Four Carbapenemase Genes Found Coexisting in One Patient’s Lung Sample

In an intensive care unit in Chongqing, China, a male patient with aspiration pneumonia and acute respiratory distress syndrome received intravenous imipenem/cilastatin and vancomycin throughout hospitalization and subsequently died. A bronchoalveolar lavage fluid sample collected from this patient has now become the focus of a remarkable genomic investigation. Researchers at The Second Affiliated Hospital of Chongqing Medical University detected not one but four distinct carbapenemase genes—blaAFM-1, blaNDM-1, blaIMP-1, and blaKPC-2—simultaneously within this single clinical specimen. The finding, published as a Brief Research Report in Frontiers in Cellular and Infection Microbiology, underscores how polymicrobial infections can harbor overlapping and synergistic resistance mechanisms that complicate antimicrobial therapy worldwide.

Carbapenems have long been considered the most reliable last line of defense against severe Gram-negative bacterial infections. Yet the growing prevalence of carbapenemase-producing organisms, particularly among Enterobacteriaceae and non-fermenting Gram-negative bacilli, has severely restricted treatment options and is associated with alarming mortality rates exceeding 50% in some regions. Carbapenem resistance is primarily mediated by carbapenemase enzymes encoded by genes such as blaKPC, blaNDM, blaIMP, blaVIM, and blaOXA-48. These resistance determinants are frequently located on mobile genetic elements like plasmids, transposons, and integrons, facilitating horizontal gene transfer among diverse bacterial species. While the co-occurrence of multiple carbapenemase genes within individual bacterial isolates represents an escalating global concern, the simultaneous detection of multiple carbapenemase genes within a single clinical specimen remains rare.

To characterize the pathogens and resistance genes present in the lavage fluid, the team employed metagenomic next-generation sequencing (mNGS), a culture-independent approach that enables identification of pathogens and resistance genes directly from clinical samples. The BALF sample was processed within 24 hours of collection, with DNA extracted using a magnetic bead-based kit and libraries sequenced on the MGI200 platform using DNA Nanoball technology, generating over 10 million single-end 50-bp raw reads. After quality filtering and removal of human sequences, the remaining reads were aligned to reference databases containing genomes of viruses, bacteria, fungi, and parasites. Microbial identification required at least three non-overlapping reads per organism that were either undetected in negative controls or present at counts at least 10 times higher than in controls.

The metagenomic analysis detected Pseudomonas stutzeri, Klebsiella pneumoniae, Shewanella algae, and Elizabethkingia anophelis in the sample. Beyond the four carbapenemase genes, the specimen contained a diverse array of antimicrobial resistance genes spanning multiple categories, including beta-lactamase genes such as blaCTX-M-65, blaSHV-66, blaOXA-55, and blaOXA-246; aminoglycoside-modifying enzymes; sulfamethoxazole resistance genes sul1 and sul2; trimethoprim resistance determinants dfrA27, dfrA6, and dfrB3; tetracycline, chloramphenicol, rifampicin, fluoroquinolone, and macrolide resistance genes. Various virulence-associated genes were also detected, including those involved in iron acquisition, fimbriae formation, and outer membrane protein synthesis, suggesting a virulence-associated genetic profile within the polymicrobial community.

Using selective culturing techniques with LB agar plates containing 0.5 µg/mL of imipenem, three carbapenemase-producing strains were isolated from the same specimen. Klebsiella pneumoniae strain CQPMC-KpYD carried blaKPC-2, Shewanella algae strain CQPMC-SheW1 harbored blaNDM-1, and Pseudomonas stutzeri strain CQPMC-PStu co-harbored both blaIMP-1 and blaAFM-1. Targeted reanalysis of the BALF sequencing data yielded 148, 477, 232, and 325 reads compatible with blaKPC-2, blaNDM-1, blaAFM-1, and blaIMP-1, respectively, with corresponding coverage breadths exceeding 99% and mean depths ranging from 8.39× to 29.34×. All three isolates showed minimum inhibitory concentrations of ≥16 µg/mL for imipenem, meropenem, and ertapenem, confirming high-level carbapenem resistance across distinct bacterial species.

Whole-genome sequencing using Oxford Nanopore Technologies on the MinION platform revealed the genetic architecture underlying these resistance determinants. The K. pneumoniae isolate was identified as an ST11 strain, a predominant clone associated with carbapenem resistance and linked to hospital outbreaks and high mortality rates across multiple continents. The blaKPC-2 gene was located on a 132,742 bp plasmid designated pCQPMC-KpYD1, which harbors 172 predicted open reading frames and contains IncFII(pHN7A8) and IncR replicons. BLAST database comparisons showed that this plasmid shares 100% coverage and 99.99% identity with the K. pneumoniae plasmid pCRKP-35_KPC, isolated from a patient’s blood sample in China in 2019. Further searches identified 45 plasmids with over 99% identity and coverage, all from K. pneumoniae isolates across various regions in China—43 from hospital patients and 2 from animals—suggesting that related plasmid backbones may contribute to blaKPC-2 dissemination.

