Molecular characterization of two conjugative resistance plasmids from a clinical multidrug-resistant ST15 Klebsiella pneumoniae Authors: Ruixue Tao, Meng Li, Nan Zhang, Linke Cao, Jie Zhang, Hanqi Li, Wenbo Yan, Yiming Yang, Tietao Wang, Yani Zhang, Ke Xing Published: 2026 ABSTRACT The increasing emergence of multidrug-resistant (MDR) Klebsiella pneumoniae strains has become a threat to public health, and it is important to understand the mechanisms underlying adaptive evolution. In this study, we characterized a clinical MDR isolate of K. pneumoniae, designated KPA1, which belongs to sequence type 15 (ST15), along with its two associated resistance plasmids, pKPA1-1 and pKPA1-2. Conjugation assays demonstrated that both plasmids were capable of self-transfer. Notably, pKPA1-1 could be transferred to both K. pneumoniae and Escherichia coli, while pKPA1-2 exhibited a restricted transfer capability, being transferable only to K. pneumoniae. Antimicrobial susceptibility testing, along with plasmid stability assays, indicated that the resistance spectrum of pKPA1-2 similar to that of pKPA1-1, and pKPA1-2 was more stable than pKPA1-1. In addition, virulence assays revealed that the acquisition of either plasmid resulted in a reduction in bacterial pathogenicity. This study elucidates the molecular characterization of conjugative resistance plasmids pKPA1-1 and pKPA1-2. These findings shed light on the adaptive mechanisms underlying multidrug resistance in K. pneumoniae.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12864-026-12953-y. Introduction Klebsiella pneumoniae is an important Gram-negative opportunistic pathogen, which causing various infections including pneumonia, bloodstream infections, urinary tract infections, and abdominal infections [1]. Over the last few decades, strains derived from classic K. pneumoniae have acquired additional genetic traits, such as virulence factors and resistance genes, facilitating the emergence of hypervirulent and antibiotic-resistant strains [2, 3]. The rise of carbapenem-resistant K. pneumoniae (CRKP) and extended-spectrum β-lactamase (ESBL)-producing strains represents a significant threat to public health [4]. ESBL genes are typically located on mobile genetic elements (MGEs), including plasmids, transposons, integrons, and insertion sequences. These MGEs play a critical role in facilitating the horizontal transfer of antibiotic resistance genes both within and between bacterial species [5]. Sequence type 15 (ST15) K. pneumoniae has recently emerged as a globally disseminated high-risk multidrug-resistant clone associated with frequent hospital outbreaks and often co-harboring virulence plasmids and ESBL genes [6, 7]. The widespread horizontal transfer of resistance plasmids across K. pneumoniae strains from diverse geographical regions has led to substantial heterogeneity in resistance profile among ST15 isolates, thereby complicating clinical treatment strategies [8]. In recent years, the increasing global prevalence of MDR and hypervirulent K. pneumoniae (hvKP) strains has raised significant public health concerns [9]. K. pneumoniae utilizes a diverse array of virulence factors to ensure survival and promote disease progression, including the polysaccharide capsule, lipopolysaccharide, siderophores, pili, and iron acquisition systems [10]. Capsule production plays a critical role in conferring hypervirulence, while biofilm formation contributes to the maintenance of virulence within the host [11]. In this study, we characterized a clinical multidrug-resistant ST15 K. pneumoniae KPA1, and its two resistance plasmids, pKPA1-1 and pKPA1-2, to elucidate the genetic basis of antimicrobial resistance and adaptive evolution. Notably, pKPA1-1 and pKPA1-2 exhibited similar resistance profiles. We hypothesis that the acquisition of plasmid pKPA1-2 may serve as a compensatory mechanism, allowing the retention of resistance traits that could potentially be lost due to the instability of pKPA1-1. Our findings contribute to a deeper understanding of adaptive mechanisms underlying multidrug resistance in K. pneumoniae. Materials and methods Bacterial strain and identification Klebsiella pneumoniae strain KPA1 was isolated from wound secretion in Shaanxi Province, China, in November 2023. The bacterial species was confirmed using matrix-assisted laser desorption ionization-time of flight mass spectrometry (Bruker, Germany). Strains were cultured in Luria-Bertani (LB) broth at 37 °C. Antimicrobial susceptibility testing Antimicrobial susceptibility testing was assessed using the microdilution method. The antibiotics tested included cefepime, ceftriaxone, cefotaxime, levofloxacin, cefoperazone/sulbactam, piperacillin-tazobactam, meropenem, tigecycline, amikacin, and colistin. Minimum inhibitory concentrations (MICs) were interpreted based on the Clinical and Laboratory Standards Institute (CLSI) of 2024. DNA sequencing and bioinformatics analysis Genomic DNA was extracted using the PureLink Genomic DNA Mini Kit (Invitrogen, United States) according to the manufacturer’s instructions. DNA libraries were prepared using a Nextera XT DNA Library Preparation Kit (Illumina, United States) following the manufacturer’s protocol. Whole-genome sequencing (WGS) was performed a hybrid approach both Illumina Novaseq6000 (short-read: 250 bp paired-end reads) and Oxford Nanopore PromethION (long-read: 4 Mbp paired-end reads) platforms. The quality of the raw reads was assessed using FastQC v0.12.1 (https://github.com/s-andrews/FastQC). Bacterial species were confirmed using Kraken2 v2.1.3 (https://github.com/DerrickWood/kraken2) [12]. Reads were de novo assembled using Spades v3.15.5 (https://github.com/ablab/spades) [13]. Draft genomes were assessed using CheckM v1.2.2 (https://github.com/Ecogenomics/CheckM) [14]. Assembled genome sequences were annotated with Rapid Annotation