Expanding the spectrum of antibiotics capable of killing multi-drug resistant Staphylococcus aureus and Pseudomonas aeru-ginosa

Abstract

Infections from antibiotic resistant Staphylococcus aureus and Pseudomonas aeruginosa are a serious threat because reduced antibiotic efficacy complicates treatment decisions and prolongs the disease state in many patients. To expand the arsenal of treatments against antimicrobial resistant (AMR) pathogens, 600-Da branched polyethylenimine (BPEI) can overcome antibiotic resistance mechanisms and potentiate β-lactam antibiotics against Gram-positive bacteria. BPEI binds cell wall teichoic acids and disables resistance factors from penicillin binding proteins PBP2a and PBP4. The present study describes a new mechanism of action for BPEI potentiation of antibiotics generally regarded as agents effective against Gram-positive pathogens but not Gram-negative bacteria. 600-Da BPEI is able to reduce the barriers to drug influx and facilitate the uptake of a non-β-lactam co-drug, erythromycin, that targets the intracellular machinery. Also, BPEI can suppress production of the cytokine interleukin IL-8 by human epithelial keratinocytes. This enables BPEI to function as a broad-spectrum antibiotic potentiator which expands the opportunities to improve drug design, antibiotic development, and therapeutic approaches against pathogenic bacteria, especially for wound care.

Graphical Abstract

Erythromycin is a macrolide antibiotic with efficacy against Gram-positive staphylococci but not Gram-negative pathogens. Nevertheless, MRSA upregulates efflux pumps that render erythromycin ineffective. Resistance is demonstrated in clinical MRSA isolates. Fortunately, small non-toxic amounts of 600-Da BPEI increase erythromycin uptake and restore antibiotic efficacy. Likewise, we demonstrate that 600-Da BPEI creates erythromycin susceptibility in a clinical isolate of multi-drug resistant Pseudomonas aeruginosa.

Introduction

Crossing the bacterial membrane is a difficult task for many antimicrobial drugs that must reach their intracellular targets of Gram-positive and Gram-negative bacteria. Improving antibiotic efficacy can be accomplished with potentiators and adjuvants comprising a vast array of different compounds and targets.^1–16 A common theme is weakening the cell envelope framework. The outermost portions of the cell envelope of Gram-negative and Gram-positive bacterial pathogens are under exploited weaknesses in antimicrobial resistance mechanisms.^17–21 Many efforts are focused on inhibitors to the cytoplasmic expression and/or the membrane translocation of essential proteins, enzymes, and precursors required for the assembly of molecules required for the cell-envelope machinery and architecture. These approaches may suffer from deleterious protein binding effects or have low solubility from hydrophobic properties necessary to cross the membrane barriers. Additionally, methods to overcome resistance are different depending on whether the pathogen is a Gram-positive or Gram-negative bacterium.

The divergent approaches to overcome resistance arise from the inherent nature of bacterial cell walls and their differing mechanisms of antibiotic resistance. The cytoplasm of Gram-positive bacterial cells is surrounded by a single phospholipid bilayer and this membrane is surrounded by a thick layer of peptidoglycan interlaced with anionic teichoic acids (Figure 1). However, the phospholipid membrane of Gram-negative bacteria is encased by a periplasm region that contains a thin peptidoglycan layer attached to an asymmetric outer membrane bilayer (Figure 1). The inner leaflet of the outer member contains phospholipids, but the outer leaflet contains lipopolysaccharides. Together, these layers comprise a formidable barrier to the influx and/or diffusion of antibiotics into the periplasm and cytoplasm to reach their drug targets. Approaches to disable resistance from β-lactamase enzymes and efflux pumps with inhibitors are often tailored for Gram-negative bacteria, such as multi-drug resistant Pseudomonas aeruginosa (MDR-PA), and rarely work against Gram-positive bacteria that lack these primary resistance mechanisms. Resistance in Gram-positive bacteria, for example methicillin-resistant Staphylococcus aureus (MRSA), is dominated by alternative means to continue the assembly and synthesis of peptidoglycan; thereby bypassing the activity of β-lactams.

Figure 1.

