conducted utilizing the 2^−ΔΔCT technique.The experiments were conducted three times.

Statistical Analysis

The comparison of statistical significance between the two strains was conducted using a two-tailed Student’s t-test with Bonferroni correction in the GraphPad Prism 6 program. A p-value less than 0.05 was deemed to be statistically significant.

Pulsed Field Gel Electrophoresis (PFGE) and S1 Pulsed-Field Gel Electrophoresis (S1-PFGE)

Genetic relatedness was assessed through molecular typing with PFGE. Samples with genomic DNA digested by XbaI were loaded into wells and subjected to electrophoresis for 17.5 hours at 6 volts per centimeter at 14 degrees Celsius using the Bio-Rad^® CHEF-Mapper XA apparatus (Bio-Rad, California, USA). The pulse durations at the beginning and end were 6 seconds and 36 seconds, respectively. Salmonella enterica strain Braenderup H9812 was used as the indicator.

Plasmid profiles were obtained from 10 CRKP isolates producing OXA-232 using S1-PFGE. In short, the samples were enclosed in a Seakem Gold gel solution at a concentration of 10 g/L, treated with endonuclease S1 nuclease from Takara in Dalian, China, and then analyzed using pulsed-field gel electrophoresis under specific conditions (14°C, 6 V/cm, 120◦ electric field angle, 4.0–40 s conversion time, and 19 h electrophoresis). Genomic DNA from the Salmonella enterica serovar Braenderup H9812 strain was digested with XbaI and utilized as a reference strain and molecular weight marker.

Whole Genome Sequencing (WGS) and Bioinformatics Analysis

Bacterial genomic DNA was isolated using the UltraClean Microbial Kit (Qiagen, NW, Germany) and sequenced from a sheared DNA library with average size of 15 kb (ranged from 10 to 20 kb) on a PacBio RSII sequencer (Pacific Biosciences, CA, United States), as well as a paired-end library with an average insert size of 350 bp (ranged from 150 to 600 kb) on a HiSeq sequencer (Illumina, CA, United States). Sequencing libraries were constructed using the NEBNext RUltraTM II DNA Library Prep Kit for Illumina R (second-generation sequencing) and the SMRTbell RExpress Template Prep Kit 2.0 (third generation sequencing) and then loaded onto NovaSeq S4 flowcell and SMRT Cell 8 M DNA sequencing chip, respectively. Paired-end short Illumina reads were employed for correcting the long PacBio reads with proovread, followed by de novo assembly of the corrected PacBio reads using SMARTdenovo (https://github.com/ruanjue/smartdenovo). ResFinder 3.2 from the Center for Genomic Epidemiology (https//cge.cbs.dtu.dk/services/ResFinder/) was used to detect antimicrobial resistance genes. PlasmidFinder 2.1, pMLST 2.0, and MLST 2.0 software were used to determine the plasmid incompatibility groups, pMLST, and multilocus sequence typing (MLST). We checked if the plasmid was a conjugative plasmid by analyzing the conjugation module using the oriT Finder website (https//bioinfo-mml.sjtu.edu.cn/oriTfifinder/). The IS elements can be identified directly from the established website (https//www-is.biotoul.fr/). We utilized blast (https//blast.ncbi.nlm.nih.gov/Blast.cgi) to identify plasmids with similarities through comparison of their coverages and identities. The circular representation of pKpn20 was generated with CGview (http://stothard.afns.ualberta.ca/cgview_server/). The plasmid linear graph was analyzed by Easyfig software (http://mjsull.github.io/Easyfifig/). Prophages, repeated regions, and recombination regions were screened for single nucleotide polymorphisms (SNP), and the core SNP variations were identified with SNP-dists (https//github.com/tseemann/snp-dists).

Conjugation Experiment

Conjugation assay was utilized to investigate the transferability of bla_OXA-232-encoding plasmids horizontally. The Kpn20 served as the donor strain, while the recipient strain was E. coli J53 (resistant to sodium azide). Both the contributors and beneficiaries were grown to the logarithmic stage (OD600 = 0.4–0.6), combined in equal parts, spun at 12,000 rpm for 1 minute, reconstituted in 20 µL of LB broth, and left to grow overnight at 37°C. Next, the spots were moved to 15 mL tubes for centrifugation and rinsed with 3 mL of LB broth. Subsequently, the serial dilutions were plated onto MH agar plates containing Ampicillin (100 µg/mL) and sodium azide (100 µg/mL). Donor and recipient cells were each utilized as controls to confirm the screening plate antibiotics’ efficacy. The K. pneumoniae 1632 was the Positive control in the test.¹³ PCR was used to confirm the presence of bla_OXA-232 in all transconjugants. The transconjugants underwent susceptibility testing. Conjugation frequency was determined by counting the amount of transconjugants that were produced for each recipient.

