Optimized Zebrafish AP2M1A-Derived Decapeptide AP10RW with Robust Stability Suppresses Multidrug-Resistant Bacteria

TL;DR

Scientists optimized a peptide called AP10RW to create a stable and effective treatment against drug-resistant bacteria.

Contribution

The paper introduces AP10RW, a redesigned antimicrobial peptide with enhanced stability and broad-spectrum efficacy against multidrug-resistant bacteria.

Findings

AP10RW shows broad-spectrum activity against drug-sensitive and multidrug-resistant bacterial pathogens.

AP10RW is stable under various environmental stresses like serum exposure, pH changes, and high salt concentrations.

AP10RW is safe for mammalian cells with low hemolysis and cytotoxicity.

Abstract

The increasing crisis of antimicrobial resistance requires innovative therapeutic strategies that can overcome the limitations of conventional antibiotics. Based on our previous finding that AP10 (a derivative of AP29) possesses antimicrobial activity but lacks thermal stability, we rationally redesigned ten new AP10 analogues to enhance functional robustness while maintaining efficacy. Among these, AP10RW is identified as the optimal candidate due to its exceptional broad-spectrum activity against both drug-sensitive and multidrug-resistant (MDR) bacterial pathogens. Structural analysis reveals that AP10RW adopts an environmentally responsive conformation, transitioning from random coil to amphiphilic α-helix in membrane-mimicking environments, while demonstrating remarkable stability under challenges including serum exposure, varying pH, high salt concentrations, and thermal stress. Mechanistic studies indicate that AP10RW exerts its effects through multiple bactericidal mechanisms involving initial high-affinity binding to bacterial characteristic molecules (LTA, LPS and PGN), followed by rapid membrane depolarization, ultrastructural damage and the induction of lethal oxidative stress. Notably, this potent antimicrobial efficacy is coupled with exceptional biosafety, demonstrating little hemolysis and negligible cytotoxicity against mammalian cells. This systematic optimization represents a significant advancement in antimicrobial peptide engineering. We have successfully transformed a thermally unstable peptide into a robust therapeutic candidate and positioned AP10RW as a promising clinical candidate for addressing the growing threat of multidrug-resistant infections.

Figures

Full text

1. Introduction

The widespread and inappropriate use of traditional antibiotics has led to drug-resistant pathogens emerging, which poses a threat to human health. The rise of drug-resistant bacteria poses a significant worldwide public health threat, given its potential to limit the efficacy of currently available antibiotics [1]. In 2019, approximately 5 million people died from conditions linked to drug-resistant bacteria. If action is not accelerated, this figure is expected to increase significantly by 2050 [2]. The multiple drug resistance mechanisms exhibited by bacteria (e.g., utilization of efflux pumps, reduction in membrane permeability, modification of antibiotic targets, enzymatic degradation of antibiotics, and biofilm formation) compromise the efficacy of nearly all known antibiotics [3]. To address therapeutic failures, it is essential to pioneer antimicrobial agents that employ alternative mechanistic strategies [4].

Antimicrobial peptides (AMPs), which are natural defense molecules found in most living organisms, are promising candidates. They typically exhibit broad-spectrum antibacterial activity without inducing drug resistance, as they attack inherent bacterial components through non-specific mechanisms [5,6].

The amphiphilic character of AMPs is a vital aspect of how they work. Their unique structure incorporates both hydrophobic and hydrophilic amino acid residues, forming distinct regions within the sequence [7]. Hydrophobic residues, such as leucine, isoleucine, methionine, and tryptophan, are distributed among hydrophilic amino acids (including cationic amino acids) [8]. This property enables them to interact with cell membranes, especially those of microbial cells. Current research findings indicate that AMPs exhibit a variety of antimicrobial mechanisms, including disruption of membrane permeability (a primary mode of action for most cationic AMPs), targeting of the cytoplasmic membrane and intracellular targets (DNA, RNA, organelles or protein synthesis), inhibition of cell wall biosynthesis, and the induction of excessive levels of reactive oxygen species (ROS), which has the potential to further delay the development of resistance [2,9,10].

While this is a powerful approach, natural AMPs often have problems that limit their use as drugs. They can be broken down by enzymes, be harmful to human cells (like red blood cells), or lose their activity in the presence of salts or blood proteins [11]. To resolve these issues, researchers are currently developing new synthetic peptides that are more stable, less toxic, and high effective.

