Combination of 7-hydroxyflavone and colistin: antibacterial activity and mechanism against multidrug-resistant Pseudomonas aeruginosa and Acinetobacter baumannii

Abstract

Background

With the continuous escalation of resistance of Pseudomonas aeruginosa and Acinetobacter baumannii to first line antibiotics such as colistin (COL), the development of novel therapeutic strategies is urgently required.

Results

This study demonstrates for the first time that 7-hydroxyflavone (7-HF) can restore COL’s bactericidal activity against clinically resistant Pseudomonas aeruginosa and Acinetobacter baumannii, and the synergistic effect of 7-HF combined with COL was confirmed by both checkerboard and time-kill curves. The anti-biofilm effect of COL combined with 7-HF was evaluated by biofilm inhibition assay. Safety assessments, including cytotoxicity and erythrocyte hemolysis tests, indicated that both monotherapy and combination regimens were well-tolerated. In addition, the combined use of COL and 7-HF reduced the bacterial load in a mouse thigh infection model.

Conclusions

Our findings may provide a clinically safe and effective strategy for combating multidrug-resistant Pseudomonas aeruginosa and Acinetobacter baumannii infections.

Background

Pseudomonas aeruginosa (P. aeruginosa) and Acinetobacter baumannii (A. baumannii) belong to non-fermentative Gram-negative bacteria (GNB), which are the main pathogens causing chronic nosocomial infections. Although there are differences in epidemiological characteristics, both of them are closely associated with respiratory tract colonization and infection. The ability of these organisms to adapt to various environments in hospitals makes them a special concern in clinical practice and a significant threat to public health. Owing to the irrational and extensive use of antibiotics [1], P. aeruginosa and A. baumannii have developed extensive drug resistance, resulting in a critically narrow spectrum of effective therapeutic agents. The mechanisms of drug resistance include: target sites mutations, outer-membrane pore deletion, efflux pumps, and expression of β-lactamase [2]. Infections caused by multidrug-resistant bacteria complicate clinical management and increase mortality rates, highlighting the urgent need for novel antimicrobials and synergistic drug combinations.

COL serves as a last-line defense medication for treating infections induced by multidrug-resistant GNB (MDR-GNB). In clinical practice, COL is typically administered either alone or in combination with other antimicrobial agents to treat severe infections in critically ill patients [3]. However, with the massive emergence of Colistin-Resistant (COL-R) GNB, developing innovative antimicrobial strategies has become a critical priority. Among these, non-traditional antibiotics in combination with COL seems to be one of the most promising alternative strategies to break through the resistance dilemma. Previous studies have shown that flavonoids such as kaempferol [4] and myricetin [5], plant extracts such as thymol [6] and α-Terpineol [7], as well as non-steroidal anti-inflammatory drugs such as diclofenac sodium [8] and flufenacil sodium [9] can enhance the activity of COL against GNB, providing new ideas for reversing the situation of drug resistance in clinical practice.

Flavonoids, a class of polyphenols, are attracting increasing attention for their antibacterial, anti-inflammatory, antioxidant, antiviral, anticancer and other biological activities [10]. Notably, certain flavonoids can suppress multidrug-resistant bacteria, suggesting their potential as antibacterial agents or adjuvants [11]. The antibacterial mechanism of these compounds mainly includes the following aspects: inhibiting inflammation by suppressing the release of cytokines including tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6) and interleukin-1β (IL-1β); destroying the cell walls and changing cell membrane permeability; targeting de novo nucleic acid synthesis for inhibition and so on. However, the clinical development of flavonoids is difficult due to their low bioavailability [12]. The combination of flavonoids with traditional antibiotics, such as COL, has become an important method to improve drug utilization and restore drug sensitivity [13].

In this study, the antibacterial effect of the combination of 7-HF and COL, along with its mechanism of action, was investigated. The findings demonstrated that 7-HF has certain antibacterial activity. Notably, when used in combination with COL, a synergistic bactericidal effect was observed against COL resistant P. aeruginosa and A. baumannii. Further investigations revealed that the combination of 7-HF and COL was capable of inhibiting bacterial biofilm formation, disrupting its structure, and destroying the bacterial biofilm barrier. It also boosted cell membrane permeability, thereby promoting the accumulation of COL in bacterial cells; interfere with cell oxidative phosphorylation, depleting energy supply; inhibit the release of inflammatory factors and alleviate host immune damage. Thus, it enhances the bactericidal effect of COL in multiple dimensions. This finding provides a new strategy to overcome bacterial COL resistance, and also provides important support for improving the treatment of refractory infections.

