Development of Xanthoangelol-Derived Compounds with Membrane-Disrupting Effects against Gram-Positive Bacteria

Infections caused by multidrug-resistant pathogens have emerged as a serious threat to public health. To develop new antibacterial agents to combat such drug-resistant bacteria, a class of novel amphiphilic xanthoangelol-derived compounds were designed and synthesized by mimicking the structure and function of antimicrobial peptides (AMPs). Among them, compound 9h displayed excellent antimicrobial activity against the Gram-positive strains tested (MICs = 0.5–2 μg/mL), comparable to vancomycin, and with low hemolytic toxicity and good membrane selectivity. Additionally, compound 9h demonstrated rapid bactericidal effects, low resistance frequency, low cytotoxicity, and good plasma stability. Mechanistic studies further revealed that compound 9h had good membrane-targeting ability and was able to destroy the integrity of bacterial cell membranes, causing an increase in intracellular ROS and the leakage of DNA and proteins, thus accelerating bacterial death. These results make 9h a promising antimicrobial candidate to combat bacterial infection.


Introduction
The emergence and spread of multidrug-resistant bacteria pose a significant threat to global public health, leading to increased morbidity and mortality [1][2][3].In particular, the most worrying ESKAPE pathogens, namely, Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacteriaceae, have limited the efficacy of most of frontline antibiotics as they have developed resistance to most of the conventional antibiotics [4][5][6].In addition, the decreasing investment from pharmaceutical companies in the discovery and development of new antibacterial agents has led to a continuous decline in the number of new antibiotic approvals, which further exacerbates the incidence of antibacterial resistance [7][8][9].Currently, almost 700,000 people die from bacterial infections each year, and the deaths are expected to rise to 10 million annually by 2050 if current trends continue [10,11].The alarmingly spread of MDR pathogens, alongside the decline in the effectiveness of existing antibiotics, has compelled researchers to look for new alternatives with unique mechanisms of action to address antimicrobial resistance.
Antimicrobial peptides (AMPs), also called host-defense peptides, have been used by almost all multicellular organisms as an initial defense mechanism against pathogenic microbes [12,13].AMPs are typically short peptides consisting of 10-50 amino acids [14].Despite a wide range of structural diversity in terms of sequence and secondary structure, these AMPs share common characteristics: being amphipathic in nature, with net cationic charge [15][16][17].This amphipathic nature of AMPs is believed to be the structural basis for their activity, and the bacterial membrane is considered as the main target for AMPs [18,19].The cationic groups in AMPs play an essential role for their initial attachment to the negatively charged bacterial membranes via electrostatic attraction, which is their activity, and the bacterial membrane is considered as the main target for AMPs [18,19].The cationic groups in AMPs play an essential role for their initial attachment to the negatively charged bacterial membranes via electrostatic attraction, which is followed by the interaction of the hydrophobic side chains of AMPs with the lipid-rich cytoplasmic membrane, causing the disruption of membrane integrity, and ultimately causing cell death [20,21].Unlike conventional antibiotics, AMPs offer several advantages as potential antibacterial agents as they exhibit rapid and broad-spectrum antibacterial activity, reduced susceptibility to resistance, high selectivity towards microbial cells, and good synergy with other antimicrobials [22,23].Although natural AMPs show potential as innovative antimicrobials, their clinical development has been hindered by challenges such as poor bioavailability, low proteolytic stability, unknown in vivo toxicity, and high production costs [24,25].To address these challenges, researchers have increasingly focused on the development of non-natural small-molecule antimicrobial peptidomimetics.These peptidomimetics can be defined as compounds bearing unnatural backbones that mimic the amphipathic structure and biological function of the natural AMPs [26,27].These small-molecule peptidomimetics can surmount the limitations associated with natural AMPs and display improved metabolic stability while retaining the similar activity and selectivity properties of AMPs.Their recent success is highlighted by three peptidomimetics (CSA-13, XF-73, and PMX-30063; Figure 1) in clinical trials [28][29][30][31].Natural products are a fascinating resource for drug development due to their structural diversity, offering a wide range of new scaffolds for antimicrobials and other bioactive components.Xanthoangelol, a natural geranylated hydroxy chalcone, is isolated from the roots of Angelica keiskei, and it has shown diverse therapeutic and biological properties, including anticancer [32,33], anti-inflammatory [34,35], and anti-microbial activities [36,37].Recent research has shown that xanthoangelol exhibited good antibacterial activity against Gram-positive bacteria by destroying cell membranes, a mechanism similar to that of AMPs [38].The geranyl chain of xanthoangelol meets the hydrophobic tail requirements observed in the structure of AMPs, facilitating its insertion into the lipid bilayer's hydrophobic core.Nevertheless, xanthoangelol also has some drawbacks, such as poor water solubility, susceptibility to metabolism, and a narrow antibacterial spectrum.The hydrophobic chalcone scaffold and geranylgeranyl chain of xanthoangelol likely contribute to its poor water solubility, while the phenolic hydroxyl groups of xanthoangelol readily form covalent bonds with endogenous molecules like glucuronic acid or sulfuric acid, thereby increasing first-pass metabolism.Herein, to improve the water solubility, reduce the metabolic rate, and expand the antibacterial spectrum of xanthoangelol, we designed and synthesized a series of amphiphilic xanthoangelol-derived compounds as antimicrobials by mimicking the structure and function of AMPs.It has been widely realized that cationic charge, hydrophobicity, and molecular weight seem to be key structural parameters that control the antimicrobial potency and hemolytic activity of the antimicrobial peptidomimetics.The presence of highly cationic groups allows for adherence to the bacterial Natural products are a fascinating resource for drug development due to their structural diversity, offering a wide range of new scaffolds for antimicrobials and other bioactive components.Xanthoangelol, a natural geranylated hydroxy chalcone, is isolated from the roots of Angelica keiskei, and it has shown diverse therapeutic and biological properties, including anticancer [32,33], anti-inflammatory [34,35], and anti-microbial activities [36,37].Recent research has shown that xanthoangelol exhibited good antibacterial activity against Gram-positive bacteria by destroying cell membranes, a mechanism similar to that of AMPs [38].The geranyl chain of xanthoangelol meets the hydrophobic tail requirements observed in the structure of AMPs, facilitating its insertion into the lipid bilayer's hydrophobic core.Nevertheless, xanthoangelol also has some drawbacks, such as poor water solubility, susceptibility to metabolism, and a narrow antibacterial spectrum.The hydrophobic chalcone scaffold and geranylgeranyl chain of xanthoangelol likely contribute to its poor water solubility, while the phenolic hydroxyl groups of xanthoangelol readily form covalent bonds with endogenous molecules like glucuronic acid or sulfuric acid, thereby increasing first-pass metabolism.Herein, to improve the water solubility, reduce the metabolic rate, and expand the antibacterial spectrum of xanthoangelol, we designed and synthesized a series of amphiphilic xanthoangelol-derived compounds as antimicrobials by mimicking the structure and function of AMPs.It has been widely realized that cationic charge, hydrophobicity, and molecular weight seem to be key structural parameters that control the antimicrobial potency and hemolytic activity of the antimicrobial peptidomimetics.The presence of highly cationic groups allows for adherence to the bacterial cell surfaces through electrostatic interactions; however, these groups are unlikely to penetrate the hydrophobic interior of the membrane, which restricts their activity [39].On the other hand, peptidomimetics with excessive hydrophobicity can cause non-specific toxicity to both human and bacterial cells [40][41][42].In addition, high molecular weights are detrimental to druggability.Therefore, it is crucial in the design of antimicrobial peptidomimetics to meticulously adjust the ratio of the cationic charge and hydrophobic characteristics, as well as to ensure a suitable molecular weight.Initially, we selected the left-hand-side structure of xanthoangelol for cyclization to obtain the pyranochromene scaffold (Figure 2); different cationic groups then were introduced on the scaffold to yield a range of amphipathic structures (Series 5).However, the resulting derivatives displayed only moderate antibacterial activity, which might have been due to the insufficient hydrophobic chain that failed to promote pore formation and cell death.Next, two lipophilic chains were incorporated into the pyranochromene scaffold to investigate their antibacterial activity (series 9).Compared with compounds in series 5, these compounds in series 9 showed significant improvement in the antibacterial activity.After the in vitro antibacterial evaluation of xanthoangelol-derived compounds, a promising compound 9h with excellent antibacterial activity, low hemolytic activity, and good membrane selectivity was obtained.Plasma stability, bactericidal kinetics, drug resistance, and inhibition and disruption of bacterial biofilms and antibacterial mechanisms were also investigated.
Antibiotics 2024, 13, x FOR PEER REVIEW 3 of 21 cell surfaces through electrostatic interactions; however, these groups are unlikely to penetrate the hydrophobic interior of the membrane, which restricts their activity [39].On the other hand, peptidomimetics with excessive hydrophobicity can cause non-specific toxicity to both human and bacterial cells [40][41][42].In addition, high molecular weights are detrimental to druggability.Therefore, it is crucial in the design of antimicrobial peptidomimetics to meticulously adjust the ratio of the cationic charge and hydrophobic characteristics, as well as to ensure a suitable molecular weight.Initially, we selected the lefthand-side structure of xanthoangelol for cyclization to obtain the pyranochromene scaffold (Figure 2); different cationic groups then were introduced on the scaffold to yield a range of amphipathic structures (Series 5).However, the resulting derivatives displayed only moderate antibacterial activity, which might have been due to the insufficient hydrophobic chain that failed to promote pore formation and cell death.Next, two lipophilic chains were incorporated into the pyranochromene scaffold to investigate their antibacterial activity (series 9).Compared with compounds in series 5, these compounds in series 9 showed significant improvement in the antibacterial activity.After the in vitro antibacterial evaluation of xanthoangelol-derived compounds, a promising compound 9h with excellent antibacterial activity, low hemolytic activity, and good membrane selectivity was obtained.Plasma stability, bactericidal kinetics, drug resistance, and inhibition and disruption of bacterial biofilms and antibacterial mechanisms were also investigated.

