Repeated insertions at positions 261–280 in KPC-2 highlight a ceftazidime-avibactam resistance hotspot

Summary

We report the emergence of KPC-261, a KPC-2 variant identified in a Klebsiella pneumoniae strain exhibiting high-level resistance to ceftazidime-avibactam (CZA). Comparative sequence analysis revealed a repeated insertion event between amino acid positions 261–280, highlighting a mutational hotspot linked to reduced avibactam binding. Our findings uncover a structural mechanism driving CZA resistance and broaden our understanding of β-lactamase evolution.

Graphical abstract

Highlights

• •KPC-261, a new KPC-2 variant, confers ceftazidime-avibactam resistance
• •Insertion-duplication at residues 261–280 is a CZA resistance hotspot
• •KPC-261 weakens avibactam binding (−7.1654 to −6.15 kcal/mol) via altered residues
• •KPC-261 shows a trade-off: 88% reduced ceftazidime efficiency and 7.2-fold higher IC₅₀

clinical microbiology

Introduction

Carbapenem-resistant Klebsiella pneumoniae (CRKP) remains a significant global health concern due to its resistance to a wide range of antibiotics, especially carbapenems.¹ Ceftazidime-avibactam (CZA) has become a critical therapeutic option against infections caused by Klebsiella pneumoniae carbapenemase (KPC, K. pneumoniae)-producing strains.² However, increasing reports of CZA resistance, often mediated by bla_KPC genes, are compromising its clinical effectiveness.³ Notably, resistance can emerge both prior to clinical use of CZA and during therapy, highlighting an urgent challenge.⁴ Of particular concern is the emergence of high-level CZA resistance among strains harboring KPC-2 variants.^5,6 This phenotype is driven by the fact that resistance in KPC-producing K. pneumoniae is frequently associated with amino acid substitutions, insertions, and deletions in the bla_KPC gene, primarily arising from bla_KPC-2 or bla_KPC-3.⁷ To date, more than 280 KPC variants have been reported, with mutations clustering in hotspot regions such as the Ω-loop (amino acid positions 164–179), loop 237–243, and loop 267–275.⁸ Protein engineering and clinical surveillance have revealed a remarkable diversity of Ω-loop alterations. The substitution of aspartate at position 179, a key residue for stabilizing the Ω-loop structure, to tyrosine (D179Y, corresponding to KPC-33) is one of the most common and well-characterized mechanisms of CZA resistance observed in vitro and in clinical settings.⁹

Collectively, these diverse Ω-loop mutations drive a characteristic resistance phenotype: they confer high-level resistance to CZA by enhancing ceftazidime affinity or hydrolysis while reducing the inhibitory potency of avibactam.¹⁰ Concurrently, many of these mutations significantly impair carbapenemase activity, often restoring susceptibility to carbapenems like meropenem and imipenem.¹¹ This “trade-off” in resistance profiles presents a complex challenge for clinical management. For instance, a newly KPC-135 variant has been reported in the ST11-K47 hypervirulent strain, featuring a deletion of Glu168 and Leu169, which enhanced ceftazidime binding while reducing avibactam susceptibility, thereby conferring resistance to CZA.¹² Moreover, the evolution of bla_KPC-2 into bla_KPC-189, via alanine-threonine and aspartate-tyrosine substitutions, led to decreased CZA susceptibility due to altered drug binding affinity.¹³ They represent evolutionary pathways where bla_KPC-2, under antibiotic pressure, acquires CZA resistance through diverse genetic alterations in the Ω-loop and other key regions.

Results

Clinical isolation and genomic features of a CZA-resistant K. pneumoniae strain

We isolated an ST11-KL64 K. pneumoniae strain L3859 from the stool of a 70-year-old male patient with severe pneumonia and septicemia, who had received CZA therapy for 1 week prior to isolation. Antimicrobial susceptibility testing (AST) confirmed high-level resistance to CZA (Table 1). Whole-genome sequencing revealed that strain L3859 harbored a new bla_KPC allele, designated bla_KPC-261, along with multiple other resistance genes including rmtB, bla_CTX-M-65, bla_SHV-12, bla_TEM-1B, bla_LAP-2, fosA6, catA2, qnrS1, sul2, tetA, and dfrA14. Furthermore, two virulence gene clusters encoding siderophores (iutA/iucABCD) and polysaccharide virulence genes (rmpA2) were identified, which are often associated with a hypervirulent phenotype. To investigate the resistance mechanism, we performed gene knockout experiments, plasmid construction, growth curve analysis, and AST. Whole-genome sequencing was performed, and plasmid profiles were examined using S1-PFGE, Southern blotting, and bioinformatics tools.