Methods

Bacteria and reagents

Sixteen COL-R GNB strains—comprising 8 COL-R P. aeruginosa and 8 COL-R A. baumannii isolates—were collected from the First Affiliated Hospital of Wenzhou Medical University [14]. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry was utilized for bacterial identification. The isolates were stored at −80 °C in Luria-Bertani (LB) broth with 30% glycerol for long-term preservation. Quality control strains in this study were A. baumannii ATCC19606 and P. aeruginosa ATCC27853. 7-HF was acquired from MCE Co., LTD., and COL from Wenzhou Kangtai Biotechnology Co., LTD (Zhejiang Province) [15]. All solvents were selected and diluted in accordance with the Clinical and Laboratory Standards Institute (CLSI M100 S31 [16]) guidelines [9].

Antimicrobial susceptibility testing

Broth microdilution assay was employed to measure the minimal inhibitory concentrations (MICs) of 7-HF and COL against 16 COL-R clinical isolates [17]. Briefly, bacteria cultured overnight were adjusted to a 0.5 McFarland standard using sterile physiological saline and then further diluted at a 1:100 ratio in cation-adjusted Mueller-Hinton broth (CAMHB). A 100 µL portion of bacterial suspension was dispensed into each well in the 96-well plate preloaded with 100 µL of serially diluted drug, with the final volume per well reaching 200 µL. After incubation at 37 °C for 16–18 h, the MIC was identified as the minimum concentration of the antimicrobial agent that completely inhibited bacterial growth. The outcomes of the antimicrobial susceptibility tests were interpreted in accordance with the breakpoint criteria specified in CLSI M100-S31 (2025 version), and all experiments were conducted in triplicate.

Checkerboard experiment

The checkerboard method is a classic technique in in vitro antimicrobial susceptibility testing for evaluating the interactions between combinations of antimicrobial agents. Its core principle lies in determining the synergistic, antagonistic, additive, or indifferent effects among the agents through the cross-combination of multiple concentration gradients [18]. Based on the results of antimicrobial susceptibility testing, we employed the checkerboard method to evaluate the in vitro synergistic antibacterial efficacy of the combination of 7-HF and COL against 16 COL-R clinical isolates [19]. In brief, serial dilutions of COL and 7-HF were prepared separately and then mixed in 96-well plates. Each resulting combination contained 50 µL of each drug (COL and 7-HF) [20]. Bacteria cultured overnight were adjusted to a 0.5 McFarland standard with sterile physiological saline and then diluted 1:100 in CAMHB. A 100 µL portion of this diluted bacterial suspension was dispensed into every well, bringing the final volume per well to 200 µL. Following incubation under 37 °C for 16–20 h, the results were examined. Fractional Inhibitory Concentration (FIC) indices were computed as below: FIC (drug A) = (MIC of drug A in combination)/(MIC of drug A alone); FIC (drug B) = (MIC of drug B in combination)/(MIC of drug B alone); and FICI = FIC (drug A) + FIC (drug B). As reported in previous studies, FICI values < 0.5, 0.5–4, and > 4 indicate synergistic, indifferent, and antagonistic effects, respectively [9]. All experiments were conducted in triplicate.