Chemistry
The synthetic routes of the target compounds are outlined in Schemes 1 and 2. Scheme 1 shows the synthetic route for compounds 5a-5h.

Chemistry
The synthetic routes of the target compounds are outlined in Schemes 1 and 2. Scheme 1 shows the synthetic route for compounds 5a-5h.First, intermediate 3 was prepared by the cyclization reaction of commercial citraldimethylacetal 2 and ketone 1 in pyridine under reflux conditions.Then, intermediate 3 reacted with 1,3-dibromopropane in acetonitrile at 60 • C for 8 h to afford 4. Finally, treatment of 4 with various amines obtained compounds 5a-5h.
Antibiotics 2024, 13, x FOR PEER REVIEW 3 of 21 cell surfaces through electrostatic interactions; however, these groups are unlikely to penetrate the hydrophobic interior of the membrane, which restricts their activity [39].On the other hand, peptidomimetics with excessive hydrophobicity can cause non-specific toxicity to both human and bacterial cells [40][41][42].In addition, high molecular weights are detrimental to druggability.Therefore, it is crucial in the design of antimicrobial peptidomimetics to meticulously adjust the ratio of the cationic charge and hydrophobic characteristics, as well as to ensure a suitable molecular weight.Initially, we selected the lefthand-side structure of xanthoangelol for cyclization to obtain the pyranochromene scaffold (Figure 2); different cationic groups then were introduced on the scaffold to yield a range of amphipathic structures (Series 5).However, the resulting derivatives displayed only moderate antibacterial activity, which might have been due to the insufficient hydrophobic chain that failed to promote pore formation and cell death.Next, two lipophilic chains were incorporated into the pyranochromene scaffold to investigate their antibacterial activity (series 9).Compared with compounds in series 5, these compounds in series 9 showed significant improvement in the antibacterial activity.After the in vitro antibacterial evaluation of xanthoangelol-derived compounds, a promising compound 9h with excellent antibacterial activity, low hemolytic activity, and good membrane selectivity was obtained.Plasma stability, bactericidal kinetics, drug resistance, and inhibition and disruption of bacterial biofilms and antibacterial mechanisms were also investigated.