using Subsystem Technology (RAST) v2.0 [15]. The Multilocus sequence typing (MLST), resistance genes, and plasmid replicon types were identified using MLST 2.0 [16], ResFinder 4.6 [17] and PlasmidFinder 2.1 [18] from Center for Genomic Epidemiology (http://www.genomicepidemiology.org). The insertion sequences (ISs) were detected using Isfinder [19]. Comparative alignment of plasmid sequences was visualized using the BLAST Ring Image Generator (BRIG) v0.95 [20]. The phylogenetic trees were constructed from single-nucleotide polymorphism (SNP) using the CSI Phylogeny 1.4 software [21]. The phylogenetic trees were visualized and edited using Interactive Tree of Life (iTOL) v7.2.2 [22]. Biofilm formation assays Biofilm formation was assessed as described previously [23]. Briefly, overnight cultures grown in LB medium were diluted (1:1000) into 1 ml of fresh LB medium in glass tube and incubated statically at 25 °C for 22 h. Biofilms attached to the sides of the glass tubes were gently washed twice with sterile water and stained with crystal violet (CV). The biofilm biomass was quantified by dissolving the CV-stained biofilm in 95% ethanol, and the absorbance of the CV solution was measured at 590 nm. Conjugation experiments To evaluate the transferability of the resistance plasmid pKPA1-1 and pKPA1-2, E. coli strain DH5α (pBBR1MCS-5) and K. pneumoniae strain ATCC700603 were used as recipients. Briefly, a 10:1 ratio of donor to recipient bacteria was combined in 1 mL of LB broth and cultured overnight. Subsequently, the culture mixture was separately spread on LB agar plates containing both gentamicin (150 µg/mL) and trimethoprim (600 µg/mL), as well as LB agar plates containing both gentamicin (150 µg/mL) and levofloxacin (0.5 µg/mL). The ability of plasmid pKPA1-1 to transfer conjugatively was assessed by counting colony growth on LB agar plates containing both trimethoprim and gentamicin. Similarly, the ability of plasmid pKPA1-2 to transfer conjugatively was evaluated by counting colony growth on LB agar plates containing both levofloxacin and gentamicin. Conjugative transfer frequencies were calculated as the number of transconjugants per donor. All transconjugants were further verified by PCR using the following primer sequences: the gene dfrA12 of plasmid pKPA1-1 (F: ATGAACTCGGAATCAGTACGCATTT, R: TTAGCCGTTTCGACGCGCATAAA) and the gene qnrS1 of plasmid pKPA1-2 (F: ATGGAAACCTACAATCATACATATC, R: TTAGTCAGGATAAACAACAATACCC). Plasmid stability assays The plasmid stability of pKPA1-1 and pKPA1-2 in the K. pneumoniae ATCC700603 was assessed by passage on antibiotic-free LB broth. The K. pneumoniae ATCC700603/pKPA1-1 and K. pneumoniae ATCC700603/pKPA1-2 were cultured with shaking at 37 °C and diluted 1:1000 in LB broth without antibiotics, and passaged continuously for 10 days (approximately 200 generations). Culture samples were collected every two days and stored in glycerol at -20 °C. After 10 days, the collected cultures were serial dilution and spreading onto LB agar plates without antibiotics, as well as LB agar plates containing trimethoprim (600 µg/mL) and levofloxacin (0.5 µg/mL). The retention rate of pKPA1-1 activity was calculated by dividing the number of colonies on trimethoprim-containing LB agar by the number of colonies on antibiotic-free LB agar. Similarly, the retention rate of pKPA1-2 activity was calculated by dividing the number of colonies on levofloxacin-containing LB agar by the number of colonies on antibiotic-free LB agar. Each experiment was performed in triplicate with three independent cultures. Error bars represent the mean ± SD from three independent experiments. Galleria mellonella killing assays The G. mellonella infection model is a widely accepted animal model, and the experiments were performed as described previously [23]. Cells were grown in LB broth to an optical density of 1.0 at OD600, washed three times with sterile phosphate buffer saline (PBS), and then diluted in PBS to a final concentration of 104 CFU/mL. A 100-µl Hamilton syringe was used to inject 20 µL of bacterial suspension into the last left proleg of G. mellonella via the last left proleg. The control group was injected with 20 µL of sterile PBS. Infected larvae were incubated and monitored in the dark at 37 °C in an incubator. The number of dead caterpillars was recorded at each time point. Caterpillars were considered dead when they showed no movement in response to touch. Results KPA1 exhibited a multidrug-resistant phenotype and harbored two plasmids K. pneumoniae strain KPA1 exhibited a MDR phenotype, showing resistance to all tested β-lactam antibiotics and quinolones, including cefepime, ceftriaxone, cefotaxime, and levofloxacin, while remaining susceptible to tigecycline, amikacin, meropenem, and colistin (Table 1). A string test was performed on LB agar by stretching bacterial colonies with an inoculation loop showed no stickiness (Data not shown), indicating that KPA1 is not a hypermucoviscous strain. Biofilm formation assays revealed that KPA1 produced biofilms at levels comparable to K. pneumoniae ATCC70060 (Fig. 1A). In a Galleria mellonella infection model, KPA1 caused significantly lower larval mortality rate than K. pneumoniae ATCC70060 strain (Fig. 1B), suggesting a reduced virulence phenotype. Table 1Phenotypic and genotypic characteristics antimicrobial susceptibility of K. pneumoniae strain KPA1 and its transconjugantsStrain IDMIC (µg/mL)TIGLVXFEPCROCXMTZPCSLAMKMEMCSTKPA11 (S)> 8 (R)16 (R)64 (R)32 (R)16/4 (S)32/16 (I)8 (S)2 (I)1 (S)ATCC7006034 (I)0.5 (S)1 (S)2 (I)16 (I)8/4 (S)16/8 (S)8 (R)1 (S)1 (S)ATCC700603/pKPA1-18 (R)0.5 (S)16 (R)64 (R)32 (R)16/4 (S)16/8 (S)8 (S)1 (S)1 (S)ATCC700603/pKPA1-28 (R)> 8 (R)16 (R)64 (R)32 (R)16/4 (S)32/16 (I)8 (S)1 (S)1 (S)TGC Tigecycline, LVX Levofloxacin, FEP Cefepime, CRO Ceftriaxone, CXM Cefuroxime, TZP Piperacillin-tazobactam, CSL