Ambroxol hydrochloride as an antibiofilm agent synergizes with tetracycline antibiotics against mature biofilms of multidrug-resistant Klebsiella pneumoniae

Abstract

Multidrug-resistant Klebsiella pneumoniae (MDR-KP) is a major pathogen responsible for hospital-acquired infections, associated with high morbidity and mortality. Biofilm formation plays a key role in the pathogenicity of MDR-KP and contributes significantly to its antibiotic resistance, substantially impairing the effectiveness of antimicrobial therapies. To enhance the efficacy of existing antibiotics, this study investigates a biofilm-targeting synergistic strategy inspired by the structural similarity between sputum and biofilm matrices. In this study, 87 clinical isolates of MDR-KP were initially screened for biofilm-forming capacity, and strong biofilm producers were selected to establish an in vitro model for systematic evaluation of the anti-biofilm efficacy of six mucolytic agents. Ambroxol hydrochloride (ABH) emerges as the optimal effective, disrupting biofilm structure at 0.7 mg/mL and achieving 50 % clearance within 8 h. ABH enhanced the anti-biofilm activity of tetracycline and doxycycline in vitro, reducing their IC₅₀ values by 98.9 % and 98.6 %, respectively, against preformed biofilms of MDR-KP compared to monotherapy. Additionally, the excellent physical and chemical compatibility between ABH and tetracycline or doxycycline provides a stable basis for in vivo co-administration. In vivo, the combination alleviates pulmonary inflammation, reduces bacterial load and inflammatory factor levels, and shows no tissue toxicity. In conclusion, ABH combined with tetracycline antimicrobials enhanced their efficacy against MDR-KP infections, especially biofilm-associated infections, in both in vitro and in vivo models, and possessed a favorable physicochemical compatibility and safety profile. These findings suggested that ABH-tetracycline therapy could represent a translationally promising and effective strategy for combating clinical MDR-KP infections.

Highlights

• •ABH effectively dispersed preformed biofilms of MDR-KP.
• •ABH and tetracyclines synergistically inhibited established MDR-KP biofilms in vitro.
• •ABH combined with tetracyclines synergizes against MDR-KP lung infection in vivo.
• •Repositioning of available drugs against MDR-KP.

1. Introduction

Klebsiella pneumoniae (K. pneumoniae), a gram-negative opportunistic pathogen, has emerged as a major cause of hospital-acquired infections, leading to serious clinical conditions such as respiratory, urinary tract, and bloodstream infections [1,2]. The widespread overuse of antibiotics has led to the global dissemination of multidrug-resistant K. pneumoniae. MDR-KP poses a significant public health threat due to its extensive drug resistance and high mortality rates. This threat is largely attributed to MDR-KP's ability to form structurally stable biofilms on medical devices and respiratory mucosal surfaces, which further enhances its pathogenicity and resistance to antimicrobial agents [3]. Biofilms are composed of three-dimensional aggregates formed by extracellular DNA (eDNA), proteins, polysaccharides, lipids, and other extracellular polymeric substances (EPSs), which contribute to the structural stability and drug resistance of the biofilm [4]. Within the biofilm, bacteria reduce antibiotic susceptibility and increase drug tolerance through a low metabolic state and a permeability barrier formed by the extracellular matrix [5]. Notably, biofilm-embedded bacteria are reported to be up to 1000 times more resistant to antimicrobial agents than their planktonic cells [6].

Bacterial biofilm-associated infections affect millions of people worldwide and pose a serious threat to public health. In recent years, various innovative therapeutic strategies have been proposed, including antibiotic combinations [[7], [8], [9]], antimicrobial peptides [[10], [11], [12]], nano-delivery systems [13], natural organic compounds [14,15], and phage therapy [[16], [17], [18]]. These approaches primarily aim to enhance antibiotic penetration and disrupt the structural integrity of biofilms by targeting key components such as extracellular polysaccharides, proteins, and eDNA. For example, the enzymatic degradation of polysaccharide matrices can compromise the physical barrier of the biofilm, thereby destabilizing the structure, promoting antibiotic diffusion, and improving therapeutic efficacy [15]. Therefore, identifying potentiators capable of disrupting biofilm architecture represents a promising and feasible strategy to enhance antimicrobial activity while reducing resistance.

Sputum, a biofluid secreted by the mammalian respiratory tract, lines the surface of the respiratory mucosa [19]. Although composed primarily of water (∼95 %), it contains various macromolecules that confer distinct physical properties. In pathological states, however, additional constituents, such as DNA, proteoglycans, non-mucin proteins, lipids, and cellular debris, significantly contribute to the solid phase [20]. These components form a hydrogel-like structure through intermolecular interactions and cross-linking [21,22], a structural feature that closely resembles that of bacterial biofilms [23].

