Introduction
Klebsiella pneumoniae is an opportunistic Gram-negative pathogen commonly found in the bloodstream, respiratory tract, and urinary tract. It can cause bloodstream infections, pneumonia, and urinary tract infections and is a frequent cause of healthcare-associated infections [
1–
4]. Over the past 3 decades, the emergence of antibiotic-resistant
K. pneumoniae has increased substantially and has been associated with more invasive disease and higher mortality rates [
3,
5].
K. pneumoniae is known to exhibit resistance to several antibiotic classes, including penicillins, cephalosporins, and quinolones [
3]. Antibiotic resistance can arise through multiple mechanisms, including point mutations in target proteins, production of β-lactamases, and multidrug resistance mediated by the overexpression of efflux pumps [
6,
7].
In Gram-negative bacteria, the resistance–nodulation–division (RND) superfamily plays a critical role in multidrug efflux systems that transport a wide range of compounds, including fatty acids, antiseptics, detergents, virulence factors, toxins, and antibiotics [
8]. The
acrAB-
tolC system belongs to the RND family and is one of the most important multidrug efflux pump systems protecting Gram-negative bacteria against toxic compounds [
8,
9]. RND-type multidrug efflux pumps typically consist of 3 components: an inner membrane transporter, a periplasmic adaptor protein, and an outer membrane channel [
9]. Additionally, clustered regularly interspaced short palindromic repeats (CRISPR) loci have been reported to contribute to multidrug resistance in bacteria [
10]. Clinical isolates of
K. pneumoniae have frequently been reported to overexpress RND-type efflux pumps, particularly
acrAB and
tolC, which substantially contribute to multidrug resistance [
8,
9]. Such overexpression may increase the risk of severe clinical outcomes, including higher morbidity and mortality [
11]. Consequently, the increasing global prevalence of antimicrobial-resistant
K. pneumoniae has led the World Health Organization to classify this organism as a critical priority pathogen.
Given the urgency of this issue, generating data on the presence and genetic characteristics of efflux pump genes is essential for improving our understanding of resistance patterns and supporting antimicrobial stewardship strategies. However, molecular data from low- and middle-income regions, including Indonesia, remain limited, particularly regarding efflux-mediated resistance mechanisms. These gaps highlight the need to characterize efflux pumps, especially through genomic profiling of
tolC diversity in
K. pneumoniae isolates.
TolC is the major outer membrane efflux channel in Enterobacterales and facilitates the extrusion of a wide range of molecules exported by bacteria [
12,
13].
To the best of our knowledge, this study represents the first report of genomic characterization of the tolC efflux pump gene in clinical K. pneumoniae isolates from Indonesia. These findings provide region-specific molecular data on multidrug resistance mechanisms.
Materials and Methods
Study Population
A cross-sectional study was conducted from June to November 2025. A total of 70 clinical specimens from patients diagnosed with K. pneumoniae infection at Dr. Soedarso Hospital in Pontianak, West Kalimantan, Indonesia, were collected during the study period. Multiple specimens from the same patient, subcultures not consistent with the phenotypic characteristics of K. pneumoniae, and isolates negative on molecular detection were excluded. Of these, 62 isolates were included in the final data analysis.
Specimens
Samples were collected from various sources, including blood, pus, tissue swabs, tissue, sputum, central venous catheters, feces, endotracheal tubes, pharyngeal swabs, and bronchoalveolar lavage, depending on the participant’s diagnosis. All specimens were placed in sterile containers and transported to the laboratory at 4 °C within a maximum of 1 hour. The samples were processed within 2 hours of collection [
14].
Phenotypic Identification
Specimens were inoculated onto blood agar (Merck, USA) and MacConkey agar (Merck, USA) and then incubated at 37 °C for 18 hours. Single colonies grown on blood agar or MacConkey agar were characterized by Gram staining to determine bacterial morphology and Gram reaction. Gram-negative rods were subsequently subjected to species identification using the BD Phoenix system (Becton Dickinson, Canada). A 0.5 McFarland bacterial suspension was prepared and inoculated into the BD identification panel according to the manufacturer’s protocol, followed by incubation at 37 °C for 18 hours. K. pneumoniae isolates with an identification probability of 80% or higher were included in the study.
