Genomic Characterization of Multidrug-Resistant Klebsiella pneumoniae Strains from Bovine Mastitis Reveals Extensive Resistome and Virulome Arsenals

doi:10.21203/rs.3.rs-7912731/v1

Download PDF

Research Article

Genomic Characterization of Multidrug-Resistant Klebsiella pneumoniae Strains from Bovine Mastitis Reveals Extensive Resistome and Virulome Arsenals

https://doi.org/10.21203/rs.3.rs-7912731/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Klebsiella pneumoniae, a major bovine clinical mastitis (CM) pathogen, carries multidrug resistance (MDR) and virulence factor genes (VFGs), posing serious animal and public health threats. This study screened 27 K. pneumoniae isolates (19 from CM milk and 8 from feces) through culture, biochemical tests, and 16S rRNA-gene sequencing. An overall prevalence of K. pneumoniae was 22.5% (27/120), with a higher rate in milk (27.14%) than feces (16.0%). Antibiogram profiling revealed that all isolates were multidrug-resistant, with high resistance to doxycycline, tetracycline, nalidixic acid, and ampicillin. Three highly resistant isolates (MBBL2, MNH_G2C5, MNH_G2C5F) underwent whole-genome sequencing for comprehensive genomic analysis. Sequence typing (ST), phylogenetic and pangenome analyses assigned MBBL2 and MNH_G2C5F to ST273 and MNH_G2C5 to ST101, clustering with global human- and animal-derived K. pneumoniae strains, and carrying notable strain-specific accessory genes (MNH_G2C5:123; MBBL2/MNH_G2C5F:826). Functional annotation identified abundant genes for carbohydrate metabolism (~ 10%), amino acid transport (~ 9%), and transcription (~ 9%). Resistome analysis identified 29–41 resistance genes, covering 12 antibiotic classes, metals, biocides, and acid stress. Virulence profiling identified 44–60 VFGs involved in adherence, biofilm formation, effector delivery, immune modulation, and metabolism. Genomic plasticity analysis revealed 27–34 variable regions, multiple prophages, 46–58 insertion sequences, and four plasmid replicons. Conserved exopolysaccharide/capsule clusters, secondary metabolites, and high pathogenicity scores (> 0.9) underscored both animal and human pathogenic potential. This study demonstrates that dairy cattle are a reservoir for high-risk MDR clones of K. pneumoniae carrying an extensive resistome and virulome arsenal, highlighting the urgent need for strengthened surveillance and control measures.

Animal Science

General Microbiology

Bacteriology

Epigenetics & Genomics

Bioinformatics

Bovine mastitis

K. pneumoniae

sequence type

pangenome

resistome

virulence

Bovine mastitis, characterized by the inflammation of the mammary gland, stands as one of the most prevalent and economically devastating diseases affecting the global dairy industry [1]. The condition leads to significant reductions in milk yield and quality, increased veterinary costs, and premature culling, imposing substantial financial losses to the dairy farmers [2, 3]. The etiology of bovine mastitis is complex and multifactorial, involving interactions between host, environment, and a diverse range of pathogens [1, 4]. Among these pathogens, bacterial agents are the most predominant, with contagious and environmental bacteria being the primary culprits [5]. Within the spectrum of bacterial pathogens, Gram-negative coliforms, particularly Escherichia coli and K. pneumoniae, are frequently isolated from CM cases [6, 7]. While E. coli is often reported as the most common cause, infections with K. pneumoniae are typically more severe, leading to massive udder inflammation, tissue necrosis, and a significantly prolonged recovery period for milk production [8–10]. The host immune response to K. pneumoniae is also notably intense, marked by elevations in pro-inflammatory cytokines [8]. The challenges in managing K. pneumoniae mastitis are compounded by its ubiquitous presence in the dairy environment, including soil, water, feed, bedding materials, and feces, which serve as reservoirs for udder contamination [11–15]. Furthermore, the bacterium can transmit laterally from infected to healthy cows, facilitating herd-level outbreaks [16, 17].

The extensive and often indiscriminate use of antibiotics in livestock has fueled the selection of MDR strains, rendering conventional therapies ineffective [18]. A critical factor exacerbating the threat of K. pneumoniae is its exceptional ability to develop AMR, which greatly elevates treatment costs and imposes a significant economic burden [15, 19, 20]. While the recent development of a K. pneumoniae mastitis vaccine marks a significant step toward prevention, the continued emergence and persistence of MDR strains highlight the urgent need for comprehensive genomic surveillance and a deeper understanding of their evolution and transmission dynamics [11, 14, 21, 22]. The pathogenicity of K. pneumoniae is mediated by an arsenal of virulence factors, including capsular polysaccharides, fimbriae for adhesion, and efficient iron-scavenging systems [15, 23–25]. Of particular concern is the global emergence of MDR and hypervirulent K. pneumoniae clones, posing a serious challenge to both animal and human health. Sequence types ST273 and ST101 are recognized as high-risk lineages combining MDR with enhanced transmissibility and virulence [26, 27]. For instance, ST273 strains frequently carry carbapenemases and other resistance genes on mobile plasmids, facilitating rapid dissemination [28]. Simultaneously, the bacterium’s capsular/exopolysaccharide (EPS/CPS) layer plays a central role in pathogenesis by shielding cells from phagocytosis, complement killing, and antimicrobial peptides, and by promoting persistence and tissue invasion [29]. Despite extensive research on K. pneumoniae, the genetic basis of virulence and resistance in bovine strains remains poorly understood. This gap is mainly due to their high genomic diversity, frequent gene acquisition and mutation events, and limited inclusion in livestock genomic surveillance, all of which hinder effective monitoring and control measures. Recent advances in high-throughput sequencing, particularly WGS, have provided unprecedented resolution for exploring bacterial genomes. These technologies enable comprehensive analyses of antimicrobial resistance genes (ARGs), virulence factors, sequence types, metabolic pathways, genomic diversity, pangenome dynamics, and mobile genetic elements (MGEs), essential for understanding pathogen evolution, transmission, and adaptation [30–32]. Therefore, this study aimed to comprehensively characterize the genomes of K. pneumoniae isolates associated with bovine CM in Bangladesh. By integrating conventional microbiological techniques with whole-genome sequencing, we sought to (i) determine the prevalence and AMR profiles of K. pneumoniae from milk and fecal samples, and (ii) conduct in-depth genomic analyses to investigate their relatedness, plasticity, resistome, and virulome. The findings will offer valuable insights into the evolution, dissemination, and pathogenic potential of MDR K. pneumoniae within dairy herds, contributing to the development of more effective surveillance and targeted control strategies against this critical mastitis pathogen.

Sample collection, processing and identification of K. pneumoniae

A total of 120 samples comprising 70 milk and 50 fecal samples were collected from 70 lactating cows diagnosed with CM. Samples were collected based on visible clinical symptoms, including changes in milk color (reddish or yellowish), udder swelling, redness, and increased temperature, along with a California Mastitis Test (CMT) score of 3, indicating numerous clots [33]. Sampling was conducted across 50 small-holding dairy farms (each with 5–15 cows) in the Gazipur district (24.09° N, 90.41° E), Bangladesh. The study was conducted during January 2022 to June 2024. The study strictly complied with animal welfare guidelines during sample collection. All procedures were carried out with minimal stress to the animals and under aseptic conditions to ensure hygiene and prevent discomfort. Before milk sampling, teat ends were disinfected to minimize infection risks. Samples (milk and feces) were collected non-invasively, ensuring no harm or undue stress to the cows [30]. Although formal ethical approval was not required under local regulations in Bangladesh for non-invasive sampling, internationally recognized standards for animal welfare, including the OIE Terrestrial Animal Health Code and FAO guidelines on responsible antimicrobial use in livestock, were followed. Verbal consent was obtained from farmers prior to sample collection, and no interventions beyond routine farm practices were applied. During the morning milking period (8:00–10:00 a.m.), all the samples were collected randomly and approximately 10 mL of milk was aseptically collected from each cow, while 5 g of feces was collected from each farm, using sterile Falcon tubes. Teat ends were disinfected prior to milk sampling, and strict hygiene protocols were maintained throughout the process to ensure sample integrity and minimize contamination [34]. The fecal samples were homogenized in distilled water at 1:1 ratio. For the isolation and phenotypic identification of K. pneumoniae, samples were first enriched in nutrient broth (Oxoid, UK) and incubated overnight at 37°C. The enriched broth was serially diluted up to 10⁻³ and plated onto Muller Hinton agar (Oxoid, UK) for overnight incubation at 37°C. Subsequently, a single pure colony was streaked onto MacConkey Agar (HiMedia, India) at 37°C for 24 h. Based on colony morphology (pink color, circular) the phenotypic characterization was conducted [18, 35]. Gram negative bacilli were identified through Gram straining and series of biochemical tests (Table S1) such as indole, catalase, oxidase, citrate, methyl red, lactose fermentation and Voges-Proskauer (VP) test [31, 36]. Gram positive Staphylococcus warneri (Accession: JAQPCE000000000 used as a positive control for biochemical test. Species level identification was performed by using 16S rRNA gene sequencing using 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and U1492R (5′-CTA-CGGCTACCTTGTTACGA-3′) primers [18, 37].

