Antibacterial and Antibiofilm Activity and Mode of Action of Magainin 2 against Drug-Resistant Acinetobacter baumannii

Antimicrobial peptides (AMPs) are promising therapeutic agents for treating antibiotic-resistant bacterial infections. Previous studies showed that magainin 2 (isolated from African clawed fogs Xenopus laevis) has antimicrobial activity against gram-positive and gram-negative bacteria. The present study was conducted to investigate the antibacterial activity of magainin 2 against Acinetobacter baumannii. Magainin 2 showed excellent antibacterial activity against A. baumannii strains and high stability at physiological salt concentrations. This peptide was not cytotoxic towards HaCaT cells and showed no hemolytic activity. Biofilm inhibition and elimination were significantly induced in all A. baumannii strains exposed to magainin 2. We confirmed the mechanism of magainin 2 on the bacterial outer and inner membranes. Collectively, these results suggest that magainin 2 is an effective antimicrobial and antibiofilm agent against A. baumannii strains.


Introduction
The use of antibiotics for treating infections has resulted in the emergence of resistant strains [1]. New antibiotic-resistant strains are becoming a problem worldwide because conventional antibiotics cannot be administered. A recent report using data from the Infectious Diseases Society of America and hospital-based surveillance research refers to the pathogens as "ESKAPE" [2,3]. The ESKAPE pathogens include Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter, Pseudomonas aeruginosa, and Enterobacter species [4]. The ability of these bacteria groups to withstand antibiotic treatment is a major cause of hospital infections worldwide [5]. Notably, Acinetobacter baumannii has been reported as one of the most serious ESKAPE organisms resistant to antibiotics [6].
Acinetobacter baumannii, a gram-negative bacterium, is an opportunistic bacterial pathogen and has a high morbidity and mortality, especially in intensive care units with high pathogen infections. Infectious diseases include urinary tract infection [7], meningitis [8], skin infection [9], bacteremia [10], and pneumonia [11]. Many strains causing infection show resistance to antibiotics such as

Circular Dichroism Measurements
Based on the predicted results shown in Figure 1B; we measured the secondary structure of magainin 2 by circular dichroism (CD) spectroscopy. Structure analysis of magainin 2 was performed in various concentrations of trifluoroethanol (TFE) and sodium dodecyl sulfate (SDS) solutions ( Figure 2). In 10 mM sodium phosphate buffer; the CD spectra of the peptide displayed a random coil. However; the peptide adopted a typical α-helical conformation with increasing TFE and SDS concentrations. The calculated α-helical contents of the peptide are shown in Table 2. As the TFE and SDS concentration increased; the α-helical contents increased from 6.4% to 50.8% and from 21.8% to 33.3%; respectively. These results demonstrate that magainin 2 forms a strong α-helix structure.

Antimicrobial Assay
The antimicrobial activity of the peptides against A. baumannii strains are summarized in Table  3. We compared the antibacterial effect of magainin 2 with those of melittin, buforin 2, ciprofloxacin,

Circular Dichroism Measurements
Based on the predicted results shown in Figure 1B; we measured the secondary structure of magainin 2 by circular dichroism (CD) spectroscopy. Structure analysis of magainin 2 was performed in various concentrations of trifluoroethanol (TFE) and sodium dodecyl sulfate (SDS) solutions ( Figure 2). In 10 mM sodium phosphate buffer; the CD spectra of the peptide displayed a random coil. However; the peptide adopted a typical α-helical conformation with increasing TFE and SDS concentrations. The calculated α-helical contents of the peptide are shown in Table 2. As the TFE and SDS concentration increased; the α-helical contents increased from 6.4% to 50.8% and from 21.8% to 33.3%; respectively. These results demonstrate that magainin 2 forms a strong α-helix structure. The projection was obtained from http://heliquest.ipmc.cnrs.fr/cgibn/ComputParam.py. Positively charged residues are represented in blue, while hydrophobic residues are shown as yellow circles. The N-terminal and C-terminal parts are represented in red letters "N" and "C". (B) Three-dimensional structure of magainin 2.

Circular Dichroism Measurements
Based on the predicted results shown in Figure 1B; we measured the secondary structure of magainin 2 by circular dichroism (CD) spectroscopy. Structure analysis of magainin 2 was performed in various concentrations of trifluoroethanol (TFE) and sodium dodecyl sulfate (SDS) solutions ( Figure 2). In 10 mM sodium phosphate buffer; the CD spectra of the peptide displayed a random coil. However; the peptide adopted a typical α-helical conformation with increasing TFE and SDS concentrations. The calculated α-helical contents of the peptide are shown in Table 2. As the TFE and SDS concentration increased; the α-helical contents increased from 6.4% to 50.8% and from 21.8% to 33.3%; respectively. These results demonstrate that magainin 2 forms a strong α-helix structure.

