Dual‐Functional Antibiotic Adjuvants Based on Antibacterial Polyesters With Guanidinium and Boronic Acid Motifs

ABSTRACT

This study addresses the challenge of antimicrobial resistance in Gram‐negative bacteria by developing a novel class of biodegradable polymeric antibiotic adjuvants. We have engineered cationic polyesters functionalized with guanidinium and phenylboronic acid (PBA) groups through the one‐pot thiol‐Michael addition post‐modification to create a dual‐targeting mechanism against the bacterial outer membrane (OM). This design combines the membrane‐disruptive action of guanidinium with the specific binding capability of PBA to lipopolysaccharides. The relationships between structure of the polymers and their antibacterial ability or cytocompatibility were explored. The optimized polymer, G95‐BA5, exhibited potent intrinsic bactericidal activity and good biocompatibility. It also demonstrated remarkable in vitro synergy with rifampicin against multidrug‐resistant pathogens, achieving fractional inhibitory concentration indices as low as 0.06 by significantly enhancing the outer membrane permeability. In a lethal murine peritonitis model, the combination of G95‐BA5 and rifampicin provided 100% survival, drastically reduced bacterial burden in vital organs, and controlled systemic inflammation. This work establishes an effective polymer platform for overcoming the OM permeability‐based resistance and revitalizing existing antibiotics.

Dual‐functional antibacterial polyesters with pendent guanidinium and phenylboronic acid groups were efficiently synthesized by the facile one‐pot procedure based on the thiol‐Michael addition post‐modification. Benefiting from the membrane‐disruptive action of guanidinium and the specific binding capability of phenylboronic acid to lipopolysaccharides, these cationic polyesters exhibit remarkable antibacterial synergy with rifampicin against a variety of clinically isolated Gram‐negative pathogens.

1. Introduction

The escalating prevalence of antimicrobial‐resistant Gram‐negative bacteria including Klebsiella pneumoniae, Acinetobacter baumannii, and Pseudomonas aeruginosa, poses a severe global public health crisis, characterized by high mortality rates and increasingly limited therapeutic options [1, 2]. The resilience of these pathogens is largely attributable to their unique outer membrane (OM) architecture [3, 4, 5]. This formidable barrier composed of lipopolysaccharide (LPS) exhibits a highly negatively charged and densely hydrophobic surface that effectively excludes a broad spectrum of antimicrobial agents [5, 6]. Crucially, it impedes the penetration of hydrophobic antibiotics such as rifampicin, thereby establishing a robust foundation for intrinsic multidrug resistance [5]. Furthermore, antibacterial efficacy is substantially undermined by the activity of efflux pump systems and the production of antibiotic‐degrading enzymes [6, 7, 8]. Confronted with a stagnant pipeline for novel antibiotic discovery, the research paradigm is shifting toward synergistic antimicrobial strategies [7, 9, 10, 11, 12, 13]. These approaches leverage complementary mechanistic interactions between agents to achieve enhanced efficacy at reduced doses, offering a promising pathway to overcome antimicrobial resistance (AMR) and revitalize our therapeutic arsenal [10, 14, 15, 16, 17].

Inspired by the membrane‐disrupting mechanisms of natural cationic antimicrobial peptides (AMPs), synthetic cationic polymers have emerged as a promising class of antibiotic adjuvants or synergistic antimicrobials [18, 19, 20, 21, 22, 23, 24, 25, 26]. These polymers emulate the core function of AMPs by utilizing cationic charges to electrostatically target the anionic LPS in outer membrane of the Gram‐negative bacteria, thereby compromising its integrity and facilitating the penetration of co‐administered antibiotics [22, 27, 28]. Despite their conceptual promise, most current cationic polymers are challenged to practical applications. A primary concern is their suboptimal synergistic potency, which mainly results from over‐reliance on a single action mode of these adjuvant polymers. Many of these polymers depend predominantly on non‐specific electrostatic and hydrophobic interactions, which often prove inadequate to inflict decisive and persistent membrane damage. Consequently, the adjuvant effect that is typically quantified by the fold‐reduction in the minimum inhibitory concentration (MIC) of antibiotics usually remains marginal and demonstrates inconsistent performance across different bacterial strains or physiological conditions [19, 20, 22, 29, 30, 31, 32, 33, 34]. Moreover, this simplistic mechanism predisposes bacteria to develop adaptive resistance, for instance, by modulating their membrane surface charge [35]. Therefore, the development of next‐generation cationic polymers capable of disrupting the outer membrane through dual targeting mechanisms, rather than relying exclusively on electrostatic interaction, represents a crucial direction for achieving robust and potent synergistic antibacterial activity.