Based on the mechanism of action of mucolytic expectorant drugs, the hypothesis was proposed that this class of agents may disrupt the structural integrity of MDR-KP biofilms by depolymerizing the extracellular matrix, thereby enhancing the penetration and antimicrobial efficacy of conventional antibiotics. To verify this hypothesis, a systematic experimental protocol was designed (Scheme 1): (1) A standardized in vitro MDR-KP biofilm model was established, and expectorant drugs with significant anti-biofilm activity were screened. (2) The synergistic effects of expectorants and antibiotics against planktonic and biofilm-associated cells were evaluated in vitro using checkerboard assays. (3) A preliminary evaluation of the compatibility and pharmacokinetic characteristics of the combination therapy was conducted to inform its in vivo dosing strategy. (4) A murine MDR-KP lung infection model was used to validate the in vivo antimicrobial potentiation of the combination therapy. This study aimed to identify an anti-biofilm agent capable of disrupting established biofilms, thereby weakening the physical barrier of the biofilm structure and enhancing the efficacy of existing antibiotics against MDR-KP biofilm-associated infections. The approach may offer a feasible basis for rapidly developing improved combination therapeutic strategies in clinical settings.

Scheme 1.

Schematic illustration of ABH synergizing with tetracycline antibiotics against preformed biofilms of Klebsiella pneumoniae in vitro and in vivo, and determining the dosing regimen of the two in combination.

2. Materials and methods

2.1. Strains, reagents, and animals

All 87 MDR-KP isolates were provided by the cooperating institutions and were isolated from animal sources in Huixian and Zhoukou, Henan Province, and from clinical specimens (blood, sputum, alveolar lavage) at the First Affiliated Hospital of Zhengzhou University. Isolates were confirmed by 16S rRNA PCR and sequencing. Escherichia coli ATCC 25922 served as the quality control. Strains were recovered from −80 °C stocks and cultured at 37 °C on LB broth or MacConkey agar.

Antibiotics (tetracycline(TCY), doxycycline(DOX), minocycline(MNO), tigecycline(TGC), polymyxin E(POL), meropenem(MEM), amikacin(AMK), cefotaxime(CTX)) and six mucolytic agents (N-acetylcysteine, ABH, DNase I, d-limonene, Ödostein, α-chymotrypsin) were sourced from commercial suppliers (McLean, Puxitang, Hynes) and prepared as sterile aqueous stock solutions stored at −20 °C. SPF-grade Kunming mice(KM) (∼24 g) from Hunan Slake Jinda Laboratory Animal Co. were acclimated for 7 days and housed at the Southwest University Pharmaceutical Animal Experiment Center. All animal experimental protocols in this study were approved by the Experimental Animal Ethics Review Committee of Southwest University (Approval No IACUC-20250421-02) and were subject to its supervision.

2.2. Biofilm formation and quantification

The optimal biofilm cultivation conditions were adapted from the method described by Stepanovic et al. [24] to suit Klebsiella pneumoniae. Briefly, biofilms were formed in sterile, flat-bottom 96-well polystyrene plates. A single colony was inoculated into Tryptic Soy Broth (TSB) and incubated at 37 °C with shaking for 12 h. The bacterial suspension was then inoculated into fresh TSB (without additional glucose or NaCl) in 96-well plates at a final concentration of 10⁷ CFU/mL. Each sample was prepared in six replicates and incubated statically at 37 °C for 24 h (The optimized results for the optimal conditions for biofilm formation were shown in Fig. S1). Wells containing sterile TSB without bacterial inoculation served as the negative control.

After 24 h of static incubation at 37 °C, the 96-well plates were washed with 200 μL sterile Phosphate-Buffered Saline (PBS), fixed with 200 μL methanol for 15 min, dried, and stained with 150 μL 0.1 % crystal violet for 30 min. After rinsing with ultrapure water and drying, biofilms were observed under an inverted microscope. Bound dye was solubilized in 150 μL 33 % acetic acid for 10 min, and OD₅₈₀ was measured. Biofilm formation was semi-quantitatively classified using the cutoff value (OD_C), defined as the mean OD₅₈₀ of negative controls plus three times their standard deviation: no biofilm (OD₅₈₀ ≤ OD_C); weak (OD_C < OD₅₈₀ ≤ 2 × OD_C); moderate (2 × OD_C < OD₅₈₀ ≤ 4 × OD_C); and strong (OD₅₈₀ > 4 × OD_C).

2.3. Polymerase chain reaction (PCR)

PCR amplification, combined with Sanger sequencing, was used to verify the major resistance genes carried by the MDR-KP 2126 strain. The primer sequences were shown in Table S1, and the PCR amplification procedure was shown in Table S2. The amplified products were sequenced and confirmed by nucleic acid sequence comparison through the NCBI database.

2.4. Confocal laser scanning microscopy (CLSM) analysis

Biofilms were cultured in confocal dishes with ABH at final concentrations of 0, 0.7, or 1.0 mg/mL, as previously described. After incubation, biofilms were washed with PBS and stained using SYTO 9 and propidium iodide (PI) following the LIVE/DEAD® Biofilm kit protocol (Thermo Fisher, USA). After 30 min of staining in the dark, excess dye was removed, and samples were imaged using a SpinSR confocal laser scanning microscope (Olympus, Japan) equipped with a 10 × oil immersion objective lens (NA 0.4) for high-resolution imaging. Fluorescence signals were captured at Ex/Em = 488/518 nm (green) and 561/656 nm (red). Following channel splitting, background subtraction, and threshold adjustment in ImageJ, regions of interest were selected for measurement of the mean fluorescence intensity in each channel. The intensity of yellow fluorescence, representing colocalization of SYTO 9 and PI, was then quantified based on their spectral overlap. After subtracting the yellow fluorescence signal, the ratio of red (PI) to green (SYTO 9) fluorescence intensity was defined as the dead/live ratio to evaluate the extent of cellular damage under different ABH treatment concentrations.