Antimicrobial Assay
The antimicrobial activity of the peptides against A. baumannii strains are summarized in Table  3. We compared the antibacterial effect of magainin 2 with those of melittin, buforin 2, ciprofloxacin,

Antimicrobial Assay
The antimicrobial activity of the peptides against A. baumannii strains are summarized in Table 3. We compared the antibacterial effect of magainin 2 with those of melittin, buforin 2, ciprofloxacin, and gentamicin. Magainin 2 showed strong antibacterial activity with minimum inhibitory concentrations (MICs) of 4 and 2 µM against the standard strain (Korea Collection for Type Cultures (KCTC) 2508) and drug-resistant strains, respectively. This is similar to the MIC of melittin, which is known to have strong antibacterial activity. The activity of buforin 2 was 8-fold lower, ranging from 8 to 16 µM. Antibiotics showed activity against drug-resistant strains at 128 µM. Particularly, gentamicin showed low antibacterial activity with an MIC of 256 µM.

Cytotoxicity Assay
Hemolysis and cell viability assays were conducted to measure the toxicity of the peptides in mammalian cells. As shown in Figure 3A, melittin, a positive control peptide, induced more than 50% hemolysis at concentrations of 1-2 µM. In contrast, magainin 2-treated cells showed no hemolysis at 64 µM. We confirmed the cytotoxicity of magainin 2 in HaCaT cells. The curve in Figure 3B shows that high concentrations of magainin 2 exhibited low cytotoxicity of 0%, while melittin exhibited 100% toxicity at 2 µM. These results confirm that magainin 2 is not toxic to cells.

Cytotoxicity Assay
Hemolysis and cell viability assays were conducted to measure the toxicity of the peptides in mammalian cells. As shown in Figure 3A, melittin, a positive control peptide, induced more than 50% hemolysis at concentrations of 1-2 µM. In contrast, magainin 2-treated cells showed no hemolysis at 64 µM. We confirmed the cytotoxicity of magainin 2 in HaCaT cells. The curve in Figure  3B shows that high concentrations of magainin 2 exhibited low cytotoxicity of 0%, while melittin exhibited 100% toxicity at 2 µM. These results confirm that magainin 2 is not toxic to cells.

Biofilm Inhibition Assay
To investigate whether magainin 2 can inhibit biofilm formation, the degree of biofilm formation by A. baumannii strains was confirmed in Mueller-Hinton broth (MHB) supplemented with glucose. As shown in Figure 4A, all strains formed biofilms and the A. baumannii 907233 strain had a biofilm mass of 3 at OD 595 nm. Compared to A. baumannii KCTC 2508 in the crystal violet staining assay, A. baumannii 907233 showed a 57.6% higher biofilm mass after 24 h of incubation ( Figure 4A). The ability of magainin 2 and antibiotics (ciprofloxacin, gentamicin) to prevent biofilm formation was compared. Magainin 2 and antibiotics inhibited the biofilm formation of A. baumannii KCTC 2508 at concentrations of 2-8 µM, whereas buforin 2 showed inhibition at 64 µM. In resistant strains, magainin 2 inhibited biofilm formation at a low concentration of 4 µM, while the antibiotics did not exhibit biofilm formation inhibition activity until 32 µM ( Figure 4B). The biofilm biomass was determined using the green dye SYTO9 and detected with the EVOS AUTO2 fluorescence

Biofilm Inhibition Assay
To investigate whether magainin 2 can inhibit biofilm formation, the degree of biofilm formation by A. baumannii strains was confirmed in Mueller-Hinton broth (MHB) supplemented with glucose. As shown in Figure 4A, all strains formed biofilms and the A. baumannii 907233 strain had a biofilm mass of 3 at OD 595 nm. Compared to A. baumannii KCTC 2508 in the crystal violet staining assay, A. baumannii 907233 showed a 57.6% higher biofilm mass after 24 h of incubation ( Figure 4A). The ability of magainin 2 and antibiotics (ciprofloxacin, gentamicin) to prevent biofilm formation was compared.
Magainin 2 and antibiotics inhibited the biofilm formation of A. baumannii KCTC 2508 at concentrations of 2-8 µM, whereas buforin 2 showed inhibition at 64 µM. In resistant strains, magainin 2 inhibited biofilm formation at a low concentration of 4 µM, while the antibiotics did not exhibit biofilm formation inhibition activity until 32 µM ( Figure 4B). The biofilm biomass was determined using the green dye SYTO9 and detected with the EVOS AUTO2 fluorescence microscope. Acinetobacter baumannii 907233 formed biofilm showing strong green fluorescence ( Figure 5). Following treatment of the biofilm with magainin 2, SYTO9 fluorescence decreased in a concentration-dependent manner, while this was not observed following treatment with the other antibiotics. This data shows that magainin 2 has strong antibiofilm activity against drug-resistant A. baumannii. microscope. Acinetobacter baumannii 907233 formed biofilm showing strong green fluorescence ( Figure 5). Following treatment of the biofilm with magainin 2, SYTO9 fluorescence decreased in a concentration-dependent manner, while this was not observed following treatment with the other antibiotics. This data shows that magainin 2 has strong antibiofilm activity against drug-resistant A. baumannii.