Time-kill assays

The checkerboard method is centered on the static evaluation of bacteriostatic effects, whereas the time-kill assays represent a dynamic in vitro antimicrobial susceptibility testing technique [21]. This assay enables quantitative analysis of the bactericidal rate and activity of antimicrobial drugs by tracking the fluctuating changes in viable bacterial counts across various time points. According to the preliminary results of the checkerboard method, we selected 8 COL-R clinical isolates (4 strains for each bacterial species) and conducted subsequent experiments in accordance with the previously established research protocol [22] to clarify the dynamic synergistic antibacterial activity of the 7-HF and COL combination. In this study, MIC concentrations were adopted, as the MIC represents the lowest concentration of an antimicrobial agent that inhibits bacterial growth. This concentration allows for a direct visualization of the bactericidal kinetic characteristics of the agent against the target strains. As a classic concentration setting for time-kill assays, MIC can provide direct data support for evaluating the time-dependent bactericidal effects of the antimicrobial [23]. The detailed procedures are described as follows: Test groups: ① blank control; ② 7-HF 64 µg/mL; ③ COL 2 µg/mL; ④ 7-HF 64 µg/mL + COL 2 µg/mL. In brief, 30 µL of a 0.5 McFarland bacterial suspension was added to 3 mL LB broth with the indicated drug and incubated at 37 °C with shaking. Samples were collected at 0, 2, 4, 6, 12, and 24 h; 10 µL aliquots were serially diluted 10-fold in PBS and spotted onto LB agar plates. After being incubated overnight at 37 °C, the colonies were enumerated. Synergistic bactericidal activity was defined as a reduction of ≥ 2 log₁₀ CFU/mL in bacterial count in the combination group versus the most potent single-drug group within 24 h, with all experiments performed in triplicate.

Biofilm formation inhibition assay

Previous studies have confirmed that the 0.5 MIC concentration avoids the destruction of biofilm structure caused by bactericidal effects, making it more suitable for biofilm-related research [24]. Therefore, sub-MIC concentrations were prioritized as the experimental concentrations in this study. However, considering the inherent heterogeneity in the antimicrobial susceptibility phenotypes among different strains, as well as the observed fluctuations in susceptibility results of some strains during subculture and cultivation, MIC concentrations were appropriately employed for certain strains to ensure the reliability and integrity of the experimental data and conclusions [25]. The detailed methodology is described as follows: blood agar plates were inoculated with bacteria and incubated overnight. After adjusting individual plate colonies to 0.5 McFarland turbidity and diluting them 1:100 in LB broth, 100 µL of the diluted bacteria was transferred to a 96-well plate. 7-HF, COL, or their mixture—at sub-MIC/MIC concentrations from the checkerboard method—was then added to achieve 200 µL per well, followed by 24 h incubation at 37 °C. The medium was aspirated, and wells were gently rinsed twice with 200 µL PBS to remove non-adherent planktonic bacterial populations. Staining involved adding 150 µL of 1% crystal violet; after air-drying at room temperature, the 96-well plate was incubated statically at 37 °C for 15 min. The crystal violet was decanted, the plate air-dried again, and 150 µL of 95% ethanol-5% acetic acid solution added to solubilize biofilm-bound dye. Finally, OD₅₉₅ absorbance was read with a Multiskan FC microplate reader [26], with three replicates per experiment.

In vitro analysis of cytotoxicity

Human kidney cells (HK-2), purchased from Wuhan Baosai Life Technology Co., Ltd. (China), were cultured in Dulbecco’s modified Eagle’s medium (DMEM) fortified with 10% heat-inactivated fetal bovine serum (FBS). At 37 °C, the culture was maintained with 5% CO₂ until the cells reached 80% confluence. After trypsin digestion, cells were plated into 96-well plates (1 × 10⁵ cells/100 µL) for overnight adherence. Ten microliters of 7-HF serial solutions (ultimate concentrations: 4, 8, 16, 32, 64, 128 µg/mL) or 7-HF/COL combinations was added per well, followed by 24 h of incubation. CCK-8 agent (10 µL, Solarbio, Beijing) was then added; post 1 h of 37 °C dark incubation, OD₄₅₀ was read via a Multiskan FC microplate readerr [27], with three replicates per experiment.

In vitro analysis of hemolytic activity

Fresh human red blood cells (RBCs) were collected from healthy volunteers and washed three times with PBS. The RBCs were then resuspended in PBS to a final concentration of 5% (v/v). Subsequently, 500 µL of the RBC suspension was mixed with 500 µL of PBS (negative control), 0.1% Triton X-100 (positive control), 7-HF or the combination of 7-HF and COL. The mixtures were incubated at 37 °C for 2 h, then centrifuged at 3000 rpm for 5 min. The supernatant was transferred to a 96-well plate, and the absorbance at 540 nm was measured using a microplate reader [28]. All experiments were performed in triplicate.