Table 1.

Antimicrobial drug susceptibility profiles

	MIC (mg/L)/antimicrobial susceptibility
Antibiotic	L3859	L3859/ΔblaKPC	Stbl2/pHSG398	Stbl2/pHSG398:blaKPC-261_p-	Stbl2/pHSG398:blaKPC-261_p+
Amoxicillin/clavulanic acid	64/32/R	64/32/R	4/2/S	4/2/S	16/8/I
Piperacillin/tazobactam	>128/4//R	>128/4//R	8/4/S	0.5/4/S	4/4/S
Ceftazidime	>128/R	>128/R	0.5/S	16/R	128/R
Ceftazidime/avibactam	>128/4/R	2/4/S	0.25/4/S	≤0.03/4/S	8/4/S
Ceftriaxone	>128/R	>128/R	0.06/S	1/S	16/R
Cefotaxime	>128/R	>128/R	0.125/S	0.5/S	8/R
Cefepime	128/R	64/R	0.125/S	0.5/S	8/SDD
Aztreonam	>128/R	>128/R	0.25/S	0.25/S	1/S
Imipenem	1/S	1/S	0.25/S	0.25/S	0.5/S
Meropenem	2/I	0.25/S	≤0.008/S	≤0.008/S	0.03/S
Amikacin	>128/R	>128/R	4/S	0.5/S	0.5/S
Gentamicin	>128/R	>128/R	2/S	0.5/S	≤0.25/S
Ciprofloxacin	>64/R	>64/R	≤0.004/S	0.03/S	≤0.004/S
Levofloxacin	64/R	64/R	≤0.008/S	0.125/S	≤0.008/S
Trimethoprim/sulfamethoxazole	8/152/R	8/152/R	≤0.125/2.375/S	≤0.125/2.375/S	≤0.125/2.375/S
Fosfomycin	256/R	256/R	0.5/S	≤0.25/S	2/S
Tigecycline	0.5/S	0.5/S	0.125/S	0.125/S	0.25/S
Chloramphenicol	>128/R	>128/R	128/R	>128/R	>128/R
Polymixin B	0.5/I	0.25/I	0.25/I	0.25/I	0.25/I

S, susceptible; R, resistont; I, intermediate; SDD, susceptible-dose dependent.

Plasmid localization and sequence analysis of bla_KPC-261

Sequence analysis, S1-PFGE, and hybridization (Figure 1A), revealed the presence of six plasmids. The bla_KPC gene was located on a ∼134 kb IncFII plasmid, showing high homology to CP054728, CP095365, CP082767, and CP033956 (Figure 1B), all plasmids shared similar IS26-tnpR-ISKpn27-bla_KPC-klcA-hin-TnAs1 configurations. Sequence alignment revealed that the bla_KPC gene exhibited high similarity to bla_KPC-58, differing by a G to T substitution at nucleotide position 532 and a six-nucleotide deletion at positions 810–815 (Figure 1C). In contrast, when compared to bla_KPC-2, both bla_KPC-261 and bla_KPC-58 carried repeated insertions, indicating a shared structural alteration contributing to CZA resistance. This unique mutation pattern led to the classification of this newly allele as bla_KPC-261, which has been submitted to the NCBI database.

Figure 1.

Genetic, phenotypic, and structural comparisons of KPC-261 and related insertion-duplication variants

(A) Plasmid profiling by S1-PFGE and Southern blotting.

(B) Comparative analysis of plasmid and mapping.

(C) DNA sequence alignment diagram of bla_KPC-2, bla_KPC-261, and bla_KPC-58.

(D) Protein sequence alignment diagram of KPC-2, KPC-261 and KPC-58.

(E) Growth curve assay, data are represented as mean ± SEM.

(F) Protein sequence alignment of KPC-261 with 72 insertion-duplication variants.

(G) Docking simulation of KPC-2 with avibactam.

(H) Docking simulation of KPC-261 with avibactam.

(I) DNA sequence alignment of bla_KPC-261 with 72 insertion-duplication variants.