2.5. Determination of minimum inhibitory concentration of planktonic cell(MIC) and biofilms(MIC-B)

MICs against planktonic K. pneumoniae were determined by CLSI-recommended broth microdilution. A 0.5 McFarland bacterial suspension was diluted 1:1000, and 100 μL was combined with 100 μL Mueller–Hinton Broth (MHB) containing serial two-fold dilutions of each drug (tigecycline: 64–0.03 μg/mL; other antibiotics: 256–0.125 μg/mL) or ABH (4–0.00195 mg/mL). Positive control wells contained 200 μL of cell suspension without drug; negative control wells contained sterile MHB. After 20 h at 37 °C, MICs were recorded.

For sessile bacteria, biofilms were formed for 24 h, washed three times with PBS to remove planktonic cells, and 200 μL TSB with drug (tigecycline: 64–0.03 μg/mL; other antibiotics: 256–0.125 μg/mL) or ABH (4–0.00195 mg/mL) dilutions was added. Following 12 h incubation at 37 °C with antibiotics or ABH, 90 μL from each well was transferred to a black 96-well plate and mixed with 10 μL of resazurin solution (15 μg/mL). After 30 min at 37 °C in the dark, fluorescence (Ex 530 nm/Em 590 nm) was measured. The undosed biofilm control defined 100 % viability, and the MIC-B was the lowest concentration inhibiting ≥80 % activity.

2.6. Synergy testing by checkerboard assay

The checkerboard dilution method was adapted from Giovanna et al. [25] to evaluate synergy between antimicrobials and ABH against planktonic and biofilm-embedded MDR-KP. For planktonic cells, 90 μL of serial two-fold dilutions of each antibiotic were added across rows, and 90 μL of ABH dilutions down columns in a 96-well plate. Each well received 20 μL of a 10⁵ CFU bacterial suspension. Controls included wells with ABH or antibiotic alone. After 20 h at 37 °C, MICs were recorded, and the fractional inhibitory concentration index (FICI) was calculated by the following equation.

$$\text{FICI}=\left(\text{Ac}/\text{Aa}\right)+\left(\text{Bc}/\text{Ba}\right)$$

Where Ac and Bc were MICs in combination, and Aa and Ba are MICs alone. Synergy, additivity, indifference, and antagonism were defined as FICI ≤0.5, 0.5 < FICI ≤1, 1 < FICI ≤4, and FICI >4, respectively [26].

For biofilm-embedded cells, biofilms were formed for 24 h, washed with PBS, and treated in a checkerboard format with ABH and antibiotics for an additional 12 h. MIC-B values and FICI were determined as above to classify the combined effects.

In the checkerboard assay, the maximum concentration of each drug was determined based on its minimum inhibitory concentration (MIC) in the corresponding model, typically set at 4 × MIC. However, in the biofilm model, the maximum concentration of certain drugs (such as POL, CTX, and TGC) was slightly elevated above 4 × MIC-B to effectively evaluate the efficacy of combination therapy. Across both the conventional checkerboard and biofilm models, the concentration range of ABH remained consistent at 2–0.2 mg/mL.

2.7. Optimization of ABH–antibiotics administration sequence in vitro

To determine the optimal administration sequence of ABH and antibiotics, four treatment groups were designed: (1) PBS pre-treatment for 8 h plus antibiotics to continue to be treated for 12 h (1. Control); (2) PBS pre-treatment for 8 h plus simultaneous treatment of ABH and antibiotics for 12 h (2. Co-treatment); (3) ABH pre-treatment for 8 h followed by antibiotics for 12 h (3. ABH pre-treatment); (4) antibiotic treatment for 12 h followed by ABH intervention for 8 h (4. ABH post-treatment). Biofilms were formed by incubating MDR-KP 2126 for 12 h and washing with PBS to remove planktonic cells. TSB medium containing ABH (0.7 mg/mL) and antibiotic gradients was added according to the treatment schedule, with drug-free medium as a reference control (n = 6). After treatment, bacterial viability was assessed via resazurin staining. Fluorescence intensity was used to calculate relative viability, and IC₅₀ values were determined using a logistic regression model in GraphPad Prism 8.0.

2.8. Compatibility assessment

Physical compatibility of ABH combined with TCY or DOX was evaluated over 7 days at 4 °C by: (i) visual inspection for color change, turbidity, or precipitation; (ii) pH change (>5 % deemed incompatible); and (iii) detection of visible particles.

For chemical compatibility, HPLC methods for ABH, TCY, and DOX were validated (specificity, linearity, LOQ, recovery, precision; Table S3). Chemical stability was assessed under simulated storage (saline, 25 °C, 24 h) and physiological (plasma, 37 °C, 24 h) conditions. In plasma, ABH was combined with TCY or DOX at 40 μg/mL, then protein-precipitated and diluted to 10 μg/mL; in saline, each drug was prepared at 10 μg/mL. Controls consisted of each individual drug prepared separately in blank plasma or saline. After 24 h incubation, residual drug concentrations were measured by HPLC to determine stability.