Biofilm Reduction Assay
The biofilm reduction activity of magainin 2 was confirmed against A. baumannii 907233, which easily forms biofilm. Figure 6 shows the effect of the peptide and antibiotics on biofilm formed for 24 h. In the crystal violet staining assay, A. baumannii 907233 showed strong staining after incubation for 24 h ( Figure 6B). Antibacterial treatment with magainin 2 and ciprofloxacin exhibited similar abilities to eradicate biofilm formation. The biofilm inhibition rates of magainin 2 were 33.3%, 53.4%, and 66.2% at 128, 192, and 256 µM, respectively. In contrast, gentamicin showed no biofilm inhibition effects at 256 µM ( Figure 6A). Next, the biofilm was observed using SYTO9 (green dye). After 24 h, microscope. Acinetobacter baumannii 907233 formed biofilm showing strong green fluorescence ( Figure 5). Following treatment of the biofilm with magainin 2, SYTO9 fluorescence decreased in a concentration-dependent manner, while this was not observed following treatment with the other antibiotics. This data shows that magainin 2 has strong antibiofilm activity against drug-resistant A. baumannii.

Biofilm Reduction Assay
The biofilm reduction activity of magainin 2 was confirmed against A. baumannii 907233, which easily forms biofilm. Figure 6 shows the effect of the peptide and antibiotics on biofilm formed for 24 h. In the crystal violet staining assay, A. baumannii 907233 showed strong staining after incubation for 24 h ( Figure 6B). Antibacterial treatment with magainin 2 and ciprofloxacin exhibited similar abilities to eradicate biofilm formation. The biofilm inhibition rates of magainin 2 were 33.3%, 53.4%, and 66.2% at 128, 192, and 256 µM, respectively. In contrast, gentamicin showed no biofilm inhibition effects at 256 µM ( Figure 6A). Next, the biofilm was observed using SYTO9 (green dye). After 24 h,

Biofilm Reduction Assay
The biofilm reduction activity of magainin 2 was confirmed against A. baumannii 907233, which easily forms biofilm. Figure 6 shows the effect of the peptide and antibiotics on biofilm formed for 24 h. In the crystal violet staining assay, A. baumannii 907233 showed strong staining after incubation for 24 h ( Figure 6B). Antibacterial treatment with magainin 2 and ciprofloxacin exhibited similar abilities to eradicate biofilm formation. The biofilm inhibition rates of magainin 2 were 33.3%, 53.4%, and 66.2% at 128, 192, and 256 µM, respectively. In contrast, gentamicin showed no biofilm 6 of 14 inhibition effects at 256 µM ( Figure 6A). Next, the biofilm was observed using SYTO9 (green dye). After 24 h, the biofilm stained with SYTO9 showed strong green fluorescence. The fluorescence was low in the presence of 256 µM magainin 2, indicating that it effectively disrupted the biofilm. Gentamicin showed no decrease in biofilm ( Figure 6C). These results suggest that magainin 2 not only inhibits but also effectively eliminates biofilm. the biofilm stained with SYTO9 showed strong green fluorescence. The fluorescence was low in the presence of 256 µM magainin 2, indicating that it effectively disrupted the biofilm. Gentamicin showed no decrease in biofilm ( Figure 6C). These results suggest that magainin 2 not only inhibits but also effectively eliminates biofilm.

Salt Sensitivity Assay
To investigate the effects of salts on antimicrobial activity, antimicrobial activity was measured at a physiological salt concentration ( Table 4). As shown in Table 3, magainin 2 exhibited similar stability in the presence of physiological salts compared to buforin 2. The MIC values of A. baumannii KCTC 2508 and A. baumannii 907233 strain were 4 and 2 µM, respectively. In the presence of NaCl and FeCl3, magainin 2 activity was nearly maintained in both strains. In contrast, magainin 2 activity was slightly reduced to 8 µM in the presence of MgCl2.