Our previous study showed that a 29-amino acid peptide AP29 (^1^WKIKRMAGMKESQISAEIELLPTNDKKKW^29^) of zebrafish AP-2 complex subunit mu-A (AP2M1A) bound to lipopolysaccharide (LPS) and lipoteichoic acid (LTA), leading to growth inhibition in Gram-negative (Escherichia coli) and Gram-positive (Staphylococcus aureus) bacteria [12]. The peptide is composed of 2 β-sheets (^1^WKIKRMA^7^ and ^14^ISAEIE^19^) and 2 heparin-binding motifs (^1^WKIKRM^6^ and ^25^DKKKW^29^) [12]. The N-terminal decapeptide of AP29, named AP10 (WKIKRMAGMK), retains antibacterial activity [13], and pilot experiments showed that AP10 was not functionally tolerant to high temperatures.

A series of AP10-derived decapeptides was designed and evaluated for their antibacterial activity and stability. Of these, a novel synthetic peptide named AP10RW (WKRKRWRIWK) demonstrated exceptional stability while retaining most of its antibacterial activity. We methodically tested its capacity to inhibit a range of bacteria, including drug-resistant strains. Our core finding is that AP10RW kills bacteria through a powerful combination of actions. After binding to bacterial signature molecules, it rapidly disrupts the cell membrane and also causes a lethal burst of ROS inside the cell. Significantly, it has been demonstrated that AP10RW remains highly active under challenging conditions like different salt concentrations, varied pH levels, high temperature, and even in the presence of serum, while being safe for mammalian cells. This combination of strong, multi-mechanism antibacterial activity, high stability, and low toxicity makes AP10RW an exciting candidate for further development.

2. Materials and Methods

2.1. Peptide Design and Antimicrobial Property Prediction

This study focused on the development of antimicrobial peptides exhibiting robust resistance to stress conditions. A total of ten AP10 analogues were designed, numbered 1-10. During the design process, we ensured that the amphiphilicity of the parent peptide was maintained, with charges ranging from +2 to +7 and hydrophobicity ratios ranging from 40% to 60% (Table 1). Among these, peptide number 1, named AP10RW, has great antimicrobial activity and stress resistance. In this peptide, M^6^ and M^9^ of AP10 are substituted by W, I^3^ and A^7^ by R, and G^8^ by I (WKRKRWRIWK). The Expasy ProtParam tool (https://web.expasy.org/protparam/, accessed on 3 January 2026) was employed to determine peptide molecular weight (MW) and isoelectric point (pI). Subsequently, the Allpeptide Polypeptide Parameter Calculator (https://www.allpeptide.com/canshu.html, accessed on 3 January 2026) was used to compute the total hydrophobic ratio and net charge (EMBOSS-based, at pH 7.0). Antimicrobial properties were predicted using the CAMP server (http://www.camp.bicnirrh.res.in/predict/, accessed on 3 January 2026), which integrates three classifiers: Random Forest (RF), Support Vector Machine (SVM), and Artificial Neural Network (ANN). Predictions with a probability ≥ 0.5 were classified as antimicrobial peptides (AMPs), while those with a probability < 0.5 were defined as non-AMPs (NAMPs).

2.2. Peptide Synthesis

All peptides were custom-synthesized by Shanghai Sangon Biological Engineering Technology & Services Co., Ltd. (Shanghai, China), employing the standard Fmoc solid-phase method, and used for functional examination. The peptides were C-terminally amidated to stabilize their secondary structure, as this modification is critical for maintaining the bioactive conformation [14]. Following synthesis, peptides underwent HPLC purification, achieving >95% purity, and were verified by mass spectrometry (Icms-2010a, Shimadzu, Kyoto, Japan). The purified peptides were subsequently stored at −80 °C till used. The peptides were dissolved in ultrapure water at a stock concentration of 1 mg/mL, and this stock solution was used for all subsequent experiments unless otherwise specified.

2.3. Structure Analysis

Circular dichroism (CD) spectra of AP29 and AP10RW were acquired on a Jasco J-1500 spectrometer (Jasco, Tokyo, Japan) at 25 °C using a 1 mm path length cuvette, following the method of Gong et al. [13]. Peptides were prepared at 200 μg/mL in 10 mM PBS (pH 7.4), 50% TFE, or 30 mM SDS micelles. Spectral scans from 190 to 250 nm were performed at 500 nm/min with a 1 nm bandwidth, with each data point representing the average of three values.

The online tool K2D3 (https://cbdm-01.zdv.uni-mainz.de/~andrade/k2d3/, accessed on 1 October 2025) was applied to quantify the α-helical content of the peptides [15]. We employed AlphaFold3 (https://alphafoldserver.com, accessed on 4 November 2025) to predict the peptides’ three-dimensional (3D) structures [16]. The amphiphilicity of the predicted α-helical segment in AP10RW was analyzed using a helical wheel projection. The analysis was performed online with the heliQuest tool (https://heliquest.ipmc.cnrs.fr, accessed on 3 January 2026).