Cefoperazone/sulbactam, AMK Amikacin, MEM Meropenem, CST Colistin, S Susceptible, I Intermediate, R Resistant Table 1Phenotypic and genotypic characteristics antimicrobial susceptibility of K. pneumoniae strain KPA1 and its transconjugantsStrain IDMIC (µg/mL)TIGLVXFEPCROCXMTZPCSLAMKMEMCSTKPA11 (S)> 8 (R)16 (R)64 (R)32 (R)16/4 (S)32/16 (I)8 (S)2 (I)1 (S)ATCC7006034 (I)0.5 (S)1 (S)2 (I)16 (I)8/4 (S)16/8 (S)8 (R)1 (S)1 (S)ATCC700603/pKPA1-18 (R)0.5 (S)16 (R)64 (R)32 (R)16/4 (S)16/8 (S)8 (S)1 (S)1 (S)ATCC700603/pKPA1-28 (R)> 8 (R)16 (R)64 (R)32 (R)16/4 (S)32/16 (I)8 (S)1 (S)1 (S)TGC Tigecycline, LVX Levofloxacin, FEP Cefepime, CRO Ceftriaxone, CXM Cefuroxime, TZP Piperacillin-tazobactam, CSL Cefoperazone/sulbactam, AMK Amikacin, MEM Meropenem, CST Colistin, S Susceptible, I Intermediate, R Resistant Phenotypic and genotypic characteristics antimicrobial susceptibility of K. pneumoniae strain KPA1 and its transconjugants TGC Tigecycline, LVX Levofloxacin, FEP Cefepime, CRO Ceftriaxone, CXM Cefuroxime, TZP Piperacillin-tazobactam, CSL Cefoperazone/sulbactam, AMK Amikacin, MEM Meropenem, CST Colistin, S Susceptible, I Intermediate, R Resistant Fig. 1Phenotypes and pathogenicity of K. pneumoniae KPA1. A Biofilm formation of K. pneumoniae KPA1 and K. pneumoniae ATCC700603 was assessed by crystal violet staining (top) and quantified by optical density measurement (bottom). Error bars represent the means ± SD from three independent experiments. ns, no significance. B Virulence of K. pneumoniae KPA1 and K. pneumoniae ATCC700603 was evaluated using the Galleria mellonella infection model. Each G. mellonella larvae were injected with 20 µL of K. pneumoniae suspension (1 × 10⁴ CFU/mL) or phosphate-buffered saline (PBS, negative control). Data are expressed as mean ± SD from different experiments. P-values were calculated using the Mantel–Cox test; ***P < 0.001. n = 20 biological replicates Phenotypes and pathogenicity of K. pneumoniae KPA1. A Biofilm formation of K. pneumoniae KPA1 and K. pneumoniae ATCC700603 was assessed by crystal violet staining (top) and quantified by optical density measurement (bottom). Error bars represent the means ± SD from three independent experiments. ns, no significance. B Virulence of K. pneumoniae KPA1 and K. pneumoniae ATCC700603 was evaluated using the Galleria mellonella infection model. Each G. mellonella larvae were injected with 20 µL of K. pneumoniae suspension (1 × 10⁴ CFU/mL) or phosphate-buffered saline (PBS, negative control). Data are expressed as mean ± SD from different experiments. P-values were calculated using the Mantel–Cox test; ***P < 0.001. n = 20 biological replicates To investigate the resistance mechanism of KPA1, we performed whole-genome sequencing using a hybrid approach combining Illumina (short-read) and Nanopore (long-read) sequencing technology. The strain KPA1 was identified as ST15 with a KL24 serotype based on MLST and the capsule synthesis loci analysis. The genome comprises a single chromosome of 5,413,199 base pairs (bp) and two plasmids of 222,035 bp (designated pKPA1-1) and 81,653 bp (designated pKPA1-2), respectively (Table 2). BLAST analysis against the resistance gene database revealed that the chromosome carries multiple resistance determinants, including blaCTX−M−15, blaSHV−106, fosA6, and oqxAB. Notably, the β-lactams resistance gene blaCTX−M−15 was flanked upstream by an IS1380 family insertion sequence and downstream by a Tn3-like transposon. BLASTn analysis showed that this genetic context is commonly found on plasmids from K. pneumoniae and E. coli, suggesting that blaCTX−M−15 may have originated from such plasmids. Table 2Genomic characteristics of K. pneumoniae strain KPA1Genetic materialLength (bp)CDSsG C contentMLSTPlasmid typeAntimicrobial resistance geneAccession numberChromosome5,413,1995,32857.3%ST15, KL24NAblaCTX−M−15,blaSHV−106, fosA6, oqxABCP150854.1pKPA1-1222,03521650.2%NAIncFII/IncFIBaph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP150855.1pKPA1-281,65318553.4%NAIncFIIaadA1, aadA2b, blaLAP−2, qnrS1, cmlA1, tet(A), sul3CP150856.1NA Not available Table 2Genomic characteristics of K. pneumoniae strain KPA1Genetic materialLength (bp)CDSsG C contentMLSTPlasmid typeAntimicrobial resistance geneAccession numberChromosome5,413,1995,32857.3%ST15, KL24NAblaCTX−M−15,blaSHV−106, fosA6, oqxABCP150854.1pKPA1-1222,03521650.2%NAIncFII/IncFIBaph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP150855.1pKPA1-281,65318553.4%NAIncFIIaadA1, aadA2b, blaLAP−2, qnrS1, cmlA1, tet(A), sul3CP150856.1NA Not available Genomic characteristics of K. pneumoniae strain KPA1 NA Not available Both pKPA1-1 and pKPA1-2 contain a MDR region Plasmid pKPA1-1 (GC content: 50.2%) comprised 216 coding sequences (CDS) and is classified as an IncFII/IncFIB hybrid plasmid. This plasmid harbored multiple resistance genes, including the aminoglycoside resistance genes aph(3’)-Ia and aadA2, the β-lactams resistance genes blaCTX−M−15, blaOXA−1, and blaTEM−1B, the quinolones resistance gene aac(6’)-Ib-cr, the chloramphenicol resistance gene catB3, the tetracycline resistance genes and tet(A), the trimethoprim resistance gene dfrA12, the macrolide resistance genes mph(A), and the sulfonamide resistance genes sul1 (Table 2). These resistance determinants are clustered within a ∼25 kb MDR region (a specific DNA fragments that harboring multidrug resistant genes and mediate multidrug resistance), which is flanked by various IS elements, including IS26-aph(3’)-Ia-IS26, IS26-catB3-blaOXA−1-aac(6’)-Ib-cr-IS26, blaCTX−M−15-ISEcp1-IS607-blaTEM-1B, int-dfrA12-aadA2-sul1, and IS6100-mphR-mph(A)-IS26 (Fig. 2A). BLAST analysis indicated that pKPA1-1 is nearly identical to plasmid pDA33144-220 (NZ_CP029591.1, 100% identity and 99% coverage), pUUH239.2 (NC_016966.1, 99.98% identity and 99% coverage), pDA33141-217 (NZ_CP029588.1, 99.98% identity and 98% coverage), and p1-5208.51 (NZ_CP083033.1, 99.98% identity and 84% coverage). Notably, the MDR cassette IS26-aph(3’)-Ia-IS26 is unique to pKPA1-1, whereas the IS6100-mphR-mph(A)-IS26 is also found in pDA33144-220 and pUUH239.2. These findings indicate that occurred different recombination events during plasmids’ evolution. Fig. 2Circular alignment of the plasmids harbored by K. pneumonia KPA1 with similar plasmids in the NCBI database. A Circular alignment of plasmid pKPA1-1 with plasmids p1-5208.51 (NZ_CP083033.1), pDA33141-217 (NZ_CP029588.1), pUUH239.2 (NC_016966.1), and pDA33144-220 (NZ_CP029591.1) using the BLAST Ring Image Generator (BRIG). B Circular alignment of plasmid pKPA1-2 with plasmids pKP20194c-p3 (CP054759.1), pCRKP_44 − 2 (CP107299.1), and pOT206-1 (CP101018.1) using BRIG. The outermost ring indicates the locations of antimicrobial resistance genes (red), mobile genetic elements (purple), plasmid replication genes (blue), conjugal transfer genes (green) genes, and other feature genes (gray) in plasmids pKPA1-1 and pKPA1-2 Circular alignment of the plasmids harbored by K. pneumonia KPA1 with similar plasmids in the NCBI database. A Circular alignment of plasmid pKPA1-1 with plasmids p1-5208.51 (NZ_CP083033.1), pDA33141-217 (NZ_CP029588.1), pUUH239.2 (NC_016966.1), and pDA33144-220 (NZ_CP029591.1) using the BLAST Ring Image Generator (BRIG). B Circular alignment of plasmid pKPA1-2 with plasmids pKP20194c-p3 (CP054759.1), pCRKP_44 − 2 (CP107299.1), and pOT206-1 (CP101018.1) using BRIG. The outermost ring indicates the locations of antimicrobial resistance genes (red), mobile genetic elements (purple), plasmid replication genes (blue), conjugal transfer genes (green) genes, and other feature genes (gray) in plasmids pKPA1-1 and pKPA1-2 The smaller plasmid pKPA1-2 (GC content: 53.40%) contained 85 CDS and is classified as an IncFII type. This plasmid carried several resistance genes, including aadA1, aadA2b, blaLAP−2, qnrS1, cmlA1, tet(A) and sul3, which confer resistance to aminoglycoside, β-lactam, quinolone, chloramphenicol, tetracycline, and sulfonamide, respectively (Table 2). These genes are clustered within a∼24 kb MDR region, including Int-aadA2b-cmlA1-aadA1-sul3-IS26 and ISKpn19-qnrS1-blaLAP−2-TnAs1 (Fig. 2B). Sequence comparison revealed that pKPA1-2 shares 99.99% identity and 87% coverage with plasmid pOT206-1 and pCRKP_44 − 2 (Accession: CP101018.1 and CP107299.1), as well as plasmid pKP20194b2-p3 (CP054765.1). Importantly, pKPA1-2 uniquely contains a class 1 integron insertion that is absent in those reference plasmids. The result suggests that the structure Int-aadA2b-cmlA1-aadA1-sul3-IS26 may have been recently acquired. Phylogenetic analysis of pKPA1-1 and pKPA1-2 To analyze the evolutionary relationship between pKPA1-1 and related plasmids, we selected 16 plasmids exhibiting over 99.98% identity and 84% coverage from both our study and the GenBank database to construct a maximum likelihood phylogenetic tree. A close phylogenetic relationship was observed between pKPA1-1 and plasmids pMB6487_1 (CP103619.1), pNMBU-W07E18_01 (CP042883.1), and p51015_CTX_M_15 (CP050379.1) (Fig. 3A). The majority of these plasmids were isolated from clinical or environmental strains of K. pneumoniae, specifically ST15 and ST16. These plasmids were collected between 2006 and 2023 and have been widely disseminated across several countries, including the USA, Sweden, Germany, India, and China (Fig. 3A). All plasmids were classified into three types: IncFII/IncFIB, IncFIA/IncFIB, and IncFIB, with the majority categorized as the IncFII/IncFIB hybrid type (Table 3). Furthermore, we observed that plasmids of the same type can harbor different antimicrobial resistance genes. Fig. 3Phylogenetic analysis of plasmid pKPA1-1 and pKPA1-2. A Phylogenetic tree of plasmid pKPA1-1 and 16 related plasmids, constructed with pKPA1-1 as the reference. Plasmid pKPA1-1 is highlighted with a red background. Details of the plasmids, including collection year, host, MLST, regions/countries, and accession numbers, are provided. B Phylogenetic tree of plasmid pKPA1-2 and 16 related plasmids, constructed with pKPA1-2 as the reference. Plasmid pKPA1-2 is highlighted with a red background. Details of the plasmids, including collection year, host, MLST, bacterial species, and accession numbers, are provided Phylogenetic analysis of plasmid pKPA1-1 and pKPA1-2. A Phylogenetic tree of plasmid pKPA1-1 and 16 related plasmids, constructed with pKPA1-1 as the reference. Plasmid pKPA1-1 is highlighted with a red background. Details of the plasmids, including collection year, host, MLST, regions/countries, and accession numbers, are provided. B Phylogenetic tree of plasmid pKPA1-2 and 16 related plasmids, constructed with pKPA1-2 as the reference. Plasmid pKPA1-2 is highlighted with a red background. Details of the plasmids, including collection year, host, MLST, bacterial species, and accession numbers, are provided Table 3pKPA1-1 and related plasmids details about plasmid type, species, identity, and antimicrobial resistance genesPlasmid accessionPlasmid typeSpeciesQuery coverageIdentityAntimicrobial resistance geneCP150855.1(pKPA1-1)IncFII/IncFIB Klebsiella pneumoniae 100%100%aph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1LR025089.1IncFII/IncFIB Klebsiella pneumoniae 100%99.98%aph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1LR025093.1IncFII/IncFIB Klebsiella pneumoniae 100%99.98%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP140500.1IncFII/IncFIB Klebsiella pneumoniae 100%99.98%aph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP055211.1IncFII/IncFIB Klebsiella pneumoniae 100%99.98%aph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP036188.1IncFIB Klebsiella pneumoniae 100%99.97%aph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP029591.1IncFII/IncFIB Klebsiella pneumoniae 