Antimicrobial susceptibility profile of the K. pneumoniae isolates

Antimicrobial susceptibility testing (AST) of K. pneumoniae isolates (n = 27) was performed using the Kirby–Bauer disc diffusion method [1], following Clinical and Laboratory Standards Institute (CLSI) guidelines, M100, 34th edition [38]. In brief, pure cultures of K. pneumoniae isolates were swabbed onto Mueller–Hinton agar plates (Oxoid, UK), and antibiotic discs were aseptically placed on the inoculated agar. Plates were incubated at 37°C overnight, and the diameters of inhibition zones were measured [39]. Each assay was conducted in triplicate. The isolates were tested against 15 antibiotics commonly used to treat bacterial infections in humans and animals, including bovine mastitis in Bangladesh. The tested antibiotics included: Tetracyclines (doxycycline, 30 µg; tetracycline, 30 µg), Aminoglycosides (gentamicin, 10 µg; streptomycin, 10 µg), Beta-lactams (ampicillin, 10 µg; oxacillin, 1 µg), Monobactams (aztreonam, 30 µg), Fluoroquinolones (ciprofloxacin, 10 µg; nalidixic acid, 30 µg), Nitrofurans (nitrofurantoin, 300 µg), Carbapenems (imipenem, 10 µg), Chloramphenicol (30 µg), Macrolides (azithromycin, 15 µg), Sulphonamides (sulphonamide compound, 300 µg), and Cephalosporins (cefoxitin, 30 µg). Isolates were classified as resistant (R), intermediate (I), or susceptible (S) according to CLSI guidelines (File S1). Isolates exhibiting resistance to three or more antibiotic classes were defined as multidrug-resistant (MDR).

Whole genome sequencing, assembly and annotation of the K. pneumoniae isolates

Based on the AST results, three MDR isolates of K. pneumoniae comprising two from milk (MNH_G2C5 and MBBL2) and one from feces (MNH_G2C5F) were selected for subsequent genomic investigation. The high-quality genomic DNA was obtained from nutrient broth cultures using the QIAamp DNA Mini Kit (QIAGEN, Germany) in accordance with the manufacturer’s instructions. DNA libraries were constructed from 1 ng of input DNA using the Nextera XT Kit (Illumina, USA). Whole-genome sequencing (WGS) was then performed on the Illumina MiSeq platform with a 2 × 250 bp paired-end run [40, 41]. The quality of raw sequencing reads was assessed using FastQC v0.11.7 [42], and Illumina adapters, known sequencing artifacts, and phiX reads were removed using Trimmomatic v0.39 [43]. Genome assembly was carried out with SPAdes v3.15.5 [44], and annotation was performed using the NCBI Prokaryotic Genome Annotation Pipeline (PGAP) v6.6 [45] and RAST v2.0 server [46]. Genome completeness was estimated using CheckM v1.2.3 [47], while PathogenFinder v1.1 [48] was employed to estimate the probability of the isolates being human pathogens. Functional genome visualization and annotation were conducted using Genovi v0.2.16 [49]. All software tools were executed with default parameters unless stated otherwise.

Sequence typing and phylogenetic analysis of the K. pneumoniae strains

Multi-locus sequence typing (MLST) and core genome MLST (cgMLST) analyses were performed in-silico using BacWGSTdb 2.0 [50] to determine sequence types and trace the bacterial origin. A grapeTree (http://achtman-lab.github.io/GrapeTree/MSTree_holder.html) was used to produce and visualize a minimum spanning tree (MST) based on findings of the cgMLST. Based on cgMLST results, three study genomes (MBBL2, MNH_G2C5, and MNH_G2C5F), along with 49 reference K. pneumoniae genomes from NCBI (File S2), were isolated from 41 humans, 4 domestic dogs, 3 bovines, 3 environmental sources, and 1 pangolin, used for phylogenetic analysis. Single nucleotide polymorphism (SNP) based phylogenetic analysis was performed using Parsnp [51], where K. pneumoniae strain 11227-1 (Accession: JDWN00000000.1) served as the reference genome. The phylogenetic tree was visualized, and annotated using the Interactive Tree of Life web-tool [52]. Furthermore, average nucleotide identity (ANI) was calculated using ANIclustermap [53], with the study genomes (MBBL2, MNH_G2C5, and MNH_G2C5F) and nine closely related K. pneumoniae genomes identified from the phylogenetic analysis (Table S2).

Pangenome analysis of the K. pneumoniae strains

To explore the complete genetic repertoire of K. pneumoniae, we performed pangenome analysis using Roary v3.11.2 [54] on 12 phylogenetically related genomes (Table S2). The genes were categorized into four groups: core (present in 99–100% of strains), soft core (95–99%), shell (15–95%), and cloud (< 15%). Furthermore, pangenome and core genome accumulation curves were constructed these genomes using the PanGP v1.0.1 software [55]. The analysis modeled new gene discovery as genomes were sequentially added, applying the distance guide algorithm with 100 replicates and 3,000 random genome order permutations. To estimate pangenome size, PanGP [55] employs a power-law regression model (y = Ax^B + C), where y denotes the total number of gene families, x is the number of genomes, and A, B, and C are model parameters. A value of B between 0 and 1 indicates an open pangenome. For core genome estimation, an exponential decay model (y = Ae^(Bx) + C) was used, with y representing the core genome size and x the number of genomes. Similarly, the rate of new gene acquisition was fitted using a least-squares power-law model (y = Ax^B), where y corresponds to the number of newly identified genes, x to the number of genomes, and A and B are fitting parameters [56].

Genome plasticity analysis of the K. pneumoniae strains

To investigate the mobile genetic elements (MGEs) in MBBL2, MNH_G2C5, MNH_G2C5F, and nine closely related K. pneumoniae genomes (Table S2), multiple bioinformatics tools were employed. Regions of genomic plasticity (RGPs) were identified using panRGP [57], insertion sequences (ISs) were detected with ISEscan [58], while prophage regions and plasmids were predicted with PHASTEST [59] and Platon [60], respectively.

Prediction of antimicrobial resistance and virulence in K. pneumoniae strains

To characterize the antimicrobial resistance and virulence potential of the 12 closely related K. pneumoniae genomes (Table S2), including the study genomes, multiple specialized databases and tools were applied. Antibiotic resistance genes (ARGs) were predicted using the Comprehensive Antibiotic Resistance Database (CARD) [61] and AMRFinderPlus [62]. Virulence-associated genes (VFGs) were identified against the Virulence Factor Database (VFDB) [63] implemented in ABRicate (https://github.com/tseemann/abricate) with default settings. Genomic islands, representing horizontally acquired regions, were detected using IslandViewer4 [64]. In addition, plasmid replicons were identified with the staramr tool [65] which integrates PlasmidFinder [66] to scan genome assemblies for plasmid-derived sequences.