2.4. Bacterial Strains and Culture

A total of eleven bacterial strains (including four MDR strains) were used in the antibacterial assays. The drug-sensitive strains used were as follows: Gram-positive bacteria—Staphylococcus aureus ATCC 25923, Micrococcus luteus ATCC 49732, Bacillus subtilis ATCC 6633, and Enterococcus faecalis ATCC 29212; and Gram-negative bacteria—Escherichia coli ATCC 25922, Vibrio anguillarum ATCC 43308, and Klebsiella oxytoca ATCC 700324. We obtained the following MDR bacterial strains: E. coli ATCC 577, Klesiella pneumoniae ATCC 2182, and S. aureus USA500 from Dr. Yashuo Wang (Qingdao University, China); and Stenotrophomonas maltophilia ATCC 13637 from Haiguang Wu (Shouyang County People’s Hospital, Shouyang, China). See Table S1 for details. For experiments, cultures were incubated in LB medium at 37 °C for 16 h to mid-log phase.

2.5. Antibacterial Activity Assay

Antibacterial activity was evaluated according to Wang et al. [17] with minor modifications. Briefly, in a sterile 96-well plate, 10 μL of bacterial suspension (1 × 10^5^ cells/mL) was combined with 90 μL of fresh LB medium and 100 μL of peptide solution per well. This produced final peptide concentrations ranging from 5 to 300 μg/mL. Plates were incubated at 37 °C for 9 h, and growth inhibition was monitored by hourly measurement of the absorbance at 600 nm using a Thermo Fisher Multiskan MK3 plate reader (Thermo Fisher Scientific, Waltham, MA, USA). Blank controls containing sterile water were processed in parallel. The minimum inhibitory concentration (MIC) was defined as the lowest concentration capable of completely inhibiting the growth of the respective bacterium. The experiments were performed in triplicate and repeated three times.

2.6. Transmission Electron Microscopy (TEM)

TEM was employed to assess AP10RW-induced bacterial cell damage, adapting the method of Gong et al. [12]. In brief, 300 μL bacterial suspensions (S. aureus or E. coli, 1 × 10^9^ cells/mL) were exposed to 200 μL of AP10RW (1× MIC) or PBS for 1 h at room temperature. Following incubation, samples were fixed for 2 h at 4 °C using 2.5% (v/v) glutaraldehyde in 100 mM phosphate buffer (pH 7.4), and then washed three times with the same buffer. Subsequently, the fixed cells were placed on 400-mesh carbon-coated copper grids for 3 min. The grids were negatively stained with 2% phosphotungstic acid (pH 7.0) for 1 min, and excess stain was blotted away with filter paper. Finally, the grids were air-dried before being examined under a JEOL JSM-840 TEM (JEOL, Tokyo, Japan).

2.7. Ligand Binding Assay

To evaluate the binding of AP10RW to LPS, LTA, and peptidoglycan (PGN), an enzyme-linked immunosorbent assay (ELISA) was employed. First, LPS, LTA, and PGN (all from Sigma-Aldrich, St. Louis, MO, USA) were individually biotinylated with biotin hydrazide (Sigma-Aldrich) as described by Zhang et al. [18]. Following this, each well of a 96-well plate was coated with 50 µL of either AP10RW or BSA (control), both at 50 µg/mL, and dried overnight at 16 °C. After blocking with 1% BSA/PBS (pH 7.4, 37 °C, 2 h) and PBST rinse, each well received 50 μL of PBS with 0.1% BSA plus biotinylated LTA (0–16 μg/mL), LPS (0–64 μg/mL), or PGN (0-64 μg/mL). Plates were incubated with these ligands at 25 °C for 3 h. Post-wash, streptavidin-HRP (CWBIO) was introduced and incubated at 25 °C for 1 h. Following five washes with PBST, 75 µL of 0.4 mg/mL O-phenylenediamine (OPD; Amresco) in development buffer (51.4 mM Na_2_HPO_4_, 24.3 mM citric acid, 0.045% H_2_O_2_, pH 5.0) was added to each well. The plates were then incubated at 37 °C for 20 min to allow the reaction. Finally, the reaction was terminated by adding 25 µL of 2 M H_2_SO_4_ per well, and the absorbance was read at 492 nm using a microplate reader.

The thermodynamic characteristics of the AP10RW-LPS interaction were investigated by isothermal titration calorimetry (ITC) using a MicroCal PEAQ-ITC instrument (Malvern, UK), following previously described methods [19,20] with modifications. We prepared LPS and AP10RW in an identical titration buffer composed of 20 mM HEPES, 150 mM NaCl (pH 7.4). LPS (50 μM) was loaded into the sample cell, and AP10RW (500 μM) was loaded into the syringe. The reference cell was filled with H_2_O. This experimental method will make a single 0.4 μL injection, followed by 24, 1.5 μL injections at 15 °C, and the stirring speed was set at 750 rpm. Data derived from a minimum of two ITC runs were subsequently analyzed via the MicroCal PEAQ-ITC (Version 1.41) analysis software, employing a single-site binding fitting model.