Recent years have witnessed a growing interest in exploring versatile applications of the dynamic boronic ester bonds [36, 37, 38]. Benefiting from its ability to specifically form the dynamic covalent bonds with 1,2‐ or 1,3‐diol structures, such as LPS and peptidoglycans on the external surface of bacteria, phenylboronic acid (PBA) motifs have been widely used to construct a variety of bacterial detection systems [39]. The PBA‐containing polymers, supramolecular hydrogels, or hybrid nanoparticles are also prepared for the treatment of bacterial keratitis [40, 41], combating intracellular bacterial infections [42], promoting wound healing [43, 44], and antibacterial photodynamic or photothermal therapies [45, 46, 47, 48]. Recently, several PBA‐containing cationic polymers have been reported, showing moderate antibacterial efficacy. However, these polymers are non‐degradable and the effect of the boronic acid group on their antibacterial performance is not clearly elucidated [49, 50, 51, 52].

Given the dual electrostatic and bidentate hydrogen bonding interactions with the anionic membrane components of bacteria, guanidinium group has been extensively applied to fabricate antimicrobial polymers or molecular transporters that show remarkable membrane‐disrupting or membrane‐penetrating capability [53, 54, 55, 56, 57, 58]. In addition, the guanidinium‐containing polymers or amphiphiles are reported to synergistically potentiate antibiotics against the multidrug‐resistant (MDR) Gram‐negative bacteria [59, 60, 61]. Building upon our previously established platform polymer PMDXO that was synthesized by the controlled ring‐opening polymerization of the seven‐membered lactone monomer, 3‐methylene‐1,5‐dioxepan‐2‐one (MDXO), we have recently developed the guanidinium‐functionalized degradable polyesters that show the good biocompatibility and excellent bactericidal capability against a broad spectrum of bacteria, including the clinically isolated multidrug‐resistant pathogens [62]. Herein, we have engineered a series of biodegradable antibacterial polyesters by incorporating both pendent guanidinium and PBA groups onto the PMDXO backbone through a one‐pot procedure of the thiol‐Michael addition (TMA) post‐modification (Scheme 1). This design is based on their respective action mode of the two functionalities. The guanidinium moieties provide the initial membrane attachment and subsequent membrane disturbance through electrostatic and hydrogen bonding interactions with the negatively charged bacterial envelope components, while the PBA groups enable additional specific binding through the dynamic complexation with 1,2‐ or 1,3‐diol structures in the bacterial envelope. We hypothesize that this dual‐targeting approach will generate a synergistic effect that enhances both binding avidity and membrane disturbing efficiency, thereby improving the permeation of the co‐administered hydrophobic antibiotics across the bacterial envelope, and revitalizing their activity against the MDR Gram‐negative pathogens.

SCHEME 1.

One‐pot synthesis of dual‐functionalized polyesters by the thiol‐Michael addition post‐modification of PMDXO.

2. Results and Discussion

2.1. Design and Synthesis of Polymers

The dual‐functional polymers containing both guanidinium and PBA groups were efficiently synthesized via a facile one‐pot procedure (Scheme 2). The design of these polymers was guided by two key structural criteria. First, building on our previous finding that a shorter spacer between the guanidine moiety and the polymer backbone enhances antibacterial potency while reducing hemolytic activity [62], we selected the Boc‐guanidine‐containing thiol compound T2C featuring a minimal two‐carbon spacer to modify PMDXO. Second, the acidity of PBA group is usually correlated with its binding affinity toward 1,2‐ or 1,3‐diols [63]. To explore how the binding affinity of PBA group toward LPS influence the antibacterial profiles of the dual‐functional polymers, we synthesized six thiol compounds containing different phenylboronic acid pinacol esters (BAm, m = 1–6, Scheme 2), which will be used to modify PMDXO to obtain the dual‐functional polymers with pendent phenylboronic acid of different pKa. The structures of T2C and BAm were confirmed by their ¹H NMR spectra (Figures S1–S4).