2.9. Pharmacokinetic evaluation of ABH combined with tetracyclines

Female KM mice (∼24 g) were randomly divided into three groups (n = 6) and received tail vein injections of ABH (25 mg/kg), TCY (40 mg/kg), or DOX (15 mg/kg) after 12 h fasting. Blood samples (∼200 μL) were collected via retro-orbital plexus at scheduled time points, centrifuged to obtain plasma, and treated with methanol for protein precipitation. After further centrifugation and filtration, drug concentrations were measured by HPLC. Sampling times (h) were: ABH: 0.083–3; TCY: 0.083–4; DOX: 0.083–7. Plasma levels were calculated from peak areas using standard curves.

2.10. Murine model of MDR-KP pneumonia for evaluating ABH-antibiotic combination efficacy and tissue pathology

Female KM mice (∼24 g) were randomly assigned (n = 6 per group) to six groups: high-dose antimicrobial (HA), therapeutic-dose antimicrobial (TA), ABH only (ABH), combined treatment (TA + ABH), model control (Model), and blank control (Control). Specifically, the doses administered were as follows: the HA group received 90 mg/kg TCY or 30 mg/kg DOX; the TA group received 30 mg/kg TCY or 15 mg/kg DOX; the ABH group received 10 mg/kg ABH; and the TA + ABH group received a combination of 10 mg/kg ABH with either 30 mg/kg TCY or 15 mg/kg DOX. Under isoflurane anesthesia, mice received 0.1 mL of a 10⁷ CFU/mL bacterial suspension intranasally (or saline for Control and Model) and were held inverted for 30 s to ensure inhalation. Twenty-four hours later, treatments (antibiotics and/or ABH; doses in Table S4) were administered once daily by tail vein injection for three days; controls received saline. Three days later, mice were euthanized by cervical dislocation.

Lungs were excised, visually examined, and weighed to calculate lung coefficient

by the following equation:

$$\mathrm{L}\text{ung}\text{coefficient}=\text{lung}\text{wet}\text{weight}\left[\mathrm{g}\right]/\text{body}\text{weight}\left[\text{kg}\right]\times 100\%$$

Under sterile conditions, half of the lung tissue was excised, weighed, and placed in a sterile test tube containing 1 mL of PBS. The tissue was homogenized using a mechanical tissue grinder to form a uniform suspension. The homogenate was then serially diluted tenfold in sterile PBS, and 100 μL of the appropriate dilutions were spread evenly on MacConkey agar plates in triplicate. Plates were incubated at 37 °C for 18–24 h, and colonies were counted to calculate the bacterial load, expressed as log₁₀ CFU per gram of lung tissue.

Lung tissue was homogenized in PBS (1:9, weight/volume) and centrifuged at 12,000 rpm/min for 10 min at 4 °C. The supernatant was collected, and the levels of IL-6, IL-8, and TNF-α were determined using the following commercial ELISA kits (Shanghai Enzyme-Linked Biotech Co., Ltd., China). The experiments were performed according to the manufacturer's instructions. The absorbance was read at 450 nm using a microplate reader. Remaining lung samples were fixed in 4 % paraformaldehyde, embedded in paraffin, sectioned, and stained with H&E to evaluate inflammation and alveolar injury. Adjacent sections underwent Gram staining to assess bacterial colonization.

An additional 18 KM mice were divided into six groups: Control, TA (TCY), TA (DOX), ABH, TA (TCY) + ABH, TA (DOX) + ABH. After three consecutive days of tail vein injections, mice were euthanized, and heart, liver, spleen, lung, and kidney tissues were harvested. All tissues underwent H&E and Gram staining for histopathological assessment of drug toxicity.

2.11. Data processing and statistical analysis

Statistical analysis was conducted using IBM SPSS Statistics 23.0 and GraphPad Prism 8. Experimental data were presented as mean ± standard deviation values (Mean ± SD). The two-tailed Student's t-test was employed to compare two experimental groups, while one-way ANOVA was utilized for statistical analysis of multiple comparisons. Statistical significance was identified by ∗ p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.

3. Results

3.1. Screening of expectorants for disruptive activity against preformed biofilms of MDR-KP in vitro

All 87 K. pneumoniae isolates formed biofilms, with 48.3 % classified as strong, 20.7 % as moderate, and 31.0 % as weak producers (Fig. S2). The predominance of strong biofilm producers suggested a critical role in drug resistance and pathogenicity. MDR-KP 2126, the strongest biofilm former (Fig. 1A), was resistant to tetracyclines (TCY, DOX, MNO, TGC), carbapenems (MEM), aminoglycosides (AMK), and β-lactams (CTX), but remained sensitive to POL. PCR analysis confirmed the presence of tet(A), bla_NDM-1, rmtB, and bla_CTX-M, correlating genotype with phenotype (Fig. 1B). Consequently, MDR-KP 2126 was selected as the model strain to assess six expectorants (N-acetylcysteine, ABH, DNase I, d-limonene, Ödostein and α-chymotrypsin) for biofilm disruption. After 12 h of treatment, only N-acetylcysteine and ABH exhibited significant activity (Fig. 1C–H), with N-acetylcysteine removing ∼42 % of the biofilm at 20 mg/mL and ABH achieving ∼89 % clearance at 1.0 mg/mL (Fig. 1C and D).

Fig. 1.