We then analyzed the molecular docking of the AP10RW and LPS. The initial structure for AP10RW was predicted using AlphaFold3. The LPS structure was sourced from PubChem [21] and geometrically optimized using the B3LYP functional with the 6-31G(d,p) basis set. Molecular docking was performed with GNINA 1.0 [22], which integrates convolutional neural networks (CNNs) to predict the optimal binding pose. In these simulations, AP10RW was held rigid while semi-flexible docking was conducted.

2.8. Membrane Depolarization Assay

The assay for membrane depolarization activity of AP10RW was performed with the membrane potential-sensitive dye 3,3′-dipropylthiadicarbocyanine iodide (DiSC_3_(5), Sigma-Aldrich). Following Ni et al. [23], seven strains were evaluated: four drug-sensitive bacteria (E. coli 25922, V. anguillarum 43308, S. aureus 25923, M. luteus 49732) and three MDR bacteria (E. coli 577, K. pneumoniae 2182, S. aureus USA500). Mid-log phase bacterial cells were harvested, washed in 5 mM HEPES buffer (pH 7.4) containing 20 mM glucose, and resuspended in the same buffer supplemented with 100 mM KCl to a density of 1 × 10^5^ cells/mL. For the assay, 100 µL of suspension per well was treated with DiSC_3_(5) (0.5 µM final concentration). After a 30-min incubation at room temperature to stabilize the signal, 100 µL of AP10RW solution (50 µg/mL) was added to initiate the measurement. The control consisted of HEPES buffer with 20 mM glucose. Measurements were performed by recording the fluorescence intensity every minute for 30 min at excitation/emission wavelengths of 622/670 nm. Establish a fluorescence baseline based on the control, and measure the specific response to AP10RW relative to this baseline.

2.9. Reactive Oxygen Species Assay

Intracellular ROS levels in representative bacteria, including four drug-sensitive and three MDR bacterial strains, were assessed using DCFH_2_-DA [24]. Bacterial cells at mid-logarithmic phase were adjusted to 1 × 10^7^ cells/mL in 10 µM DCFH_2_-DA and incubated for 30 min at 37 °C. Subsequently, the cells were pelleted by centrifugation (3500× g, 10 min) and subjected to three PBS washes.

Cells were treated by resuspension in AP10RW (50 μg/mL), PBS with Rosup (50 μg/mL, Beyotime; positive control), or PBS (blank control). Rusop is a compound mixture provided as a positive control in the Reactive Oxygen Species Assay Kit (Beyotime, Shanghai, China) to induce intracellular ROS generation. Following a 1-h incubation at 25 °C, fluorescence was immediately measured at excitation/emission wavelengths of 488/525 nm.

2.10. Assay for Effects of Serum, Salt, pH and Heating on Antibacterial Activity

To evaluate serum effects, suspensions of S. aureus USA500 or E. coli 577 (1 × 10^8^ cells/mL) were mixed with LB and AP10RW prepared in 25% human serum, achieving final concentrations of 1×, 1.5×, and 2× MIC, based on Maystrenko et al. [25] with adjustments. The 96-well plates with the above mixtures were incubated at 37 °C for 9 h. Bacterial growth kinetics were followed by hourly absorbance at 600 nm readings. The data represent the mean of three biological replicates, each performed in technical triplicate.

The salt tolerance of AP10RW was tested against S. aureus USA500 and E. coli 577 using LB medium and peptide solutions in 0.9% saline at 1×, 1.5×, and 2× MIC, with subsequent steps identical to the standard assay.

To examine the pH-dependent activity of AP10RW, assays were conducted against E. coli 577 or S. aureus USA500. The peptide was prepared in PBS at pH 5.5 or 8.5, mixed with bacterial suspensions and LB medium, and then processed as previously described.

To assess thermal stability, AP10RW was heated to 55 or 85 °C for 30 min [26]. After cooling, the peptide was tested for antibacterial activity as described above.

2.11. Hemolytic Activity Assay

According to Sæbø et al. [27] with minor adjustments, hemolytic assays were performed using red blood cells (RBCs). Fresh blood was collected in EDTA tubes. Subsequently, RBCs were pelleted by centrifugation (1000× g, 10 min), washed with PBS, and finally resuspended to 4% (v/v). Aliquots (200 μL) of the RBC suspension were mixed with equal volumes of AP10RW at various concentrations (ranging from 12.5 to 200 μg/mL). After incubation (37 °C, 1 h), the samples were centrifuged, and the absorbance of the supernatants was measured at 540 nm. The percentage of hemolysis was calculated using the following formula: % Hemolysis = [(OD_sample − OD_blank)/(OD_positive − OD_blank)] × 100. PBS, BSA (100 μg/mL), and 0.1% Triton X-100 served as the blank (0% hemolysis), negative, and positive (100% hemolysis) controls, respectively. According to the widely accepted criterion for in vitro hemocompatibility assessment, a hemolysis percentage below 5% is considered to indicate negligible hemolytic activity and is generally acceptable for biomaterials and therapeutic candidates.