PI experiment

Sub-MIC concentrations do not directly induce bacterial cell death, thereby avoiding the interference of bacterial lysis with experimental detection indicators. Therefore, sub-MIC concentrations were prioritized as the experimental concentrations in this study [29]. The detailed procedures are described as follows: Membrane integrity was determined using an adapted protocol referenced in prior research [30]. Bacterial suspensions with an OD₆₀₀ of 0.3–0.4 were incubated with the selected concentrations of the various drugs for 2 h, then centrifuged at 4000 rpm for 5 min. The precipitate was washed twice with PBS, and finally resuspended in 1 mL of PI solution (50 µg/mL as the final concentration), followed by incubation in a water bath for 30 min in the dark. The mixture was then centrifuged at 4000 rpm for 5 min, washed twice with PBS to remove excess dye, and finally resuspended and mixed thoroughly in 1 mL of PBS. Aliquot 200 µL of the suspension into a light-proof 96-well plate. Fluorescence intensity in cell samples was then read using a Bio Tek microplate reader (Ex 535 nm/Em 615 nm) [30]. Results were presented as relative fluorescence units (RFU), and all experiments were performed in triplicate.

ATP determination

The enhanced ATP Assay kit (Beyotime, Shanghai) was employed to measure intracellular ATP levels, as per the manufacturer’s instructions. In short, bacterial cells cultured overnight were washed and resuspended in PBS to achieve an OD₆₀₀ of 0.3–0.4. The bacterial suspending solution after treatment was exposed to the selected concentrations of the various drugs (sub-MIC concentrations were prioritized) and treated for 2 h. The bacteria were lysed by boiling followed by centrifugation, and intracellular ATP levels were determined using the supernatant. 100 µL ATP assay working solution was transferred to a 96-well plate and equilibrated at ambient temperature for 5 min. The supernatant was loaded into the well for rapid mixing and read using a Synergy H1 luminescence module [31]. All experiments were repeated three times.

Cytokine concentrations were determined by ELISA

LPS (1 µg/mL)-challenged RAW 264.7 cells, and drugs at relevant concentrations were added thereafter. Following 4 h of incubation, the cell supernatant was collected. TNF-α and IL-6 levels were measured with ELISA kits (J&L Biological, Inc.) according to the manufacturer’s protocols, and the levels were expressed in pg/mL. All experiments were performed in triplicate.

In vivo analysis of synergistic antibacterial effects in mice

To appraise the in vivo synergistic action of 7-HF and COL, a mouse thigh infection model was set up. Male ICR mice aged 5–6 weeks and weighing 22–25 g without specific-pathogen-free were used in the experiment (purchased from Zhejiang Vital River Laboratory Animal Technology Co., LTD., China).This study secured approval from the Animal Ethics Committee of the First Affiliated Hospital of Wenzhou Medical University (approval number: SYXK 2021-0017), and all mice that survived were euthanized when the experiment concluded [6]. Neutropenia was first induced by intraperitoneally administering cyclophosphamide (Shanghai Yuanye Biotechnology Co., LTD., China) at 150 mg/kg to mice for 3 consecutive days. P. aeruginosa TL7333 (a COL-R strain) was used as the model bacterium, and 100 µL of its bacterial suspension in the exponential growth phase was administered via injection into the thigh muscle of mouse’s hind limb. After two hours infection, the mice were assigned to 4 groups: ① blank control; ② COL 5 mg/kg; ③ 7-HF 20 mg/kg; ④ COL 5 mg/kg + 7-HF 20 mg/kg. After 24 h, the mice were first anesthetized using sodium pentobarbital (30 mg/kg) and then rapidly euthanized via CO₂ inhalation. Following this, cervical dislocation was carried out to ensure death. Then, thigh muscles were excised under aseptic conditions, processed to remove nerves, and weighed. Each tissue sample was homogenized with 3 mL of saline and the grinding beads via a tissue grinder. The resulting homogenate was subjected to 10-fold serial dilution. Subsequently, 10 µL aliquots of the dilutions were dropped onto LB agar plates and cultured overnight at 37 °C for subsequent colony counting [32]. Results were expressed as log₁₀ CFU/g muscle. All experiments were repeated three times.

