1. Introduction

Antimicrobial resistance (AMR) in Gram-negative bacilli represents a major clinical concern, particularly in carbapenem- and polymyxin-resistant Klebsiella pneumoniae (CPR-Kp), which has been recognized by the World Health Organization as a critical priority pathogen for the development of new antibiotics [1,2]. The convergence of plasmid-borne resistance determinants (e.g., bla_KPC, bla_NDM, bla_OXA-48-like) with chromosomal alterations in porins (ompK35/ompK36) and envelope regulatory systems (e.g., PhoPQ, PmrAB) underpins multidrug-resistant phenotypes, drastically reducing therapeutic options [3].

In this context, pharmacological combinations have been explored to assess potential antimicrobial synergy and to investigate genetic responses that might partially reverse extreme resistance profiles [4]. Among β-lactams, ceftibuten (CTB) occupies a peculiar role: although its stability is limited in the presence of extended-spectrum β-lactamases (ESBLs) and AmpC enzymes, its action on cell wall targets may remodel the bacterial envelope and, theoretically, modulate permeability to amphipathic agents such as polymyxin B (PMB) [5].

Resistance to polymyxins in K. pneumoniae is mainly associated with mgrB inactivation by insertion sequences or to mutations in the phoPQ and pmrAB systems, which induce lipopolysaccharides (LPS) modifications through the arnBCADTEF and eptA operons, ultimately reducing drug affinity [6]. However, how multidrug-resistant K. pneumoniae modulates gene expression when simultaneously exposed CTB and PMB remains poorly understood [7]. Understanding these adaptive responses is essential to elucidate key mechanisms underlying cell wall synthesis, envelope remodeling, and antimicrobial resistance regulation. In this study, we aimed to evaluate the effect of CTB/PMB combination on the expression of genes involved in the regulation of cell wall synthesis and antimicrobial resistance in CPR-Kp.

2. Methodology

2.4. qRT-PCR

In order to identify the possible mechanisms of action of CTB/PMB and their role in modulating cell wall synthesis, qRT-PCR was employed to evaluate gene expression. For this purpose, total RNA from CPR-Kp cultures treated with the combination was obtained by using RNA affinity columns (PureLink™ RNA Mini Kit, Thermo Fisher Scientific, Waltham, MA), being subsequently subjected to DNA digestion (PureLink™ DNase Kit, Thermo Fisher Scientific, Waltham, MA) to avoid any interference. Subsequently, cDNA synthesis was carried out using the High-Capacity™ cDNA Reverse Transcription Kit (Thermo Fisher Scientific, Waltham, MA). qRT-PCR was performed using PowerUp™ SYBR™ Green Master Mix (TaqMan™ system, Thermo Fisher Scientific).

The analysis was performed on a StepOne Plus™ Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA), and relative gene expression was calculated using the 2^(-ΔΔCt) method, normalized to the 16S endogenous control gene, where:

mRNA expression values were expressed relative to the control group as fold change.

The 16S rRNA gene was used as the endogenous control for normalization. Its stability was not formally validated under all experimental conditions used in this study.

3. Results

The analysis of conserved domains and motifs revealed that pagP and pagL possess complete and highly conserved domains directly associated with the lipid A palmitoyltransferase PagP. AmpC and ftsL presented domains directly related to class C β-lactamases, with domain coverage across the entire protein. For dacA and dacC, the domain PRK13482 was identified; although it lacks a directly interpretable short name, it represents a specific protein within the cl44310 superfamily. It was observed that pagP and pagL proteins (PagP palmitoyl transferase and Lipid-A 3-O-deacylase, respectively) are located in the outer membrane, while AmpC is localized in the periplasmic space, and FtsL, DacA, and DacC are anchored in the cytoplasmic membrane (Fig 1).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Schematic representation of the predicted subcellular localization of proteins associated with antimicrobial resistance (A) and cell wall synthesis (B) in CPR-Kp.