In vitro MDR-KP biofilm modeling and expectorant destruction analysis. (A) Biofilm-forming capacity of 87 clinical isolates (n = 3); (B) antimicrobial resistance phenotype and corresponding resistance genes in MDR-KP 2126; (C–H) biofilm biomass remained after treatment with expectorants (N-acetylcysteine, ABH, DNase I, d-limonene, Ödostein, α-chymotrypsin; n = 6); (I) CLSM images showed live/dead bacteria distribution in biofilms (scale bar = 40 μm); (J) dead-to-live cell ratio in biofilms increased after exposure to increasing ABH concentrations (n = 3); (K–L) ABH treatment optimization: (K) concentration-dependent biofilm clearance; (L) time-dependent biofilm clearance; (M) biofilm morphology by light microscopy (crystal violet) after ABH exposure at 0, 4, 8, and 12 h (scale bar = 50 μm). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

CLSM was employed to assess ABH's impact on biofilm ultrastructure. The dead/live bacterial ratio increased from 0.21 to 3.59 as ABH concentration rose from 0 to 1 mg/mL (Fig. 1I and J). Three-dimensional reconstructions revealed that densely packed aggregates became progressively dispersed, indicating that ABH destabilized the formed biofilm and might enhance antibiotic penetration and bactericidal efficacy (Fig. 1I and J). Dose–time optimization used 50 % clearance as the threshold: ABH's minimum effective concentration was 0.7 mg/mL (Fig. 1K), achieving 50 % clearance after 8 h (Fig. 1L and M), establishing a dosing regimen for subsequent combination therapies.

3.2. ABH modulates antibiotics activity against MDR-KP planktonic and biofilm in vitro

Three strong biofilm-forming clinical isolates were used: MDR-KP 2108 (AMK), 2126 (TCY, DOX, MNO, TGC, POL, MEM), and 2186 (CTX). MIC and MIC-B values were determined under planktonic and biofilm conditions. Biofilm formation led to a 4–2048-fold increase in MIC-B values (Table S5), indicating substantial antibiotic resistance. MIC-Bs of tetracyclines (TCY, DOX, MNO) and CTX reached 256 μg/mL. Meropenem's MIC-B exceeded solubility limits, likely due to biofilm-related resistance mechanisms such as polymeric barriers, metabolic suppression, and persister cells. ABH alone showed consistent MIC and MIC-B values (0.5 and 0.9 mg/mL, respectively) across all strains but remained above clinically relevant concentrations, limiting its monotherapy potential.

Checkerboard assays were performed to assess ABH's potentiation of various antibiotics against planktonic and biofilm‐embedded MDR-KP (Fig. 2A and B). MICs and FICI values indicated that ABH enhanced the activity of several drugs, demonstrating synergistic or additive effects (Fig. 2C–F). In planktonic assays, ABH combined with tetracyclines (TCY, DOX, MNO, TGC) and CTX yielded synergy (FICI = 0.5, 0.5, 0.5, 0.375, 0.5), with the ABH + TGC pair reducing TGC's MIC eight-fold (Fig. 2C). POL and AMK combinations were additive (Fig. 2E). In contrast, ABH + MEM was antagonistic (FICI = 4.25), despite MEM's MIC decreasing four-fold (128–32 μg/mL) and ABH's MIC increasing four-fold (0.5–2 mg/mL) (data not shown).

Fig. 2.

Evaluation of ABH synergy with various antimicrobial agents against MDR-KP. (A–B) Experimental schematics for synergy assays in planktonic (A) and biofilm (B) models; (C–D) MICs of seven antimicrobials, alone or combined with ABH, in planktonic (C) and biofilm (D) conditions; (E–F) FICI values for ABH combined with seven antimicrobials in planktonic (E) and biofilm (F) assays; (G–J) checkerboard results for ABH combined with four tetracyclines (TCY, DOX, MNO, TGC) in biofilms, which showed inhibition rates of bacterial resuscitation (n = 6); ≥80 % inhibition was considered synergistic.

In the biofilm model, ABH retained synergistic activity with tetracyclines (FICI 0.389–0.403) and showed additive effects with POL, AMK, and CTX (FICI 0.681–0.860) (Fig. 2F–S3). MEM was excluded due to undefined MIC-B and antagonism in planktonic cells. Combination therapy also reduced ABH's effective concentration from 0.9 mg/mL alone to 0.13 mg/mL with TCY or TGC, and 0.25 mg/mL with DOX or MNO (Fig. 2G–J). Notably, ABH + TGC lowered TGC's MIC-B to 2 μg/mL, matching its planktonic susceptibility threshold (Fig. 2D), suggesting that ABH disrupted biofilm integrity and enhanced antibiotic penetration to kill persister cells.

3.3. Optimization of in vivo co-administration strategy for ABH and tetracyclines via multidimensional assessment

To achieve optimal synergistic antibacterial efficacy between ABH and tetracyclines, we evaluated multiple aspects including administration sequence, physicochemical compatibility, and pharmacokinetic characteristics to determine a feasible in vivo dosing regimen.

First, four treatment regimens were compared to evaluate antibiotic inhibition against MDR-KP biofilms under different dosing sequences. Under co-treatment, the IC₅₀ values of TCY, DOX, MNO, and TGC (0.63, 0.56, 0.46, 0.04 μg/mL) were markedly lower than those following ABH pre-treatment or post-treatment and the control groups (55.55, 39.90, 17.15, 1.28 μg/mL) (Fig. 3A–H). Co-treatment reduced IC₅₀ values by 97.0–98.9 % compared to control, with the greatest decreases for TCY and DOX (>98 %). These results demonstrated that simultaneous administration produced the optimum synergistic effect and that the timing of drug exposure was a critical determinant of antibacterial efficacy.