SCHEME 2.

Synthesis of Gx‐BAm polymers and the chemical structures of the thiol compounds with Boc‐guanidine group (T2C) and various phenylboronic acid pinacol esters (BAm, m = 1–6). TMA denotes the thiol‐Michael addition.

In our previous work we found that the guanidinium‐containing polyesters with a degree of polymerization (DP) of 10 or 20 and hydrophilic hydroxy end groups possess the optimal antibacterial selectivity [62]. Therefore, HG‐P20 with a DP of 20 was selected as the platform polymer, which was synthesized by the ring‐opening polymerization of MDXO using 2‐tert‐butyloxyethanol as the initiator (Figure S5 and Table S1). The functionalization of HG‐P20 was carried out via the one‐pot two‐step thiol‐Michael addition reaction by sequentially adding T2C and BAm, respectively, under a mild condition and catalyzed by 1,8‐diazabicyclo[5.4.0]undec‐7‐ene (DBU). Upon treatment with trifluoroacetic acid (TFA), the Boc‐ and tert‐butyl protection groups were removed thorouphly, furnishing the final dual‐functinal polyesters Gx‐BAm, wherein x denotes the percent content of guanidinium group in the polymers and m represents the chemical structure of phenylboronic esters (Scheme 2). The guanidinium content (x, ranging from 95 to 80 mol%) was determined by analyzing the in situ ¹H NMR spectra of the intermediate polymer (Boc‐Gx) after T2C‐functionalization (Figure S6 and Table S2). In addition, all the intermediate polymers (Boc‐Gx and Boc‐Gx‐BAm) prior to deprotection possess narrowly distributed SEC curves, indicating that the thiol‐Michael addition post‐modification exerts a negligible influence on the polymer backbone (Figure S7 and Table S3). It is noted that the pinacol protecting group could not be efficiently removed by TFA treatment as indicated by the ¹H NMR analyses of Gx‐BAm polymers in DMSO‐d ₆ (Figures S8–S13). However, the phenylboronic ester underwent spontaneous and rapid hydrolysis in aqueous buffer (pH 7.4), liberating the active phenylboronic acid moiety (Figure S14). All Gx‐BAm polymers are water‐soluble, forming clear solutions at concentrations up to 10.0 mg/mL.

2.2. Binding Affinity Test of PBA With Diol

The fluorometric assay with Alizarin Red S (ARS) as a probe was applied to qualitatively examine the binding affinity of these dual‐functional polymers with different PBA groups toward diols [63]. ARS itself exhibits weak fluorescence in a pH 7.4 buffer, but upon complexation of its catechol group with PBA the fluorescence intensity was significantly increased. As shown in Figure 1a, incubation of the G80‐BAm polymer series with ARS resulted in a great fluorescence enhancement. Under the identical condition, the fluorescence intensity follows an order of G80‐BA6> G80‐BA5> G80‐BA4> G80‐BA3> G80‐BA1> G80‐BA2. In contrast, the control polymer HG‐P20‐2C that contains only pendent guanidinium group but no PBA group did not induce such an increase in fluorescence (Figure S15). The variation in binding affinity of the polymers indicated roughly by the increment in fluorescence intensity is correlated well with the electronic effect of the substituents. It is generally accepted that the electron‐withdrawing substituents on the phenyl ring of a PBA enhance its acidity, resulting very likely in an increased binding affinity [63, 64, 65]. Furthermore, the fluorescence intensity gradually decreased with decreasing the content of PBA in the polymers, demonstrating the binding specificity of the 1,2‐diol to PBA (Figure 1b). The addition of LPS isolated from K. pneumoniae caused a concentration‐dependent decrease in fluorescence intensity of the G80‐BA5/ARS complex (Figure 1c). This competitive fluorescence quenching clearly demonstrated that the PBA groups in the polymers are able to specifically bind to the 1,2‐ or 1,3‐diol structures of bacterial LPS [50].

FIGURE 1.

Fluorescence spectra of (a) G80‐BAm polymers and (b) Gx‐BA5 polymers upon complexation with ARS. (c) Fluorescence spectra of G80‐BA5 complexed with ARS in the presence of different LPS concentrations.