99%100%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1NC_016966IncFII/IncFIB Klebsiella pneumoniae 99%99.98%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP029588.1IncFII/IncFIB Klebsiella pneumoniae 98%99.98%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP103619.1IncFII/IncFIB Klebsiella pneumoniae 97%100%aph(3’)-Ia, aadA2, blaCTX−M−15, blaTEM−1B, aac(6’)-Ib-cr, dfrA12, mph(A), sul1CP067589.1IncFIA/IncFIB Klebsiella pneumoniae 96%99.99%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, dfrA12, mph(A), sul1, aph(6)-Id, aph(3’’)-Ib, dfrA14, sul2CP067900.1IncFIA/IncFIB Klebsiella pneumoniae 96%99.98%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, dfrA12, mph(A), sul1, aph(6)-Id, aph(3’’)-Ib, dfrA14, sul2CP036337.1IncFIB Klebsiella pneumoniae 96%99.96%aph(3’)-Ia, blaCTX−M−15, blaOXA−1, aac(6’)-Ib-cr, catB3, tet(A), mph(A)CP050379.1IncFII/IncFIB Klebsiella pneumoniae 94%99.98%aph(3’)-Ia, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, mph(A), sul2, aac(3)-IId, aph(6)-Id, aph(3’’)-IbCP144132.1IncFII/IncFIB Klebsiella pneumoniae 90%99.9%blaCTX−M−15, blaOXA−1, aac(6’)-Ib-cr, catB3, mph(A)CP042883.1IncFII/IncFIB Klebsiella pneumoniae 90%99.98%blaCTX−M−15, blaTEM−1BCP083033.1IncFII/IncFIB Klebsiella pneumoniae 84%99.98%NAAntimicrobial resistance genes: aminoglycoside: aph(3’)-Ia, aadA2, aac(6’)-Ib-cr, aph(6)-Id, aph(3’’)-Ib, aac(3)-IId; beta-lactam: blaCTX−M−15, blaOXA−1, blaTEM−1B; folate pathway antagonist: dfrA12, dfrA14, sul1, sul2; macrolide: mph(A) ; tetracycline: tet(A); amphenicol: catB3NA Not available Table 3pKPA1-1 and related plasmids details about plasmid type, species, identity, and antimicrobial resistance genesPlasmid accessionPlasmid typeSpeciesQuery coverageIdentityAntimicrobial resistance geneCP150855.1(pKPA1-1)IncFII/IncFIB Klebsiella pneumoniae 100%100%aph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1LR025089.1IncFII/IncFIB Klebsiella pneumoniae 100%99.98%aph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1LR025093.1IncFII/IncFIB Klebsiella pneumoniae 100%99.98%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP140500.1IncFII/IncFIB Klebsiella pneumoniae 100%99.98%aph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP055211.1IncFII/IncFIB Klebsiella pneumoniae 100%99.98%aph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP036188.1IncFIB Klebsiella pneumoniae 100%99.97%aph(3’)-Ia, aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP029591.1IncFII/IncFIB Klebsiella pneumoniae 99%100%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1NC_016966IncFII/IncFIB Klebsiella pneumoniae 99%99.98%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP029588.1IncFII/IncFIB Klebsiella pneumoniae 98%99.98%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, tet(A), dfrA12, mph(A), sul1CP103619.1IncFII/IncFIB Klebsiella pneumoniae 97%100%aph(3’)-Ia, aadA2, blaCTX−M−15, blaTEM−1B, aac(6’)-Ib-cr, dfrA12, mph(A), sul1CP067589.1IncFIA/IncFIB Klebsiella pneumoniae 96%99.99%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, dfrA12, mph(A), sul1, aph(6)-Id, aph(3’’)-Ib, dfrA14, sul2CP067900.1IncFIA/IncFIB Klebsiella pneumoniae 96%99.98%aadA2, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, dfrA12, mph(A), sul1, aph(6)-Id, aph(3’’)-Ib, dfrA14, sul2CP036337.1IncFIB Klebsiella pneumoniae 96%99.96%aph(3’)-Ia, blaCTX−M−15, blaOXA−1, aac(6’)-Ib-cr, catB3, tet(A), mph(A)CP050379.1IncFII/IncFIB Klebsiella pneumoniae 94%99.98%aph(3’)-Ia, blaCTX−M−15, blaOXA−1, blaTEM−1B, aac(6’)-Ib-cr, catB3, mph(A), sul2, aac(3)-IId, aph(6)-Id, aph(3’’)-IbCP144132.1IncFII/IncFIB Klebsiella pneumoniae 90%99.9%blaCTX−M−15, blaOXA−1, aac(6’)-Ib-cr, catB3, mph(A)CP042883.1IncFII/IncFIB Klebsiella pneumoniae 90%99.98%blaCTX−M−15, blaTEM−1BCP083033.1IncFII/IncFIB Klebsiella pneumoniae 84%99.98%NAAntimicrobial resistance genes: aminoglycoside: aph(3’)-Ia, aadA2, aac(6’)-Ib-cr, aph(6)-Id, aph(3’’)-Ib, aac(3)-IId; beta-lactam: blaCTX−M−15, blaOXA−1, blaTEM−1B; folate pathway antagonist: dfrA12, dfrA14, sul1, sul2; macrolide: mph(A) ; tetracycline: tet(A); amphenicol: catB3NA Not available pKPA1-1 and related plasmids details about plasmid type, species, identity, and antimicrobial resistance genes CP150855.1 (pKPA1-1) Antimicrobial resistance genes: aminoglycoside: aph(3’)-Ia, aadA2, aac(6’)-Ib-cr, aph(6)-Id, aph(3’’)-Ib, aac(3)-IId; beta-lactam: blaCTX−M−15, blaOXA−1, blaTEM−1B; folate pathway antagonist: dfrA12, dfrA14, sul1, sul2; macrolide: mph(A) ; tetracycline: tet(A); amphenicol: catB3 NA Not available The evolutionary relationship of pKPA1-2 and 16 related plasmids was determined based on over 99.99% identity and 85% coverage. A close phylogenetic relationship was observed between pKPA1-2 and the plasmid unnamed2 (CP055221.1) (Fig. 3B). All plasmids were classified as IncFII type and were collected from China between 2016 and 2024. Plasmids unnamed1 (CP101018.1) and unnamed2 (CP055221.1) were isolated from Klebsiella aerogenes OT206 and Shigella flexneri STEFF_12, respectively, while the remaining plasmids were derived from K. pneumoniae, the majority of which belonged to K. pneumoniae ST11 (Fig. 3B). The resistance genes present in these plasmids were categorized into seven distinct types, with most plasmids containing the structural module qnrS1-blaLAP−2-sul2-dfrA14-tet(A)-catA2 (Table 4). Notably, pKPA1-2 possesses a unique structural module that sets it apart from the other plasmids. Table 4pKPA1-2 and related plasmids details about plasmid type, species, identity, and antimicrobial resistance genesPlasmid accessionPlasmid typeSpeciesQuery coverageIdentityAntimicrobial resistance geneCP150856.1(pKPA1-2)IncFII Klebsiella pneumoniae 100%100%aadA1, aadA2b, blaLAP−2, qnrS1, cmlA1, tet(A), sul3CP101018.1IncFII Klebsiella aerogenes 87%99.99%aac(6’)-Ib-cr, aadA16, blaLAP−2, sul1, dfrA27, mph(A), tet(A), floR, arr-3, qnrS1, qnrB52CP107293.1IncFII Klebsiella pneumoniae 87%99.99%rmtB, qnrS1, blaKPC−2, blaCTX−M−65, blaTEM−1B, sul2, dfrA14, tet(A), blaSHV−12, blaLAP−2, fosA3CP180333.1IncFII Klebsiella pneumoniae 87%99.99%rmtB, qnrS1, blaKPC−2, blaCTX−M−65, blaTEM−1B, sul2, dfrA14, tet(A), blaSHV−12, blaLAP−2, fosA3CP107299.1IncFII Klebsiella pneumoniae 87%99.99%rmtB, qnrS1, blaKPC−2, blaCTX−M−65, blaTEM−1B, sul2, dfrA14, tet(A), blaSHV−12, blaLAP−2, fosA3CP055221.1IncFII Shigella flexneri 86%99.99%aac(6’)-Ib-cr, aadA1, blaTEM−1 A, tet(A), floR, blaLAP−2, blaOXA−9, qnrS1CP118694.