Graphical presentation of 600-Da BPEI’s mechanisms of action on Gram-positive and Gram-negative cell wall and membrane. Cationic BPEI not only binds anionic wall teichoic acid (WTA) to indirectly disable penicillin binding proteins PBP2a/4 (which only function properly by localization of WTA), it also electrostatically binds the phosphate heads of the lipid membrane, causing a partial loss of the permeability barrier. Consequently, BPEI can potentiate both β-lactams and non-β-lactams (those target intracellular machinery) against MRSA (Gram-positive). In Pseudomonas aeruginosa (Gram-negative), BPEI binds anionic LPS, creating new hydrophilic conduits to enhance drug-influx.

First-line antibiotics include the β-lactam class of antibiotics, considered to be among the safest antibiotics to use.^22–25 In 2015, the US had 269.3 million antibiotic prescriptions given by healthcare providers, which is equivalent to 838 prescriptions per 1000 people. Among them, β-lactams (i.e. penicillin, oxacillin, and amoxicillin) were the most popular prescribed antibiotics with amoxicillin at the top of the chart at 171 prescriptions per 1000 people.²⁶ Although MRSA is β-lactam resistant, disabling penicillin binding protein PBP2a and PBP4 renders MRSA susceptible. We reported that cationic 600-Da branched polyethylenimine (600-Da BPEI) accomplishes this objective by interacting with wall teichoic acid (WTA) that is essential for PBP2a/4 functionality.^27–30 For those with penicillin allergies, erythromycin and other broad-spectrum macrolides are prescribed as standard of care antibiotics, but clinical isolates of MRSA do not respond to erythromycin treatment. Here, we show that MRSA clinical isolates with erythromycin resistance can be rendered drug-susceptible when 600-Da BPEI is used to reduce the barriers to drug-influx. We also show that 600-Da BPEI potentiates erythromycin against clinical isolates of MDR-PA. This is noteworthy because erythromycin is regarded as an antibiotic without efficacy against Gram-negative bacteria, including those without antimicrobial resistance. The mechanism of action (MOA) involves binding with anionic sites of the bacterial cell envelope (lipopolysaccharide (LPS), wall teichoic acid (WTA), and lipoteichoic acid (LTA)) to create new hydrophilic conduits for erythromycin to reach the cytoplasm (Figure 1). 600-Da BPEI is hydrophilic and targets anionic sites on the cell envelope away from the alkyl chains of membrane bilayers. It reduces diffusion barriers to increase drug uptake and enables broad-spectrum efficacy against different bacterial species. Instead of acting as an antimicrobial agent itself, low concentrations of 600-Da BPEI potentiate erythromycin efficacy against clinical isolates of MRSA and MDR-PA. Additionally, BPEI reduces interleukin-8 (IL-8) cytokine production by primary human epithelial keratinocytes (HEKa) cells, suggesting another therapeutic application for wound care. These data also show that improving the efficacy of standard of care antibiotics, such as β-lactams and macrolides, against AMR Gram-positive and Gram-negative bacteria creates new opportunities to improve patient health and well-being.

Results and Discussion

Since the 1950s, erythromycin—a macrolide antibiotic—has been widely used as a substitute for β-lactams for penicillin-allergic patients. It is a first-line treatment for many pediatric infections.³¹ Because erythromycin targets protein synthesis instead of the cell wall, it could be effective against methicillin-resistant staphylococci if the drug could reach the cytoplasm. The ability of 600-Da BPEI to increase erythromycin susceptibility was determined with in vitro checkerboard assays in 96-well microtiter plates. Two clinical MRSA isolates have erythromycin MICs over 2000 μg/mL. This demonstrates the strong resistance of MRSA. Because erythromycin targets intracellular targets, resistance could lie with hindered drug uptake imposed by the cell wall peptidoglycan and teichoic acids. This barrier to drug uptake has been described for polymyxins.³² However, 600-Da BPEI binds to these sites^27–30 and this action improves the MRSA susceptibility to erythromycin. The MIC is 250 times lower in the presence of 16 μg/mL of 600-Da BPEI (Figure 2 and Table 1). This broadens the spectrum of potential anti-MRSA drugs because, as previously reported,^{27, 28, 33} 600-Da BPEI was able to eliminate β-lactam resistance in these MRSA isolates and their biofilms. Against the MDR-PA clinical isolate OU19, 16 μg/mL BPEI lowers the erythromycin MIC from 256 to 2 μg/mL (Figure 2 and Table 1). This demonstrates antibiotic potentiation against a formidable Gram-negative pathogen. As shown in Figure 1, potentiation by 600-Da BPEI relies on its interaction with different bacterial targets due to the different cell envelope architecture of MRSA and MDR-PA. Nevertheless, 600-Da BPEI can overcome both resistance barriers, and erythromycin potentiation by 600-Da BPEI is characterized as synergistic (Table 1). According to the EUCAST guidelines, a fractional inhibitory concentration index (FICI) is used to identify synergistic effects. FICI values can indicate synergy (FICI ≤ 0.5), additivity (0.5 < FICI < 1), or indifference (FICI ≥ 1).³⁴ Erythromycin and 600-Da BPEI have synergy against MRSA (FICI = 0.26 for OU6, 0.31 for OU11) and MDR-PA OU19 (FICI = 0.26).