2.12. Cell Viability

The CCK-8 assay (Beyotime Biotech, Shanghai, China) was employed to evaluate the cytotoxicity of AP10RW (0–200 μg/mL) towards HEK293T and RAW264.7 cells, based on Wu et al. [28] with modifications. Briefly, cells (5 × 10^3^/well) were seeded in 96-well plates with 10% FBS medium overnight. After replacing the medium with 2% FBS medium containing the peptide, cells were treated for 12 h. After adding CCK-8 solution and incubating for 2 h, the absorbance was read at 450 nm. Relative to PBS-treated controls (100% viable), cell viability was calculated by: (A_AP10RW_ − A_Blank_)/(A_Control_ − A_Blank_) × 100%, in which A_Blank_ is the absorbance of cell-free wells.

2.13. Statistical Analysis

All experiments were conducted with three technical replicates and repeated three times independently (biological replicates). Data are presented as means ± standard deviation (SD). Statistical analysis was performed using GraphPad Prism 5 software. Differences were assessed by Student’s t-test or one-way analysis of variance (ANOVA), with a p-value of less than 0.05 considered statistically significant.

3. Results

3.1. Characteristics of Synthesized Peptides

AP10RW was purified (>95%, HPLC; Figure S1) and verified by MS (Figure S2). Computational analysis indicated that AP10RW possesses a net charge of +6.97 and a hydrophobic ratio of 40% (Table 1). Furthermore, prediction using the CAMP server classified AP10RW as an antimicrobial peptide (AMP) in all three algorithms (Table 2). Pilot experiments suggest that AP10RW is the most stable while retaining its antimicrobial efficacy; therefore, subsequent experiments will focus on this peptide.

3.2. Peptide Structure

Analysis of the circular dichroism (CD) spectra illustrates the conformational behavior of the maternal peptide AP29 and its derivative AP10RW in different solvent environments: 10 mM PBS (aqueous buffer), 50% TFE (helix-inducing buffer), and 30 mM SDS micelles (membrane-like environment) (Figure 1A,B). In 10 mM PBS, both peptides exhibited random coil conformations, with their CD spectra displaying a characteristic negative peak at <200 nm. Additionally, AP10RW exhibits a positive peak at 225–230 nm in aqueous solution, a characteristic feature of π–π stacking, consistent with its sequence rich in tryptophan [29,30]. In 50% TFE solution, the CD spectra of AP29 and AP10RW showed double minima at approximately 208 and 222 nm, confirming the formation of a typical α-helical structure. Notably, in SDS micelles, neither AP29 nor AP10RW adopted a stable α-helical conformations, retaining largely disordered structure. The characteristic helical signals (minima at 208 and 222 nm) were absent in SDS. This lack of helical stabilization in SDS environment may arise from two interrelated factors. First, the strong electrostatic attraction between the cationic peptides and the negatively charged SDS surface inhibits deep insertion and ordering [31]. Second, the potential destabilizing effect of the central, bulky arginine residue on helical folding in an SDS-solvated context. Research on transmembrane peptides suggests that such centrally located, positively charged arginine can disrupt helical conformation and even promote alternative structures and aggregation [32]. The α-helical content calculation was performed using the K2D3 online software. The results showed that, compared to the PBS solution, the corresponding α-helical content of AP29 and AP10RW is the highest in a 50% TFE solution (Table 3), which suggests their functional conformation transitions have the capacity to occur. The α-helix content calculation formula is only directly proportional to the absolute value of the ellipticity of the residues at 222 nm [33]. AP10RW exhibits relatively low ellipticity at this wavelength (at a concentration of 200 μg/mL), resulting in a calculated value that reflects a relatively low α-helix content.

AlphaFold3 prediction of the 3D structure also indicates that AP10RW adopts an α-helical conformation. Further structural analysis revealed that the helical segment (encompassing residues Lys4 to Trp9) exhibits a facial amphiphilicity. As depicted in Figure 1D,E, the side chains of Trp6, Ile8, and Trp9 cluster to form a well-defined hydrophobic face, while those of Lys4 and Arg7 constitute a distinct polar/positively charged face. This spatially segregated arrangement is characteristic of membrane-active antimicrobial peptides and supports the observed functional properties.