Genomic functional analysis in K. pneumoniae strains

The genomes of MBBL2, MNH_G2C5, and MNH_G2C5F were functionally annotated, and their metabolic pathways were classified using the KEGG database [67] via the KEGG Automatic Annotation Server [68]. Secondary metabolite biosynthetic gene clusters were identified with antiSMASH v7.0 [69]. In addition, antimicrobial proteins and their corresponding genes were predicted using the BAGEL4 web server [70], and the predicted bacteriocin domains were validated by BLASTp [71] searches against the non-redundant protein (nr) database.

Prediction of the EPS gene operons

The sequence of K. pneumoniae genomes was aligned with known EPS coding manipulators through the Basic Local Alignment Search Tool (BLAST) program (https://blast.ncbi.nlm.nih.gov/Blast.cgi) to identify the presence of the genes and map the EPS synthesis gene cluster.

Statistical analysis

We randomly selected 120 samples and prevalence was determined from the observed frequencies without applying any statistical analyses. Descriptive statistics were used to examine the distribution of antimicrobial resistance profile of the study isolates, ARGs and VFGs repertoire of the MBBL2, MNH_G2C5 and MNH_G2C5F genomes. The data from the ASTs were analyzed using one-way analysis of variance (ANOVA) followed by Tukey’s multiple-comparison test. Both ARGs and VFGs data were normalized by Total Sum Scaling (TSS) that uses the total read count for each gene in each sample [72]. Statistical significance was set for all tests at p ≤ 0.05.

Prevalence of K. pneumoniae associated with bovine mastitis

A total of 27 isolates were confirmed as K. pneumoniae using selective culture media, a series of biochemical tests and 16S rRNA gene sequencing. In biochemical assays, the isolates were positive for catalase, citrate utilization, and lactose fermentation, and negative for indole, oxidase, and methyl red identified as K. pneumoniae. Source-specific evaluation showed a higher prevalence in milk samples (27.1%, 19 isolates) compared to fecal samples (16.0%, 8 isolates). The prevalence was calculated directly from observed frequencies.

Antibiogram profile of the K. pneumoniae isolates

A total of 27 isolates were tested against fifteen antimicrobial agents, and all isolates demonstrated MDR, exhibiting resistance to 3 ≥ antibiotics (File S1). The highest resistance was recorded for doxycycline (DO) with 21/27 (77.8%) resistant isolates, followed by ampicillin (AMP) and tetracycline (TE), each showing 20/27 (74.1%) resistance. Moderate resistance frequencies of 18/27 (66.7%) were observed for ciprofloxacin (CIP), nalidixic acid (NA), and sulfonamides (S3). Resistance to chloramphenicol (C) and cefoxitin (FOX) was also substantial, observed in 17/27 (63.0%) isolates. Similarly, azithromycin (AZM) and ceftriaxone (CRO) displayed resistance in 16/27 (59.3%) isolates each, while nitrofurantoin (F), oxacillin (OX), and imipenem (IPM) exhibited resistance in 15/27 (55.6%) isolates. Moderate resistance was seen with Streptomycin (S) (14/27, 51.9%), whereas gentamicin (CN) showed the lowest resistance level (9/27, 33.3%) and the highest susceptibility (14/27, 51.9%) (File S1). Notably, three isolates (MBBL2, MNH_G2C5, and MNH_G2C5F) exhibited extensive multidrug resistance, with resistance to 13 of the 15 antibiotics tested. These three MDR isolates were selected for detailed genomic characterization, and their resistance profiles are summarized in Fig. 1.

Genomic features of K. pneumoniae strains

The assembled genomes of K. pneumoniae strains MBBL2, MNH_G2C5, and MNH_G2C5F were approximately 5.39 Mbp, 4.59 Mbp, and 5.38 Mbp in size, respectively, each with a GC content of 57.5%. Genome quality assessment indicated high completeness (> 98%) and low contamination (0.82%) for MBBL2 and MNH_G2C5F, whereas MNH_G2C5 exhibited reduced completeness (80.93%) and slightly elevated contamination (2.35%). We predicted 5,341 coding sequences in MBBL2, 4,478 in MNH_G2C5, and 5,311 in MNH_G2C5F. Additionally, the genomes contained 89, 63, and 87 RNA genes, respectively. Further genomic characterization revealed the presence of a single CRISPR array in MBBL2 and MNH_G2C5F, while no CRISPR elements were detected in MNH_G2C5. Functional subsystem classification assigned 393 subsystems to MBBL2, 341 to MNH_G2C5, and 395 to MNH_G2C5F. A detailed summary on the genomic features of these genomes is provided in Table S3.

The circular genome representations of K. pneumoniae strains MBBL2, MNH_G2C5, and MNH_G2C5F, together with their COG-based functional annotations, are shown in Figure S1A and S1B. COG classification assigned genes into 27 functional categories across four major groups. The most abundant category was G (carbohydrate transport and metabolism), comprising 10.3%, 10.0%, and 10.4% of the genes in MBBL2, MNH_G2C5, and MNH_G2C5F, respectively. This was followed by category E (amino acid transport and metabolism; 8.8–9.4%), category K (transcription; 8.9–9.0%), category R (general function prediction only; 6.9–7.2%), and category P (inorganic ion transport and metabolism; ~6%). Additional enriched categories included C (energy production and conversion; 5.0–5.9%), M (cell wall/membrane/envelope biogenesis; 5.8%), and J (translation, ribosomal structure, and biogenesis; 4.7–5.4%).

Molecular typing and genomic epidemiology of K. pneumoniae strains

To elucidate the global genomic epidemiology of K. pneumoniae, we reconstructed the phylogenetic relationships among three study isolates (MBBL2, MNH_G2C5, and MNH_G2C5F) and 591 global genomes from NCBI GenBank using a seven-gene MLST scheme (gapA, infB, mdh, pgi, phoE, rpoB, tonB) (Fig. 2). K. pneumoniae strains MBBL2 and MNH_G2C5F were assigned to sequence type 273 (ST273), differing by 97 core genome MLST (cgMLST) alleles, whereas MNH_G2C5 formed a distinct clade within ST101, separated by 747 alleles (Figs. 2A–C). cgMLST analysis revealed that MBBL2 and MNH_G2C5F clustered with 18 MDR ST273 strains from nine countries, while MNH_G2C5 showed close relatedness to 273 ST101 strains from 32 countries. All study strains exhibited genetic links to both human and animal-derived isolates (File S1). An SNP-based phylogeny, constructed from 50,138 core SNPs across 52 genomes (including 49 reference genomes), placed MBBL2, MNH_G2C5F, and MNH_G2C5 in a clade predominantly composed of human-derived strains, except for one environmental isolate (STIN_94). MBBL2 and MNH_G2C5F co-clustered with clinical strains KP_NORM_BLD_116392 and ARLG-3223, while MNH_G2C5 grouped with C17KP0052, a human bloodstream infection isolate (Fig. 3, File S2). This topology suggested recent common ancestry for MBBL2 and MNH_G2C5F, with MNH_G2C5 representing an earlier divergence.

Genomic diversity and strain-specific genes in K. pneumoniae strains

To substantiate the phylogenetic relationships and provide high-resolution genomic demarcation of the study isolates, we performed ANI analysis on a curated panel of 12 closely related strains (including study strains), selected based on their phylogenetic proximity from an initial set of 52 global genomes (Table S2, Figure S2). This analysis robustly confirmed the species-level identity of all three study strains (MBBL2, MNH_G2C5, and MNH_G2C5F), each demonstrating > 98% ANI with established K. pneumoniae reference strains, far exceeding the 95% threshold for species delineation [53]. The ANI results provided precise quantification of the genetic distances inferred from phylogeny. MBBL2 and MNH_G2C5F were nearly identical (99.9% ANI), whereas MNH_G2C5 showed slightly lower similarity (98.9%), reflecting its more distant phylogenetic position. MBBL2 and MNH_G2C5F exhibited the highest ANI (99.8%) with Kpn223 and KP_NORM_BLD_116392. In contrast, MNH_G2C5 was most closely related (98.9% ANI) to a broader set of strains, including ARLG-3223, Kpn223, KP_NORM_BLD_116392, MNH_G2C5F, MBBL2, 032D, 7008.20, and 5844 (Table S2, Figure S2).