Figure 2.

Checkerboard data presentation of bacterial growth inhibition from the combination of erythromycin and 600-Da BPEI. The MICs in these assays can be used to show synergy in the clinical isolates: MRSA OU6 (A, FICI = 0.26), MRSA OU11 (B, FICI = 0.31), and PA OU19 (C, FICI = 0.26). Red = cell growth; White = no growth. Each assay was performed in triplicate and the data presented above are the average of these assays.

Table 1:

Synergy of 600-Da BPEI and antibiotics against MRSA and MDR-PA clinical isolates

Strain	Antibiotic	MIC[a] BPEI (μg/mL)	MIC[a] Antibiotic (μg/mL)	MIC[a] Antibiotic with 16 μg/ml BPEI	FICI[b]	Outcome
MRSA OU6	Erythromycin	64	>2000	8	0.26	Synergy
MRSA OU11	Erythromycin	64	>2000	128	0.31	Synergy
PA OU19	Erythromycin	64	256	2	0.26	Synergy

^[a]MIC, Minimum Inhibitory Concentration;

^[b]FICI, the Minimum Fractional Inhibitory Concentration Index

Synergy between 600-Da BPEI and erythromycin is attributed to increased drug influx. Resistance to non-β-lactam antibiotics (such as macrolides, tetracyclines, and fluoroquinolones) involves membrane-bound efflux-pump proteins. These protein assemblies expel toxic substances (e.g. antibiotics), hindering accumulation of antibiotics in bacterial cells.^35–37 To examine increased drug influx, we tested the ability of 600-Da BPEI to increase the intracellular concentration of a fluorescence probe molecule, Hoechst 33342 bisbenzimide (H33342). H33342 fluorescence increases when it penetrates the cell-membrane and binds to intracellular DNA. Greater accumulation of H33342 in the cells creates a higher fluorescence intensity. Fluorescence measurements were taken for untreated and BPEI-treated samples immediately after adding the H33342 dye.

As shown in Figures 3, 600-Da BPEI enhanced dye uptake in MRSA OU6 and MRSA OU11. The BPEI-treated cells had much higher fluorescence intensity by approximately 10,000 fluorescence units compared to their untreated controls. Similar trends are observed for the influx of H33342 into cells of the MDR-PA clinical isolate OU19 (Figure 4). P. aeruginosa is well-known for a powerful drug-efflux system and thus the low accumulation of H33342, compared to Gram-positive MRSA, is not unexpected. Nonetheless, as with MRSA, 600-Da BPEI increases the uptake of H33342 in PA OU19.

Figure 3.

H33342 permeation curves show the addition of BPEI (128 μg/mL) enhances the cell-membrane permeability of MRSA OU6 and OU11 as the fluorescence of H33342 increased, compared to their untreated control. Error bars denote standard deviation (n = 5).

Figure 4.

P. aeruginosa PA OU19 dye uptake data show the addition of 128 μg/mL polymyxin B (PmB) increases the uptake H33342 and the effect over three times greater than that caused by 128 μg/mL BPEI. Error bars denote standard deviation (n = 6).