These results suggest that the structures of AP29 and AP10RW are flexible, allowing them to adopt different conformations in response to varying microenvironments. The transformation of AP29 and AP10RW from random coil to α-helix provides further evidence that they function as antimicrobial peptides (AMPs).

3.3. Antibacterial Activity

We examined the antimicrobial activities of AP10RW against the drug-sensitive strains of B. subtilis 6633, E. faecalis 29212, M. luteus 49732, and S. aureus 25923 (Gram-positive bacteria), and E. coli 25922, V. anguillarum 43308, and K. oxytoca 700324 (Gram-negative bacteria).

As shown in Figure 2A–G, AP10RW exhibits good antimicrobial activity towards these strains, with the MIC values all ≤ 100 μg/mL (Figure 2L). Because of the strong antimicrobial activities of AP10RW against seven drug-sensitive strains, we were wondering if AP10RW could inhibit MDR bacteria growth. The antimicrobial activity of AP10RW was evaluated against four MDR bacterial strains (E. coli 577, K. pneumoniae 2182, S. aureus USA500, and S. maltophilia 13637) using a standard assay. Results demonstrated that AP10RW inhibited the growth of each strain (Figure 2H–K), with MIC values ≤ 100 μg/mL (Figure 2M).

3.4. AP10RW Induces Ultrastructural Damage in Bacterial Cells

To ascertain whether AP10RW causes direct physical damage to bacterial cells, we examined its effects on E. coli 25922 and S. aureus 25923 using TEM. As shown in Figure 3, treatment with AP10RW resulted in severe and distinct morphological alterations in both strains. In E. coli cells, we observed the formation of intracellular voids and increased electron transparency. In S. aureus, the treated cells exhibited a widened periplasmic space, apparent cell envelope rupture, loss of cellular integrity, and leakage of cytoplasmic contents. In contrast, PBS-treated control cells of both species maintained their intact morphology. These findings clearly demonstrate that AP10RW exerts its bactericidal activity by directly disrupting the structural integrity of bacterial cell envelopes.

3.5. Binding of AP10RW to LPS, LTA, and PGN

The binding capacity of AP10RW to key bacterial surface molecules (LPS, LTA, and PGN) was assessed by ELISA. The results revealed dose-dependent interactions between AP10RW and all three targets (Figure 4). In contrast, the negative control protein BSA showed negligible binding. These findings indicate that AP10RW specifically recognizes diverse bacterial signature molecules, supporting its potential role as a multivalent pattern recognition molecule.

We then performed isothermal titration calorimetry (ITC) to quantitatively assess the thermodynamics of the interaction between the AP10RW and LPS. A titration isotherm (Figure 5A) revealed a multiphasic LPS-AP10RW binding profile. Notably, the initial injections were endothermic, with the heat change decreasing over the first seven injections. This was followed by an exothermic phase, which exhibited a V-shaped profile where the released heat first increased to a maximum before gradually decreasing towards the baseline as the titration reached saturation. This sequential endothermic-to-exothermic pattern is indicative of a multi-step binding mechanism. Consequently, a more detailed discussion of these points can be found in the following section. Control experiments titrating the peptide into the buffer alone confirmed that the observed heats were not attributable to dilution effects.

For the molecular docking results, the top nine poses were selected (Table S2) based on their calculated binding affinities using the CNNs score function in GNINA 1.0. Among these, pose 04 exhibited the strongest binding affinity (−6.41 kcal/mol). As a representative binding mode, the key interactions between AP10RW (pose 04) and LPS are illustrated in Figure 5B,C. Three hydrogen bonds were formed between Lys2 (backbone oxygen atom), Arg5 (side chain N atom), Trp9 (backbone oxygen atom) of AP10RW and three hydroxyl oxygen atoms at different sites of LPS, respectively. Additionally, AP10RW engaged in hydrophobic interactions with LPS by Trp1, Trp6, Ile8, and Trp9, as well as electrostatic interactions through Arg3, Lys4, Arg7, and Lys10. These computational findings are consistent with the ITC and ELISA results, providing further support for the idea that AP10RW can recognize the bacterial-specific component LPS and carry out its antimicrobial function.

3.6. Membrane Depolarization by AP10RW

The membrane depolarization activities of AP10RW were assayed using a potential-dependent distributional fluorescent dye DiSC_3_(5). The fluorescence intensity was monitored for 30 min after peptide addition to capture the kinetic profile of membrane potential changes. Upon exposure to AP10RW for 30 min, all tested bacterial cells exhibited an increase in fluorescence compared to the buffer control, indicating a potent depolarizing effect on the bacterial cytoplasmic membrane (Figure 6).