2.3. Antibacterial Activity and Cytocompatibility

A clinically isolated multidrug‐resistant E. coli strain (DC2769) was used to examine the antibacterial ability of Gx‐BAm polymers. All 24 polymer variants (Table S3) demonstrated potent activity with MIC and minimum bactericidal concentration (MBC) values ranging from 1.0–4.0 and 2.0–16.0 µg/mL, respectively (Figure 2a). Notably, neither the chemical structure nor the content of PBA groups exerted a significant influence on MIC or MBC values. We speculate that the bactericidal activity of Gx‐BAm polymers, when used alone, is predominantly governed by the guanidinium moiety. For these “dual‐targeting” polymers, while the PBA group provides a specific binding to bacterial envelope, the guanidinium group acts as the primary executor of membrane disruption and lethal damage of the bacteria. This interpretation is consistent with the recent reports which revealed that the polymers with pendent PBA moiety alone, without sufficient cationic components, lack intrinsic bactericidal activity [49, 51]. It is rational to speculate that the strong membrane‐disruptive capability of the guanidinium group could effectively mask any subtle contribution of the PBA structural variations to the direct killing outcome.

FIGURE 2.

(a) MIC and MBC values of Gx‐BAm polymers. (b) Hemolysis curves of Gx‐BA1 polymers. (c) Hemolysis rate of Gx‐BAm polymers at 5000 µg/mL. (d) Cytotoxicity of Gx‐BA4 polymers to 293T cells.

The hemolytic profile of Gx‐BAm polymers was tested by using rabbit red blood cells. As shown in Figure 2b, the introduction of PBA group led to a moderate increase in hemolytic activity compared to the previously reported guanidinium‐containing polymers such as HG‐p20‐2C [62]. This observation is consistent with the well‐established structure‐activity relationship for amphiphilic cationic polymers. The increased hemolytic activity of Gx‐BAm polymers is most probably attributed to the enhancement in their overall hydrophobicity imparted by the aromatic PBA group [66, 67]. It is noted that the hemolytic profile was markedly influenced by the content of PBA but largely independent of its individual structure (Figure 2c). This suggests that hemolysis is primarily governed by the amphiphilic character of polymer conferred by the combination of cationic guanidinium moiety and hydrophobic aromatic rings, rather than by the electronic character of the substituents. Nevertheless, all Gx‐BAm polymers maintained good hemocompatibility with the hemolysis rates remaining below 40% even at a high concentration of 5000 µg/mL (Figure 2c). Furthermore, the polymers exhibited a moderate cytotoxicity toward the mammalian 293T cells with the half‐maximal inhibitory concentration (IC₅₀) values ranging from 150 to 200 µg/mL (Figure 2d and Figure S16). The IC₅₀ values were estimated based on the polymer concentration corresponding to approximately 50% cell viability (see Supporting Information for details). The IC₅₀ values of these Gx‐BAm polymers are substantially higher than their corresponding MICs, yielding a promising antibacterial selectivity index (IC₅₀/MIC) of approximately 40‐ to 150‐fold.