Phylogenetic analysis reveals conserved insertion hotspots in CZA-resistant KPC variants

Phylogenetic analysis of 232 known bla_KPC alleles revealed 72 variants containing insertion-duplication mutations. Most of these were linked to CZA resistance. Comparative alignment of these alleles with KPC-2 and KPC-261 identified conserved insertion hotspots, including a 6-bp duplication of GAGCTG at position 493, and tandem duplications between positions 808 and 820 (Figure 1I). At the protein level, this corresponds to the duplication of Glu165-Leu166 (Glu165_Leu166dup) in the Ω-loop and diverse insertions at residues 270–273 in the C-terminal domain (Figure 1F).

Structural modeling and docking predict altered avibactam binding in KPC-261

Protein structural modeling via Swiss-Model revealed a D179Y substitution in the Ω-loop and a six-residue insertion (residues 270–275) in the C-terminal domain of KPC-261 (Figure 1D), where Asp was replaced to Tyr. These alterations are predicted to influence the active site geometry and the electrostatic environment of the substrate-binding cleft, potentially enhancing catalytic activity against β-lactam/β-lactamase inhibitor combinations, such as CZA. Molecular docking simulations using AutoDock Vina 1.2.0 showed that avibactam formed hydrogen bonds with GLU166, ASN132, SER70, THR237, and ASN170 in KPC-2 (Figure 1G) with a binding energy of −7.1654 kcal/mol. In contrast, avibactam interaction with KPC-261 involved residues SER130, ASN132, LYS73, and SER70, with a weaker binding affinity of −6.15 kcal/mol (Figure 1H), suggesting reduced inhibitor efficacy due to structural alterations.

Genetic complementation confirms the role of bla_KPC-261 in CZA resistance

To confirm the role of bla_KPC-261 in CZA resistance, we created a gene knockout mutant (L3859/Δbla_KPC) using the suicide plasmid pCVD442AX and Escherichia coli (E. coli) β2155. The knockout strain exhibited restored susceptibility to CZA. To further validate this, we cloned bla_KPC-261 into the pHSG398 vector with (pHSG398 bla_KPC-261_p+) and without (pHSG398 bla_KPC-261_p−) its native promoter, and introduced them into E. coli Stbl2 by electroporation.

Minimum inhibitory concentration (MIC) testing showed that the strain harboring pHSG398b bla_KPC-261_p+ exhibited significantly higher resistance to CZA than the promoterless counterpart. Growth curve assays demonstrated no substantial fitness differences between L3859 and its Δbla_KPC isogenic mutant, while the E. coli Stbl2 strains with KPC-261 constructs showed slightly increased biomass yield compared to the empty vector control (Figure 1E).

Enzymatic kinetics demonstrate altered substrate hydrolysis and reduced avibactam susceptibility

Enzymatic kinetic analysis revealed that the KPC-261 variant induced distinct and substrate-dependent alterations in catalytic efficiency and inhibitor susceptibility compared to wild-type KPC-2 (Table 2). Against ceftazidime, KPC-261 exhibited a pronounced ∼88% reduction in catalytic efficiency (k_cat/K_m), primarily driven by a marked decrease in turnover number (k_cat from 27 to 2.54 s⁻¹). A more moderate decline was observed against meropenem (∼55% reduction in k_cat/K_m). Hydrolysis of imipenem was severely compromised, with an ∼85% reduction in efficiency. Notably, the variant demonstrated significantly diminished susceptibility to avibactam inhibition, with the IC₅₀ value increasing approximately 7.2-fold compared to KPC-2.

Table 2.

Kinetic parameters and IC₅₀ of purified KPC-2 and KPC-261

Drug	KPC-2				KPC-261
	kcat (s−1)	Km (μM)	kcat/Km (μM−1s−1)	IC50 (μM)	kcat (s−1)	Km (μM)	kcat/Km (μM−1s−1)	IC50 (μM)
Meropenem	69.6	330	0.210	/	10.7	113	0.095	/
Imipenem	153.3	540	0.280	/	11.6	273	0.042	/
Ceftazidime	27.0	370	0.073	/	2.5	270	0.009	/
Avibactam	/	/	/	5.2	/	/	/	37.4

