Resistance to ceftazidime-avibactam in a KPC-2–producing Klebsiella pneumoniae caused by the extended-spectrum beta-lactamase VEB-25

Carbapenem-resistant Enterobacterales, including KPC-producing Klebsiella pneumoniae, represent a major threat to public health due to their rapid spread. The beta-lactam/beta-lactamase inhibitor (BL/BLI) combination ceftazidime-avibactam (CAZ-AVI) has recently been introduced and shown to exhibit excellent activity toward multidrug-resistant KPC-producing Enterobacterales strains. However, CAZ-AVI-resistant K. pneumoniae isolates are being increasingly reported, mostly corresponding to producers of KPC variants that confer resistance to CAZ-AVI but at a cost of carbapenem resistance. We have characterized here, both phenotypically and genotypically, a clinical CAZ-AVI- and carbapenem-resistant KPC-2 K. pneumoniae isolate co-producing the inhibitor-resistant extended-spectrum beta-lactamase VEB-25.


Introduction
VEB-1 (Vietnamese extended-spectrum β-lactamase) was initially described in 1999 in an Escherichia coli isolate obtained from a Vietnamese patient hospitalised in Paris [1]. The enzyme was found to confer resistance to broad-spectrum cephalosporins and be plasmid-encoded and located within a class 1 integron [1]. Since then, this Ambler class A enzyme along with its variants have been predominantly described in Enterobacterales, Pseudomonas spp. and Acinetobacter spp., and have been reported globally [2,3]. To date, 31 VEB variants have been identified (http:// bldb. eu/).
Avibactam, a non-β-lactam-based diazabicyclooctane (DBO) molecule, used in combination with the third-generation cephalosporin ceftazidime, has been shown to have potent inhibitory activity against Ambler classes A and C and to some class D β-lactamases [4]. Hence, the ceftazidime-avibactam (CAZ-AVI) combo is being used as lastresort option for the treatment of serious infections caused by carbapenem-resistant Enterobacterales, especially those producing KPC-type carbapenemases against which it usually exhibits excellent activity [5]. However, in recent years, resistance to CAZ-AVI has emerged among KPC-producing Klebsiella pneumoniae (KPC-Kp), usually mediated via the production of KPC variants and at the cost of carbapenem resistance that is subsequently almost fully lost [6]. VEB variants are known to be effectively inhibited by avibactam [7]. However, a novel VEB variant, namely VEB-25, which was initially reported in 2019 from two KPC-2-producing K. pneumoniae isolates obtained from patients hospitalized in two Greek hospitals, has been reported to be a source of CAZ-AVI resistance [8]. This study identified these two isolates as belonging to Sequence Types ST147 and ST258, harbouring the bla VEB-25 gene on ca. 150 kb InC plasmid and a ca. 170 kb IncC/IncR plasmid, respectively. A subsequent study reported an outbreak within two ICUs in a single hospital in Athens, which affected seven patients within a 5-week period in 2019, and the strain responsible was identified as an ST147 K. pneumoniae isolate co-producing KPC-2 [9]. To our knowledge, this variant has only been reported in Greece so far.
We report here the phenotypic and genotypic characterisation of a CAZ-AVI-resistant K. pneumoniae isolate coproducing KPC-2 and VEB-25, recovered from a patient in Switzerland.

Bacterial isolates
A carbapenem-resistant K. pneumoniae isolate, N3418, was sent to the Swiss National Reference Centre for Emerging Antibiotic Resistance (NARA) for further investigation. Species identification was confirmed using the EnteroPluri-Test (Liofilchem https:// www. liofi lchem. com) and UriSelect 4 agar (Bio-Rad, Cressier, Switerland). The isolate was subjected to the detection of carbapenemase activity by the Rapidec Carba NP test (bioMérieux) and then to NG-Test CARBA 5 test (NG Biotech), according to the manufacturer's instructions. KPC alleles were identified by PCR amplification and Sanger sequencing.

Cloning experiments
The bla VEB alleles were amplified using the above primers, and cloned into pCR-Blunt II-TOPO (Invitrogen, ThermoFisher), before transformation into Escherichia coli Top10. Transformants were selected on plates supplemented with kanamycin (50 mg/L). Successful transformants were confirmed by amplification and sequencing of the bla VEB alleles.

Conjugation experiments and plasmid transformation
Mating-out assays of the bla VEB-25 -harbouring plasmid was attempted using a sodium azide-resistant E. coli J53 as recipient, as previously described [11], with selection on sodium azide at 100 mg/L and ceftazidime-avibactam at 8/4 mg/L. Following failure of conjugation attempts, plasmid pVEB-25_IncC was transformed into E. coli Top10 by electroporation and selected on LB agar plates containing ceftazidime-avibactam at 8/4 mg/L. Transformants were confirmed by the amplification and subsequent sequencing of the bla VEB-25 and IncC repA gene.