Antibiotic Susceptibility Test
A 0.5 McFarland bacterial suspension was inoculated into the BD Antibiotic Susceptibility Test CPO panel and analyzed using EpiCenter ver. 6.61A (BD Diagnostic Systems). The antibiotics tested represented 11 antibiotic classes and included ampicillin, cefazolin, colistin, gentamicin, amikacin, cefepime, ertapenem, imipenem, meropenem, sulfamethoxazole-trimethoprim, ceftazidime, cefotaxime, minocycline, piperacillin-tazobactam, tigecycline, ampicillin-sulbactam, and ceftazidime-avibactam. In addition to classifying isolates as susceptible, intermediate, or resistant, the system also identified extended-spectrum β-lactamase (ESBL) production and carbapenem-resistant Enterobacterales (CRE).
Efflux Pump Gene Detection and Sequencing
Identification of the
acrAB and
tolC genes was performed using polymerase chain reaction (PCR). The
khe gene was used as a housekeeping gene for
K. pneumoniae.
K. pneumoniae isolates were processed for DNA extraction according to the manufacturer’s protocol (Presto Mini gDNA Bacteria Kit; GeneAid). PCR conditions were as follows: initial denaturation at 94 °C for 5 minutes, followed by 40 cycles for
khe and
tolC and 35 cycles for
acrAB. Each cycle consisted of denaturation at 94 °C for 1 minute, annealing for 30 seconds (
acrAB, 58 °C;
tolC, 51 °C; khe, 55 °C), and extension at 72 °C for 60 seconds. Final extension was performed at 72 °C for 10 minutes for
khe and
tolC and for 7 minutes for
acrAB. PCR products were subjected to 3% agarose gel electrophoresis and visualized using a gel documentation system. Primers used to detect efflux pump and housekeeping genes are listed in
Table 1 [
15,
16]. All isolates positive for efflux pump genes were subjected to Sanger sequencing. Sequencing results were analyzed against reference sequences from
K. pneumoniae ATCC 4386 (Gene Accession CP064352),
K. pneumoniae KP 52.145 (Gene Accession FO834906), and
K. pneumoniae HS21886 (Gene Accession YP_005228876.1).
Data Analysis
Descriptive statistical analysis was performed to summarize specimen sources, antibiotic resistance patterns, and the presence of efflux pump-related genes. Variables were presented as frequencies and percentages. Data were entered into Microsoft Excel ver. 16.91 (Microsoft Corp.), and the results were presented in tabular form. Phylogenetic analysis and visualization were performed using MEGA ver. 12 for Macintosh and Interactive Tree of Life (iTOL).
Ethics Approval
The study was approved by the Health Research Ethics Committee of Doctor Soedarso Regional General Hospital, Pontianak, Indonesia (No: 36/RSUD/KEPK/IV/2025).
Results
A total of 62 clinical isolates of
K. pneumoniae were analyzed in this study from 70 patients clinically diagnosed with
K. pneumoniae infection. The most common specimen source was sputum (19.4%), whereas tissue and pharyngeal swab specimens were the least common sources (1.6% each) (
Table 2).
Phenotypic Detection of Antibiotic Resistance Characteristics
This study showed that 41 of 62 clinical isolates (66.1%) were classified as ESBL producers, whereas 4 of 62 (6.5%) were classified as CRE. An additional 4 isolates (6.5%) were identified as both ESBL and CRE, and 13 isolates (21.0%) were classified as non-resistant (
Table 3). Antimicrobial resistance patterns were further characterized on the basis of antibiotic susceptibility results, as shown in
Table 3. We found that 98.4% (61/62) of isolates were resistant to ampicillin. In addition, high proportions of isolates were resistant to cefazolin (90.3%, 55/62, ceftazidime (88.7%, 55/62), ceftriaxone (85.5%, 53/62), ciprofloxacin (83.9%, 52/62), and cefepime (83.9%, 52/62).
Efflux Pump Gene Detection and Sequencing
All 62 isolates (100%) were positive for the
acrAB operon and
tolC gene. Isolates positive for
tolC were subsequently subjected to Sanger sequencing and analyzed using phylogenetic methods. The circular phylogenetic tree illustrates the genetic relatedness of the
tolC gene among
K. pneumoniae clinical isolates (labeled KP-xT) in comparison with the reference strains KP ATCC 43816, KP 52.145, and KP HS11286 (
Figure 1). The phylogenetic tree demonstrated that the clinical isolates clustered into distinct clades. One isolate (KP-2T) showed similarity to the
tolC gene sequence of the hypervirulent KP 52.145 strain, whereas the remaining clinical isolates formed separate clusters, suggesting possible local genetic variation.