The concentration of BPEI used in fluorescence assays, 128 μg/mL, would appear to be a lethal concentration as it is greater than the MIC of each isolate. However, an important consideration is that generating fluorescent signals above the detection limit requires a higher cell density (~ 7 × 10⁹ CFU/mL) than in checkerboard assays (~ 5 × 10⁵ CFU/mL). Thus, while 128 μg/ml of 600-Da BPEI is lethal in the checkerboard assays, it is sub-lethal in the fluorescence studies. This is shown by measuring cell viability using a resazurin cellular metabolism assay. MRSA OU11 was grown until it reached the same cell density in H33342 assays (~ 7 × 10⁹ CFU/mL) before BPEI treatment. Resazurin was then added and, after 1 hour of incubation, the fluorescence intensity was measured. Cellular metabolic product NADH irreversibly reduces resazurin into resorufin, which emits strong fluorescence at 580–590 nm, indicating cell viability. As shown in Figure 6, MRSA cells were slightly less viable with higher concentrations of 600-Da BPEI, suggesting that the membrane remains intact. Previous studies showed that BPEI attached to the surface of MRSA cells.^{28, 30} Additionally, scanning electron micrographs of MRSA show that sub-lethal amount of 600-Da BPEI altered the cell wall morphology.^{27, 28, 30} These data support a mechanism of action (MOA) involving the ability of 600-Da BPEI to weaken the cell envelope rather than lysing the bacteria. This is a different MOA than that of polymyxin-B, a cationic antibiotic known to cause widespread disruption of bacterial membranes.

Figure 6.

Resazurin assay data used to evaluate cell viability of MRSA OU11 (at the cell density of ~ 7 × 10⁹ CFU/mL) treated with either BPEI or polymyxin B (PmB). Resazurin is converted to resorufin by cellular metabolism product NADH/H⁺ and thus provide an indication of cell viability. Error bars denote standard deviation (n = 8).

Using the resazurin assay, corresponding concentrations of polymyxin-B were more lethal to MRSA OU11 (Figure 6) than 600-Da BPEI. In fact, 64 μg/mL of polymyxin-B (PmB) caused greater cell death than 512 μg/mL 600-Da BPEI, and the highest concentration of PmB (512 μg/mL) resulted in complete loss of MRSA OU11 viability. These experiments highlight the low antibiotic propensity, but high potentiation ability, of 600-Da BPEI. These data also support the paradigm that PmB is considered a Gram-negative selective drug due to its low MICs (≤ 2 μg/mL), while Gram-positive bacteria require much higher concentration of PmB (≥ 32 μg/mL) due to the diffusion barrier imposed by the thick bacterial cell wall.³⁸

The different MOAs of 600-Da BPEI and PmB can be examined by gauging their effect on the influx of H33342.¹ The clinical isolate PA OU19 (Figure 4) and MRSA OU11 (Figure 5) were exposed to similar amounts of 600-Da BPEI and PmB. As shown, PmB dramatically increased the intracellular concentration of H33342 by disrupting the membrane bilayer using its hydrophobic alkyl tail. In contrast, 600-Da BPEI is hydrophilic and lacks the energetic driving force to penetrate the membrane. Thus, BPEI reduces drug diffusion barriers within the peptidoglycan layer of MRSA and LPS of PA without damaging the membrane. This MOA also explains why the rate of H33342 influx during the first few minutes is much higher for PmB than 600-Da BPEI. The ability of PmB to disrupt membrane layers aligns with the literature precedent that PmB is nephrotoxic and neurotoxic towards human cells.³⁹ In contrast, 600-Da BPEI is unlikely to damage the membranes^{1, 29, 40, 41} but instead reduces drug-influx barriers that allows faster diffusion through the bacterial membrane that allows enhanced H33342 accumulation in the MRSA and PA cells.

Figure 5.

H33342 permeation curves show the addition of 128 μg/mL polymyxin B (PmB) increases the dye uptake by MRSA OU11, more than twice of that caused by 128 μg/mL BPEI addition. Error bars denote standard deviation (n = 5).