Statistical analysis

In this study, data are presented as the mean ± standard deviation (SD) and analyzed using GraphPad Prism 8.0.2 (GraphPad Software, La Jolla, California, USA) with one-way analysis of variance (ANOVA). Differences were considered statistically significant at p < 0.05. Statistical significance in the figures is denoted by asterisks (ns, P > 0.05; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P<0.0001).

Results

7-HF can restore the susceptibility of P. aeruginosa and A. baumannii to COL

Antimicrobial susceptibility test

The findings demonstrated that 7-HF exhibited a certain degree of antibacterial efficacy against the selected COL resistant strains, but its antibacterial activity was low. None of the strains were susceptible (i.e., all were resistant) to COL, with MIC values ranging from 4 µg/mL to 32 µg/mL (Table 1).

Table 1.

Summary of MIC values and FICIs for 16 COL-R GNB clinical isolates

Species	Strains	Monotherapy MIC (µg/mL)		Combination MIC (µg/mL)		FICI	Interpretation
		COL	7-HF	COL	7-HF
Pseudomonas aeruginosa	TL7333	32	256	2	32	0.1875	Synergistic
	TL7929	32	256	2	64	0.1875	Synergistic
	TL8126	32	256	2	64	0.3125	Synergistic
	TL7548	16	256	2	64	0.375	Synergistic
	TL2314	32	256	2	32	0.1875	Synergistic
	TL1671	32	256	2	64	0.3125	Synergistic
	TL2917	8	256	2	32	0.375	Synergistic
	TL7505	16	256	2	64	0.375	Synergistic
Acinetobacter baumannii	BM2431	8	256	1	8	0.15625	Synergistic
	BM2349	4	256	1	8	0.28125	Synergistic
	BM1595	4	256	1	8	0.28125	Synergistic
	BM2370	16	256	0.5	8	0.0625	Synergistic
	BM7994	16	256	1	64	0.3125	Synergistic
	BM8014	16	256	1	64	0.3125	Synergistic
	BM7962	16	256	1	16	0.3125	Synergistic
	BM1412	32	256	1	16	0.09375	Synergistic

Checkerboard assay

The potential synergistic effect of 7-HF on the selected strains was investigated by checkerboard assay. The results demonstrated that adding 7-HF reduced the MIC of COL, indicating that 7-HF could enhance COL’s antimicrobial activity; meanwhile, the Fractional Inhibitory Concentration Index (FICI) ranged from 0.0625 to 0.375 (Table 1).

Time-kill assays

A time-kill assay with eight chosen isolates was employed to further assess the synergism between 7-HF and COL [14] (Fig. 1). The experimental concentrations were determined by the checkerboard assay (Table 1). As presented in Fig. 1, the combined treatment potently inhibited bacterial growth, resulting in a reduction of >2log₁₀ CFU/mL within 24 h, thereby confirming its bactericidal activity [20]. The upward growth trend may be associated with drug concentrations, as well as the degree and duration of the effect of drugs on different bacterial species [14, 33]. An attempt may be made to re-administer the drug after 24 h to eliminate this trend. In addition, this phenomenon may also be related to the presence of persisters (drug-sensitive dormant bacteria). Persisters themselves do not have drug resistance gene mutations, but can survive under stress conditions such as antimicrobial exposure, acidic environments, and nutrient deprivation through phenotypic adaptations like metabolic dormancy. Once the stress is eliminated, they can quickly recover and resume growth [34].

Fig. 1.

Time-kill curve results regarding COL and 7-HF, administered alone or in combination, against the selected COL-R GNB. Bacterial suspensions were incubated at 37 °C with shaking, and viable bacterial counts (log₁₀ CFU/mL) were quantified at 0, 2, 4, 6, 12, and 24 h by serial dilution and plating on LB agar

Biofilm formation-inhibiting effect of the 7-HF and COL combination

Crystal violet staining experiment

The ability of different treatments to form biofilms against selected COL resistant bacterial strains was next assessed by crystal violet staining technique. The findings showed that at 0.5×MIC concentrations determined by checkerboard method, all strains showed a decrease in absorbance at 595 nm in contrast to the control group and monotherapy group (Fig. 2). This provided further evidence that the COL and 7-HF combination potently suppressed biofilm formation in the selected bacterial strains.