Discussion

The resistance patterns of CZA in KPC-producing K. pneumoniae are frequently attributed to alterations in the enzyme’s active site, particularly in the Ω-loop. Variants such as bla_KPC-33, bla_KPC-86, and bla_KPC-129, affects the enzyme’s ability to hydrolyze β-lactams and mediate resistance.¹⁴ Although these mutations often impair carbapenemase activity and reduce bacterial fitness,^15,16 insertions or duplications in secondary structural regions may confer resistance without compromising catalytic efficiency. Nevertheless, the molecular basis and clinical implications of such structural rearrangements remain insufficiently explored. Here, we characterize the strain L3859 carrying the newly KPC-261 variant, which displays high-level resistance to CZA while maintaining partial resistance to imipenem. Sequence analysis revealed that most KPC-2 variants associated with CZA resistance carry repeated insertions in the 261–280 amino acid region.

The profile of KPC-261 aligns with the evolutionary pattern of KPC variants under CZA selective pressure, characterized by a functional trade-off to evade inhibition.^17,18 The drastically impaired ceftazidime hydrolysis is a core mechanism for CZA resistance, commonly seen in variants with mutations in key loops.¹⁹ The concurrent reduction in carbapenem hydrolysis efficiency, particularly for imipenem, suggests a potential “resistance trade-off,” reminiscent of variants like KPC-151, which regain carbapenem susceptibility while resisting CZA.²⁰ The markedly reduced avibactam sensitivity (elevated IC₅₀) is the direct biochemical basis for CZA failure, a phenomenon consistently reported in newly variants such as KPC-176 and KPC-78.^21,22 This kinetic compromise underscores that resistance evolution can involve collateral damage to enzymatic function.²³ Clinically, while CZA is compromised, this broad functional impairment may inform the evaluation of alternative β-lactam/β-lactamase inhibitor combinations.^18,24

Importantly, the strain L3859 also harbors two virulence gene clusters encoding siderophores (iutA/iucABCD) and polysaccharide virulence genes (rmpA2), which often confer a hypervirulent phenotype in K. pneumoniae.²⁵ This co-occurrence of a CZA-resistant KPC variant with hypervirulence markers in a clinical isolate is of particular concern, as it may lead to infections that are both difficult to treat and highly pathogenic.

We report KPC-261 as a newly bla_KPC variant responsible for CZA resistance. Genetic and functional evidence support its direct role in conferring this phenotype. Our analysis highlights conserved tandem insertion-duplication patterns within structural domains as a key resistance mechanism. These findings reveal diagnostic targets and reinforce the need to monitor structural evolution of bla_KPC variants to inform therapeutic strategies against resistant infections.

Limitations of the study

This single-center, single-isolate study requires further validation for generalizability; additional research is needed on in vivo evolution and fitness cost; molecular docking results need experimental confirmation; and the variant was not tested against other new β-lactamase inhibitor combinations.

Resource availability

Lead contact

Further information or requests can be directed to the lead contact, Hao Xu (xuhao0523@zju.edu.cn).

Materials availability

This study did not generate new unique reagents. There are restrictions to the availability of the K. pneumoniae clinical isolate L3859 used in this study because it was stored in the hospital system that restricts distribution of the bacteria, but we can guide you to the proper contact.

Data and code availability

• •Sequences mentioned in the present study were submitted to the GenBank nucleotide database with BioProject accession numbers: PRJNA390758. In addition, the bla_KPC-261 gene was deposited in the NCBI database under accession no. PV107385.1.
• •This paper does not report original code.
• •This paper does not report any other items.

Acknowledgments

This work was supported by the Natural Science Foundation of Zhejiang Province (grant number LQ24H260001) and Medical Science and Technology Project of Zhejiang Province (grant number 2023KY721).