2.4. Potentiation of Antibiotics Against Gram‐Negative Bacteria

Rifampicin is a hydrophobic antibiotic typically ineffective against most Gram‐negative bacteria due to its poor outer membrane permeability [68, 69]. It was used as a model antibiotic to demonstrate the adjuvant potential of Gx‐BAm polymers against a variety of clinically isolated Gram‐negative bacteria. We first assessed the potentiation of rifampicin by G90‐BAm polymers against MDR K. pneumoniae (FK2452) using the checkerboard assay (Figure 3a and Figure S17). All G90‐BAm polymers demonstrated remarkable synergistic effects when they were used in combination with rifampicin. Among the tested polymers, G90‐BA4, G90‐BA5 and G90‐BA6 exhibited the more potent synergy with fractional inhibitory concentration (FIC) indices ranging from 0.06 to 0.09 (Figure 3b), outperforming most previously reported polymeric antibiotic adjuvants [20, 22, 29, 34, 68, 70]. This superior synergistic antibacterial performance is well associated with the stronger binding affinity of the corresponding PBA groups as indicated in Figure 1a. Next, we expanded the checkerboard assays covering the Gx‐BA4, Gx‐BA5 and Gx‐BA6 polymer series to clarify the effect of PBA content on the potentiation of rifampicin against MDR K. pneumoniae (Figures S17 and S18). The FIC indies summarized in Figure 3c revealed that the polymers containing 5% or 10% of PBA unit exhibited the stronger synergy, while the incorporation of more PBA units led to a reduction in synergistic effect. We also validated the broad‐spectrum synergistic antibacterial capability of the optimized polymers (G95‐BAm and G90‐BAm, where m = 4–6) combined with rifampicin against other clinically relevant MDR Gram‐negative pathogens, including E. coli, E. cloacae, A. baumannii, and P. aeruginosa (Table 1 and Figures S19–S22). G95‐BA5 and G90‐BA5 consistently demonstrated the better synergistic antibacterial effect across all the tested strains. The control polymer HG‐P20‐2C that does not contain PBA unit yielded FIC indices of 0.31 and 0.5 against MDR E. coli and P. aeruginosa, respectively, showing a slightly weaker synergy compared to the PBA‐containing dual‐functional polymers (Figure S23). The remarkable synergistic potentiation of rifampicin by the polymers as observed in the checkerboard assays prompted us to further investigate the killing kinetics of the rifampicin/G95‐BA5 combination against MDR K. pneumoniae. As shown in Figure 3d, the combination of rifampicin (1/16 x MIC) with G95‐BA5 (1/32 x MIC) achieved a reduction of more than three orders in viable bacterial counts within just 1 h of exposure, indicating the rapid and potent bactericidal activity. In stark contrast, neither rifampicin nor G95‐BA5 alone, at the same concentrations as used in the combination, exhibited any detectable bactericidal activity over a 12 h period. This dramatic enhancement in both the rate and extent of killing underscores the potency of the synergistic interaction and its potential to rapidly suppress bacterial populations, which represents a critical advantage in the treatment of severe infections.

FIGURE 3.

(a) Checkerboard broth microdilution assays. The bacterial density was measured by optical density at 600 nm. Darker regions represent the higher bacterial density. FIC indices of (b) G90‐BAm (m = 1–6) polymers and (c) the Gx‐BAm (m = 4–6) polymers in combination with rifampicin against MDR K. pneumoniae. (d) Killing kinetics of rifampicin (1.0 µg/mL), G95‐BA5 (0.5 µg/mL), and their combination against MDR K. pneumoniae.

TABLE 1.

FIC indices of Gx‐BAm (m = 4–6) combined with rifampicin against different MDR Gram‐negative bacterial strains.

Strains	G95‐BA4	G90‐BA4	G95‐BA5	G90‐BA5	G95‐BA6	G90‐BA6
K. pneumoniae	0.09	0.09	0.06	0.06	0.09	0.09
E. coli	0.28	0.19	0.14	0.12	0.19	0.19
P. aeruginosa	<0.31	<0.38	0.25	0.25	<0.31	<0.5
E. cloacae	<0.18	0.19	0.14	0.25	0.19	0.19
A. baumannii	<0.28	0.38	0.25	0.38	0.38	0.38

Based on the findings that the dual‐functional polymers derived from BA5 exhibited the stronger interaction with LPS and the superior synergistic antibacterial potency when combined with rifampicin against K. pneumoniae, we selected G95‐BA5 as a representative polymer for in‐depth investigation of its antibacterial mechanism against this pathogen. The NPN uptake assay revealed that G95‐BA5 initiated outer membrane (OM) perturbation at a concentration as low as 0.1 µg/mL and this perturbation was significantly boosted with increasing the polymer concentration to 1.0 µg/mL (Figure 4a). However, at this concentration, G95‐BA5 only induced mild cytoplasmic membrane depolarization and triggered the generation of reactive oxygen species (ROS) (Figure 4b and Figure S24). The results of propidium iodide (PI) uptake assay demonstrated minimal membrane disruption at the same polymer concentration (1.0 µg/mL), indicating that the cytoplasmic membrane remained largely intact (Figure 4c). Critically, the potent synergistic effect with rifampicin was observed at a much lower polymer concentration (0.2 µg/mL), where no significant membrane depolarization, ROS production, or cytoplasmic membrane damage was detected. Given above results, we speculate that the superior potentiation of rifampicin by G95‐BA5 is most probably due to its significant enhancing effect on the outer membrane permeability, which could facilitate the cellular uptake of rifampicin.

FIGURE 4.

Results of (a) NPN assay, (b) DiSC₃(5) assay, (c) PI assay, and (d) EtBr assay for G95‐BA5 against K. pneumoniae.