Chemistry
The synthetic routes of the target compounds are outlined in Schemes 1 and 2. Scheme 1 shows the synthetic route for compounds 5a-5h.Compounds 9a-9h were synthesized using a similar method to that of compounds 5a-5h, as illustrated in Scheme 2. The synthesis of intermediate 7, in contrast to intermediate 3, necessitated a higher quantity of citraldimethylacetal (4.0 eq) and a longer reaction time (18 h).Following this, compound 7 underwent alkylation with 1,3-dibromopropane and amination with different amines to yield compounds 9a-9h.

In Vitro Antibacterial and Hemolytic Activity
All target derivatives were assessed for their in vitro antibacterial activities using a broth microdilution assay to test their minimum inhibitory concentrations (MICs) against six Gram-positive strains (S. aureus ATCC31007, S. aureus ATCC25923, S. aureus ATCC43300, E. faecalis ATCC29212, B. subtilis ATCC9372, and S. epidermidis ATCC12228) and four Gram-negative strains (A.baumannii ATCC19606, K. pneumonia ATCC10031, E. coli ATCC25922, and P. aeruginosa ATCC27853).These compounds were also evaluated for their hemolytic activities and membrane selectivity by determining their HC50 (the concentration required to lyse 50% of red blood cells) and SI (HC50/MICs) values, respectively, and vancomycin was used as the positive drug.
To begin with, compounds 5a-5h consisting of an unsaturated lipid chain and a pyranochromene scaffold were synthesized and tested.Compounds 5a-5h showed moderate or weak antibacterial activity against the eight Gram-positive bacterial strains tested (MICs ≥ 4 µg/mL) and poor antibacterial activity against the Gram-negative bacterial strain tested.As shown in Tables 1 and 2, compounds 5a-5e containing one tertiary amine or quaternary amine exhibited weak antimicrobial activity (MICs = 8->128 µg/mL), while compound 5f-5h containing two cationic amine groups showed moderate antibacterial activity against the Gram-positive strains tested (MICs = 8-32 µg/mL).This result suggests that incorporation of cationic groups was beneficial for improving the activity.Among them, compounds 5g and 5h were best, and they exhibited moderate activity against the three S. aureus tested (MICs = 8 µg/mL) and good activity against S. epidermidis ATCC12228 with an MIC value of 4 µg/mL.The unsatisfactory antibacterial performance in this series could be attributed to the insufficient hydrophobic chain that failed to insert into the phospholipid bilayer and promote pore formation.
To further explore the effect of hydrophobicity from unsaturated lipid chains on antibacterial activity, pyranochromene derivatives with dual unsaturated chains (9a-9h) were synthesized.The results revealed that almost all derivatives in this series exhibited significantly improved antimicrobial properties against eight Gram-positive bacteria when compared to the compounds in series 5, which only possessed one unsaturated lipid chain.For example, compounds 5a and 9a, as well as 5c and 9c, shared the same scaffold and cationic group.The MICs against S. aureus ranged from 2 to 4 µg/mL for 9a and 9c, while compounds 5a and 5c exhibited weak antibacterial activity (MICs = 8-128 µg/mL), which further highlighted the importance of unsaturated lipid chains in enhancing activity.These derivatives in series 9, except 9d and 9e, displayed excellent antibacterial activity against the Gram-positive bacteria.For example, the MIC values of compound 9d and 9e against the eight tested positive bacteria were almost all higher than 64 µg/mL, suggesting their poor antimicrobial activity, while the MIC values of other compounds in this series 9 ranged from 0.5 to 32 µg/mL, indicating their potent antimicrobial properties.In particular, compound 9h displayed the most potent antibacterial activity, demonstrating Compounds 9a-9h were synthesized using a similar method to that of compounds 5a-5h, as illustrated in Scheme 2. The synthesis of intermediate 7, in contrast to intermediate 3, necessitated a higher quantity of citraldimethylacetal (4.0 eq) and a longer reaction time (18 h).Following this, compound 7 underwent alkylation with 1,3-dibromopropane and amination with different amines to yield compounds 9a-9h.