Notably, the kinetics of this response varied across bacterial strains. While some showed an initial prompt increase, others displayed a more gradual rise, reflecting differences in the time required for AP10RW to interact with and disrupt membranes of varying strains. This strain-dependent kinetic behavior underscores the importance of continuous monitoring over a single time point, as it reveals both the rate and the ultimate extent of depolarization.

3.7. Induction of ROS Production by AP10RW

The treatment of bacterial cells with AP10RW led to a significant increase in intracellular ROS levels (Figure 7). Since elevated ROS can induce cell death, we propose that AP10RW exerts its bactericidal effect by provoking lethal ROS production.

3.8. Functional Antimicrobial Peptide Stability of AP10RW

For clinical application, it is vital that AMPs maintain their activity within a complex physiological state [34]. To evaluate the environmental stability of AP10RW, we monitored changes in its MIC against the multidrug-resistant strains E. coli 577 and S. aureus USA500.

3.8.1. Antibacterial Activity of Incubation in Serum

While serum components often compromise AMP activity [26], AP10RW retained its full potency in the presence of 25% human serum, with 1× MIC sufficient for complete bacterial inhibition (Figure 8A).

3.8.2. Salt Resistance Assay

In contrast to many AMPs whose activity is diminished by physiological salt concentrations [35], AP10RW maintained nearly constant activity in the presence of physiologically relevant, high-ionic-strength conditions (~150 mM NaCl in physiological saline), with 1× MIC achieving complete growth inhibition (Figure 8B).

3.8.3. pH Resistance Assay

AP10RW displayed pH tolerance, maintaining consistent antibacterial activity across both pH 5.5 and pH 8.5 conditions. At 1× MIC, it fully inhibited bacterial growth in all tested pH environments (Figure 8C).

3.8.4. Heat Resistance Assay

AP10RW also exhibited high thermostability. Even after heating at 85 °C for 30 min, its activity was fully retained, and 1× MIC completely inhibited bacterial growth (Figure 8D). This high thermal stability indicates that AP10RW does not require stringent temperature control throughout the entire process, potentially simplifying its storage and handling in various settings.

3.9. Cytocompatibility of AP10RW with Mammalian Cells

The cytocompatibility of AP10RW was investigated through an assessment of its hemolytic activity towards human RBCs. As shown in Figure 9, RBCs treated with AP10RW exhibited minimal hemolysis, with values remaining below 5% (a widely accepted threshold for negligible hemolytic activity in therapeutic candidates) and comparable to the controls. The results demonstrate that AP10RW exhibits a favorable hemocompatibility profile at its effective antimicrobial concentrations.

Furthermore, the cytotoxicity of AP10RW toward human HEK293T cells and murine RAW264.7 macrophages was examined using the CCK-8. As summarized in Table 4, HEK293T cells and RAW264.7 cells maintained high viability after exposure to AP10RW across all concentrations tested, confirming its negligible cytotoxic effect.

Taken together, these data indicate that AP10RW exhibits minimal toxicity towards human RBCs, human HEK293T macrophages, and murine RAW264.7, indicating high selectivity for bacterial over mammalian cell membranes.

4. Discussion

4.1. Structural Basis of AP10RW as a Potential Antimicrobial Peptide

Structural characterisation of AP10RW reveals key features that underlie its antimicrobial potential. This peptide has an optimal amphiphilic charge balance (+6.97) and moderate hydrophobicity (40%), supporting its electrostatic targeting of bacterial membranes, followed by hydrophobic interactions.

AP10RW contains three Trp residues, potentially classifying it within the tryptophan-rich antimicrobial peptide family. The indole ring of Trp exhibits unique amphiphilic properties, predisposing it to localize at the interface region of the lipid bilayer, which is a critical step for initial peptide binding and anchoring to membranes. Furthermore, Trp cooperates with abundant cationic residues, particularly arginine (Arg). After the positively charged Arg undergoes electrostatic attraction with the negatively charged phospholipid headgroups in the membrane, the Trp residue can further stabilize the peptide-membrane complex through cation-π interactions. This significantly enhances membrane affinity and disruption efficiency [36,37].

Circular dichroism analysis confirms AP10RW’s environmentally responsive conformation. Under TFE conditions, AP10RW adopts an alpha-helical structure, showing the ability to fold that is key to disrupting membranes. Intriguingly, unlike in TFE, AP10RW did not adopt a canonical α-helical structure in anionic SDS micelles. This is likely due to electrostatic attraction between the highly cationic peptide and the negatively charged SDS surface, which restricts the peptide to a “surface-bound” state and prevents deep insertion and helical folding within the micelle core [31], and the intrinsic destabilizing influence of its central arginine residue on helix formation in this environment [32].