The localization was inferred using bioinformatic analysis (PSORTb v3.0). This figure is provided for illustrative purposes only and does not represent experimental validation.

https://doi.org/10.1371/journal.pone.0349583.g001

Functional annotation analyses using KEGG revealed that the genes are associated with important metabolic pathways related to physiology and AMR in K. pneumoniae, with ftsI, dacC, ampC, and rtxB directly involved in general metabolic pathways (map01100); peptidoglycan biosynthesis (map00550); and β-lactam resistance (map01501). These analyses provide functional context for the selected genes but do not directly explain the transcriptional responses observed.

In contrast, ampC expression remained stable under all conditions, suggesting that β-lactamase-mediated resistance in this strain is not transcriptionally modulated by CTB/PMB exposure. Expression analysis showed that ampC remained stable across all groups after 2 and 4 h of treatment (Fig 2A). Similarly, the dacA gene, which encodes D-alanine carboxypeptidase (PBP5), gene associated with peptidoglycan synthesis, exhibited no significant changes in expression in response to the Test Items compared with the vehicle group in 2 h, however had increased expression in all treatment groups within 4 h (Fig 2B). A similar pattern was observed for the ftsL gene, a crucial component of the bacterial divisome, which has been previously associated with septum formation during cytokinesis during cytokinesis and consequently affect cell viability (Fig 2D). The dacC gene, encoding PBP6, another enzyme involved in peptidoglycan biosynthesis, displayed an increased expression in CTB/PMB 1 × MIC group after 4 h (Fig 2C).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Experimental qRT-PCR analysis of gene expression in CPR-Kp after 2 and 4 h of exposure to CTB/PMB.

The experimental groups included: CPR-Kp + vehicle (C−); CPR-Kp + CTB/PMB 0.5 × MIC; CPR-Kp + CTB/PMB 1 × MIC; CPR-Kp + CTB 0.5 × MIC; and CPR-Kp + PMB 0.5 × MIC. mRNA expression levels were normalized to the endogenous 16S gene, and relative expression was calculated using the 2^ − ΔΔCt method, expressed as fold change relative to the control group. Gene expression profiles are shown for (A) ampC, (B) dacA, (C) dacC, (D) ftsL, (E) pagL, and (F) pagP. Data represent three biological replicates. The Ct values for the 16S reference gene were 23.9 at 2 h and 22.2 at 4 h. MIC: minimum inhibitory concentration.

https://doi.org/10.1371/journal.pone.0349583.g002

Regarding genes involved in outer membrane remodeling, pagL, which encodes an enzyme that modifies LPS and has been linked to polymyxin resistance, showed significant variation in expression between CTB/PMB 1 × MIC group and the control after 4 h (Fig 2E). Conversely, pagP, which also modifies LPS and has been implicated in adaptation to polymyxin exposure, demonstrated a consistent trend toward increased transcription in all groups exposed to the Test Items after 4 h (Fig 2F).

4. Discussion

The study evaluated the effects of CTB and PMB under sub-inhibitory and inhibitory concentrations to assess modulation of key pathways, including genes associated with expression of porins, efflux systems, and envelope remodeling. Results obtained in this study showed a biphasic and time-dependent response to the antimicrobial pressure exerted by CTB and PMB. Within the first 2 h, transcriptional changes were minimal, suggesting predominance of basal homeostasis and rapid post-transcriptional adjustments predominate. However, after 4 h, clear modulation patterns emerged, characterized by the induction of dacA and ftsL in all treated groups, increased expression of dacC and pagL in the CTB/PMB 1 × MIC group, and pagP was consistently upregulated. This finding showed that the transcriptional adaptation of resistant K. pneumoniae requires sustained exposure, corroborating previous studies describing the temporal plasticity of bacterial responses to antimicrobial combinations [9,10].

The in silico analyses provided the structural and functional foundation to interpret these findings. The predicted subcellular localization of pagP and pagL proteins in the outer membrane reinforces their role in lipid A modification, an essential component of LPS. In this study, the increased expression of pagP and pagL after 4 h suggests activation of transcriptional responses associated with outer membrane remodeling under CTB/PMB exposure. This pattern is consistent with previously described roles of these genes in lipid A modification, although no direct functional assessment was performed, similar mechanisms have been described in Pseudomonas aeruginosa [11] and Escherichia coli [12]. Although not directly investigated in the present study, the late induction of these genes may indicate activation of LPS remodeling pathways, potentially driven by regulatory systems such as PhoPQ/PmrAB, which has been described as a key pathway in the adaptation of K. pneumoniae to polymyxin pressure [6]. These analyses provide functional context for the selected genes but do not directly explain the transcriptional responses observed.