Previous studies have indicated that membrane depolarization is likely associated with the bacterial efflux pump dysfunction and ethidium bromide (EtBr) could be used as a broad efflux pump substrate to examine if an antibiotic adjuvant affect the function of bacterial efflux pump or not [26, 71, 72]. Considering that G95‐BA5 exhibited a moderate depolarization capability on the bacterial cytoplasmic membrane at MIC (8.0–16.0 µg/mL) (Figure 4b), we further explored the influence of G95‐BA5 treatment on efflux pump of K. pneumoniae using EtBr assay (Figure 4d). The cellular uptake of EtBr was dramatically facilitated by G95‐BA5 even at the sub‐inhibitory concentrations (2.0–4.0 µg/mL), which indicates that the polymer‐induced membrane depolarization may partially disrupt the bacterial efflux systems and subsequently enhance the intracellular antibiotic accumulation. However, it should be noted that this efflux pump inhibition occurred at polymer concentrations substantially higher than those required for synergistic potentiation. Therefore, while we cannot rule out a possible auxiliary contribution of efflux pump inhibition at higher adjuvant doses, it is unlikely to be the primary mechanism underlying the synergy observed at low adjuvant concentrations.

To clarify if the PBA group have a crucial contribution to the synergistic antibacterial activity of G95‐BA5 in combination with antibiotics, we carried out a series of competitive binding assays using sorbitol to selectively bind with PBA in G95‐BA5 and block its binding activity with LPS or other diol‐containing bacterial membrane components. An E. coli O157:H7 strain, which does not metabolize sorbitol, was selected to maintain a consistent solute concentration. We found that the addition of sorbitol (2.0 mg/mL) to the culture medium significantly attenuated the synergistic effect between G95‐BA5 and rifampicin, increasing the FIC index from 0.25 to 0.38 (Figure 5a,b). The NPN uptake assay revealed that the presence of sorbitol markedly reduced the polymer's ability to perturb the outer membrane (Figure 5c). Moreover, the membrane depolarization capability of G95‐BA5 was slightly retarded by sorbitol (Figure 5d), and the influx of propidium iodide (PI) that is often used as an indicator of cytoplasmic membrane damage was also dramatically reduced in the presence of sorbitol (Figure 5e). In contrast, the G95‐BA5 induced ROS generation remained largely unaffected by sorbitol (Figure S25). These results suggest that the PBA group in G95‐BA5 played a pivotal role for the enhancement of outer membrane permeability and the perturbation of cytoplasmic membrane integrity, but did not interfere with the pathways of ROS generation. This speculation was supported by the control experiments using HG‐P20‐2C that contains pendent guanidinium group only but no PBA group to treat E. coli O157:H7. As shown in Figures S26a, b, the FIC index of the HG‐P20‐2C and rifampicin combination remained unchanged upon sorbitol addition, and the outer membrane‐disrupting capability of HG‐P20‐2C was not influenced by sorbitol. In addition, the bactericidal rate of G95‐BA5 was significantly slowed by sorbitol (Figure 5f) but HG‐P20‐2C was not (Figures S26c), demonstrating that the specific biding of the PBA group with the bacterial envelope components is crucial for achieving a rapid bacterial killing (Scheme 3).

FIGURE 5.

Checkerboard assays of the G95‐BA5 and rifampicin combination against E. coli O157:H7 (a) without or (b) with sorbitol. (c) Effect of sorbitol competition on the outer membrane perturbation capacity of G95‐BA5 against E. coli O157:H7, as measured by NPN uptake. (d) Effect of sorbitol competition on the membrane depolarization level of E. coli O157:H7 induced by G95‐BA5, as measured by DiSC₃(5) fluorescent probe. (e) PI uptake assay showing membrane disruption capacity of G95‐BA5 against E. coli O157:H7 with or without sorbitol. (f) Bactericidal kinetics of G95‐BA5 against E. coli O157:H7 with or without sorbitol. The concentration of sorbitol for all above assays is 2.0 mg/mL.

SCHEME 3.

Schematic illustration of the proposed dual‐targeting synergistic mechanism of G95‐BA5 with rifampicin against Gram‐negative bacteria. The PBA moiety binds specifically to LPS diols on the outer membrane (OM), while the guanidinium groups interact with the negatively charged LPS through electrostatic attraction and hydrogen bonding. These dual interactions lead to OM permeabilization, allowing rifampicin to penetrate the OM and exert its antibacterial effect.