Moreover, SDS micelles represent a simplified membrane model that lacks key biological features such as lipid diversity and transmembrane potential. Therefore, the absence of a stable helix in SDS does not contradict the potent, rapid membranolytic activity observed in biological assays. Rather, it highlights that AP10RW’s functional conformational transition is highly context-dependent. The specific lipid composition and electrochemical gradient of a live bacterial membrane are likely required to overcome the initial electrostatic and structural barriers, guiding the peptide toward its final disruptive conformation. Its inherent helical propensity, confirmed in TFE, supports this latent capability.

Thus, AP10RW may belong to a class of AMPs whose final disruptive conformation is not pre-formed but is dynamically shaped by complex membrane interactions. The initial “unfolded” or “surface-bound” state in simplified models like SDS may even represent a functional intermediate, where specific residue-level interactions (e.g., arginine destabilization) must be carefully managed during the process of membrane recognition and disruption.

4.2. Functional Efficacy and Mechanistic Insights

AP10RW demonstrates potent and broad-spectrum antimicrobial activity against both drug-sensitive and multidrug-resistant bacterial strains, with MIC values consistently below 100 μg/mL (64 μM). This notable efficacy against clinically relevant MDR pathogens highlights its potential as a promising therapeutic candidate in the era of antimicrobial resistance.

Mechanistic investigations reveal that AP10RW employs a diverse attack strategy against bacterial cells. The peptide initially establishes specific interactions with key bacterial components (LPS, LTA, and PGN) as evidenced by dose-dependent binding. ITC analysis further reveals that sequential endothermic-to-exothermic pattern is indicative of a multi-step binding mechanism of AP10RW-LPS. The initial endothermic phase is likely dominated by entropically favorable processes, such as the disruption of the LPS aggregate structure and/or the release of ordered water molecules from the interaction surfaces (dehydration). The subsequent, dominant exothermic phase is enthalpically driven, reflecting the establishment of direct, favorable non-covalent interactions, such as electrostatic and hydrophobic contacts, between the peptide and LPS. The increasing exothermicity observed in the early part of the V-shaped profile suggests a potential positive cooperativity, where initial binding events facilitate subsequent interactions, possibly by inducing a conformational change in LPS that exposes additional high-affinity binding sites. Once all the binding sites have been released, the thermal effect returns to baseline. The results of the molecular docking are consistent with the findings of the above-mentioned binding experiment, providing further confirmation that AP10RW can interact with LPS. This interaction forms the basis of its potential biological functions.

This initial binding results in rapid membrane depolarization, indicating immediate disruption of the critical proton motive force. TEM observations confirm extensive ultrastructural damage, including membrane rupture, cytoplasmic leakage, and formation of intracellular voids. Notably, AP10RW additionally induces substantial ROS accumulation, imposing lethal oxidative stress on bacterial cells.

The combined effects of these mechanisms (surface binding, membrane disruption, and ROS-mediated damage) likely explain AP10RW’s strong bactericidal effects and may also reduce the likelihood of resistance development through target multiplicity.

4.3. Stability and Safety Profile of AP10RW

The clinical translation potential of AP10RW is strongly supported by its remarkable stability under diverse physiological conditions. Unlike many antimicrobial peptides that lose efficacy in biological environments, AP10RW maintained potent antibacterial activity despite exposure to serum, physiological salt concentrations, varying pH levels (5.5–8.5), and even high-temperature treatment (85 °C).

Complementing its stability, AP10RW demonstrated excellent biosafety. At concentrations effective against MDR bacteria, the peptide exhibited minimal hemolytic activity against human red blood cells and negligible cytotoxicity toward murine RAW264.7 macrophages and human HEK293T cells. This selective toxicity, which is potent against bacterial membranes but harmless to mammalian cells, is likely due to its specific interactions with bacterial membrane components rather than the neutral, cholesterol-rich membranes of eukaryotic cells. The combination of maintained antimicrobial efficacy in challenging environments together with minimal cytotoxicity addresses two major barriers to the antimicrobial peptides translation (stability in physiological conditions and biosafety). This makes AP10RW a promising candidate for further therapeutic development.

5. Conclusions

Based on comprehensive characterization, AP10RW emerges as a promising candidate to address multidrug-resistant bacterial infections. Its rational design, balancing a strong positive charge with controlled hydrophobicity, allows it to effectively target bacterial membranes. AP10RW exerts potent, broad-spectrum bactericidal activity through a unique multi-mechanism action: initial high-affinity, multi-step binding to bacterial surface components triggers rapid membrane depolarization and ultrastructural damage, compounded by lethal ROS accumulation. Crucially, AP10RW overcomes key translational hurdles, maintaining robust activity under physiological stressors (serum, physiological salt, pH, heat) while exhibiting minimal toxicity against mammalian cells due to its selective membrane interaction. This combination of efficacy, stability, and safety strongly supports its potential for further therapeutic development.