Whole genome sequencing (WGS)
WGS was performed using both short and long-read sequencing technologies as previously described, using either a MiSeq (Illumina) or MinION Mk1C (Oxford Nanopore Technologies, Oxford, UK) platforms [11].

IC 50 measurements
IC50 measurements were performed as previously described [17]. Briefly, β-lactamases from crude extracts were prepared and used for the measurement of the specific activity in 100 mM sodium phosphate (pH 7.0). Measurements were performed in a Genesys 10S UV/VIS spectrophotometer (Thermo Scientific) spectrophotometer using a wavelength of 262 nm for cephalothin. The 50% inhibitory concentration (IC 50 ) for KPC variants was determined as the concentration of clavulanic acid (CLA), tazobactam (TAZ), REL, AVI, or VAB, that reduced the hydrolysis rate of 100 μM cephalothin (CEF) by 50%. Extracts were preincubated with inhibitor for 3 min prior to the addition of CEF. All measurements were performed in triplicate.

Results and discussion
Phenotypic and genotypic profiling of N3418 K. pneumoniae isolate N3418 was obtained from the urine of an 84-year-old male who had previously been hospitalised in Greece but had not previously received CAZ-AVI treatment. Susceptibility testing showed that N3418 was resistant to all beta-lactams tested, including the carbapenems and cefiderocol (FDC), in addition to the BL/BLI combination CAZ-AVI. The isolate remained susceptible to IPM-REL and MEM-VAB, but exhibited resistance to colistin (COL; MIC of 16 mg/L) ( Table 1). N3418 was also resistant to ATM-AVI (using a provisional breakpoint of 4 mg/L), and high MICs to unconventional BL/BLI combinations CAZ-REL (128 mg/L, CAZ-CLA (256 mg/L), and CAZ-VAB (32 mg/L). PCR and Sanger sequencing identified the KPC-2 variant that could therefore not explain the resistance to CAZ-AVI. To ascertain the mechanism of CAZ-AVI resistance, the isolate was subjected to WGS which subsequently led to the identification of bla VEB-25 . WGS identified that this isolate belonged to ST323 and harboured the bla KPC-2 , bla VEB-25 , and bla OXA-10 beta-lactamase genes ( Table 2). ST323 has been identified as one of the most dominant lineages associated with ESBL phenotype in Australia, but is not related to the ST147 and ST258 previously identified as VEB-25 producers in Greece [17]. Plasmid replicon type analyses revealed a number of replicon types present, comprising Col440II, ColRNAI, IncC, IncFIB(pKPHS1), IncFIB(pQIL), IncFII(K), IncFII(pKP91), and IncR. Using a combination of long and short read sequencing, we could identify that the bla KPC-2 gene was encoded on a ~ 100 kb FIB(pQIL) plasmid and bla VEB-25 was encoded on a ~ 114 kb IncC plasmid. A deletion of the mgrB gene (involved in lipopolysaccharide synthesis) and disruption into the fiu gene were also found, likely contributing to COL and FDC resistance, respectively [19,20].

Phenotypes of the recombinant E. coli strains producing VEB-25 and VEB-1, and relative enzyme kinetic measurements
Susceptibility testing of recombinant E. coli strains producing VEB-1 and VEB-25 showed that, relative to VEB-1, the VEB-25 recombinant strain exhibited increased MICs for AMC, PTZ, CAZ-AVI, CAZ-REL, CAZ-TAZ, CAZ-CLA, and ATM-AVI (Table 1). Fold changes for each BLBI ranged from 2-to 128-fold, with the greatest change being observed between the BL/BLI combination encompassing the betalactams (CAZ and ATM) and the DBOs, AVI and REL.
IC 50 assays showed that the inhibitory activity of AVI, REL, clavulanic acid and tazobactam were considerably decreased against VEB-25 compared to VEB-1, correlating with the susceptibility testing results ( Table 3). The Lys234 residue is conserved in most class A betalactamases [21], and has been indicated as critical for the inhibitory activity of avibactam toward KPC enzymes. Accordingly, this finding correlates with what has been observed from in vitro mutation studies in KPC enzymes [22]. Noteworthy, in silico Genbank analysis identified the VEB-20 variant exhibiting the same Lys234Arg substitution as VEB-25, encoded on the chromosome of a Pseudomonas aeruginosa recovered from France (GenBank accession number: NG_063894). However, the phenotype conferred by this enzyme remains unknown.

Conclusions
We characterized here a VEB variant responsible for CAZ-AVI resistance that has previously only been rarely reported in Greece. The amino acid change, Lys234Arg, that differentiates VEB-1 from VEB-25, clearly plays an important role in the activity of beta-lactamase inhibitors -most notably the DBO compounds. Overall, VEB-mediated resistance to CZ-AVI in K. pneumoniae has been reported rarely but should be considered when encountering resistant isolates.

Conflict of interest
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.