The therapeutic potential for 600-Da BPEI is strong and has a foundation in previous work of its antibiotic^{40, 41} and drug-delivery characteristics. First reported in 1995, large-molecular-weight BPEIs (> 25,000 Da) strongly interaction with lipid bilayers and high transfection activity for gene delivery both in vitro and in vivo due to high N/P ratios (polycationic nitrogen binding to anionic phosphates in DNA or RNA).^42–46 Large BPEIs have higher transfection efficiency than small BPEIs (< 5 kDa). Although large PEIs are more beneficial than viral-vectors in gene therapy, larger molecular weights correspond to higher cytotoxicity due to more interactions with blood components.^47–49 Unlike large-molecular-weight BPEIs, 600-Da BPEI is non-cytotoxic²⁹ and furthermore it lacks the size and high charge ratio of N/P to be an effective gene delivery candidate. Unlike polypeptide antibiotics (i.e. polymyxin-B), 600-Da BPEI lacks the hydrophobic region necessary to -disrupt through the cytoplasmic membrane. However, as an antibiotic potentiator, the surface charge of 600-Da BPEI is sufficient to attach to anionic teichoic acids of the MRSA cell wall^27–30 and lipid phosphate groups of the membrane to create hydrophilic regions for drug intake, which explains its antibiotic synergy with erythromycin against the clinical isolates of MRSA. In a similar attraction force, 600-Da BPEI binds to anionic LPS of P. aeruginosa and thus opens more pathways for a co-drug to easily pass through the bacterial membrane.¹ Because of this new MOA, 600-Da BPEI may have broader applications than originally envisioned.^{27–30, 40, 41}

Staphylococci are notorious for their skin and soft-tissue infections that often lead to more complicated diseases. Each year, millions of acute skin and soft tissue infections (SSTIs) become chronic wound infections.^50–53 Instead of taking 3–6 weeks to heal, chronic wounds persist for 3–6 months. Delays in healing acute SSTIs are often due to a prolonged inflammatory phase of healing caused by bacterial debris, such as peptidoglycan from S. aureus, which is a Pathogen-Associated Molecular Pattern (PAMP) molecule. Preventing S. aureus peptidoglycan from triggering the release of inflammatory cytokines will restore the optimal inflammatory response.⁵⁴ However, successful drugs are elusive because the cell wall debris has a large variation in size and shape, making it virtually impossible to target peptidoglycan with monoclonal antibodies that recognize specific polysaccharide units. Instead, 600-Da BPEI binds the anionic sites of peptidoglycan and inhibits its ability to stimulate cytokine production. As shown in Figure 7, S. aureus peptidoglycan promotes the release of interleukin-8 (IL-8) from primary human epithelial keratinocytes (HEKa) cells. IL-8 is a cytokine and chemokine molecule involved with neutrophil recruitment to the wound site.⁵⁵ Its production is stimulated when peptidoglycan binds to, and is mainly recognized by, toll-like receptor 2 (TLR2).^56–60 In contrast, 600-Da BPEI does not promote IL-8 production by the HEKa cells (Figure 7A). However, when 600-Da BPEI and S. aureus peptidoglycan are added to HEKa cells, the amount of IL-8 diminishes (Figure 7B), suggesting another promising therapeutic benefit of BPEI for wound care. Studies to evaluate wound healing with in vivo and in vitro models are underway.

Figure 7.

ELISA data show the amount of cytokine IL-8 released by human epithelial keratinocytes (HEKa cells) in responses to: peptidoglycan (PGN) and 600-Da BPEI (A); combinations of PGN and 600-Da BPEI (B). Data are shown as average of triplicate trials. Error bars denote standard deviation. Statistical analysis with the student’s t-test generates p-values of < 0.05% (95% confidence, denoted by *) and <0.01 (99% confidence, denoted by **). nd = no statistical difference.

CONCLUSION

Experts predict that, by 2050, antimicrobial resistance (AMR) will be the leading cause of death, claiming 10 million lives a year—a figure that exceeds the number of deaths caused by cancer today. A swift global response is required to prevent this alarming scenario,⁶¹ but pharmaceutical companies are facing significant market pressures that hinder their ability to meet this need. The cost of bringing a drug to market is extraordinary, up to one billion dollars, yet there are little or no incentives for clinicians to use the new drug. New antibiotics are held in reserve as drugs of last resort to prevent the emergence of resistance. Instead, the paradigm of antibiotic potentiators has emerged to overcome resistance barriers and restore efficacy to existing antibiotics; thereby providing an opportunity to kill drug-resistant and drug-susceptible bacteria with the same formulation. However, antibiotic + potentiator combinations are being developed against Gram-negative pathogens or Gram-positive pathogens rather than a broad-spectrum formulation against both.