In Vitro Antibacterial and Hemolytic Activity
All target derivatives were assessed for their in vitro antibacterial activities using a broth microdilution assay to test their minimum inhibitory concentrations (MICs) against six Gram-positive strains (S. aureus ATCC31007, S. aureus ATCC25923, S. aureus ATCC43300, E. faecalis ATCC29212, B. subtilis ATCC9372, and S. epidermidis ATCC12228) and four Gramnegative strains (A.baumannii ATCC19606, K. pneumonia ATCC10031, E. coli ATCC25922, and P. aeruginosa ATCC27853).These compounds were also evaluated for their hemolytic activities and membrane selectivity by determining their HC 50 (the concentration required to lyse 50% of red blood cells) and SI (HC 50 /MICs) values, respectively, and vancomycin was used as the positive drug.
To begin with, compounds 5a-5h consisting of an unsaturated lipid chain and a pyranochromene scaffold were synthesized and tested.Compounds 5a-5h showed moderate or weak antibacterial activity against the eight Gram-positive bacterial strains tested (MICs ≥ 4 µg/mL) and poor antibacterial activity against the Gram-negative bacterial strain tested.As shown in Tables 1 and 2, compounds 5a-5e containing one tertiary amine or quaternary amine exhibited weak antimicrobial activity (MICs = 8->128 µg/mL), while compound 5f-5h containing two cationic amine groups showed moderate antibacterial activity against the Gram-positive strains tested (MICs = 8-32 µg/mL).This result suggests that incorporation of cationic groups was beneficial for improving the activity.Among them, compounds 5g and 5h were best, and they exhibited moderate activity against the three S. aureus tested (MICs = 8 µg/mL) and good activity against S. epidermidis ATCC12228 with an MIC value of 4 µg/mL.The unsatisfactory antibacterial performance in this series could be attributed to the insufficient hydrophobic chain that failed to insert into the phospholipid bilayer and promote pore formation.
To further explore the effect of hydrophobicity from unsaturated lipid chains on antibacterial activity, pyranochromene derivatives with dual unsaturated chains (9a-9h) were synthesized.The results revealed that almost all derivatives in this series exhibited significantly improved antimicrobial properties against eight Gram-positive bacteria when compared to the compounds in series 5, which only possessed one unsaturated lipid chain.For example, compounds 5a and 9a, as well as 5c and 9c, shared the same scaffold and cationic group.The MICs against S. aureus ranged from 2 to 4 µg/mL for 9a and 9c, while compounds 5a and 5c exhibited weak antibacterial activity (MICs = 8-128 µg/mL), which further highlighted the importance of unsaturated lipid chains in enhancing activity.These derivatives in series 9, except 9d and 9e, displayed excellent antibacterial activity against the Gram-positive bacteria.For example, the MIC values of compound 9d and 9e against the eight tested positive bacteria were almost all higher than 64 µg/mL, suggesting their poor antimicrobial activity, while the MIC values of other compounds in this series 9 ranged from 0.5 to 32 µg/mL, indicating their potent antimicrobial properties.In particular, compound 9h displayed the most potent antibacterial activity, demonstrating comparable activity (MICs = 0.5-8 µg/mL) to that of the positive control vancomycin against the Gram-positive bacteria.Additionally, it also exhibited good activity against the Gram-negative bacteria tested (MICs = 4-16 µg/mL).To further detect the hemolytic toxicity and therapeutic potential of these derivatives, we proceeded to examine the hemolytic activities and membrane selectivity of these derivatives.The HC 50 value was defined as the concentration needed for a compound to lyse 50% of red blood cells, and it is often used to evaluate the hemolytic toxicity to eukaryotic cells.As shown in Table 1, compounds 5a-5h showed poor hemolytic activities, with HC 50 values of ranging from 62.3 to 746.5 µg/mL, suggesting that they were unlikely to lyse blood cells at MIC concentrations.In contrast, compounds in series 9 except for 9c-9e showed moderate hemolytic activity with HC 50 values of 26.6-57.9µg/mL, indicating that these compounds have some hemolytic toxicity to eukaryotic cells.The selectivity index (SI) was calculated by HC 50 /MICs of S. aureus ATCC43300 and was used to assess the potency and safety of the compounds.As illustrated in Table 1, these xanthoangelolderived compounds except for 9c and 9h possessed low membrane selectivity.Among these derivatives, compounds 9h showed the most potent antibacterial activity with a MICs of 0.5-2 µg/mL against Gram-positive bacteria and exhibited low hemolytic toxicity (HC 50 = 98.0 µg/mL) and the best membrane selectivity (SI = 49.0).Thus, compound 9h was selected as a representative molecule for the in-depth in vitro and in vivo studies.

Plasma Stability and Bactericidal Activity in Mammalian Fluids
Due to the presence of various biological enzymes in plasma, natural AMPs are susceptible to enzymatic degradation upon entering the bloods, leading to a loss of their antimicrobial activity [43,44].To evaluate the plasma stability of active compound 9h, we conducted a stability assay by determining their MBC (minimum bactericidal concentration) values in plasma and different blood components.As shown in Figure 3A, the MBC value of 9h dissolved in Mueller Hinton broth (MHB) against S. aureus ATCC4330 was 16 µg/mL.Subsequent treatment with 50% plasma resulted in an increase in the MBC value for 9h to 32 µg/mL.Notably, this MBC value remained stable and unchanged after both 3 and 6 h.The observed initial increase in the MBC value, followed by remaining unchanged, could potentially be attributed to the initial binding of compound 9h with plasma proteins, causing a reduction in its activity.Once this binding became saturated, the bactericidal activity of the compound remained stable, further indicating the good stability of compound 9h in plasma.Additionally, compound 9h was also found to retain its antibacterial activity in 50% serum (MBC = 32 µg/mL), 50% plasma (MBC = 32 µg/mL), and 50% blood (MBC = 32 µg/mL) against S. aureus ATCC4330 (Figure 3B).The above results demonstrated that compound 9h had good stability, even in serum, plasma, and blood.

In Vitro Cytotoxicity Evaluation
A key component in the development of antimicrobial agents for clinical application is the assessment of their potential toxicity to mammalian cells.To assess the cytotoxic impact, compound 9h was chosen to undergo in vitro testing for its cytotoxic effects on LO2 cells.As shown in Figure 3C, at a concentration of 8 µg/mL, compound 9h exhibited a low impact on LO2 cells, maintaining their viability by 82.61%.Furthermore, when the concentration of 9h increased to 32 µg/mL, LO2 cells still maintained 71.43% viability.These findings underscored that compound 9h (IC50 = 55.05 µg/mL) had high safety toward towards mammalian cells.

In Vitro Cytotoxicity Evaluation
A key component in the development of antimicrobial agents for clinical application is the assessment of their potential toxicity to mammalian cells.To assess the cytotoxic impact, compound 9h was chosen to undergo in vitro testing for its cytotoxic effects on LO2 cells.As shown in Figure 3C, at a concentration of 8 µg/mL, compound 9h exhibited a low impact on LO2 cells, maintaining their viability by 82.61%.Furthermore, when the concentration of 9h increased to 32 µg/mL, LO2 cells still maintained 71.43% viability.These findings underscored that compound 9h (IC 50 = 55.05 µg/mL) had high safety toward towards mammalian cells.