Fig. 2.

0.5×MIC of COL and/or 7-HF was used to inhibit biofilm formation in selected COL-R GNB. Bacterial biofilm formation was assessed by crystal violet staining after 24 h incubation at 37 °C, with biofilm biomass quantified as OD₅₉₅ absorbance using a microplate reader. (*, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001)

Safety tests in vitro

The combination regimen also exhibited a promising safety profile. In human kidney cells—a model mammalian cell line—even a high dose of 7-HF (64 µg/mL) was well-tolerated and did not increase cell death relative to control treatment groups or dimethyl sulfoxide (DMSO) (Fig. 3), supporting its potential for safe in vitro application [7]. Furthermore, previous studies have also demonstrated that 7HF exhibits no human hepatotoxicity and has a low risk of cardiotoxicity [35].

Fig. 3.

Assessment of the cytotoxicity of COL/7-HF combination at different concentrations and DMSO on human kidney cells. Cell viability was detected by the CCK-8 assay (OD₄₅₀). (ns, P > 0.05)

To further assess the safety profile of the 7-HF and COL combination, we evaluated its hemolytic effect on red blood cells (RBCs). With PBS as the negative control and 0.1% Triton X-100 as the positive control [36], we found that neither 7-HF alone nor its combination with COL induced additional hemolysis compared to the negative control (Fig. 4). These findings supply extra in vitro evidence to confirm the safety of the combination therapy.

Fig. 4.

Hemolytic activity of 7-HF and COL across a range of concentrations. PBS served as the negative control, while 0.1% Triton X-100 served as the positive control. (ns, P > 0.05)

Mechanistic basis for synergistic antimicrobial activity

PI membrane permeability assay

To explore the mechanism of synergism between 7-HF and COL, membrane integrity was evaluated using propidium iodide (PI). PI is a cell membrane-impermeable fluorescent dye that accumulates exclusively in cells exhibiting damaged membranes, thus serving as a marker for assessing membrane integrity [31]. As shown in Fig. 5, the combination of 7-HF and COL yielded a significantly elevated fluorescence signal when compared to PBS control and monotherapy groups, indicating a greater disruption to bacterial cell membrane integrity [20] (Fig. 5).

Fig. 5.

Bacterial membrane integrity was assessed by the PI experiment. Fluorescence intensity of different groups treated with 7-HF and/or COL. (*, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001)

ATP production

We also investigated how 7-HF influences intracellular ATP levels in COL-treated bacterial cells to determine its role in the synergistic interaction [31]. As illustrated in Fig. 6, when relative to the blank control group, the use of 7-HF alone resulted in a significantly lower ATP level within bacterial cells. Likewise, the combination of 7-HF and COL led to a dramatic decline in the ATP level in bacterial cells (Fig. 6).

Fig. 6.

Intracellular ATP levels in cells following treatment using 7-HF, COL, or their combination. (**, P < 0.01; ***, P < 0.001; ****, P < 0.0001)

The 7-HF/COL combination can reduce the expression of inflammatory factors and exert antibacterial effects in vivo

Cytokine concentrations measured by ELISA

Bacterial infection can induce the production of endotoxin, which further induces the secretion of elevated cytokine levels (TNF-α and IL-6 included) in macrophages [37]. In order to further explore 7-HF and COL synergy mechanism, we use PBS as the negative control group and 1 g/mL LPS (endotoxin) as the positive group. LPS is the major element of stimulating RAW 264.7 cells produce cytokines. Then, we treated the cells with varying concentrations of 7-HF and COL as single agents or in combination. As shown in Fig. 7, the combination of 7-HF and COL reduced the intracellular expression levels of TNF-α and IL-6 relative to the control group (Fig. 7).

Fig. 7.