A pangenome analysis was performed on the study genomes (MBBL2, MNH_G2C5, and MNH_G2C5F) together with nine reference genomes (Table S2) to explore their gene repertoire. The pangenome dendrogram revealed that MBBL2 and MNH_G2C5F were closely related to reference K. pneumoniae strains 5844, 032D, and 7008.20, while MNH_G2C5 showed the highest similarity to strains 6072, 11383, and 4254 (Fig. 4A). We identified 7,873 genes across the 12 K. pneumoniae genomes (Fig. 4B, Table S4). Strain-specific analysis further showed that MBBL2 possessed 1 unique gene and 826 accessory genes, MNH_G2C5 carried 123 unique genes and 234 accessory genes, whereas MNH_G2C5F, notably, contained no unique genes but shared 826 accessory genes (Fig. 4C, Table S4). The new gene accumulation curve (Fig. 4D) followed a power-law decay (y = 312.947x^− 1.19, R² = 0.981963), indicating that the number of novel genes declines steadily with the addition of more genomes but does not reach a plateau, consistent with an open genome concept. Similarly, the pangenome expansion curve (Fig. 4E) also supported an open pangenome model, as the total number of gene clusters continued to rise with each additional genome (y = 200124x⁰, R² = 0.1), reflecting high genetic diversity. In contrast, the core genome displayed a progressive reduction, best fitted by an exponential decay model (y = 1358.75e^− 0.08x + 2343.19, R² = 0.994518), suggesting gene loss with the inclusion of additional genomes.

Genomic plasticity and mobile genetic elements in K. pneumoniae strains

To assess the genomic plasticity of 12 phylogenetically related K. pneumoniae genomes, we employed a pangenome-based approach to predict regions of genomic plasticity (RGPs), defined as gene clusters located within highly variable genomic regions. The number of RGPs varied widely across genomes, ranging from 11 in KP_NORM_BLD_116392 to 34 in MNH_G2C5, which harbored the highest count. Notably, MBBL2 and MNH_G2C5F carried an equal number of RGPs (27 each) (Fig. 5). The sizes of these regions spanned from approximately 0.26 Mb to 1.061 Mb (File S3). To further explore contributors to genomic plasticity, we examined MGEs, including plasmids, prophages, and ISs. Plasmid profiling revealed four distinct replicon types among the three isolates. Both MBBL2 and MNH_G2C5F harbored IncFIB(K)(pCAV1099-114), IncFII(K), and IncHI1B(pNDM-MAR) plasmids while MNH_G2C5 carried only an IncR plasmid. Prophage counts varied across strains, with some harboring only intact prophages and others carrying a mixture of intact and questionable ones. MNH_G2C5 contained four prophage regions, whereas MBBL2 and MNH_G2C5F each carried two. Across the 12 K. pneumoniae genomes, a total of 816 IS elements were detected, of which 58, 52, and 46 were present in MBBL2, MNH_G2C5F, and MNH_G2C5, respectively (File S3).

Resistome repertoire in K. pneumoniae strains

Comprehensive resistome analysis of 12 closely related K. pneumoniae genomes highlighted the genomic diversity and adaptability of these strains to multiple stressors, including acid, AMR, biocides, and metals. The three study genomes viz. MNH_G2C5, MBBL2, and MNH_G2C5F carried 41, 30, and 29 resistome genes, respectively (Fig. 6, File S4), all of which included a strong core of acid resistance genes (e.g., asr). A total of 48 AMR genes, conferring resistance to 12 antibiotic classes, were identified across the three genomes. The AMR gene count was highest in MNH_G2C5F (18), followed by MBBL2 (17) and MNH_G2C5 (13). These genes conferred resistance to aminoglycosides (e.g., aac(3)-IId, aadA5), beta-lactams (e.g., blaCTX-M-15, blaLAP-2, blaSHV-1, blaSHV-11, blaTEM-1), trimethoprim (e.g., dfrA1, dfrA14, dfrA17), phenicols (e.g., oqxA, oqxB, oqxB20), tetracyclines (e.g., tet(A), tet(D)), sulfonamides (e.g., sul1, sul2), fluoroquinolones (e.g., qnrS1), fosfomycin (e.g., fosA), rifamycin (e.g., arr), macrolides (e.g., mph(A)), and efflux pumps (e.g., emrD), representing a shared resistance core (Fig. 6). The qacEdelta1 gene, conferring resistance to quaternary ammonium compounds, was detected in all three strains, suggesting tolerance to common chemical disinfectants. In addition, 34 metal resistance genes were detected in the three genomes. Arsenic resistance genes (e.g., arsC) were present in all strains, while mercury resistance genes (e.g., merA/C/P/R/T) were found in MBBL2 and MNH_G2C5F. Notably, copper resistance genes (e.g., pcoA/B/C/D/R/S) and silver resistance genes (e.g., silA/B/C/E/F/P/R/S) were exclusive to MNH_G2C5. Furthermore, MNH_G2C5 harbored nine heat resistance genes (e.g., psi-GI, kefB-GI, trxLHR, hdeD-GI, yfdX1/2, shsP, clpK, hsp20) (Fig. 6, File S4).

Virulence gene arsenal and mechanistic insights in K. pneumoniae strains

Virulence factor gene (VFG) profiles of the three study genomes were analyzed to assess their pathogenic potential. MBBL2 and MNH_G2C5F each harbored 60 VFGs, whereas MNH_G2C5 contained 44. These VFGs were found to facilitate virulence through several distinct mechanisms, including nutritional/metabolic factor, regulation, biofilm, adherence, effector delivery system, antimicrobial activity, immune modulation (Figure S3, File S4). In all genomes, adherence-related genes represented the largest category, accounting for 26.67% in MBBL2 and MNH_G2C5F and 34.1% in MNH_G2C5. Moreover, MBBL2 and MNH_G2C5F genomes possessed 14 effector delivery system genes (23.33%), 13 nutritional/metabolic genes (21.67%), eight biofilm-related genes (13.33%), four immune modulation genes (6.67%), three regulatory genes (5%), and two antimicrobial activity genes (3.33%). In contrast, MNH_G2C5 contained 13 effector delivery system genes (29.54%), eight biofilm-related genes (18.18%), four genes each for regulation and antimicrobial activity (4.54% each), and lacked immune modulation genes (Figure S3, File S4).

Prediction of genomic islands in K. pneumoniae strains

Genomic islands (GIs), which are gene clusters frequently acquired through horizontal gene transfer (HGT), were predicted in all three genomes and were found to harbor multiple VFGs and pathogen-associated genes, underscoring their potential role in pathogenicity. In the MBBL2 strain, 37 putative GIs were identified, of which 25 were predicted by SIGI-HMM and 12 by IslandPath-DIMOB, with sizes ranging from 4,035 to 291,727 bp. Among them, ten GIs carried genes associated with either virulence factors (VFs) or AMR. Notably, MBBL2 harbored curated AMR genes, including aac(3)-IId (WP_000557454.1), dfrA17 (WP_001389366.1), mrxA (WP_000004159.1), sul1 (WP_000259031.1), sul2 (WP_001043260.1), ACHL6M_RS25720 (WP_001516695.1), ACHL6M_RS25995 (WP_000503573.1), and ACHL6M_RS26045 (WP_000219391.1). In addition, homologs of resistance genes such as alaS (WP_004174700.1), floR (WP_023300759.1), kexD (WP_004200176.1), tetA (WP_000804064.1), ugd (WP_023328477.1), and ACHL6M_RS08150 (WP_023341379.1) were also identified (Fig. 7).