The K. pneumoniae isolate also harbored virulence genes including iucABCD, iutA, rmpA, rmpD, and rmpC, located on a 200,970 bp IncHI1B-IncFIB hybrid plasmid lacking additional antibiotic resistance genes. Genetic analysis revealed that the core structure of the blaKPC-2 gene region, termed variable region 1, comprises IS26-ΔTn3-ISKpn27-blaKPC-2-ISKpn6-korC-orf-klcA-ΔrepB. Two additional resistance gene clusters adjacent to blaKPC-2 were identified: a mercury resistance gene cluster flanked by IS5057 and TnpA, and a cluster containing blaSHV-12 enclosed by TnAs1 and IS26. These genetic contexts suggest potential roles for transposition and recombination in the genetic diversity and mobility of blaKPC-2-carrying plasmids.

The Shewanella algae strain CQPMC-SheW1 presented an equally striking genomic picture. Its 4,891,834 bp chromosome assembly contained two blaNDM-1 copies in the same orientation, with start positions separated by 6,399 bp in repeated ISCR1-associated contexts. The related context was ISCR1-partial trpF-bleMBL-blaNDM-1-ΔISAba125-catB3-arr-3-qacEΔ1-sul1. Remarkably, a similar ISCR1-NDM-1 structure is observed in E. coli strain Y5, which shares 100% homology with the chromosomal ISCR1-NDM-1 element in CQPMC-SheW1. Downstream of blaNDM-1, an IntI1 integron with a cassette of six resistance genes was identified. Comparative analysis indicates that the blaNDM-1 gene and its surrounding resistance cluster share up to 99% homology with plasmid p6061604-KPC from E. coli strain 2016061604, highlighting potential cross-species transmission. While that plasmid contains only one complete ISCR1-NDM-1 element, CQPMC-SheW1 exhibits a more complex composition with two same-orientation blaNDM-1 copies, possibly resulting from recombination or transposition events.

The Pseudomonas stutzeri strain CQPMC-PStu, with a 5,242,803 bp genome and 62.2% GC content, co-harbored blaAFM-1 and blaIMP-1, genes encoding metallo-β-lactamases that confer high-level resistance to carbapenems. The blaAFM-1 gene is located within an approximately 12 kb multidrug-resistant region primarily composed of an ISCR29 element and two class 1 integrons. Comparative genomic analysis revealed that this region is highly similar to the chromosome of clinically isolated Pseudomonas qingdaonensis strain L1712hy, exhibiting 99.9% nucleotide identity. The blaIMP-1 gene is located within a truncated class 1 integron arranged as IntI1-blaIMP-1-aac(6′)-IIa-aph(3′)-VIa, a structure closely resembling integron In86. Notably, this multidrug-resistant region exhibits high homology with the chromosome of P. putida strain AHSWHJXPP1, demonstrating 100% coverage and 99% identity, and that strain also carries both blaAFM-1 and blaIMP-1. Similar genomic compositions have been discovered in plasmids of three P. aeruginosa strains, revealing related resistance-associated regions across several bacterial species and suggesting potential mobility of the associated determinants.

The coexistence of four carbapenemase genes within one clinical sample highlights the complexity of polymicrobial carbapenem resistance and underscores the importance of genomic surveillance. The authors note important limitations: the absence of direct transfer experiments and longitudinal evidence limits inference of horizontal gene transfer within the patient, the contribution of each organism and its resistance determinants to the fatal outcome could not be established, and BALF short reads do not establish isolate origin or copy-specific coverage, with independent long-read validation across the SheW1 repeat junctions not established. Nevertheless, the mobile-element-associated genetic contexts of these co-occurring genes suggest their potential mobility, and the findings emphasize the value of comprehensive genomic surveillance networks to detect and monitor the spread of highly resistant organisms, informing international infection control measures and guiding effective therapeutic strategies against the growing global threat of antibiotic resistance.

Subject of Research: Co-occurrence and genetic contexts of four carbapenemase genes in a single clinical specimen from a critically ill patient

Article Title: Co-occurrence and genetic contexts of NDM-1, KPC-2, IMP-1, and AFM-1 carbapenemase genes in a critically ill patient

Article References: Zhang, Y., Zhang, Z., Ying, S., Wang, S., & Xiang, R. (2026). Co-occurrence and genetic contexts of NDM-1, KPC-2, IMP-1, and AFM-1 carbapenemase genes in a critically ill patient. Frontiers in Cellular and Infection Microbiology, 16, Article 1973471. https://doi.org/10.3389/fcimb.2026.1973471

Image Credits: AI Generated

DOI: 10.3389/fcimb.2026.1973471

Keywords: carbapenemase, antimicrobial resistance, NDM-1, KPC-2, IMP-1, AFM-1, metagenomic sequencing, whole-genome sequencing, Klebsiella pneumoniae, Pseudomonas stutzeri, Shewanella algae, genomic surveillance