Quantitative analysis of inflammatory cytokines. Concentrations of TNF-α and IL-6 were determined by ELISA. PBS served as the negative control, while 1 g/mL LPS (endotoxin) served as the positive control. (*, P < 0.05; **, P < 0.01; ***, P < 0.001)

In vivo efficacy analysis

Having verified the synergistic antimicrobial activity of 7-HF and COL, we went on to evaluated the in vivo efficacy of therapy of the 7-HF and COL combination in a murine model of bacterial thigh infection. As presented in Fig. 8, at 24 h after injection, both 7-HF and COL administered alone suppressed the growth process of P. aeruginosa TL7333. In contrast, the combined intervention of COL and 7-HF produced a better efficacy than the monotherapy group [38] (Fig. 8).

Fig. 8.

Quantities of bacteria in a mouse model for thigh infection. Quantification of COL-R P. aeruginosa TL7333 in mouse thighs 24 h post-treatment with 7-HF, COL, or their combination. (**, P < 0.01; ***, P < 0.001)

Discussion

Prior research has indicated the rapid in vitro bactericidal effects of COL toward multidrug-resistant GNB (mainly P. aeruginosa and A. baumannii) [39]. However, the extensive use of COL in recent years has led to the emergence of a large quantity of COL resistant pathogens. Among them, drug-resistant bacterial strains of P. aeruginosa and A. baumannii spread rapidly [40]. This has led to a renewed emphasis on the use of COL. In this context, the repurposing of already approved drugs as antibacterial adjuvants for use with COL represents a promising approach. The previous adjuvant mechanisms mainly directly inhibited the drug resistance of bacteria, including reducing the membrane permeability of drugs, inhibiting the efflux pump in bacterial, and opposing the formation of bacterial biofilms, etc [41]. Few studies have addressed the intrinsic resistance mechanisms of bacteria, such as inflammation-mediated tissue damage, active bacterial resistance, and biofilm tolerance. Our research aims to fill these gaps by evaluating 7-HF as a novel COL adjuvant.

Here, we demonstrate for the first time the synergistic effect between 7-HF and COL. Bacterial biofilm is the root cause of chronic infection in multiple parts of the body, and it is strongly resistant to conventional antibiotics, which poses a significant challenge for eradication in clinical practice. Therefore, the development of novel anti-biofilm molecules that can efficiently destroy and eradicate bacterial biofilms has become the key to the field of anti-infection [42]. In this study, we demonstrated that the 7-HF/COL combination can significantly suppress the formation of bacterial biofilms. This finding not only provides a new drug combination paradigm for the clinical prevention and treatment of multidrug-resistant bacteria infections but also is expected to contribute to the development of preventive surface coatings for medical devices such as catheters and artificial joints. PI, an impermeable fluorescent dye that accumulates exclusively in cells with membrane damage, serves as an indicator of membrane integrity. As shown in our results, fluorescence intensity was significantly elevated when 7-HF and COL were used in combination. This further indicates that the intactness of the bacterial membrane was significantly impaired, which in turn led to an increase in cell membrane permeability. This finding could open up a new way to solve the problem of COL resistance [43]. ATP is an irreplaceable “energy currency” for bacteria, which supports almost all life activities of bacteria. Our study demonstrated that the 7-HF/COL combination reduced the intracellular ATP production, which resulted in the impaired energy expenditure, cell metabolism, growth and development of bacteria [44]. As illustrated in Figs. 6 and 7-HF played a dominant role in mediating this effect, whereas COL exerted a negligible impact on bacterial ATP levels. This phenomenon may be attributed to the fact that the primary target of COL (such as lipid A in the outer membrane [45]) is not directly involved in intracellular energy metabolism. Therefore, the ATP depletion observed in the 7-HF/COL combination group may represent a complementary outcome arising from the non-overlapping targets of the two drugs in the energy metabolic pathway. On the one hand, 7-HF-mediated ATP depletion attenuated the energy-dependent resistance mechanisms of COL-R GNB (e.g., efflux pump systems and lipid A modifications [2, 45], both of which require a continuous supply of ATP). This further enhanced the permeability of the bacterial membrane to COL. On the other hand, COL disrupts the integrity of the bacterial membrane, which facilitates the intracellular accumulation of 7-HF and enhances its inhibitory effect on energy metabolism. Therefore, synergistic bactericidal activity may be achieved without additional ATP depletion. Finally, we found that the combination of 7-HF and COL could decrease the secretion of IL-6 and TNF-α in macrophages, which further speculated that 7-HF and COL might exert a synergistic antibacterial effect by suppressing macrophage inflammatory pathways, including NF-κB or MAPK signaling [46]. In summary, 7-HF/COL combination exerts a synergistic effect through the “membrane disruption - energy depletion - inflammation inhibition” triple mechanism, counteracting the key drivers of COL resistance: membrane disruption directly bypasses the reduction in COL binding caused by LPS modification in resistant bacterial strains (a major COL resistance mechanism), and energy depletion further weakens the ability of bacteria to maintain the integrity of the outer membrane, which is an energy-dependent process in P. aeruginosa and A. baumannii. This dual targeting of membrane and energy metabolism solves the “drug entry barrier” that limits the efficacy of COL in resistant isolates. Inhibition of inflammation acts on host factors, further enhancing the antibacterial effect of COL.