In comparison, the MNH_G2C5 strain harbored 38 GIs, with 25 predicted by the SIGI-HMM method and 13 by IslandPath-DIMOB, ranging from 3,010 bp to 46,746 bp in length. Four of these islands contained genes associated with VFs or AMR. MNH_G2C5 carried multiple curated resistance genes, such as PGZ78_RS07210 (WP_001516695.1), PGZ78_RS07200 (WP_000027057.1), dfrA1 (WP_000777554.1), and sul1 (WP_000259031.1), as well as homologs of resistance genes, including tetA (WP_000804064.1), and PGZ78_RS00380 (WP_032419057.1). Notably, one pathogen-associated gene, PGZ78_RS16620 (WP_000376623.1) was also identified (Fig. 7). Similarly, MNH_G2C5F contained 39 GIs, comprising 25 predicted by the SIGI-HMM method and 14 by IslandPath-DIMOB, spanning 4,035–190,576 bp, with ten GIs carrying VFs or AMR-related genes. This strain encoded multiple curated resistance genes such as Q2T98_RS25690 (WP_001516695.1), Q2T98_RS25965 (WP_000503573.1), dfrA17 (WP_001389366.1), Q2T98_RS26015 (WP_000219391.1), mrxA (WP_000004159.1), aac(3)-IId (WP_000557454.1), sul1 (WP_000259031.1), and sul2 (WP_001043260.1). It also harbored homologs of resistance genes, including alaS (WP_004174700.1), floR (WP_023300759.1), kexD (WP_004200176.1), tetA (WP_000804064.1), ugd (WP_023328477.1), and Q2T98_RS08145 (WP_023341379.1) (Fig. 7).

K. pneumoniae strains utilized diverse metabolic pathways to facilitate adaptation and survival

The systemic metabolic pathways of gene products in the MBBL2, MNH_G2C5, and MNH_G2C5F genomes were analyzed to better understand how these strains utilized nutrients, interacted with their environment, and influenced health and disease processes. A total of 3,524 genes (68.7%) from MBBL2, 2,973 genes (68.9%) from MNH_G2C5, and 3,514 genes (68.8%) from MNH_G2C5F were successfully annotated and mapped to KEGG pathways, spanning six major categories and forty-one subcategories (Figure S4). Among the major categories, metabolism-related genes were the most abundant (MBBL2 = 1,712; MNH_G2C5 = 1,492; MNH_G2C5F = 1,711), followed by those associated with environmental information processing, genetic information processing, cellular processes, human diseases, and organismal systems. Additionally, 58, 56, and 58 drug (antimicrobial) resistance genes were identified in MBBL2, MNH_G2C5, and MNH_G2C5F, respectively. All three genomes carried a complete biosynthesis of amino acids pathway that contributed to biogenic amine production. Seventeen complete modules related to production of biogenic amine identified in MNH_G2C5, while both MBBL2 and MNH_G2C5F genomes have 21 complete modules which correspond to serine biosynthesis, methionine biosynthesis, histidine biosynthesis and shikimate pathway respectively (Figure S5, File S5). Further analysis revealed that the three genomes harbored multiple genes influencing cellular community, including 71 quorum sensing genes in MBBL2 and MNH_G2C5F and 63 in MNH_G2C5 (e.g., luxS, qseC, qseB, lsrA, hfq, etc.), as well as biofilm-associated genes homologous to those characterized in Vibrio cholerae (e.g., cysE, wecB, csrA, etc.), Pseudomonas aeruginosa (e.g., gacS, crp, barA, etc.), and Escherichia coli (e.g., glgA, bcsA, rpoS, etc.) (File S5). In addition, both MBBL2 and MNH_G2C5F genomes possessed 45 beta-lactam resistance genes while MNH_G2C5 contained 44 beta-lactam resistance genes. All three genomes also carried five vancomycin resistance genes and five platinum drug resistance genes. Moreover, MBBL2 and MNH_G2C5F harbored 58 cationic antimicrobial peptide (CAMP) resistance genes and seven antifolate resistance genes, compared with 57 and five, respectively, in MNH_G2C5. Moreover, all three genomes possessed 19 genes (e.g., dxr, ACAT, atoB, ispA, uppS, ispE, ispD, dxs) associated with terpenoid backbone biosynthesis, along with 15 genes (e.g., entA, entB, dhbB, vibB, mxcF, entC, entD, pptT) related to the biosynthesis of siderophore group nonribosomal peptides.

Secondary metabolite gene clusters and human pathogenic potential in K. pneumoniae strains

The genomes of MBBL2, MNH_G2C5, and MNH_G2C5F harbored multiple biosynthetic gene clusters (BGCs) linked to secondary metabolites (Fig. 8). MBBL2 and MNH_G2C5 contained a redox-cofactor gene cluster, including pqqC, pqqD, and pqqE, involved in pyrroloquinoline quinone (PQQ) biosynthesis. Both genomes also encoded two azole-containing RiPPs (ribosomally synthesized and post-translationally modified peptides) encoded by pflA and ycaO (Figs. 8A and 8C). Moreover, MBBL2 and MNH_G2C5F genomes carried an NRP (non-ribosomal peptide)-metallophore gene cluster for enterobactin biosynthesis (entF, entC, entA) and a terpene-precursor which is encoded by ispA gene (Figs. 8A and 8B). Importantly, all of three genomes possessed a RiPP-like protein dsbA involved in disulfide bond formation (Figs. 8A-C). Additionally, BAGEL4 analysis identified rlmN (BmbF) in K. pneumoniae MNH_G2C5 (Protein ID: MDB1120801.1), a conserved bifunctional rRNA/tRNA methyltransferase experimentally validated at the protein level (PE = 3). The protein showed 100% sequence identity with characterized methyltransferases across multiple K. pneumoniae strains. Besides, MNH_G2C5 genome harbored BmbF (dual-specificity RNA methyltransferase RlmN) protein (Protein Id = MDB1120801.1) (Fig. 8D).

Simultaneously, the pathogenicity prediction analysis suggested that all three K. pneumoniae genomes possess a high likelihood of being human pathogens, as evidenced by PathogenFinder scores of 0.901 for MBBL2 and MNH_G2C5F, and 0.911 for MNH_G2C5. These values corresponded to the identification of 322, 255, and 322 pathogenic gene families across these genomes, respectively.

EPS and capsule biosynthesis gene clusters in K. pneumoniae strains

Further analysis of EPS and capsule biosynthesis clusters in three K. pneumoniae genomes revealed a largely conserved architecture with minor variations (Fig. 9). All genomes harbored two EPS/capsule biosynthesis clusters, with the MBBL2 and MNH_G2C5F clusters showing relative similarity. Both MBBL2-C1 and MNH_G2C5F_C1 contained a canonical EPS/capsule cluster, including genes involved in sugar-nucleotide biosynthesis (ugd, gndA), glycosyltransferases and tailoring enzymes (GT, GT2, GT4, wbaP), regulation and signaling (wzc, wzb, wcaJ, galF), and export/translocation (wza, Wzi, EpsG) (Table S5). Notably, MNH_G2C5F_C1 shared nearly all genes with MBBL2-C1, differing only by one or two hypothetical or accessory genes. Conversely, MBBL2-C2 and MNH_G2C5F_C2 contained a distinct EPS-like cluster comprising glycosyltransferases (wecA, wecF, wecG, rffC, rffM), polymerization and chain-length control genes (wzyE, wzzE), sugar-nucleotide biosynthesis genes (rffA, rffG, rfbA, wecB, wecC), and export-associated genes (wzxE). These clusters were highly conserved between the genomes, suggesting functional equivalence in polysaccharide assembly. Additionally, MNH_G2C5_C1 carried genes for regulation and signaling (rho), glycosyltransferases (wecA, wecF, rffM), polymerization/chain-length control (wzyE, wzzE), sugar-nucleotide biosynthesis (wecB, wecC, rffG, rfbA, rffA), and export (wzxE). MNH_G2C5_C2 featured a cluster including glycosyltransferases (GT, GT2, GT4, GT9, kbl, rfaC, rfaF, waaZ, waaA), accessory enzymes (tdh), sugar-nucleotide biosynthesis (rfaD), and export (waaL) (Fig. 9, Table S5).