The data reported here demonstrate potentiation of erythromycin against Gram-positive and Gram-negative pathogens. They provide a better understanding of the 600-Da BPEI mechanisms of action against multidrug-resistant MRSA and P. aeruginosa that may lead to development of broad-spectrum antibiotic and potentiator combinations. Antibiotic combination therapy using existing drugs also reserves the newer last-resort antibiotics for use against the most serious cases of antibiotic-resistant infection. Yet, the need to couple 600-Da BPEI with an antibiotic for effective killing of AMR pathogens creates technical hurdles of reducing drug toxicity while matching the pharmacokinetics and pharmacodynamics (PK/PD) of the combination. These problems are mitigated when the combination is used as a topical agent against wound infections. We know that 600-Da BPEI dissolves biofilms^{1, 27, 33} and expands the possible classes of antibiotics for potentiation that increases therapeutic opportunities.¹ Chronic wound infections, those that have not proceeded through a reparative process in three months, affect millions of Americans each year^{50, 51} and are often caused by drug resistant bacteria, such as MRSA and MDR-PA. In the absence of a robust pipeline of new drugs, existing drugs and regimens have to be re-evaluated as combination(s) with potentiators that overcome biofilms and/or antibiotic resistance. Ideally, the potentiator should be a single compound with multi-function properties that disable biofilms and antibiotic resistance and possibly diminish inflammation. We envision wound treatment with antibiotics given topically, orally or intravenously, and external topical application of 600-Da BPEI to disable biofilms, overcome resistance mechanisms, and reduce inflammation. This mitigates concerns about toxicity and differences in the PK/PD of antibiotics vs. 600-Da BPEI. This may improve wound care outcomes by restoring potency to existing antibiotics with a single potentiator. Likewise, using an antibiotic potentiator, such as 600-Da BPEI, to lower the release of cytokines in response to peptidoglycan stimulation increases the therapeutic benefit of 600-Da BPEI. Efforts to evaluate the ability of 600-Da BPEI to modulate pro-inflammatory cytokine production in response to other PAMPs is currently underway. Reducing inflammation helps prevent many acute infections from becoming chronic wounds; and lowers the risk of recurrent infection and tissue necrosis.^{62, 63} that results in substantial morbidity, disability, hospitalization, and mortality, especially among older adults.⁶⁴

Experimental Section

Materials

Two MRSA clinical isolates (MRSA OU6 and MRSA OU11) and a clinical isolate of P. aeruginosa (PA OU19) from patient swabs were kindly provided by Dr. McCloskey from the University of Health Sciences Center with an institutional review board (IRB) approval. Chemicals (DMSO, growth media, erythromycin, polymyxin B, H33342 dye, and peptidoglycan from Staphylococcus aureus (product number 77140) were purchased from Sigma-Aldrich. 600-Da BPEI was purchased from Polysciences. HEKa cells (primary human epithelial keratinocytes), Epilife Medium, and growth supplement were purchased from Invitrogen. Human IL-8/CXCL8 Quantikine ELISA Kit was purchased from R&D.

The multi-drug resistance characteristics of the clinical isolates was confirmed using the Beckman Coulter MicroScan Walkaway™ 96plus with the PC33 gram positive panel. The MIC values are listed below.

• MRSA OU6: oxacillin > 2 μg/mL (resistant); clindamycin > 4 μg/mL (resistant); daptomycin ≤ 0.5 μg/mL (susceptible); erythromycin > 4 μg/mL (resistant); gentamicin ≤ 4 μg/mL (susceptible); linezolid 2 μg/mL (susceptible); tetracycline ≤ 4 μg/mL (susceptible); vancomycin 2 μg/mL (susceptible).

• MRSA OU11: oxacillin > 2 μg/mL (resistant); clindamycin > 4 μg/mL (resistant); daptomycin ≤ 0.5 μg/mL (susceptible); erythromycin > 4 μg/mL (resistant); gentamicin ≤ 4 μg/mL (susceptible); linezolid 2 μg/mL (susceptible); tetracycline ≤ 4 μg/mL (susceptible); vancomycin 1 μg/mL (susceptible).

• P. aeruginosa OU19: aztreonam 16 μg/mL (resistant); cefepime 16 μg/mL (resistant); ceftazidime 16 μg/mL (resistant); ciprofloxacin 2 μg/mL (resistant); gentamicin 4 μg/mL (susceptible); meropenem 8 μg/mL (resistant); piperacillin + tazobactam 64 μg/mL (resistant); tobramycin 4 μg/mL (susceptible).