In Vitro Time-Kill Kinetics
Natural AMPs have rapid bactericidal properties [45], and in order to examine whether active compound 9h have similar rapid bactericidal properties, the time-kill kinetics of 9h were carried out to test the bactericidal action against S. aureus ATCC4330 at four different concentrations.As shown in Figure 4A, compound 9h at low concentration (1×, 2× MIC) had a significant inhibitory effect on the growth of S. aureus ATCC43300, and the decreasing trend of bacterial load was obvious when compared to the negative control.As the concentrations increased, compound 9h (4×, 8× MIC) exhibited rapid bactericidal activity.For example, compound 9h (8× MIC, 16 µg/mL) achieved 5.95-log colony-forming unit (CFU) reduction (killing ≥ 99.9% of S. aureus) within 1 h compared with the control, and complete bacterial killing was observed after 4 h.These results demonstrated that compound 9h had a strong bactericidal effect against Gram-positive bacteria.

In Vitro Cytotoxicity Evaluation
A key component in the development of antimicrobial agents for clinical application is the assessment of their potential toxicity to mammalian cells.To assess the cytotoxic impact, compound 9h was chosen to undergo in vitro testing for its cytotoxic effects on LO2 cells.As shown in Figure 3C, at a concentration of 8 µg/mL, compound 9h exhibited a low impact on LO2 cells, maintaining their viability by 82.61%.Furthermore, when the concentration of 9h increased to 32 µg/mL, LO2 cells still maintained 71.43% viability.These findings underscored that compound 9h (IC50 = 55.05 µg/mL) had high safety toward towards mammalian cells.

In Vitro Time-Kill Kinetics
Natural AMPs have rapid bactericidal properties [45], and in order to examine whether active compound 9h have similar rapid bactericidal properties, the time-kill kinetics of 9h were carried out to test the bactericidal action against S. aureus ATCC4330 at four different concentrations.As shown in Figure 4A, compound 9h at low concentration (1×, 2× MIC) had a significant inhibitory effect on the growth of S. aureus ATCC43300, and the decreasing trend of bacterial load was obvious when compared to the negative control.As the concentrations increased, compound 9h (4×, 8× MIC) exhibited rapid bactericidal activity.For example, compound 9h (8× MIC, 16 µg/mL) achieved 5.95-log colony-forming unit (CFU) reduction (killing ≥ 99.9% of S. aureus) within 1 h compared with the control, and complete bacterial killing was observed after 4 h.These results demonstrated that compound 9h had a strong bactericidal effect against Gram-positive bacteria.

Resistance Development Studies
The rapid increase in bacterial resistance to antibiotics is a major challenge in clinical treatment [46], and in order to assess the drug resistance development for compound 9h, we performed a drug resistance assay against S. aureus ATCC43300.Norfloxacin was selected as the positive control.The S. aureus cells were incubated with sub-MIC (1/2 MIC) concentration of compound 9h and were serially passaged for a duration of 20 days.As shown in Figure 4B, the MIC value of 9h against S. aureus ATCC43300 was almost consistent with its original value, and no more than a 4-fold increase was observed.In contrast, norfloxacin quickly induced resistance in S. aureus, showing a significant 512-fold increase in MIC value at the 19th passage.Thus, this result suggests that compound 9h was unlikely to induce resistance in Gram-positive bacteria.

Antibiofilm Activity Studies
Bacterial biofilms created by the aggregation of bacteria is considered as one of the main reasons for the development of drug resistance [47].In order to assess the ability of these xanthoangelol-derived compounds to inhibit or disrupt biofilms, the best compound, 9h, was chosen against the biofilms of S. aureus ATCC43300.As illustrated in Figure 5A, compound 9h was able to inhibit the biofilm formation of S. aureus in a concentrationdependent manner.Compound 9h at 1× MIC (2 µg/mL) showed an 8.65% inhibition of S. aureus biofilm formation, which increased to 10.87% at 2× MIC, while the positive control LL-37 (4 µg/mL) displayed the inhibition of biofilm formation by 55.41%.Subsequently, we further assessed the efficiency of 9h to eradicate the preformed biofilm.Matured S. aureus biofilm (grown for 24 h) was treated with compound 9h, and the rate of biofilm disruption was then assessed by the crystal violet staining assay.As shown in Figure 5B, when treated with 1× MIC (2 µg/mL) of 9h, only 11.65% of preformed S. aureus biofilm was eradicated, and about 19.83% of the eradication of the established S. aureus biofilm was achieved after treatment with 8 µg/mL of compound 9h.These results suggest that compound 9h had a relatively low ability to inhibit the biofilm formation and disrupt the preformed biofilm.
sen against the biofilms of S. aureus ATCC43300.As illustrated in Figure 5A, compound 9h was able to inhibit the biofilm formation of S. aureus in a concentration-dependent manner.Compound 9h at 1× MIC (2 µg/mL) showed an 8.65% inhibition of S. aureus biofilm formation, which increased to 10.87% at 2× MIC, while the positive control LL-37 (4 µg/mL) displayed the inhibition of biofilm formation by 55.41%.Subsequently, we further assessed the efficiency of 9h to eradicate the preformed biofilm.Matured S. aureus biofilm (grown for 24 h) was treated with compound 9h, and the rate of biofilm disruption was then assessed by the crystal violet staining assay.As shown in Figure 5B, when treated with 1× MIC (2 µg/mL) of 9h, only 11.65% of preformed S. aureus biofilm was eradicated, and about 19.83% of the eradication of the established S. aureus biofilm was achieved after treatment with 8 µg/mL of compound 9h.These results suggest that compound 9h had a relatively low ability to inhibit the biofilm formation and disrupt the preformed biofilm.