Fig. 3.

Optimization of in vivo dosing regimens. (A–D) Effects of TCY, DOX, MNO and TGC on bacterial viability within the biofilm under different dosing sequences; (E–H) IC₅₀ values of the antimicrobial drugs under the corresponding conditions, respectively; the four dosing strategies included 1. Control (antimicrobial drug administered alone); 2. Co-treatment (ABH and antibiotic administered simultaneously); 3. ABH pre-treatment (ABH pre-treatment followed by addition of antibiotic); 4. ABH post-treatment (antibiotic treatment followed by addition of ABH); (I–J) physical compatibility of compounded ABH–TCY (I) and ABH–DOX (J) injections (n = 3); (K–L) chemical stability of ABH combined with TCY (K) or DOX (L) in saline and mouse plasma (n = 3); (M–N) blood concentration-time profiles of ABH with TCY (M) or DOX (N) following simultaneous intravenous administration (n = 3).

Then, to evaluate formulation feasibility, physical compatibility assays revealed that both ABH–TCY and ABH–DOX mixtures exhibited no visible color changes, turbidity, or precipitation throughout the observation period. pH variations remained within ±5 % of baseline values, ranging from 2.59 to 2.71 for ABH–TCY and 2.52–2.63 for ABH–DOX. No visible particulates were detected, confirming good physical compatibility (Fig. 3I and J). In addition, chemical stability studies demonstrated that residual concentrations of ABH, TCY, and DOX remained above 80 % in plasma and above 90 % in saline solutions during the observation period (Fig. 3K and L). These results supported the chemical stability of ABH when combined with either TCY or DOX, further confirming the feasibility of co-administration via intravenous injection.

Following this, validated HPLC methods for ABH, TCY, and DOX exhibited a promising specificity (Fig. S4) and linearity (Fig. S5). Limits of detection were 0.08, 0.04, and 0.04 μg/mL, respectively. Recoveries exceeded 80 %, and RSDs were below 5 % (Table S6). Intra and inter-day precisions in mouse plasma showed RSDs below 6 % (Table S7), meeting bioanalytical requirements. Finally, in pharmacokinetic studies involving simultaneous intravenous administration, elimination half-lives (t_1/2) of ABH, TCY, and DOX were 0.95, 0.84, and 1.19 h, respectively (Fig. 3M and N, and Table S8), indicating comparable elimination kinetics. This temporal overlap following co-administration supported the synchronized delivery of ABH with TCY or DOX to enhance antibiotic tissue penetration and maximize antimicrobial efficacy.

3.4. Establishment and evaluation of the MDR-KP lung infection model

The lung tissues of mice in the model group exhibited marked congestion and edema, with focal hemorrhage and necrosis (Fig. 4A). The lung coefficient increased significantly to 25.7 % ± 3.1 %, nearly four-fold higher than that of the blank group (6.4 % ± 0.1 %) (Fig. 4B). Bacterial load analysis revealed approximately 10⁶ CFU/g in lung tissue, representing a 10⁶-flod increase compared with the blank group (Fig. 4C and D). Histopathological examination showed widened alveolar septa accompanied by extensive inflammatory cell infiltration (Fig. 4E). Gram staining confirmed dense colonization by short, rod-shaped gram-negative bacteria within the lung tissue, consistent with MDR-KP infection (Fig. 4F). These findings collectively confirmed the successful establishment of the MDR-KP infection model.

Fig. 4.

Establishment and evaluation of MDR-KP 2126 infection model. (A) Gross morphology of lung tissue; (B) lung coefficient analysis; (C–D) bacterial colony count in lung tissue; (E) histopathological assessment by H&E staining; (F) gram staining results of lung tissue (E and F scale bars: 100 μm).

3.5. Assessment of the efficacy and safety of ABH-TCY, ABH-DOX combination therapy in a murine MDR-KP infection model

Lung tissue phenotypes (Fig. S6A and S6B) revealed that TA (TCY, DOX) and ABH monotherapy groups exhibited pronounced pulmonary congestion and hemorrhagic foci, whereas TA–ABH and high-dose antimicrobials groups(HA) maintained near-normal morphology comparable to controls. Correspondingly, lung coefficients in TA–ABH groups were reduced by 37.5 % (TCY) and 38.6 % (DOX) relative to TA monotherapy (P < 0.001) (P < 0.001), with no significant difference compared to HA or control groups (P > 0.05) (Fig. 5B and D). In line with these morphological improvements, bacterial loads in the lungs of TA–ABH-treated mice decreased by 10²–10⁴ CFU compared to the TA, ABH, and Model groups (P < 0.001), reaching levels similar to those observed in the HA group (P > 0.05) (Fig. 5C and E; S6C, S6D), while no bacterial growth was detected in the control group.

Fig. 5.