Complementarily, genes associated with peptidoglycan synthesis and remodeling also corroborated the bioinformatic predictions. Specifically, dacA and dacC, linked to the cl44310 superfamily and to cell wall biosynthesis pathways in KEGG, exhibited increased expression after 4 hours, with dacC showing particular induction in the CTB/PMB 1 × MIC group. Likewise, the upregulation of dacA, dacC, and ftsL after 4 h suggests transcriptional modulation of genes involved in cell wall synthesis and division under antimicrobial pressure. This response may reflect an adaptive adjustment to envelope stress [13], although the functional consequences of these changes were not assessed in the present study. Notably, the increased expression of ftsL further supports the involvement of cell division-associated pathways during CTB/PMB exposure. However, the absence of analysis of central divisome regulators, such as ftsZ, precludes more definitive conclusions regarding divisome dynamics [14]. Considering that CTB exerts its antibacterial effect by targeting penicillin-binding proteins involved in septal peptidoglycan cross-linking, the observed transcriptional response may be associated with cellular processes directly affected by CTB in K. pneumoniae [15]. These findings warrant further investigation to clarify whether this late transcriptional induction represents a compensatory stress response, altered divisome regulation, or a broader envelope adaptation mechanism under combined CTB/PMB exposure.

In the case of ampC, the data demonstrated expression stability at both 2 h and 4 h, regardless of the treatment condition. This result indicates that resistance mediated by class C β-lactamase is already established in the isolate and is not transcriptionally modulated by CTB/PMB, in line with reports of constitutive AmpC-mediated resistance [16]. This stability contrasts with the plasticity observed in cell wall and LPS-related genes, reinforcing that the main adaptive response under combination pressure involves structural mechanisms rather than β-lactamase amplification.

Functional annotation results from KEGG further highlighted the overlap among cell wall biosynthesis, β-lactam resistance, and LPS remodeling pathways. This convergence demonstrates the interdependence of structural maintenance, resistance mechanisms, and virulence factors, an aspect previously emphasized in integrative genomic and transcriptomic analyses of K. pneumoniae [17]. Moreover, recent studies have pointed out that the complexity of resistance in multidrug-resistant Gram-negative pathogens requires the exploration of new β-lactam/inhibitor combinations [18] and the development of innovative approaches that consider the adaptive plasticity of the bacterial envelope [19].

The KEGG-based functional annotation supports the biological relevance of the selected genes; however, these data provide contextual information and do not directly explain the transcriptional responses observed in this study.

This study has several limitations that should be considered when interpreting the findings. First, the analysis was conducted using a single K. pneumoniae strain, which restricts the extent to which these results can be generalized to other resistant isolates or strains belonging to different sequence types. Second, only two antibiotic-exposure time points were evaluated, limiting the ability to capture broader temporal dynamics and potentially transient transcriptional responses.

In addition, 16S rRNA was used as the sole endogenous control for qRT-PCR normalization, although its stability was not formally validated under all antibiotic-stress conditions examined here. This aspect should be taken into account when interpreting the normalized transcriptional data. Moreover, changes in transcript levels do not necessarily translate into corresponding alterations in protein abundance or activity, since bacterial adaptive responses may also be shaped by post-transcriptional and post-translational regulatory mechanisms. Finally, relevant physiological parameters, including bacterial viability, membrane integrity, morphological alterations, and lipid A remodeling, were not investigated and warrant further study to better contextualize the transcriptional patterns observed.

In conclusion, combined exposure to CTB and PMB was associated with transcriptional changes in multidrug-resistant K. pneumoniae, particularly in genes involved in cell wall synthesis and lipopolysaccharide modification, while ampC remained stable. These findings suggest activation of adaptive responses under antimicrobial pressure and provide preliminary molecular data for future studies on the mechanisms of this combination.

Supporting information

Acknowledgments

The authors thank the Center for Innovation and Pre-Clinical Studies (CIEnP) for the technical support and for conducting the gene expression analyses, which were essential for the development of this study.