Antimicrobial Mechanism Studies
Since the derivatives developed here were designed based on mimicking AMPs, we hypothesized that they also acted through a membrane-interrupting mechanism similar to that of AMP.Given the excellent antimicrobial properties of compound 9h, we chose it as the model molecule to further explore its antimicrobial mechanisms through a series of experiments, including fluorescence and scanning electron microscopy (SEM), membrane

Antimicrobial Mechanism Studies
Since the derivatives developed here were designed based on mimicking AMPs, we hypothesized that they also acted through a membrane-interrupting mechanism similar to that of AMP.Given the excellent antimicrobial properties of compound 9h, we chose it as the model molecule to further explore its antimicrobial mechanisms through a series of experiments, including fluorescence and scanning electron microscopy (SEM), membrane depolarization and permeability assays, ROS measurements, and DNA and protein leakage studies.

Fluorescence and Electron Scanning Microscopy
Fluorescent dyes propidium iodide (PI) and 4 ′ ,6 ′ -diamidino-2-phenylindole dihydrochloride (DAPI), as indicators of bacteria cell membrane damage, were used to perform fluorescence microscopy.DAPI was able to penetrate both living and dead bacterial cell membranes and bond with DNA to emit blue fluorescence, while PI can only pass through impaired cell membrane and integrates into double-stranded DNA to emit red fluorescence [48,49].If the bacterial cell membrane is damaged, it leads to the emission of a significant quantity of red fluorescence.As shown in Figure 6B, the control group emitted distinct blue fluorescence and only very weak red fluorescence, suggesting that the vast majority of bacteria were living and their cell membranes were intact.By contrast, obvious red and blue fluorescence was observed at the same time after treatment with compounds 9h, indicating that a lot of bacteria had died, and that the integrity of bacterial cell membranes had been disrupted.Furthermore, the surface changes in cell morphology of S. aureus ATCC43300 treated with compound 9h were observed using SEM.The surface morphology of untreated S. aureus cells was intact and smooth.However, after treatment with compound 9h, the S. aureus cells had lost their normal morphology, and they appeared wrinkled, collapsed, and ruptured (Figure 6A).This result directly demonstrated that compound 9h could effectively disrupt the cell membranes of S. aureus.
Furthermore, the surface changes in cell morphology of S. aureus ATCC43300 treated with compound 9h were observed using SEM.The surface morphology of untreated S. aureus cells was intact and smooth.However, after treatment with compound 9h, the S. aureus cells had lost their normal morphology, and they appeared wrinkled, collapsed, and ruptured (Figure 6A).This result directly demonstrated that compound 9h could effectively disrupt the cell membranes of S. aureus.

Cytoplasmic Membrane Depolarization
To explore the antibacterial mechanism of compound 9h, a cytoplasmic membrane depolarization assay was performed against S. aureus ATCC43300 using 3,3 ′ -dipropylthiadicarbocyanine iodide (DiSC3(5)) assay.In general, DiSC3(5) could penetrate the intact cell membranes and was quenched when accumulated inside the bacterial cells.Once the cell membrane was destroyed, DiSC3(5) was released from the cell membrane, leading to a significant increase in fluorescence intensity [50,51].As shown in Figure 7A, the fluorescence intensity of the blank control did not exhibit any significant variation, whereas a sharp enhancement in fluorescence intensity was observed after the addition of 9h (1×, 2×, 4×, 8× MIC).Moreover, the fluorescence intensity in S. aureus increased in a dose-dependent manner with increasing concentrations of 9h.It could be found that when treated with 16 µg/mL of 9h, the fluorescence intensity (at 20 min) increased about 5-fold compared with that of the blank control group.The result thus suggested that compound 9h had the ability to depolarize the membrane potential of cell membrane, thus increasing their permeability.
ent manner with increasing concentrations of 9h.It could be found that when treated with 16 µg/mL of 9h, the fluorescence intensity (at 20 min) increased about 5-fold compared with that of the blank control group.The result thus suggested that compound 9h had the ability to depolarize the membrane potential of cell membrane, thus increasing their permeability.

Cell Membrane Permeabilization
The cell membrane permeabilization of compound 9h against S. aureus ATCC43300 was assessed by PI uptake assay.PI could only pass through the impaired cell membrane and had been widely used to evaluate the cell membrane permeability.When the cell membranes were disrupted, PI entered into the cell membrane, leading to a significant improvement in fluorescence intensity.Therefore, the degree of membrane damage could be reflected by observing the fluorescence intensity in S. aureus.According to Figure 7B, the fluorescence intensity in the control group was weak and remained unchanged, whereas the fluorescence intensity of PI progressively increased when different concentrations of 9h were added.Compared with the control group, the fluorescence intensity (at 20 min) increased approximately 5-fold after the addition of 9h (16 µg/mL).The abovementioned results demonstrated that compound 9h had a strong ability of cell membrane permeabilization in a concentration-dependent manner.

Reactive Oxygen Species (ROS) Measurement
It is known that the disruption of bacterial cell membranes often causes the production of ROS in bacterial cells [31,52].The fluorescent probe DCFH-DA was used to monitor the level of intracellular ROS produced by S. aureus ATCC43300.DCFH-DA is initially non-fluorescent; however, it undergoes hydrolysis to DCFH upon entry into the cell.Subsequently, DCFH is converted to the fluorescent compound DCF when ROS are produced in the cells.The non-membrane-targeted antibiotic ciprofloxacin (MIC = 0.25 µg/mL, MBC = 2 µg/mL against S. aureus ATCC43300), membrane-targeted peptide melittin (MIC = 8 µg/mL, MBC = 32 µg/mL against S. aureus ATCC43300), and PBS were used as controls.