In vivo evaluation of the therapeutic efficacy of ABH combined with TCY or DOX against MDR-KP infection. (A) A schematic diagram illustrated the MDR-KP 2126-infected mouse model and treatment regimens; (B, D) lung coefficients following treatment with ABH + TCY (B) or ABH + DOX (D); (C, E) bacterial loads in lung tissues after treatment with ABH + TCY (C) or ABH + DOX (E); (H, I) colony counts from lung homogenates following ABH + TCY (H) or ABH + DOX (I) treatment; (J–L) levels of IL-6, IL-8, and TNF-α in lung tissues after ABH + TCY treatment.; (M–O) levels of IL-6, IL-8, and TNF-α in lung tissues after ABH + DOX treatment; (P, Q) H&E-stained lung sections following ABH + TCY (P) or ABH + DOX (Q) treatment; (R, S) gram-stained lung sections following ABH + TCY (R) or ABH + DOX (S) treatment; (P–S scale bars: 100 μm).

Furthermore, the combined treatment significantly mitigated pulmonary inflammation. Levels of inflammatory cytokines (IL-6, IL-8, TNF-α) were elevated in TA, ABH, and Model groups compared to controls (P < 0.05), but were significantly lower in TA–ABH: IL-6 (3.78 ± 0.32 pg/mL TCY; 2.45 ± 0.46 pg/mL DOX), IL-8 (68.73 ± 4.43 pg/mL TCY; 82.27 ± 9.74 pg/mL DOX), TNF-α (33.10 ± 3.68 pg/mL TCY; 30.75 ± 3.37 pg/mL DOX), aligning with HA and controls (P > 0.05) (Fig. 5J–L, 5M − O). Histological analysis further supported these findings. H&E staining showed that lungs from TA, ABH, and Model groups displayed inflammatory cell infiltration, interstitial congestion, and alveolar distortion. In contrast, TA–ABH-treated lungs retained structurally intact alveolar walls with minimal exudation and inflammation (Fig. 5P and Q). Gram staining confirmed substantial bacterial colonization in the TA and ABH groups, which was markedly reduced in TA–ABH-treated lungs, again resembling the HA and control groups.

Toxicology results (Fig. S7) showed no histopathological abnormalities in heart, liver, spleen, lung, or kidney tissues among ABH, TA (TCY/DOX), and TA–ABH groups compared to controls. All groups exhibited orderly cellular architecture without signs of inflammatory infiltration, hemorrhage, or necrosis, confirming that the combination regimen presented no apparent toxicity at the tested doses.

4. Discussion

MDR-KP often harbors multiple resistance genes, including bla_CTX-M, bla_NDM-1, rmtB, and tet(A), compromising conventional therapy. These strains frequently form biofilms that hinder antibiotic penetration and immune clearance, further enhancing resistance [1,27]. This study explores biofilm-targeted combination strategies as a potential approach to overcome these therapeutic challenges.

In this study, six expectorants exhibited varying abilities to disrupt MDR-KP biofilms, likely due to differences in their molecular targets. While all interfere with mucus integrity, their mechanisms diverge: N-acetylcysteine and Ödostein break the disulfide bonds of mucin in sputum [28], DNase I cleaves extracellular DNA [29], d-limonene targets lipid components [30], and α-chymotrypsin hydrolyzes peptide bonds [31]. Notably, ABH, traditionally classified as a mucokinetic rather than a purely mucolytic agent, enhanced mucociliary clearance primarily by stimulating surfactant secretion and improving ciliary activity [32,33]. While some studies described ABH as having mucolytic effects through the degradation of acidic mucopolysaccharide fibers and modification of sputum glycoprotein structure [34], these actions remained indirect and distinct from the chemical dissolution of mucus or EPS. This distinction was important when considering its role in biofilm disruption. Beyond its established mucokinetic activity, growing evidence suggested that ABH exerted broader effects on bacterial physiology and biofilm dynamics.

Li et al. [35] demonstrated that ABH altered the stability of Pseudomonas aeruginosa biofilms via modulation of quorum sensing and alginate gene expression, while Abdelaziz et al. [36] reported its inhibition of Staphylococcus aureus biofilm formation and suppression of efflux pump activity. Similarly, Lu et al. showed that ABH reduced biofilm biomass and interfered with multidrug resistance mechanisms without directly degrading the biofilm matrix [37]. More recently, Sui et al. showed that ABH acted synergistically with bacteriophages, significantly enhancing the clearance of Klebsiella pneumoniae biofilms by reducing bacterial adhesion, increasing susceptibility to host immunity, and disrupting the biofilm microenvironment and extracellular matrix [16]. In summary, these studies indicated that ABH compromised biofilm stability through multiple mechanisms, including: (i) enhancing host mucociliary clearance; (ii) disrupting bacterial quorum-sensing networks; (iii) modulating gene expression related to biofilm formation; and (iv) attenuating antibiotic resistance mediated by efflux pumps. These synergistic effects accounted for the superior performance of ABH observed in this study compared to other expectorants, both in disrupting biofilms and in potentiating antibiotic efficacy. Further mechanistic investigations, especially those incorporating transcriptomic and metabolomic approaches, were considered essential to clarify the precise pathways through which ABH interacted with biofilm-forming pathogens.