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP154395.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP054759.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP029382.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP033957.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP170703.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A)CP100074.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP146604.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP100109.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP141603.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A)CP090435.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2Antimicrobial resistance genes unique to pKPA1-2 are highlighted in bold. Antimicrobial resistance genes: aminoglycoside: aadA1, aadA2b, aac(6’)-Ib-cr, aadA16, rmtB; quinolone: qnrS1, qnrB52; beta-lactam: blaOXA−9, blaTEM−1 A, blaTEM−1B, blaLAP−2, blaKPC−2, blaCTX−M−65, blaSHV−12; folate pathway antagonist: dfrA14, dfrA27, sul1, sul2, sul3; fosfomycin: fosA3; tetracycline: tet(A); amphenicol: cmlA1, catA2; florfenicol: floR Table 4pKPA1-2 and related plasmids details about plasmid type, species, identity, and antimicrobial resistance genesPlasmid accessionPlasmid typeSpeciesQuery coverageIdentityAntimicrobial resistance geneCP150856.1(pKPA1-2)IncFII Klebsiella pneumoniae 100%100%aadA1, aadA2b, blaLAP−2, qnrS1, cmlA1, tet(A), sul3CP101018.1IncFII Klebsiella aerogenes 87%99.99%aac(6’)-Ib-cr, aadA16, blaLAP−2, sul1, dfrA27, mph(A), tet(A), floR, arr-3, qnrS1, qnrB52CP107293.1IncFII Klebsiella pneumoniae 87%99.99%rmtB, qnrS1, blaKPC−2, blaCTX−M−65, blaTEM−1B, sul2, dfrA14, tet(A), blaSHV−12, blaLAP−2, fosA3CP180333.1IncFII Klebsiella pneumoniae 87%99.99%rmtB, qnrS1, blaKPC−2, blaCTX−M−65, blaTEM−1B, sul2, dfrA14, tet(A), blaSHV−12, blaLAP−2, fosA3CP107299.1IncFII Klebsiella pneumoniae 87%99.99%rmtB, qnrS1, blaKPC−2, blaCTX−M−65, blaTEM−1B, sul2, dfrA14, tet(A), blaSHV−12, blaLAP−2, fosA3CP055221.1IncFII Shigella flexneri 86%99.99%aac(6’)-Ib-cr, aadA1, blaTEM−1 A, tet(A), floR, blaLAP−2, blaOXA−9, qnrS1CP118694.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP154395.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP054759.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP029382.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP033957.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP170703.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A)CP100074.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP146604.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP100109.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2CP141603.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A)CP090435.1IncFII Klebsiella pneumoniae 85%99.99%qnrS1, blaLAP−2, sul2, dfrA14, tet(A), catA2Antimicrobial resistance genes unique to pKPA1-2 are highlighted in bold. Antimicrobial resistance genes: aminoglycoside: aadA1, aadA2b, aac(6’)-Ib-cr, aadA16, rmtB; quinolone: qnrS1, qnrB52; beta-lactam: blaOXA−9, blaTEM−1 A, blaTEM−1B, blaLAP−2, blaKPC−2, blaCTX−M−65, blaSHV−12; folate pathway antagonist: dfrA14, dfrA27, sul1, sul2, sul3; fosfomycin: fosA3; tetracycline: tet(A); amphenicol: cmlA1, catA2; florfenicol: floR pKPA1-2 and related plasmids details about plasmid type, species, identity, and antimicrobial resistance genes CP150856.1 (pKPA1-2) Antimicrobial resistance genes unique to pKPA1-2 are highlighted in bold. Antimicrobial resistance genes: aminoglycoside: aadA1, aadA2b, aac(6’)-Ib-cr, aadA16, rmtB; quinolone: qnrS1, qnrB52; beta-lactam: blaOXA−9, blaTEM−1 A, blaTEM−1B, blaLAP−2, blaKPC−2, blaCTX−M−65, blaSHV−12; folate pathway antagonist: dfrA14, dfrA27, sul1, sul2, sul3; fosfomycin: fosA3; tetracycline: tet(A); amphenicol: cmlA1, catA2; florfenicol: floR Both pKPA1-1 and pKPA1-2 are capable of horizontal transfer As both pKPA1-1 and pKPA1-2 were shown to harbor conjugative transfer genes (tra operons) (Fig. 2A and B), we assessed their conjugative transfer potential through a series of conjugation experiments. K. pneumoniae KPA1 served as the donor strain, while gentamicin-resistant K. pneumoniae ATCC700603 and E. coli DH5α were employed as recipient strains. The trimethoprim resistance conferred by pKPA1-1 and the levofloxacin resistance conferred by pKPA1-2 were utilized as selective antibiotic markers for identifying successful transfers of these plasmids. Our results demonstrated that trimethoprim resistance mediated by pKPA1-1 was successfully transferred to both K. pneumoniae ATCC700603 and E. coli DH5α. In contrast, the levofloxacin resistance phenotype mediated by pKPA1-2 was transferred exclusively to K. pneumoniae ATCC700603 but not E. coli DH5α (Fig. 4). These findings indicate that both plasmids, pKPA1-1 and pKPA1-2, are self-transmissible. In our experimental strains, pKPA1-1 exhibited broad-host-range transfer capability, whereas pKPA1-2 showed a narrower host range, consistent with the transfer characteristics typically associated with IncFII-family plasmids. We also measured the conjugative transfer efficiency of pKPA1-1 and pKPA1-2 from KPA1 to ATCC700603. The transfer frequencies of pKPA1-1 and pKPA1-2 were 4.23 × 10⁻¹ and 2.04 × 10⁻¹ per donor cell, respectively. Furthermore, the trimethoprim resistance gene dfrA12 was used as a marker in PCR amplification to confirm its presence in the donor strain KPA1, the recipient strain