Cell Membrane Permeabilization
The cell membrane permeabilization of compound 9h against S. aureus ATCC43300 was assessed by PI uptake assay.PI could only pass through the impaired cell membrane and had been widely used to evaluate the cell membrane permeability.When the cell membranes were disrupted, PI entered into the cell membrane, leading to a significant improvement in fluorescence intensity.Therefore, the degree of membrane damage could be reflected by observing the fluorescence intensity in S. aureus.According to Figure 7B, the fluorescence intensity in the control group was weak and remained unchanged, whereas the fluorescence intensity of PI progressively increased when different concentrations of 9h were added.Compared with the control group, the fluorescence intensity (at 20 min) increased approximately 5-fold after the addition of 9h (16 µg/mL).The abovementioned results demonstrated that compound 9h had a strong ability of cell membrane permeabilization in a concentration-dependent manner.

Reactive Oxygen Species (ROS) Measurement
It is known that the disruption of bacterial cell membranes often causes the production of ROS in bacterial cells [31,52].The fluorescent probe DCFH-DA was used to monitor the level of intracellular ROS produced by S. aureus ATCC43300.DCFH-DA is initially non-fluorescent; however, it undergoes hydrolysis to DCFH upon entry into the cell.Subsequently, DCFH is converted to the fluorescent compound DCF when ROS are produced in the cells.The non-membrane-targeted antibiotic ciprofloxacin (MIC = 0.25 µg/mL, MBC = 2 µg/mL against S. aureus ATCC43300), membrane-targeted peptide melittin (MIC = 8 µg/mL, MBC = 32 µg/mL against S. aureus ATCC43300), and PBS were used as controls.As demonstrated in Figure 8A, compound 9h was able to effectively increase the level of ROS in S. aureus cells in a concentration-dependent manner.The fluorescence intensity of 9h was about 8-fold higher than the PBS-treated group when the concentration of 9h increased 16 µg/mL.Melittin was also able to dose-dependently increase the fluorescence intensity, whereas the ciprofloxacin-treated group showed little change in fluorescence intensity.The above result suggests that compound 9h could disrupt bacterial cell membranes, thus inducing the production of ROS.
ROS in S. aureus cells in a concentration-dependent manner.The fluorescence intensity of 9h was about 8-fold higher than the PBS-treated group when the concentration of 9h increased 16 µg/mL.Melittin was also able to dose-dependently increase the fluorescence intensity, whereas the ciprofloxacin-treated group showed little change in fluorescence intensity.The above result suggests that compound 9h could disrupt bacterial cell membranes, thus inducing the production of ROS.

Leakage of DNA and Protein
The destruction of bacterial cell membranes usually causes the leakage of macromolecules, including DNA and proteins [53][54][55].To investigate the effect of the compound 9h on the leakage of cytoplasmic contents, we performed DNA and protein leakage assays against S. aureus ATCC43300.The degree of damage to the bacterial cell membrane could be reflected by the concentrations of DNA and protein.The non-membrane-targeted antibiotic ciprofloxacin, membrane-targeted peptide melittin, and PBS were used as controls.As depicted in Figure 8B, compound 9h was able to cause the leakage of S. aureus contents (DNA) in a dose-dependent manner, exhibiting a similar action to that of melittin.The concentration of DNA excreted from S. aureus was significantly increased, and the concentration was approximately 5.5-fold higher than that of the blank control when the concentration of 9h was 16 µg/mL.However, the DNA concentration in the ciprofloxacintreated group changed little.A similar result were also observed in the leakage of protein with S. aureus (Figure 8C).These findings suggest that compound 9h was able to disrupt the bacterial cell membrane, leading to the leakage of DNA and protein.
Altogether, the above results indicated that the xanthoangelol-derived compound 9h acted through a membrane-interrupting mechanism and was able to destroy the integrity of bacterial cell membranes by disrupting the polarized state of the membrane and increasing membrane permeability, which is accompanied by intracellular ROS production, as well as the leakage of DNA and proteins, thus accelerating bacterial death.

Chemistry
All chemicals used in this study were of analytical grade and sourced from reputable commercial suppliers.The progress of reactions was monitored by analytical thin-layer chromatography (TLC) on silica gel GF254 plates. 1 H NMR and 13 C NMR spectra of all compounds were determined using a Bruker (Billerica, MA, USA) Avance instrument (500

Leakage of DNA and Protein
The destruction of bacterial cell membranes usually causes the leakage of macromolecules, including DNA and proteins [53][54][55].To investigate the effect of the compound 9h on the leakage of cytoplasmic contents, we performed DNA and protein leakage assays against S. aureus ATCC43300.The degree of damage to the bacterial cell membrane could be reflected by the concentrations of DNA and protein.The non-membrane-targeted antibiotic ciprofloxacin, membrane-targeted peptide melittin, and PBS were used as controls.As depicted in Figure 8B, compound 9h was able to cause the leakage of S. aureus contents (DNA) in a dose-dependent manner, exhibiting a similar action to that of melittin.The concentration of DNA excreted from S. aureus was significantly increased, and the concentration was approximately 5.5-fold higher than that of the blank control when the concentration of 9h was 16 µg/mL.However, the DNA concentration in the ciprofloxacin-treated group changed little.A similar result were also observed in the leakage of protein with S. aureus (Figure 8C).These findings suggest that compound 9h was able to disrupt the bacterial cell membrane, leading to the leakage of DNA and protein.
Altogether, the above results indicated that the xanthoangelol-derived compound 9h acted through a membrane-interrupting mechanism and was able to destroy the integrity of bacterial cell membranes by disrupting the polarized state of the membrane and increasing membrane permeability, which is accompanied by intracellular ROS production, as well as the leakage of DNA and proteins, thus accelerating bacterial death.