Author contributions

Y.L. and S.S., writing original draft and formal analysis; R.L., K.W., and L.G., visualization and data curation; T.S. and H.X., writing review and editing, resources, and conceptualization.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Bacterial and virus strains
Klebsiella pneumoniae L3859	This study	L3859
ATCC 25922	Laboratory stock	ATCC 25922
Salmonella Braenderup serotype strain (H9812)	Laboratory stock	H9812
Escherichia coli Stbl2	Laboratory stock	Stbl2
Escherichia coli BL21	Laboratory stock	BL21
Escherichia coli β2155	Laboratory stock	β2155
Klebsiella pneumoniae L3859/ΔblaKPC	This study	L3859/ΔblaKPC
Escherichia coli Stbl2/pHSG398	This study	Stbl2/pHSG398
Escherichia coli Stbl2/pHSG398:blaKPC-261_p-	This study	Stbl2/pHSG398:blaKPC-261_p-
Escherichia coli Stbl2/pHSG398:blaKPC-261_p+	This study	Stbl2/pHSG398:blaKPC-261_p+
Chemicals, peptides, and recombinant proteins
LB Broth	Sangon Biotech	Cat: A507002
LB agar	Sangon Biotech	Cat: A507003
MacConkey agar	OXOID	Cat: CM0115B
Deposited data
Sequence of L3859	This study	BioProject: PRJNA390758
Sequence of L3859/ΔblaKPC	This study	BioProject: PRJNA390758
Sequence of blaKPC-261	This study	GenBank: PV107385.1
Software and algorithms
Unicycler software	Wick et al.26	Unicycler
PlasmidFinder	Carattoli, A. et al.27	PlasmidFinder
Prokka	https://github.com/tseemann/prokka	Prokka
BLASTn	https://blast.ncbi.nlm.nih.gov/Blast.cgi	BLASTn
Kleborate	https://github.com/klebgenomics/Kleborate	Kleborate
ResFinder	https://www.genomicepidemiology.org/services/	ResFinder
BRIG	Alikhan et al.28	BRIG
Swiss-Model	https://swissmodel.expasy.org/	Swiss-Model
AutoDock Vina	Eberhardt et al.29	AutoDock Vina
PyMol	Lill et al.30	PyMol
ADFRsuite	Ravindranath et al.31	ADFRsuite
Clustal Omega	Sievers et al.32	Clustal Omega
Jalview	Waterhouse et al.33	Jalview
R	https://www.r-project.org	R
GraphPad Prism	https://www.graphpad-prism.cn	GraphPad Prism

Experimental model and study participant details

Bacterial strain

Strain L3859 was isolated from the stool sample of a 70-year-old male patient with severe pneumonia and septicemia at the First Affiliated Hospital, Zhejiang University School of Medicine. The strain was identified as Klebsiella pneumoniae by whole-genome sequencing (WGS) and was classified as sequence type ST11 and capsular type KL64. L3859 was found to harbor a newly bl_aKPC variant, bla_KPC-261, along with multiple other resistance genes and two virulence gene clusters encoding siderophores and polysaccharide virulence genes. To investigate the resistance mechanism, gene knockout experiments, plasmid construction, growth curve analysis, and antimicrobial susceptibility testing (AST) were performed. Whole-genome sequencing was conducted, and plasmid profiles were examined using S1-PFGE, Southern blotting, and bioinformatics tools.

Ethical approval and consent to participate

This study was conducted following the Declaration of Helsinki and obtained approval from the clinical research ethics committee of The First Affiliated Hospital, Zhejiang University School of Medicine [No. 2018–752]. The ethics committee waived the requirement for informed consent for the following reasons: the study was retrospective; all patient information was anonymized; and the collection and storage of clinical isolates were part of the hospital diagnostic laboratory’s routine workflow, with no impact on patient health.

Method details

Bacterial isolation and clinical context

Klebsiella pneumoniae strain L3859 was isolated from a stool sample of a 70-year-old male patient undergoing ceftazidime-avibactam therapy at The First Affiliated Hospital, Zhejiang University School of Medicine. For bacterial isolation, the stool was spread on MacConkey agar plates containing 2 μg/mL cefotaxime, then selected monoclonally and identified using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF/MS).³⁴

Antimicrobial susceptibility testing (AST)

Minimum inhibitory concentrations (MICs) were determined via broth microdilution and interpreted according to the CLSI guidelines (2026) and European Committee on Antimicrobial Susceptibility Testing (EUCAST) 2022 criteria. Escherichia coli ATCC 25922 was used as a control strain. Antibiotics tested included ceftazidime-avibactam, carbapenems, β-lactams, fluoroquinolones, aminoglycosides, polymyxins, and others.