ATCC700603, and the transconjugant ATCC700603/pKPA1-1. The results indicated that dfrA12 was present in both KPA1 and ATCC700603/pKPA1-1 (Figure S1). Similarly, the levofloxacin resistance gene qnrS1 was confirmed in KPA1 and ATCC700603/pKPA1-2 (Figure S1). Fig. 4Horizontal transfer potential of pKPA1-1 and pKPA1-2. Schematic diagram illustrating the conjugative transfer of plasmids pKPA1-1 and pKPA1-2. The donor strain is K. pneumoniae KPA1, and the recipient strains are K. pneumoniae ATCC700603 and E. coli DH5α. The transconjugants include K. pneumoniae ATCC700603/pKPA1-1, K. pneumoniae ATCC700603/pKPA1-2, and E. coli DH5α/pKPA1-1 Horizontal transfer potential of pKPA1-1 and pKPA1-2. Schematic diagram illustrating the conjugative transfer of plasmids pKPA1-1 and pKPA1-2. The donor strain is K. pneumoniae KPA1, and the recipient strains are K. pneumoniae ATCC700603 and E. coli DH5α. The transconjugants include K. pneumoniae ATCC700603/pKPA1-1, K. pneumoniae ATCC700603/pKPA1-2, and E. coli DH5α/pKPA1-1 Furthermore, we determined the MICs of tigecycline, levofloxacin, cefepime, ceftriaxone, cefuroxime, piperacillin-tazobactam, cefoperazone/sulbactam, amikacin, meropenem, and colistin for K. pneumoniae ATCC700603 and the transconjugants K. pneumoniae ATCC700603/pKPA1-1 and K. pneumoniae ATCC700603/pKPA1-2 using broth dilution tests. Compared to the recipient strain K. pneumoniae ATCC700603, the transconjugants K. pneumoniae ATCC700603/pKPA1-1 displayed different MIC patterns for tigecycline, cefepime, ceftriaxone, cefuroxime, and piperacillin-tazobactam (Table 1). Meanwhile, the transconjugant K. pneumoniae ATCC700603/pKPA1-2 demonstrated resistance to the above antibiotics, along with additional resistance to levofloxacin and cefoperazone/sulbactam (Table 1). These results suggest that both pKPA1-1 and pKPA1-2 are capable of transferring specific antibiotic resistance genes from the donor strain K. pneumoniae KPA1 to the recipient strain K. pneumoniae ATCC700603. Stability and virulence of pKPA1-1 and pKPA1-2 To evaluate the stability of pKPA1-1 and pKPA1-2 in their host bacteria, K. pneumoniae ATCC700603/pKPA1-1 and K. pneumoniae ATCC700603/pKPA1-2 were serially passaged in antibiotic-free medium. The retention of pKPA1-1 and pKPA1-2 was assessed based on antibiotic resistance phenotypes, specifically trimethoprim for pKPA1-1 and levofloxacin for pKPA1-2. After 120 generations, a significant reduction in the stability of pKPA1-1 was observed, with approximately 53.5% loss of the plasmid by the end of 200 generations (Fig. 5A). In contrast, plasmid pKPA1-2 exhibited stable maintenance throughout the 200 generations. Collectively, these findings demonstrate that pKPA1-2 is significantly more stable in its host than pKPA1-1, suggesting inherent differences in the plasmid maintenance systems. Fig. 5Stability and virulence of pKPA1-1 and pKPA1-2. A The stability of plasmids pKPA1-1 and pKPA1-2 was assessed in the K. pneumoniae ATCC700603. Error bars represent the mean ± SD from three independent experiments. B Growth of the K. pneumoniae ATCC700603, K. pneumoniae ATCC700603/pKPA1-1, and K. pneumoniae ATCC700603/pKPA1-2 were determined by a Synergy 2 Microplate Reader (BioTek). Three independent cultures of each strain were tested, and error bars indicate the standard error of the mean (n = 3). C Virulence potential of K. pneumoniae ATCC700603, K. pneumoniae ATCC700603/pKPA1-1, and K. pneumoniae ATCC700603/pKPA1-2 was evaluated using the G. mellonella infection model. Each G. mellonella larvae was injected with 20 µL of K. pneumoniae dilution (1 × 10⁴ CFU/mL) or phosphate-buffered saline (PBS, negative control). Quantified data from different experiments are expressed as mean ± SD. The P-values were determined using the Mantel–Cox test, **P < 0.01 and ***P < 0.001. n = 20 biological replicates Stability and virulence of pKPA1-1 and pKPA1-2. A The stability of plasmids pKPA1-1 and pKPA1-2 was assessed in the K. pneumoniae ATCC700603. Error bars represent the mean ± SD from three independent experiments. B Growth of the K. pneumoniae ATCC700603, K. pneumoniae ATCC700603/pKPA1-1, and K. pneumoniae ATCC700603/pKPA1-2 were determined by a Synergy 2 Microplate Reader (BioTek). Three independent cultures of each strain were tested, and error bars indicate the standard error of the mean (n = 3). C Virulence potential of K. pneumoniae ATCC700603, K. pneumoniae ATCC700603/pKPA1-1, and K. pneumoniae ATCC700603/pKPA1-2 was evaluated using the G. mellonella infection model. Each G. mellonella larvae was injected with 20 µL of K. pneumoniae dilution (1 × 10⁴ CFU/mL) or phosphate-buffered saline (PBS, negative control). Quantified data from different experiments are expressed as mean ± SD. The P-values were determined using the Mantel–Cox test, **P < 0.01 and ***P < 0.001. n = 20 biological replicates To further investigate the contribution of these plasmids to the virulence of the host strain, we employed the G. mellonella infection model. We observed that the transfer of plasmids pKPA1-1 and pKPA1-2 into K. pneumoniae ATCC700603 did not significantly affect cell growth (Fig. 5B). As Fig. 5C shown, larvae infected with K. pneumoniae ATCC700603 harboring either pKPA1-1 or pKPA1-2 exhibited significantly reduced mortality compared to those infected with the wild-type strain. Specifically, the wild-type strain resulted in a mortality rate of approximately 74.5% after 30 h, whereas strains carrying plasmid pKPA1-1 or pKPA1-2 demonstrated mortality rates of 45.5% and 27.8%, respectively. These results collectively suggest that both pKPA1-1 and pKPA1-2 play critical roles in attenuating the virulence of K. pneumoniae, with pKPA1-2 exerting a more pronounced impact. Discussion K. pneumoniae, a symbiotic opportunistic human pathogen commonly isolated from clinical settings and causes both community-acquired and nosocomial