Chemistry
All chemicals used in this study were of analytical grade and sourced from reputable commercial suppliers.The progress of reactions was monitored by analytical thin-layer chromatography (TLC) on silica gel GF254 plates. 1 H NMR and 13 C NMR spectra of all compounds were determined using a Bruker (Billerica, MA, USA) Avance instrument (500 MHz) with solvents CDCl 3 , DMSO-d 6 , or CD 3 OD.A Thermo (Waltham, MA, USA) DFS mass spectrometer was used to identify the HRMS of target compounds.The purity of all final compounds was verified to exceed 95% through HPLC analysis.

Hemolysis Assay
Sheep blood was used to isolate red blood cells (RBCs), which were then resuspended in 1× PBS (5%).The compounds were dissolved in PBS to give an initial concentration of 5.12 mg/mL, compound solution (100 µL) was added to the first well of a 96-well plate, twofold serial dilutions of compounds were prepared in 96-well plates, and then 150 µL of 5% RBC suspension was added to each well.This was followed by incubation at 37 • C for 1 h.After incubation, the plate was centrifuged (3500 rpm, 5 min), and 100 µL of supernatant solution from each well was carefully pipetted into a new plate.The absorbance values were measured at 540 nm.Erythrocyte hemolysis rate = (Absorbance compound -Absorbance negative control )/(Absorbance 1% Triton X-100 − Absorbance negative control ) × 100%.All biological experiments were repeated three times.

Bactericidal Activity in Plasma and Complex Mammalian Fluids
Compounds (1×, 2×, 4×, 8×, and 16× MIC) were mixed with 50% plasma, and the mixture was preincubated at 37 • C for 0, 3, and 6 h.MBCs of the compounds were tested.Furthermore, MBCs of compounds in 50% plasma, 50% serum, and 50% blood suspension were tested.The MBC value is defined as the concentration at which the colony count is less than 5 after 18 h of incubation at 37 • C. All biological experiments were repeated three times.

Resistance Development Assay
Resistance was defined as a more than 4-fold increase in the tested compound in MIC when compared to the beginning value.Briefly, S. aureus ATCC43300 was cultured (37 • C, 200 rpm) in MHB broth for 5 h and diluted to 1 × 10 6 CFU/mL.Then, the bacteria were cultured with compound 9h at a sub-MIC concentration (1/2 MIC) for 18 h.The MICs were determined according to this method for consecutive 20 days.All biological experiments were repeated three times.

Cell Membrane Permeabilization Assay
Mid-log S. aureus ATCC43300 was collected and then washed with PBS three times.Bacterial suspension (1.0 × 10 8 CFU/mL, 150 µL) was added to a black 96-well plate; following this, fluorescent dye PI (10 µM, 50 µL) was added to each well of plates, and they were preincubated in the dark for 30 min.After the incubation, the fluorescence intensity was detected every minute for 6 min using a microplate reader (λex/λem = 535 nm/615 nm).Then, compound 9h with different concentrations (1×, 2×, 4×, 8×, 16× MIC) was added to the wells.The fluorescence intensity was determined similarly for another 14 min.All biological experiments were repeated three times.

DAPI/PI Fluorescence Assay
Mid-log S. aureus ATCC43300 was incubated with compound 9h (4× MIC) for 2 h.Next, the mixture was centrifuged (4000 rpm, 5 min), and the supernatant was discarded and resuspended in 160 µL PBS (1 × 10 8 CFU/mL).After incubation, PI (10 µg/mL, 20 µL) and DAPI (10 µg/mL, 20 µL) were added into the bacterial suspension, and the suspension was further cultured at 4 • C in the dark for 30 min.A total of 20 µL of the suspension was taken out, pipetted on slides, and recorded using a laser confocal microscope (Axio Vert AI, Carl Zeiss AG, Oberkochen, Germany).

ROS Assay
Mid-log S. aureus ATCC43300 was collected and resuspended in PBS.Bacteria suspension (1.0 × 10 8 CFU/mL) was placed into a 96-well plate, and an equal volume of DCF-DA

Figure 2 .
Figure 2. Design concept for amphiphilic xanthoangelol-derived compounds by mimicking the structure and the biological function of AMPs.

Figure 2 .
Figure 2. Design concept for amphiphilic xanthoangelol-derived compounds by mimicking the structure and the biological function of AMPs.

Figure 2 .
Figure 2. Design concept for amphiphilic xanthoangelol-derived compounds by mimicking the structure and the biological function of AMPs.

Figure 3 .
Figure 3. Plasma stability and cytotoxicity of compound 9h.Plasma stability (A) and bactericidal activity in complex mammalian fluids (B), and in vitro cytotoxicity (C) of 9h toward LO2 cells.Data are expressed as mean ± standard deviation (n = 3).

Figure 3 .
Figure 3. Plasma stability and cytotoxicity of compound 9h.Plasma stability (A) and bactericidal activity in complex mammalian fluids (B), and in vitro cytotoxicity (C) of 9h toward LO2 cells.Data are expressed as mean ± standard deviation (n = 3).

Figure 3 .
Figure 3. Plasma stability and cytotoxicity of compound 9h.Plasma stability (A) and bactericidal activity in complex mammalian fluids (B), and in vitro cytotoxicity (C) of 9h toward LO2 cells.Data are expressed as mean ± standard deviation (n = 3).

Figure 6 .
Figure 6.Fluorescence and electron scanning microscopy.(A) SEM images of the cell membrane of S. aureus cells, scar bar: 1.00 um.(B) Fluorescence micrographs of S. aureus cells stained with DAPI and PI and treated with 9h, scar bar: 50 um.

Figure 6 .
Figure 6.Fluorescence and electron scanning microscopy.(A) SEM images of the cell membrane of S. aureus cells, scar bar: 1.00 um.(B) Fluorescence micrographs of S. aureus cells stained with DAPI and PI and treated with 9h, scar bar: 50 um.