Discussion
Antibiotic Susceptibility Test
In this study, all
K. pneumoniae isolates were obtained at Dr. Soedarso Hospital, Pontianak, from a diverse range of clinical specimen types. The antimicrobial resistance patterns observed in these isolates underscore the substantial burden posed by multidrug-resistant
K. pneumoniae in the hospital setting. Nearly all isolates were resistant to ampicillin (98.4%), which is expected given the intrinsic resistance of
K. pneumoniae mediated by chromosomal sulfhydryl variable (SHV) β-lactamase [
17,
18]. High resistance rates across first-, third-, and fourth-generation cephalosporins, ranging from 83.9% to 90.3%, further suggest a widespread presence of ESBL-producing isolates. Similar findings have been reported globally, particularly in Southeast Asia, where the prevalence of ESBL-producing organisms continues to rise [
19,
20].
In this study, carbapenem resistance, although less common than cephalosporin resistance, remains a significant clinical concern. Resistance to ertapenem (50.8%, 31/61), meropenem (33.9%, 21/62), and imipenem (29.0%, 18/62) may reflect the circulation of carbapenemase-producing strains or synergistic mechanisms involving ESBL production combined with porin loss. The relatively higher rate of ertapenem resistance is characteristic of porin-associated changes, whereas resistance to imipenem and meropenem may more strongly suggest carbapenemase activity. Ertapenem monoresistance, defined as resistance to ertapenem while retaining susceptibility to imipenem and meropenem, is a common phenotype in strains producing ESBLs or AmpC enzymes in combination with porin mutations, often in the absence of a carbapenemase [
21,
22]. These findings are consistent with regional surveillance data showing increasing carbapenem non-susceptibility among Enterobacterales in Indonesia [
23]. Other reports have also highlighted a high proportion (74.6%) of carbapenem-resistant isolates among non-Enterobacterales, suggesting a broader global increase in carbapenem-resistant bacteria [
24].
The β-lactam/β-lactamase inhibitor combinations tested showed variable activity. Resistance to ampicillin-sulbactam was high (77.4%), consistent with the limited efficacy of this agent against ESBL- and AmpC-producing strains. Resistance to piperacillin-tazobactam was lower (30.65%), although its clinical effectiveness in severe ESBL-associated infections remains uncertain, as highlighted by evidence from the MERINO trial demonstrating inferior outcomes compared with carbapenems [
25]. Ceftazidime-avibactam showed greater activity, with 43.4% resistance, which is consistent with its retained efficacy against KPC- and OXA-48-producing strains; however, the presence of metallo-β-lactamases may explain resistance in nearly half of the isolates [
26].
Resistance to non-β-lactam agents was also substantial. Fluoroquinolone resistance was high (83.9%), and trimethoprim-sulfamethoxazole resistance was also considerable (63.9%), limiting the utility of these agents for empirical therapy. These trends mirror global reports attributing fluoroquinolone resistance to plasmid-mediated
qnr genes and chromosomal
gyrA/
parC mutations [
27]. Aminoglycosides showed variable activity, with gentamicin resistance at 51.6%, whereas amikacin remained relatively active, with a resistance rate of 11.3%, consistent with its stability against most aminoglycoside-modifying enzymes [
28]. These findings support amikacin as a valuable treatment option for multidrug-resistant
K. pneumoniae infections.
Colistin resistance remained low (8.1%), indicating retained susceptibility despite global concerns regarding plasmid-mediated
mcr genes [
29]. Although colistin remains a last-resort therapy for carbapenem-resistant strains, the relatively low resistance observed in this study suggests that it may still have clinical utility, provided that it is used cautiously to minimize nephrotoxicity and limit further resistance emergence.
Overall, the resistance profile identified in this study reflects the broader challenges posed by multidrug-resistant K. pneumoniae worldwide. The high prevalence of cephalosporin and fluoroquinolone resistance, together with notable carbapenem resistance, substantially limits available therapeutic options.
Genetic Profiling
Our study showed that all 62 clinical isolates were positive for the
acrAB operon and
tolC gene. The detection rates of the
acrAB operon and tolC in this study suggest that the
acrAB-
tolC system is a dominant efflux mechanism associated with multidrug resistance. These findings are consistent with our previous study and with recent reports showing high detection rates of these genes in clinical isolates [
7,
8,
30].
acrAB and
tolC are part of the RND superfamily, which is associated with antibiotic resistance and also contributes to environmental adaptation, pathogenicity, and biofilm formation [
12]. Notably, isolates harboring these genes may export several classes of antibiotics, including β-lactams, macrolides, fluoroquinolones, and tetracyclines [
31]. This observation is consistent with our findings, in which most
K. pneumoniae isolates showed high resistance to β-lactams and fluoroquinolones (
Table 3). However, gene expression analysis was not performed, which should be considered a limitation of this study.