Results

The Clinical Characterization of Klebsiella Pneumoniae Co-Harboring Bla_oxa-232 and rmtF

Between July 1th 2020 to January 30th 2021, ten OXA-232-producing K. pneumoniae strains were detected from a tertiary hospital in China. The ten strains were isolated from nine patients. These strains were recovered from sputum (n=8) and blood (n=2).

These patients were admitted into the ICU ward or the neurosurgery ward, and most patients share the same ward at the same time (Table 1 and Figure 1). It is worth that the two wards are located on the third and fifth floors of the same building (the fourth floor no patients), and the frequent movement of humans (medical personnel, patients, and visitors) and portable medical equipment occurs between the two floors, providing many opportunities for bacterial pathogens to spread between wards in hospitals. The mean age of these patients was 68 years, the majority of whom were men (67%). It is noteworthy that most patients have severe underlying disease, including hypertension, heart disease and renal inadequacy et al. The majority of patients were treated with β-lactam antibiotics, such as piperacillin/tazobactam, meropenem, tigecycline, and cefoperazone/sulbactam. Most patients underwent invasive surgery and instrumentation, including peripherally inserted central catheterization, evacuation of meninges hematoma and tracheotomy et al (Table 1). Furthermore, some patients had a poor prognosis, and six patients died during their hospital stay. Other clinical characteristics of these nine patients were summarized and listed in Table 1.

Figure 1 The timeline of OXA-232-producing CRKP outbreak. The red circle indicates the isolation of the OXA-232-producing CRKP strains; The blue long bars indicate the duration of hospitalization phase.

Antibiotic Resistance Phenotypes

High resistance to cephalosporins, fluoroquinolone, and aminoglycoside was observed in the antibiotic susceptibility test for K. pneumoniae, with the exception of a single strain (Kpn30) that showed susceptibility to aminoglycoside. Various strains exhibited varying degrees of resistance to different types of carbapenems, including imipenem, meropenem, and ertapenem. Every strain showed resistance to ertapenem with minimum inhibitory concentrations (MICs) equal to or greater than 8 μg/mL, however, they exhibited varying degrees of resistance to meropenem (MICs between 2 and 4 μg/mL) and imipenem (MICs between 1, 2 and 4 μg/mL) as shown in Table 2.

Table 2 In vitro Activities of Antimicrobial Agents Against OXA-232-Producing K. Pneumoniae Isolates (MICs μg/Ml)

Overexpression of Efflux Pump Mediated Tigecycline-Resistant

Of the ten OXA-232-producing K. pneumoniae isolates were resistant to tigecycline (MICs ranged from 2 to 8 μg/mL). The MIC distribution is presented in Table 2. The tigecycline resistance rate for these strains was 100% (MIC>0.5 mg/L, EUCAST, 2024). Exposure of tigecycline-resistant isolates to the efflux pump inhibitor CCCP resulted in an obvious decrease (1 to 2-fold) in the MICs of tigecycline for the strains (Kpn 13 and Kpn 30) but do not restored susceptibility.

mRNA expression levels of efflux pump genes (acrA/acrB and oqxA/oqxB) and regulators (ramA) were examined by RT-qPCR. Primers used RT-qPCR studies and PCR amplification. Figure 2 depicts the results of RT-QPCR experiments on efflux pump genes in various isolates and the ATCC13883 control group. The Figure 2 comprises three parts: A, B, and C, and each panel includes statistical significance indicators. In Figure 2A, the expression levels of efflux pump genes acrA and acrB are presented. The 10 isolates are compared to control group ATCC13883, with varying expression levels indicated numerically. The statistical significance of the differences in gene expression levels between the control and each isolates is denoted (*pFigure 2B focuses on the efflux pump genes oqxA and oqxB. Similar to Figure 2A, statistical significance indicators are provided to indicate which isolates exhibit significant changes in gene expression. In Figure 2C highlights the efflux pump regulatory gene ramA. Statistical significance is also assessed for ramA expression. The findings indicated a significant correlation between mRNA levels of efflux pump genes and tigecycline MICs (p