# Investigation of carbapenem-hydrolysing Klebsiella oxytoca strains lacking detectable carbapenemase genes

## TL;DR

Researchers found Klebsiella oxytoca strains that showed carbapenem resistance without known resistance genes, highlighting the need for better diagnostic methods.

## Contribution

Identified K. oxytoca strains with carbapenem resistance despite lacking known carbapenemase genes, suggesting overproduction of OXY-2 as a resistance mechanism.

## Key findings

- K. oxytoca strains showed reduced carbapenem susceptibility but lacked detectable carbapenemase genes.
- mCIM-positive results were likely due to overproduction of the class A β-lactamase OXY-2.
- Combined susceptibility testing with clavulanic and dipicolinic acids can help distinguish resistance mechanisms.

## Abstract

Introduction. Carbapenemase-producing Enterobacterales (CPE) pose a clinical concern due to limited treatment options and plasmid-mediated spread of carbapenemase genes. The modified carbapenem inactivation method (mCIM) is increasingly used in clinical settings to detect CPE.

Gap Statement. Interpretation of mCIM results for isolates with reduced carbapenem susceptibility remains challenging.

Aim. To assess the interpretation of positive mCIM results in Klebsiella oxytoca isolates lacking detectable carbapenemase genes and exhibiting reduced carbapenem susceptibility and to propose a phenotypic testing strategy for accurate evaluation.

Methodology. Antimicrobial susceptibility testing with clavulanic or dipicolinic acid supplementation, phenotypic carbapenemase assays, carbapenem hydrolysis assays and whole-genome sequencing (WGS) were conducted on K. oxytoca strains.

Results.
K. oxytoca strains were isolated from both blood and urine samples of a patient in a Japanese hospital. Although classified as susceptible by clinical breakpoints, these isolates exhibited reduced susceptibility to carbapenems. Although these strains were mCIM-positive, the zinc-supplemented carbapenem inactivation method did not support the presence of metallo-β-lactamase (MBL) activity. MIC reductions in the presence of CVA were consistent with inhibition of a class A β-lactamase. In contrast, dipicolinic acid supplementation did not reduce the meropenem MIC, ruling out MBL production. Furthermore, WGS revealed that both K. oxytoca strains lacked known carbapenemase genes, with blaOXY-2-4 being the only identifiable β-lactamase gene encoding a class A enzyme. No novel β-lactamase genes were identified; however, an enhanced blaOXY-2-4 expression-associated point mutation was found in the promoter region. Both K. oxytoca strains exhibited carbapenem-hydrolysing activity, although to a lesser extent than that of the MBL-producing strain.

Conclusion. We identified K. oxytoca strains that lacked known carbapenemase genes but that yielded positive mCIM results, possibly due to OXY-2 overproduction. Our study findings emphasize the need for cautious mCIM result interpretation for K. oxytoca and highlight the diagnostic value of combined susceptibility testing with clavulanic and dipicolinic acids to distinguish the underlying resistance mechanisms. Combining complementary clinical laboratory tests will enable clarification of the factors underlying mCIM positivity and guide clinical decision-making.