Checkerboard assays: were conducted to identify synergy between BPEI and antibiotics against bacteria. Serial dilutions of antimicrobial agents (BPEI and antibiotic solutions) were added to a 96-well microtiter plate with 100 μL cation-adjusted Muller Hinton broth (CAMHB) per well. Untreated control and positive control (5% bleach) were also conducted. Bacterial inoculation (5 × 10⁵ CFU/mL) from an overnight culture was added to the plate (1 μL/well) and incubated at 37°C for 20 hr. The change in optical density at 600 nm (ΔOD₆₀₀) was measured using a Tecan Infinite M20 plate reader immediately after inoculation. Minimum inhibitory concentration (MIC) of each drug is determined as the lowest concentration that inhibited cell growth (ΔOD₆₀₀ < 0.05). Fractional inhibitory concentration index (FICI) and synergistic effects are determined using EUCAST guidelines: synergy (FICI ≤ 0.5), additivity (0.5 < FICI < 1), and indifference (FICI > 1).³⁴ Each assay was done in triplicate.

Cell-permeation Assay / bisBenzimide H33342 Intracellular Accumulation: Cryogenic stock of bacteria (MRSA OU6, OU11, MRSE 35984, or PA OU19) was inoculated overnight on tryptic soy agar. The culture was sub-inoculated in fresh tryptic soy broth (TSB) media for another 5–6 hours with shaking (100 rpm/min) at 37°C. Cells were pelleted by centrifugation at 7000 rpm for 40 min. The supernatant was discarded. The cells were resuspended in PBS and readjusted to OD₆₀₀ = 1.0. (which had a density of ~7 × 10⁹ CFU/mL). Aliquots of the cell suspension were transferred to a 96-well flat-bottom black plate (180 μL/well) including the controls of the solvent (PBS blank), the untreated cells, and treated samples (either BPEI treated or polymyxin-B treated). Five technical replicates of each group were conducted. Hoechst 33342 bisbenzimide (H33342) was added (20 μL) to each well (final concentration of 5 μM). Fluorescence was read right after adding the H33342 by a Tecan Infinite M20 plate reader with the excitation and emission filters of 355 and 460 nm, respectively. Fluorescence data were normalized to the emission before cells were added in the PBS control (BPEI did not change the fluorescence by itself), and they were plotted against time to show the cellular uptake of H33342 over 10 minutes.

Cell viability assays: with resazurin were performed with MRSA OU11 cells were grown in TSB as similar to the procedure of cell-permeation assays until they reach a density of ~7 × 10⁹ CFU/mL. Then the cell culture was transferred into a 96-well plate (100 μL/well) for BPEI or polymyxin B (PmB) treatments at varied concentrations from 64 – 512 μg/mL. Controls of untreated and positive control of 5% bleach were also conducted. The plate was incubated at 37°C overnight. Resazurin (50 μL; final concentration of 50 μg/mL) was then added and, after 1 hour of incubation, the fluorescence intensity was measured (λ_ex = 560 nm; λ_em = 590 nm).

IL-8 responses: HEKa cells (Invitrogen, Carlsbad, CA.) were seeded in T-75 tissue culture flasks with Epilife media supplemented with human keratinocyte growth supplement 100 ug/mL and 100 U/mL of pen/strep and incubated at 37°C with 5% CO₂. Fresh media was replaced every 2 days. When the cell confluence reached 80–90%, they were expanded into additional-75 tissue culture flasks T. To avoid cell senescent, all experiments were performed with cells at passage 3–7. HEKa cells were cultured in a new 24-well plate until 80–90% confluence (total volume = 1 mL/well). Then treatments of 600-Da BPEI (64, 128, 256, 512, and 1024 μg/mL) or S. aureus peptidoglycan (5 μg/mL) were added in triplicate cultures for 24 hr. The cell media was collected in 1.5 mL Eppendorf microtubes and stored at −20°C until ELISA assays were performed. Concentrations of IL-8 cytokine released into the media were quantified following the instructions of Quantikine Colorimetric ELISA assay kits (R&D Systems). Absorbance was measured at 450 nm and 570 nm. Final corrected absorbance was the subtraction at 450 nm from the one at 570 nm.