Here, the disrupting activity of ABH against preformed biofilms of MDR-KP was investigated, aiming to evaluate its potential as an antibiotic adjuvant to enhance the efficacy of existing antibiotics against the strong biofilm-forming capacity of MDR-KP. While enhancing the efficacy of antimicrobial agents, the effective concentration of ABH required for biofilm disruption was markedly reduced in combination therapy, from 0.9 mg/mL as a monotherapy to 0.13 mg/mL when combined with TCY or TGC, and 0.25 mg/mL when combined with DOX or MNO. These concentrations were substantially lower than those reported in previous studies. For instance, Pulcrano et al. [25] demonstrated that 7.5 mg/mL of ABH could reduce voriconazole resistance in Candida albicans biofilms, while Zhang et al. [38] showed that 1.875 mg/mL of ABH, combined with vancomycin, was effective against catheter-associated Staphylococcus epidermidis biofilms in both in vitro and in vivo models.

Compared to tetracycline antibiotics, ABH exhibited antagonistic effects when combined with meropenem, while demonstrating synergistic or additive interactions with polymyxin E, amikacin, and cefotaxime sodium, albeit with relatively modest synergistic strength. These variations suggested that even among MDR-KP strains with similar genetic backgrounds, the combinatory effects of ABH differ depending on the mechanism of action of the partnered antibiotic. For instance, in strains harboring resistance genes such as bla_CTX-M, bla_NDM-1, and rmtB, ABH showed limited ability to potentiate the activity of the associated antibiotics. Conversely, in strains carrying efflux pump genes like tet(A), which render tetracycline efficacy more dependent on intracellular drug accumulation, ABH may enhance local antibiotic effectiveness by disrupting the structural integrity of the biofilm, thereby improving drug penetration [39,40]. Although ABH does not significantly affect the expression of β-lactamase or methyltransferase genes, it has been reported to downregulate efflux pump gene expression [25]. Interestingly, other studies have observed that higher concentrations of ABH may upregulate these genes [41], possibly due to species- or strain-specific differences. Nonetheless, the precise mechanisms underlying the synergism between ABH and tetracyclines warrant further investigation.

ABH has been widely used in the treatment of respiratory diseases since its first approval in 1979. As the active metabolite of bromhexine, ABH exhibited superior pharmacokinetic properties, efficacy, and safety compared to its parent compound [42]. Both toxicological studies and clinical experience confirmed its favorable safety profile [43,44]. In recent years, to restore the susceptibility of clinical pathogens to tetracyclines, numerous combination strategies have been proposed. For instance, Faure et al. [45] reported that an iron chelator (CP762) enhanced the antibacterial activity of all tetracyclines, except minocycline, by iron chelation preserves the binding of tetracyclines to the bacterial ribosome. Although these novel agents showed promising synergistic effects, their safety and drug-forming properties remained to be fully established, thereby limiting their immediate clinical applicability. In this context, the present study demonstrated that ABH had good physicochemical compatibility and a pharmacokinetic profile comparable to that of tetracyclines. As both ABH and tetracyclines were mature, marketed drugs, their combination offered not only high drug development potential but also a clear path toward clinical translation. Moreover, it was noteworthy that ABH exhibited excellent pulmonary tissue permeability [33], which further supported its application as a practical component of combination therapy for MDR-KP-induced pulmonary infections. This study suggested the repositioning of tetracycline antibiotics and ABH.

Although this study highlighted the promising potential of ABH as an anti-biofilm agent and an adjuvant to tetracycline antibiotics, further experimental validation was required to determine whether its mechanism of action was as hypothesized in this paper, i.e., that it affects the characteristics and integrity of the biofilm matrix and the expression level of efflux pumps. Additionally, models that more closely resembled in vivo or clinical conditions were needed to further assess the therapeutic efficacy and to better elucidate the translational value and clinical potential of this strategy.

5. Conclusion

In summary, this study identified the expectorant ABH as a promising agent with significant activity in disrupting the established biofilm matrix of MDR-KP. In vitro experiments demonstrated that ABH could enhance the susceptibility of MDR-KP to tetracycline antibiotics, especially biofilm-embedded bacteria. Further investigations revealed favorable physicochemical compatibility and safety when ABH was combined with tetracyclines. In an in vivo infection model, ABH synergistically improved the therapeutic efficacy of tetracyclines against MDR-KP-induced lung infections. These findings supported the potential of ABH as a practical adjunct to tetracyclines against MDR-KP infections and suggested the potential for repositioning both ABH and tetracycline antibiotics.

CRediT authorship contribution statement

TengLi Zhang: Writing – original draft, Visualization, Data curation. XunQin Gao: Validation, Methodology, Investigation. MengTing Liu: Validation, Investigation. Chun Wen: Validation, Conceptualization. Peng Jin: Investigation. Hong Yao: Writing – review & editing. XiWang Liu: Writing – review & editing. YingLan Yu: Writing – review & editing, Funding acquisition. Hao Shao: Writing – review & editing. Lei Luo: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization.

Ethical approval

All animal experimental protocols in this study were approved by the Experimental Animal Ethics Review Committee of Southwest University (Approval No. IACUC-20250421-02) and were subject to its supervision.

Funding sources

This research was financially supported by the National Key Research and Development Program of China (2021YFD1800900), Sichuan Science and Technology Program (MZGC20240020), Chongqing Science and Technology Commission (CSTB2023NSCQ-JQX0002, CSTB2024NSCQ-MSX0547, CSTB2024TIAD-STX0038), Special Fund for Youth Team of Southwest University (SWU-XJLJ202306).

Competing interests statement

The authors declare that they have no competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article.

Data availability

The raw data supporting the findings of this study are not publicly available due to the policies and confidentiality agreements within our laboratory.