infections. In this study, we investigated a multidrug-resistant clinical ST15 isolate, KPA1, along with its two resistance plasmids, pKPA1-1 and pKPA1-2, to dissect their antimicrobial resistance mechanisms and evolutionary trajectories. Through conjugation assays and antimicrobial susceptibility testing, we demonstrated that pKPA1-1 was efficiently transferred to K. pneumoniae ATCC700603, significantly enhancing resistance to tigecycline, cefepime, ceftriaxone, cefuroxime, and piperacillin–tazobactam. In parallel, pKPA1-2 conferred resistance to the same antibiotics, with additional resistance observed against levofloxacin and cefoperazone–sulbactam (Table 1). Notably, the co-presence of both plasmids facilitated the establishment of a multidrug-resistant phenotype, promoting persistence and proliferation under antibiotic pressure in nosocomial environments. The horizontal transfer of pKPA1-1 and pKPA1-2 to other K. pneumoniae strains rapidly disseminated corresponding resistance phenotypes, highlighting their critical roles as vectors in the spread of antimicrobial resistance via conjugative plasmid exchange. Horizontal gene transfer (HGT) is a pivotal driver of rapid bacterial evolution and adaptation, primarily mediated by MGEs [24]. ISs, a subclass of MGEs, play critical roles in capturing and disseminating antibiotic resistance genes (ARGs). Among them, IS26 is frequently implicated in the mobilization of ARGs between plasmids [25]. Within the MDR region of plasmid pKPA1-1, we identified five copies of IS26, several single IS, and a single integron, indicating that this region has likely undergone multiple homologous recombination events. Notably, the genetic structure IS26-aph(3’)-Ia-IS26 was unique to pKPA1-1, while IS6100-mphR-mph(A)-IS26 also was found in plasmid pDA33144-220 and pUUH239.2. These findings suggest that the recombination events introduced both genetic structures into the backbone of pKPA1-1. Similarly, the MDR region of pKPA1-2 harbors two distinct genetic structures: Int-aadA2b-cmlA1-aadA1-sul3-IS26 and ISKpn19-qnrS1-blaLAP−2-TnAs1. In contrast, related reference plasmids contain only the latter structure, suggesting that the former was acquired through homologous recombination during the evolution of pKPA1-2. Collectively, our data support the conclusion that the MDR regions of these resistance plasmids were shaped by homologous recombination events mediated by IS elements. In China, K. pneumoniae ST11 is the predominant clone associated with hospital-acquired infections, exhibiting both hypervirulence and extensive antimicrobial resistance [26]. In these CRKP ST11 strains, the plasmid-borne blaKPC−2 is widely recognized as the primary driver of carbapenem resistance [27]. However, in our study, we found that blaKPC−2 was detected in only four plasmids, while blaLAP−2 was found in all plasmids examined. This suggests that blaLAP−2 may be emerging as a prevalent β-lactam resistance determinant in K. pneumoniae ST11. In addition, we identified qnrS1 and tet(A) in pKPA1-2 and related plasmids, which contribute to quinolone and tigecycline resistance, respectively. Notably, the majority of pKPA1-2-related plasmids were associated with K. pneumoniae ST11 strains and were exclusively identified in isolates from China, spanning geographically diverse regions including Shanghai, Wuhan, Beijing, Hangzhou, and Hong Kong. These findings support the hypothesis that pKPA1-2 likely originated in a K. pneumoniae ST11 lineage and subsequently underwent broad interregional dissemination. Plasmid transfer and stability are vital for host strains to adapt to and thrive in new environmental conditions. The host ranges of a plasmid is largely determined by it replication systems, which is closely associated with plasmid incompatibility types [28]. The plasmid pKPA1-1 was transferable to both K. pneumoniae and E. coli, whereas pKPA1-2 could only be transferable to K. pneumoniae, classifying them as broad- and narrow-host-range plasmids under our experimental strains, respectively. Additionally, pKPA1-1 was identified as a hybrid-type plasmid belonging to the IncFII/IncFIB group, while pKPA1-2 belonged to the IncFII type. In our experimental conditions, IncFII/IncFIB hybrid plasmids may possess broader host ranges compared to IncFII plasmids alone. More importantly, we observed that pKPA1-1 exhibited instability in host strains, with a higher loss rate resulting in frequent plasmid loss during cell division. In contrast, pKPA1-2 appeared to be stably maintained and capable of sustaining antibiotic resistance within bacterial populations, even in the absence of selective pressure. Intriguingly, both pKPA1-1 and pKPA1-2 conferred similar resistance profiles. However, it remains unclear which plasmid predominantly contributes to the overall resistance phenotype of strain KPA1. We suspect that pKPA1-2 was acquired as a compensatory mechanism to preserve resistance traits that may be lost due to the instability of pKPA1-1. It might be interesting to further confirm the underlying compensatory mechanism. Conclusion In summary, we identified and characterized the multidrug-resistant K. pneumoniae ST15 strain KPA1 and its two conjugative plasmids, pKPA1-1 and pKPA1-2. These plasmids contribute to the formation of a multidrug resistance region through homologous recombination, thereby conferring resistance to multiple antibiotics and facilitating the emergence of the MDR K. pneumoniae strain KPA1. Although both plasmids share similar resistance profiles, pKPA1-1 exhibited instability under non-selective conditions, a fitness cost that may appears to be compensated by the acquisition and stable maintenance of pKPA1-2. Together, these findings provide important insights into the adaptive mechanisms of multidrug-resistant bacteria. Supplementary Information Supplementary Material 1. Supplementary Material 1.