Whole genome sequencing and bioinformatics

Genomic DNA was extracted using a commercial DNeasy UltraClean Microbial Kit (Qiagen), followed by sequencing on both Illumina NovaSeq and Oxford Nanopore MinION platforms.³⁵ Reads were assembled using Unicycler v0.4.8²⁶ and annotated via Prokka v1.14.5 (https://github.com/tseemann/prokka). Plasmid identification was conducted using PlasmidFinder²⁷ and BLASTn. Kleborate (https://github.com/klebgenomics/Kleborate) was used to conduct multilocus sequence typing (MLST), determine KL capsule type, and identify virulence genes. Antimicrobial resistance genes were examined with Resfinder on the CGE server (https://www.genomicepidemiology.org/services/).

Plasmid profiling and southern blot

BLAST Ring Image Generator (BRIG v0.95)²⁸ was employed to generate the comparative circular Figures. S1-nuclease pulsed-field gel electrophoresis (S1-PFGE)³⁶ and Southern blotting were performed to estimate plasmid sizes and confirm the location of bla_KPC-261, using digoxigenin-labeled probes by Roche DIG-High Prime DNA Labeling and Detection Starter Kit. The Xba-I digested Salmonella Braenderup serotype strain (H9812) was used as the DNA size marker.

Gene cloning and mutant construction

The bla_KPC-261 gene was amplified by PCR and ligated into the pHSG398 vector. Constructs with and without native promoters were introduced into E. coli Stbl2 via electroporation. A Δbla_KPC knockout mutant was generated in L3859 using the suicide plasmid pCVD442AX³⁷ and E. coli β2155 donor strain. The suicide plasmid pCVD442AX carrying the bla_KPC-261 gene was introduced into E. coli β2155 via electroporation to create the donor strain β2155/pCVD442AX-Δbla_KPC. This strain was conjugated with a Klebsiella recipient, and apramycin-resistant colonies, indicating plasmid integration, were selected and named Kp/pCVD442AX-Δbla_KPC. These colonies were cultured and spread on LB plates with 10% sucrose to select for single-crossover resolution. PCR screening identified clones with the bla_KPC gene deleted, resulting in the strain Kp/Δbla_KPC.

Structural modeling and docking simulations

Protein models of KPC-2 (PDB ID: 5JU3) and KPC-261 were generated using Swiss-Model (https://swissmodel.expasy.org/). Docking of avibactam (Pubchem id: 98350) was performed using AutoDock Vina v1.2.0,²⁹ before docking, the receptor protein was prepared using PyMol v2.5.5³⁰ by removing water, salts, and small molecules. A grid box was set to cover the entire protein, and both the receptor and small molecules were converted to PDBQT format using ADFRsuite v1.0³¹ for AutoDock Vina docking. The exhaustiveness was set to 32, with other parameters at default. The conformation with the highest docking score was selected as the binding conformation, and interactions were visualized with PyMOL.

Phylogenetic and sequence analysis

All known bla_KPC alleles (n = 232) were downloaded from NCBI. Multiple sequence alignments were conducted with Clustal Omega,³² and insertion-duplication variants were annotated relative to the KPC-2 reference. The bla_KPC nucleotide and protein sequences were ultimately displayed through Jalview.³³

Growth curve assays

Overnight cultures of L3859, Δbla_KPC, and E. coli transformants were inoculated into LB broth and monitored in a microplate reader at OD600 every 5 min for 20 h at 37°C.¹³ Using R version 4.3.1, growth curves were plotted from OD600 readings, and relative growth rates along with the area under the curve (AUC) were computed.

KPC carbapenemase extraction and evaluation of enzyme kinetics

The bla_KPC-2 and bla_KPC-261 coding sequences were inserted into the pET-28a vector (Sangon Biotech, Shanghai, China) and introduced into E. coli BL21 (DE3) competent cells for the purpose of protein expression. After purification, the KPC carbapenemases reached a concentration of 1–1.5 mg/mL. The hydrolysis of β-lactam substrates, including ceftazidime (260 nm), imipenem (300 nm), and meropenem (298 nm), was used to determine the steady-state kinetic parameters (K_m and k_cat). Nonlinear regression was employed to fit the data to the Michaelis–Menten equation, and the IC₅₀ values for avibactam were calculated by assessing the inhibition of ceftazidime hydrolysis, using GraphPad Prism 10 (GraphPad Software, MA, USA), the data were analyzed with a four-parameter variable-slope model for dose-response inhibition.

Quantification and statistical analysis

There are no quantification or statistical analyses to include in this study.