2.5. In Vivo Therapeutic Efficacy

The in vivo therapeutic efficacy of G95‐BA5, alone or in combination with rifampicin, was evaluated using a murine model of bacterial peritonitis. Mice were infected with a lethal clinical isolate of MDR K. pneumoniae. One hour after infection, they received a single dose of G95‐BA5 (4.0 mg/kg), rifampicin (8.0 mg/kg), a combination of G95‐BA5 and rifampicin, or PBS, and were then monitored for 6 days (Figure 6a). In the PBS‐treated control group, 89% of the mice succumbed to the infection. Treatment with G95‐BA5 alone resulted in a survival rate of 55%, while rifampicin alone led to a 66% survival rate. In contrast, the combination of G95‐BA5 and rifampicin significantly enhanced survival, achieving a 100% survival rate over the same period (Figure 6b). To further assess the in vivo antibacterial efficacy of the combination therapy, we analyzed the bacterial burden in major organs and the levels of key inflammatory cytokines (IL‐6, IL‐18, IL‐1β and TNF‐α,) in blood at 24 h post‐treatment (Figure 6c,d). The results demonstrated that the combination of G95‐BA5 and rifampicin markedly reduced the levels of these inflammatory markers compared to monotherapy or PBS control. Moreover, bacterial loads in the organs such as liver, spleen, and kidney were 1–3 orders of magnitude lower in the combination group than in those treated with G95‐BA5 or rifampicin alone. Histopathological examination (H&E staining) further confirmed the therapeutic advantage of the combination, showing no significant inflammatory infiltration or tissue damage in mice receiving both agents, in contrast to clear pathological changes in the other groups (Figure S27). These findings collectively demonstrate the potent synergistic antibacterial effect of G95‐BA5 and rifampicin in vivo. Notably, this polymer‐antibiotic combination strategy effectively lowers the required concentration of the antibiotic, thereby reducing the risk of dose‐related toxicity. This dose‐sparing effect has been widely reported in the literature on polymer‐based antibiotic adjuvants, where various polymer designs have been shown to enhance antibiotic activity against multidrug‐resistant Gram‐negative bacteria [73, 74].

FIGURE 6.

(a) Schematic of the mouse peritonitis model. Cartoons were created with BioRender. (b) Survival curves of mice with peritonitis following a single intraperitoneal injection of PBS, G95‐BA5 (4.0 mg/kg), rifampicin (RIF, 8.0 mg/kg), or their combination (n = 9). (c) Bacterial burden in major organs (peritoneal cavity, heart, liver, spleen, lung, and kidney) of the mice infected with MDR K. pneumoniae and (d) in vivo immunogenicity markers in blood. One hour post‐infection, mice were treated with PBS, G95‐BA5, rifampicin, or the combination of G95‐BA5 and rifampicin. Samples were collected 24 h post‐treatment.

3. Conclusion

We have demonstrated the design and synthesis of biodegradable antibacterial polyesters that contain both pendent guanidinium and PBA functionalities and represent a new type of attractive antibiotic adjuvants. The dual‐functional cationic polyesters were efficiently prepared from a platform polymer HG‐P20 by the facile one‐pot procedure composed of two‐step TMA post‐modifications and the subsequent deprotection. They are cytocompatible and possess potent intrinsic bactericidal capability against a MDR E. coli strain. Of importance, these dual‐functional polyesters used as antibiotic adjuvants demonstrates moderate or excellent synergistic potency, dramatically revitalizing the activity of rifampicin against a broad spectrum of clinically isolated MDR Gram‐negative pathogens. Mechanism studies reveal that the superior adjuvant activity of these polyesters is mainly attributed to their capability of enhancing the outer membrane permeability and facilitating the intracellular accumulation of the co‐administered antibiotics. Using the optimized lead polymer (G95‐BA5) as an adjuvant and rifampicin as an antibiotic, we have finally validated the exceptional therapeutic efficacy of the combination strategy in a lethal murine model of bacterial peritonitis. This work may provide a new perspective for the rational integration of different functionalities in designing next‐generation polymer antibiotic adjuvants.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: mabi70217‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding authors upon reasonable request.