The emergence of K. pneumoniae as a major pathogen in bovine mastitis [1, 14, 23] represents a significant challenge to both animal welfare and public health due to its potential for zoonotic transmission and MDR [23, 73]. This study provides a comprehensive genomic characterization of three MDR K. pneumoniae strains isolated from CM milk (MBBL2, MNH_G2C5) and feces (MNH_G2C5F) in the dairy farms of Bangladesh, revealing a complex interplay of AMR, virulence, and genomic plasticity. The high prevalence of K. pneumoniae (22.5%), particularly in milk samples (27.14%), underscores its role as a primary mastitis pathogen. This is consistent with studies from China and the USA, which have reported an increasing incidence of Klebsiella species in bovine CM, often linked to environmental contamination and poor udder hygiene [11, 13]. Subsequent AST profiling revealed an alarming resistance scenario, with 86.67% of tested antibiotics ineffective and all 27 isolates classified as MDR. The high resistance to first-line antibiotics like ampicillin, tetracycline, and nalidixic acid mirrors trends observed in human and veterinary medicine worldwide, driven by the selective pressure of antimicrobial misuse [20, 74]. To unveil the genetic basis of this high MDR, three isolates (MBBL2, MNH_G2C5, MNH_G2C5F), resistant to 13 (out of 15) antibiotics, were subjected to WGS to characterize their resistome and virulence repertoire.

One of the noteworthy findings was the co-circulation of two high-risk K. pneumoniae clones, ST273 (MBBL2, MNH_G2C5F) and ST101 (MNH_G2C5), within the dairy herds of Bangladesh, underscoring their direct relevance to the One Health framework. While ST101 is a well-documented global MDR clone associated with hospital outbreaks in humans and increasingly reported in livestock [25, 75], ST273 is less commonly reported but has been linked to severe infections in both humans and animals in Asia [76]. The phylogenetic proximity of these bovine-derived ST273 isolates to human clinical strains provides compelling evidence for recent cross-species transmission [77]. Furthermore, the pangenome analysis revealed an open genome, characterized by a large accessory gene pool and a decaying core, a hallmark of K. pneumoniae that facilitates niche adaptation [24]. The substantial number of strain-specific genes in MNH_G2C5 (ST101) and the nearly identical accessory genome of the ST273 isolates (MBBL2, MNH_G2C5F) underscore how clonal background shapes genomic plasticity. This open pangenome structure promotes the acquisition of MGEs, effectively turning the farm environment into a potential crucible for the evolution of new MDR and virulent variants through HGT [27]. The diverse resistome observed among the K. pneumoniae strains through integrated phenotypic and genomic analyses reflects strong selective pressures driving genomic plasticity and adaptation. Such heterogeneity suggests frequent HGT and exposure to varied antimicrobial and environmental stressors [27]. Both MBBL2 and MNH_G2C5F genomes harbored a more extensive arsenal of AMR genes, conferring resistance to critically important beta-lactams and fluoroquinolones. This genetic repertoire is critical for successful MDR clones, and has been extensively documented in human clinical isolates [78]. The presence of oqxAB and fosA in all three genomes, conferring resistance to chloramphenicol and fosfomycin, is noteworthy as these are often considered last-resort treatments for MDR infections. A recent study from India similarly identified a high prevalence of oqxAB in bovine K. pneumoniae, suggesting a widespread and potentially overlooked resistance mechanism in livestock reservoirs [79]. Furthermore, the detection of metal and biocide resistance genes (qacEdelta1, ars, mer) indicates adaptation to farm environments where disinfectants and heavy metals are commonly used, a trait increasingly recognized as co-selected with AMR in agricultural settings [80]. Beyond their resistance traits, the pathogenic potential of these strains is notable, as all genomes were predicted with a high probability (> 0.90) of being human pathogens. The high number of VFGs, particularly those related to adherence and biofilm formation, is crucial for establishing infections in both the bovine udder and human tissues [1, 12, 15, 81]. The abundance of adherence genes in MNH_G2C5 and the effector delivery system genes in all strains suggest multiple strategies for host colonization and immune evasion. The presence of complete siderophore systems (e.g., enterobactin) is a key virulence factor for systemic infection, as recently demonstrated in hypervirulent K. pneumoniae strains of livestock origin [82].

The high genomic plasticity observed in the study strains underscores their remarkable ability to adapt and evolve in response to selective pressures. The identification of numerous AMR, drug resistance and virulence genes within predicted GIs strongly suggests their acquisition via HGT [83]. A recent study on K. pneumoniae from dairy farms similarly found that GI-carried genes were major contributors to the strain-specific adaptation and resistance profiles [22]. The metabolic versatility of the strains, evidenced by the extensive KEGG pathway annotations for amino acid biosynthesis, quorum sensing, and biofilm formation, further explains their ability to thrive in diverse niches, from the bovine gut to the udder [1, 81] and potentially the humans [12, 15]. The distinct plasmid profiles further highlight the dynamic nature of MGEs; the presence of IncHI1B(pNDM-MAR) and IncFIB(K) plasmids in the ST273 isolates is particularly concerning, as these replicon types are frequently associated with the global dissemination of carbapenem resistance, even though no carbapenem resistance was phenotypically detected in this study [26]. Analysis of EPS and capsule biosynthesis clusters in the three K. pneumoniae genomes revealed a largely conserved architecture with minor variations, emphasizing their key role in virulence and environmental adaptation [84]. Core genes involved in capsule export and regulation (e.g., wza, wzxE) are essential for assembling a protective capsule that shields the bacterium from phagocytosis, while the extensive repertoire of glycosyltransferases supports biofilm formation [1]. This is further corroborated by virulence factor analysis, which showed that all three isolates harbor numerous biofilm-associated genes. Collectively, the observed genomic plasticity within EPS clusters underscores the strains’ potential for persistent infections through enhanced immune evasion and biofilm-mediated survival [84, 85]. The genomic insights from this study are limited by the small number of sequenced isolates and the absence of in-vitro or in-vivo validation of the predicted resistance and virulence determinants. Further investigations are necessary to confirm the pathogenic potential and zoonotic risk posed by these high-risk K. pneumoniae clones.

This in-depth genomic investigation reveals that K. pneumoniae strains from bovine CM-associated milk and feces are important pathogens harboring high-risk MDR clones (ST101 and ST273). The convergence of a broad resistome, a rich virulome, and significant genomic plasticity in these strains poses a significant threat not only for animal health and production but also a potential reservoir for human infections. These findings emphasize the role of dairy animals in the evolution and dissemination of MDR and hyper-virulent superbugs like K. pneumoniae. This study underscores the urgent need for integrated One Health surveillance strategies that combine genomic epidemiology with robust antimicrobial stewardship in both veterinary and human medicine to mitigate the spread of these ever-challenging pathogens.

Ethical approval

The Animal Research Ethics Committee (AREC) of the Gazipur Agricultural University (Gazipur, Bangladesh) reviewed and approved the experimental procedures of this study (reference number FVMAS/AREC/2023/6679 [16 January 2023]). Written informed consent was obtained from each participating dairy farmer prior to the inclusion of their animals in this study, ensuring ethical compliance and voluntary participation.

Clinical trial number

Not applicable.

Consent for publication

Not applicable.

Conflict of interest

The authors declare no conflict of interest.

Funding

This research was supported by a grant from the Bangladesh Bureau of Educational Information and Statistics (BANBEIS), Ministry of Education, Government of the People’s Republic of Bangladesh (Grant No. LS20221764, duration 2023–2025).

Authors contributions

M.N.H. conceived and designed the study. K.Y.A, M.A.A.G., M.M.R., N.S. and F.H. performed experiment, curated, analyzed and visualized data, interpreted results, and wrote original draft. M.N.H. critically reviewed and edited the manuscript. The final manuscript was read and approved by all authors.

Data availability

The 16S rRNA gene sequences of K. pneumoniae strains MBBL2, MNH_G2C5 and MNH_G2C5F have been deposited in NCBI GenBank under accession numbers PX447844, PX447845 and PX447846, respectively. The whole genome shotgun sequences for these strains are available in GenBank and the NCBI Sequence Read Archive (SRA) under BioProject accessions PRJNA1171635 for MBBL2, PRJNA923580 for MNH_G2C5, and PRJNA994715 for MNH_G2C5F. The genome versions reported in this study are JBINJS000000000.1, for MBBL2, JAQLOG000000000.1 for MNH_G2C5 and JAULSG000000000.1 for MNH_G2C5F. All datasets generated for this study are included in this article and the supplementary files.