Conclusion

Taken together, the in vitro-in vivo data consistently indicate that 7-HF offers a promising prospect for overcoming COL resistance through the triple mechanism of “membrane disruption, energy depletion, and inflammation inhibition” to address the prevalent infections induced by multidrug-resistant GNB, especially P. aeruginosa and A. baumannii [47]. Despite these promising results, this study has limitations. First, the murine thigh infection model is an acute induced infection model, which is primarily used to evaluate the therapeutic efficacy of antimicrobial agents [48]. Compared with clinical chronic infections, this model lacks complex factors such as host immune regulation and microbial community interactions [49]; therefore, it cannot be employed to assess the complete eradication of bacterial infections [48]. Meanwhile, typical inflammatory responses occur at the infection site in the murine thigh infection model, creating a pathological microenvironment characterized by tissue edema and cellular infiltration [49]. This impairs the diffusion and penetration of drugs into the infected focus, leading to limited therapeutic effects of the drugs [50]. Second, only a mouse thigh model was employed to assess the efficacy in vivo; further validation in other models (e.g., pneumonia or sepsis, which are more clinically relevant for P. aeruginosa and A. baumannii) is needed. Finally, the exact molecular targets of 7-HF in membrane disruption and energy metabolism remains unclear—identification of these targets will help optimize 7-HF’s structure for better efficacy. Future studies will focus on these aspects to advance 7-HF/COL as a clinical candidate for MDR-GNB infections.

Abbreviations

GNB

Gram-negative bacteria

7-HF

7-hydroxyflavone

COL

colistin

IL-6

interleukin-6

COL-R

Colistin-Resistant

IL-1β

interleukin-1β

RBCs

red blood cells

MIC

minimum inhibitory concentration

PBS

phosphate buffered saline

TNF-α

tumor necrosis factor-α

PI

propidium iodide

LB

Luria-Bertani

CAMHB

cation-adjusted Mueller-Hinton broth

CCK-8

cell counting kit-8

FICI

Fractional Inhibitory Concentration Index

ELISA

enzyme-linked immunosorbent assay

FIC

Fraction Inhibitory Concentration

DMSO

dimethyl sulfoxide

Authors’ contributions

MJW: Conceptualization, Data curation, Formal analysis, Writing – original draft. ZYZ: Funding acquisition, Investigation. FY: Methodology, Project administration. XWL: Resources, Software. ZCY: Software. DYZ: Validation, Visualization. TLZ: Project administration, Supervision, Validation, Visualization. LJC: Supervision, Validation, Visualization.

Funding

This research was granted support from the Consortium for Infection and Innovation (CII) Research Grant of the First Affiliated Hospital of Wenzhou Medical University (Project No.: 2025WMU-X001).

Data availability

All original data produced during the conduct of this study are included in the article in the form of figures and tables.

Declarations

Ethics approval and consent to participate

All participants in this study have signed the informed consent form, consenting to their participation in the study and authorizing the collection of relevant strains as well as the conduct of associated subsequent research. This study secured approval from the Animal Ethics Committee of the First Affiliated Hospital of Wenzhou Medical University (approval number: SYXK 2021-0017). Furthermore, all study procedures strictly complied with the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.