The importance of this study lies in the sequencing of tolC, which revealed genetic variation among clinical K. pneumoniae isolates. The phylogenetic tree showed several major clusters, indicating genetic diversity in the tolC gene sequences across the isolates. Sequence comparisons demonstrated nucleotide variation among isolates relative to the reference strains. However, comprehensive single-nucleotide polymorphism and mutation analyses were not performed and should be recognized as a limitation of this study.
Isolate KP-38T clustered closely with the KP ATCC 43816 reference strain, suggesting genetic similarity between them (
Figure 1). This finding may indicate that they share a relatively recent common ancestor, as reflected by the short branch length in the phylogenetic tree. KP-2T showed an identical
tolC sequence to that of the reference strain KP 52.145 (
Figure 1).
K. pneumoniae strain 52.145 originated in Indonesia and was isolated in 1935. The KP 52.145 strain is considered a hypervirulent
K. pneumoniae strain and is frequently used in comparative studies of virulence factor detection in
K. pneumoniae [
32]. The phylogenetic tree further showed that KP-2T clustered closely with the hypervirulent reference strain KP 52.145, suggesting the presence of shared genetic features potentially related to virulence. This clinical isolate was obtained from sputum and was resistant to ampicillin, ampicillin-sulbactam, cefazolin, cefepime, ceftazidime, ceftriaxone, and ciprofloxacin and was identified as an ESBL producer. Hypervirulent
K. pneumoniae strains possess plasmids encoding genetic factors that enable evasion of host immune defenses and the development of severe infections [
33]. We also compared our sequencing results with those of the KP HS21886 strain.
K. pneumoniae KP HS21886 is a multidrug-resistant strain primarily resistant to carbapenems. In our study, no clinical isolates clustered with KP HS21886 (
Figure 1), despite the detection of 4 CRE isolates. However, this finding does not exclude the possibility that some isolates may share antimicrobial resistance mechanisms with KP HS21886.
Phylogenetic analysis of
tolC among the clinical
K. pneumoniae isolates revealed substantial genetic diversity. The reference strains were distributed across different clusters. This variability may indicate that the clinical isolates included in this study were derived from multiple ancestral lineages rather than from a single clonal population. This interpretation is consistent with the known genomic plasticity of
K. pneumoniae.
K. pneumoniae is recognized as a pathogen capable of adapting to diverse environments, largely because its genomic plasticity facilitates the acquisition and dissemination of antimicrobial resistance determinants through plasmids and other mobile genetic elements [
34,
35].
Because
tolC is associated with efflux pump activity that contributes to antibiotic resistance, genetic variation in this gene may be associated with differences in phenotypic resistance profiles. In the phylogenetic tree, branch length reflects relatedness within subclusters, and longer branches may indicate greater sequence divergence in
tolC. The observed diversity supports the need for ongoing molecular surveillance to identify emerging lineages of potential clinical importance [
36].
Article information
Ethics Approval
This study was approved by the Institutional Review Board of Doctor Soedarso Regional General Hospital, Pontianak, Indonesia (No: 36/RSUD/KEPK/IV/2025) and performed in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained for publication of this study and accompanying images.
Conflicts of Interest
The authors have no conflicts of interest to declare.
Funding
This study was supported in part by a grant from the Ministry of Higher Education, Science, and Technology, Indonesia, under the Fundamental Research program (No. 115/C3/DT.05.00/PL/2025).
Availability of Data
All data generated or analyzed during this study are included in this published article. For other data, these may be requested through the corresponding author.
Authors’ Contributions
Conceptualization: MMar, SEP, DFL; Data curation: MMar, MMah, RA; Formal analysis: MMar; Funding acquisition: MMar, SEP, DFL; Investigation: MMar, DFL, SEP, MMah, RA; Methodology: MMar, MMah; Project administration: MMar; Resources: MMar, SEP, DFL, MMah, RA; Software: MMah; Supervision: DFL; Validation: SEP; Visualization: MMar; Writing–original draft: MMar; Writing–review & editing: all authors. All authors read and approved the final manuscript.