Hoque MN et al (2020) Insights into the resistome of bovine clinical mastitis microbiome, a key factor in disease complication. Front Microbiol 11:860
Hoque MN et al (2019) Metagenomic deep sequencing reveals association of microbiome signature with functional biases in bovine mastitis. Sci Rep 9(1):13536
Wani SA et al (2022) A brief analysis of economic losses due to mastitis in dairy cattle. Indian Vet J 90:27–31
Hoque MN et al (2020) Microbiome dynamics and genomic determinants of bovine mastitis. Genomics 112(6):5188–5203
Degi J et al (2025) Antimicrobial Susceptibility Patterns of Staphylococcus spp. Isolates from Mastitic Cases in Romanian Buffaloes from Western Romania. Antibiotics 14(6):537
Erskine R et al (2002) Efficacy of systemic ceftiofur as a therapy for severe clinical mastitis in dairy cattle. J Dairy Sci 85(10):2571–2575
Zhao X, Lacasse P (2008) Mammary tissue damage during bovine mastitis: causes and control. J Anim Sci 86(suppl13):57–65
Cheng J et al (2020) Klebsiella pneumoniae isolated from bovine mastitis is cytopathogenic for bovine mammary epithelial cells. J Dairy Sci 103(4):3493–3504
Cheng J et al (2021) Klebsiella pneumoniae infection causes mitochondrial damage and dysfunction in bovine mammary epithelial cells. Vet Res 52(1):17
Goulart DB, Mellata M (2022) Escherichia coli mastitis in dairy cattle: etiology, diagnosis, and treatment challenges. Front Microbiol 13:928346
Zheng Z et al (2022) Whole-genome analysis of Klebsiella pneumoniae from bovine mastitis milk in the US. Environ Microbiol 24(3):1183–1199
Aslam B et al Transmission Dynamics and Genetic Relatedness of MDR Klebsiella pneumoniae at the Human-Animal-Environment Interface in Pakistan. Saima and Siddique, Abu Baker and Yasmeen, Nafeesa and Khurshid, Mohsin and Rasool, Muhammad Hidayat and Xia, Xueshan and baloch, zulqarnain, Transmission Dynamics and Genetic Relatedness of MDR Klebsiella pneumoniae at the Human-Animal-Environment Interface in Pakistan
Fu S et al (2022) Molecular epidemiology and antimicrobial resistance of outbreaks of Klebsiella pneumoniae clinical mastitis in Chinese dairy farms. Microbiol Spectr 10(6):e02997–e02922
Hoque MN et al (2023) Draft genome sequence of a multidrug-resistant Klebsiella pneumoniae fecal isolate from a cow with clinical mastitis. Microbiol Resource Announcements 12(11):e00730–e00723
Yang Y et al (2019) Genomic diversity, virulence, and antimicrobial resistance of Klebsiella pneumoniae strains from cows and humans. Appl Environ Microbiol 85(6):e02654–e02618
Cheng J et al (2021) Genetic diversity and molecular epidemiology of outbreaks of Klebsiella pneumoniae mastitis on two large Chinese dairy farms. J Dairy Sci 104(1):762–775
Munoz MA et al (2007) Molecular epidemiology of two Klebsiella pneumoniae mastitis outbreaks on a dairy farm in New York State. J Clin Microbiol 45(12):3964–3971
Chowdhury MSR et al (2025) Emergence of highly virulent multidrug and extensively drug resistant Escherichia coli and Klebsiella pneumoniae in buffalo subclinical mastitis cases. Sci Rep 15(1):11704
Oliver SP, Murinda SE (2012) Antimicrobial resistance of mastitis pathogens. Veterinary Clinics: Food Anim Pract 28(2):165–185
Murray CJ et al (2022) Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. lancet 399(10325):629–655
Gorden P et al (2018) Efficacy of vaccination with a Klebsiella pneumoniae siderophore receptor protein vaccine for reduction of Klebsiella mastitis in lactating cattle. J Dairy Sci 101(11):10398–10408
Han B et al (2025) Mobile genetic elements encoding antibiotic resistance genes and virulence genes in Klebsiella pneumoniae: important pathways for the acquisition of virulence and resistance. Front Microbiol 16:1529157
Kang HJ et al (2025) Genetic diversity, virulence genes, antimicrobial resistance, and biofilm formation of Klebsiella pneumoniae isolated from bovine mastitis milk in South Korea. Microbiol Spectr, : p. e01343–e01325
Bialek-Davenet S et al (2014) Genomic definition of hypervirulent and multidrug-resistant Klebsiella pneumoniae clonal groups. Emerg Infect Dis 20(11):1812
Wyres KL et al (2020) Genomic surveillance for hypervirulence and multi-drug resistance in invasive Klebsiella pneumoniae from South and Southeast Asia. Genome Med 12(1):11
Li L et al (2024) Global emergence and genomic epidemiology of bla NDM-carrying Klebsiella variicola. Infect Drug Resist, : p. 1893–1901
Wyres KL, Holt KE (2018) Klebsiella pneumoniae as a key trafficker of drug resistance genes from environmental to clinically important bacteria. Curr Opin Microbiol 45:131–139
Chou A et al (2016) Emergence of Klebsiella pneumoniae ST273 carrying bla NDM-7 and ST656 carrying bla NDM-1 in Manila, Philippines. Microb Drug Resist 22(7):585–588
Rendueles O (2020) Deciphering the role of the capsule of Klebsiella pneumoniae during pathogenesis: A cautionary tale. Mol Microbiol 113(5):883–888
Sultana T et al (2025) Genomic characterization of Pseudomonas asiatica as an emerging mastitis pathogen in dairy cows with resistance and virulence implications. J Global Antimicrob Resist
Mallick TT et al (2025) Molecular and genomic investigation unveils Pseudomonas putida as an emerging multidrug-resistant pathogen linked to bovine clinical mastitis. Microb Pathog 203:107461
Rahman MM et al (2024) Genomic insights into the probiotic potential and genes linked to gallic acid metabolism in Pediococcus pentosaceus MBBL6 isolated from healthy cow milk. PLoS ONE 19(12):e0316270
Hoque MN et al (2015) Different screening tests and milk somatic cell count for the prevalence of subclinical bovine mastitis in Bangladesh. Trop Anim Health Prod 47(1):79–86
Sultana T et al (2024) Draft genome sequencing of multidrug-resistant Pseudomonas asiatica strains isolated from dairy cows with clinical mastitis and their farm environment. Microbiol Resource Announcements 13(11):e00907–e00924
Hoque MN et al (2023) Whole-genome sequence of multidrug-resistant Klebsiella pneumoniae MNH_G2C5, isolated from bovine clinical mastitis milk. Microbiol Resource Announcements 12(5):e00079–e00023
Rahman MM et al (2025) Draft genome sequencing of a multidrug-resistant Klebsiella pneumoniae strain MBBL2 isolated from mastitic cow milk. Microbiol Resource Announcements 14(3):e01204–e01224
Masomian M et al (2016) Analysis of comparative sequence and genomic data to verify phylogenetic relationship and explore a new subfamily of bacterial lipases. PLoS ONE 11(3):e0149851
Rai S, Dash D, Agarwal N (2023) Introducing the new face of CLSI M100 in 2023: An explanatory review. Ind J Med Microbiol 46:100432
Hoque MN et al (2024) Unveiling distinct genetic features in multidrug-resistant Escherichia coli isolated from mammary tissue and gut of mastitis induced mice. Heliyon, 10(5)
Siddique N et al (2025) Probiotic potential and antimicrobial efficacy of a dairy isolate, Enterococcus faecium MBBL3. Appl Microbiol Biotechnol 109(1):1–20
Arafat KY et al (2025) Draft genome sequencing of a multidrug-resistant and virulent Escherichia coli strain isolated from milk of a cow with clinical mastitis. Microbiol Resource Announcements 14(3):e01260–e01224
Simon Andrews F et al (2010) FastQC: a quality control tool for high throughput sequence data
Bolger AM, Lohse M, Usadel B (2014) Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30(15):2114–2120
Bankevich A et al (2012) SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol 19(5):455–477
Tatusova T et al (2016) NCBI prokaryotic genome annotation pipeline. Nucleic Acids Res 44(14):6614–6624
Overbeek R et al (2014) The SEED and the Rapid Annotation of microbial genomes using Subsystems Technology (RAST). Nucleic Acids Res 42(D1):D206–D214
Parks DH et al (2015) CheckM: assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Res 25(7):1043–1055
Cosentino S et al (2013) PathogenFinder-distinguishing friend from foe using bacterial whole genome sequence data. PLoS ONE 8(10):e77302
Cumsille A et al (2023) GenoVi, an open-source automated circular genome visualizer for bacteria and archaea. PLoS Comput Biol 19(4):e1010998
Feng Y et al (2021) BacWGSTdb 2.0: a one-stop repository for bacterial whole-genome sequence typing and source tracking. Nucleic Acids Res 49(D1):D644–D650
Kille B et al (2024) Parsnp 2.0: scalable core-genome alignment for massive microbial datasets. Bioinformatics 40(5):btae311
Letunic I, Bork P (2021) Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res 49(W1):W293–W296
Musiał K, Petruńko L, Gmiter D (2024) Simple approach to bacterial genomes comparison based on Average Nucleotide Identity (ANI) using fastANI and ANIclustermap. Acta Universitatis Lodziensis Folia Biologica et Oecol 18:66–71
Page AJ et al (2015) Roary: rapid large-scale prokaryote pan genome analysis. Bioinformatics 31(22):3691–3693
Zhao Y et al (2014) PanGP: a tool for quickly analyzing bacterial pan-genome profile. Bioinformatics 30(9):1297–1299
Thomas P et al (2021) Genome sequence analysis of Clostridium chauvoei strains of European origin and evaluation of typing options for outbreak investigations. Front Microbiol 12:732106
Bazin A et al (2020) panRGP: a pangenome-based method to predict genomic islands and explore their diversity. Bioinformatics 36(Supplement2):i651–i658
Xie Z, Tang H (2017) ISEScan: automated identification of insertion sequence elements in prokaryotic genomes. Bioinformatics 33(21):3340–3347
Wishart DS et al (2023) PHASTEST: faster than PHASTER, better than PHAST. Nucleic Acids Res 51(W1):W443–W450
Schwengers O et al (2020) Platon: identification and characterization of bacterial plasmid contigs in short-read draft assemblies exploiting protein sequence-based replicon distribution scores. Microb genomics 6(10):e000398
Jia B et al (2016) CARD 2017: expansion and model-centric curation of the comprehensive antibiotic resistance database. Nucleic Acids Res, : p. gkw1004
Feldgarden M et al (2021) AMRFinderPlus and the Reference Gene Catalog facilitate examination of the genomic links among antimicrobial resistance, stress response, and virulence. Sci Rep 11(1):12728
Chen L et al (2016) VFDB 2016: hierarchical and refined dataset for big data analysis—10 years on. Nucleic Acids Res 44(D1):D694–D697
Bertelli C et al (2017) IslandViewer 4: expanded prediction of genomic islands for larger-scale datasets. Nucleic Acids Res 45(W1):W30–W35
Bharat A et al (2022) Correlation between phenotypic and in silico detection of antimicrobial resistance in Salmonella enterica in Canada using Staramr. Microorganisms 10(2):292
Carattoli A et al (2014) PlasmidFinder and pMLST: in silico detection and typing of plasmids. Antimicrob Agents Chemother 58(7):3895–3903
Kanehisa M, Sato Y, Morishima K (2016) BlastKOALA and GhostKOALA: KEGG tools for functional characterization of genome and metagenome sequences. J Mol Biol 428(4):726–731
Moriya Y et al (2007) KAAS: an automatic genome annotation and pathway reconstruction server. Nucleic Acids Res 35(suppl2):W182–W185
Blin K et al (2023) antiSMASH 7.0: new and improved predictions for detection, regulation, chemical structures and visualisation. Nucleic Acids Res 51(W1):W46–W50
van Heel AJ et al (2018) BAGEL4: a user-friendly web server to thoroughly mine RiPPs and bacteriocins. Nucleic Acids Res 46(W1):W278–W281
Lavigne R et al (2008) Unifying classical and molecular taxonomic classification: analysis of the Podoviridae using BLASTP-based tools. Res Microbiol 159(5):406–414
Huang M et al (2021) Microbiome-specific statistical modeling identifies interplay between gastrointestinal microbiome and neurobehavioral outcomes in patients with autism: a case control study. Front Psychiatry 12:682454
Hu Y et al (2021) Klebsiella pneumoniae: prevalence, reservoirs, antimicrobial resistance, pathogenicity, and infection: a hitherto unrecognized zoonotic bacterium. Foodborne Pathog Dis 18(2):63–84
Al Amin M et al (2022) Methodology for laboratory-based antimicrobial resistance surveillance in animals. Veterinary World 15(4):1066
Shankar C et al (2018) KPC-2 producing ST101 Klebsiella pneumoniae from bloodstream infection in India. J Med Microbiol 67(7):927–930
Yasmeen N et al (2023) Occurrence of extended-spectrum β-lactamase harboring K. pneumoniae in various sources: a one health perspective. Front Cell Infect Microbiol 13:1103319
Wyres KL, Lam MM, Holt KE (2020) Population genomics of Klebsiella pneumoniae. Nat Rev Microbiol 18(6):344–359
Navon-Venezia S, Kondratyeva K, Carattoli A (2017) Klebsiella pneumoniae: a major worldwide source and shuttle for antibiotic resistance. FEMS Microbiol Rev 41(3):252–275
Hassan TS, Al-Oebady MAH (2024) Prevalence of plasmid-mediated quinolone resistance (PMQR) genes in Klebsiella pneumoniae isolates from Iraqi patients. in AIP Conference Proceedings. AIP Publishing LLC
Pal C et al (2015) Co-occurrence of resistance genes to antibiotics, biocides and metals reveals novel insights into their co-selection potential. BMC Genomics 16(1):964
Hoque MN et al (2022) Induction of mastitis by cow-to-mouse fecal and milk microbiota transplantation causes microbiome dysbiosis and genomic functional perturbation in mice. Anim microbiome 4(1):43
Russo TA et al (2018) Identification of biomarkers for differentiation of hypervirulent Klebsiella pneumoniae from classical K. pneumoniae. J Clin Microbiol 56(9). p. 10.1128/jcm00776 – 18
Hoque MN et al (2024) Genomic features and pathophysiological impact of a multidrug-resistant Staphylococcus warneri variant in murine mastitis. Microbes Infect 26(3):105285
Follador R et al (2016) The diversity of Klebsiella pneumoniae surface polysaccharides. Microb genomics 2(8):e000073
Xu L et al (2024) Klebsiella pneumoniae capsular polysaccharide: Mechanism in regulation of synthesis, virulence, and pathogenicity. Virulence 15(1):2439509

The authors declare potential competing interests as follows: The authors declare no conflict of interest.

FigureS1.jpg
FigureS1
FigureS2.jpg
FigureS2
FigureS3.jpg
FigureS3
FigureS4.jpg
FigureS4
SupplementaryMaterials.docx
Supplementaty Tables S1-S5
FileS1.xlsx
File S1
FileS2.xlsx
File S2
FileS3.xlsx
File S3
FileS4.xlsx
File S4
FileS5.xlsx
File S5

Download PDF

Version 1

posted

You are reading this latest preprint version

Genomic Characterization of Multidrug-Resistant Klebsiella pneumoniae Strains from Bovine Mastitis Reveals Extensive Resistome and Virulome Arsenals

Status:

Version 1

Abstract

Figures

Introduction

Methodology

Prediction of the EPS gene operons

Statistical analysis

Results

Discussion

Conclusions

Declarations

Clinical trial number

Consent for publication

Conflict of interest

Funding

Authors contributions

Data availability

References

Additional Declarations

Supplementary Files

Status:

Version 1