Salt Sensitivity Assay
To investigate the effects of salts on antimicrobial activity, antimicrobial activity was measured at a physiological salt concentration ( Table 4). As shown in Table 3, magainin 2 exhibited similar stability in the presence of physiological salts compared to buforin 2. The MIC values of A. baumannii KCTC 2508 and A. baumannii 907233 strain were 4 and 2 µM, respectively. In the presence of NaCl and FeCl 3 , magainin 2 activity was nearly maintained in both strains. In contrast, magainin 2 activity was slightly reduced to 8 µM in the presence of MgCl 2 .

Mechanism of Action Of Magainin 2
To investigate the mechanism of magainin 2 activity towards drug-resistant bacteria, A. baumannii 907233 was used, while A. baumannii KCTC 2508 as a control strain. First, the outer membrane permeability of A. baumannii 907233 following treatment with magainin 2 was examined using NPN dye (hydrophobic fluorescent dye, exhibits fluorescence intensity in a hydrophobic environment). The bacterial outer membrane was disrupted, and the hydrophobic part of the membrane was exposed, after which NPN reacted to show fluorescence. As shown in Figure 7A,B, the intensity increased in a concentration-dependent manner, indicating that magainin 2 acts on the outer membrane. For A. baumannii KCTC 2508, the fluorescence intensity rapidly increased to 268, 246, and 209 within 2 min at 4×, 2×, and 1× MIC, respectively. In contrast, fluorescence increased to approximately 80 in A. baumannii 907233.
To investigate the mechanism of magainin 2 activity towards drug-resistant bacteria, A. baumannii 907233 was used, while A. baumannii KCTC 2508 as a control strain. First, the outer membrane permeability of A. baumannii 907233 following treatment with magainin 2 was examined using NPN dye (hydrophobic fluorescent dye, exhibits fluorescence intensity in a hydrophobic environment). The bacterial outer membrane was disrupted, and the hydrophobic part of the membrane was exposed, after which NPN reacted to show fluorescence. As shown in Figure 7A,B, the intensity increased in a concentration-dependent manner, indicating that magainin 2 acts on the outer membrane. For A. baumannii KCTC 2508, the fluorescence intensity rapidly increased to 268, 246, and 209 within 2 min at 4×, 2×, and 1× MIC, respectively. In contrast, fluorescence increased to approximately 80 in A. baumannii 907233.
The ability to depolarize the A. baumannii 907233 inner membrane in the presence of magainin 2 was further confirmed using DiSC3-5. DiSC3-5 is a dye that accumulates in the cytoplasmic membrane of bacteria; the dye is released when the bacteria membrane is affected by the peptide. As shown in Figure 7C,D, addition of magainin 2 resulted in increased fluorescence in a dose-dependent manner, indicating the ability of magainin 2 to depolarize the membrane. The fluorescence intensity of A. baumannii KCTC 2508 was higher than that of A. baumannii 907233 strain and decreased at 60 min after peptide treatment. These results suggest that magainin 2 affects the inner membranes of both A. baumannii KCTC 2508 and 907233.

Discussion
Bacterial infections are among the major causes of death worldwide [28]. Conventional antibiotics were effectively used to treat pathogenic infections for half a century, but the emergence of antibiotic-resistant bacteria has made these infections difficult to treat [29,30]. Generally, long-term exposure to antibiotics leads to resistance. Antibiotics function against proteins that are essential to bacteria cells, and mutations in bacterial proteins cause antibiotic resistance. Bacteria can also The ability to depolarize the A. baumannii 907233 inner membrane in the presence of magainin 2 was further confirmed using DiSC 3 -5. DiSC 3 -5 is a dye that accumulates in the cytoplasmic membrane of bacteria; the dye is released when the bacteria membrane is affected by the peptide. As shown in Figure 7C,D, addition of magainin 2 resulted in increased fluorescence in a dose-dependent manner, indicating the ability of magainin 2 to depolarize the membrane. The fluorescence intensity of A. baumannii KCTC 2508 was higher than that of A. baumannii 907233 strain and decreased at 60 min after peptide treatment. These results suggest that magainin 2 affects the inner membranes of both A. baumannii KCTC 2508 and 907233.

Discussion
Bacterial infections are among the major causes of death worldwide [28]. Conventional antibiotics were effectively used to treat pathogenic infections for half a century, but the emergence of antibiotic-resistant bacteria has made these infections difficult to treat [29,30]. Generally, long-term exposure to antibiotics leads to resistance. Antibiotics function against proteins that are essential to bacteria cells, and mutations in bacterial proteins cause antibiotic resistance. Bacteria can also transform antibiotics and make them inefficient by producing specific enzymes. Thus, an antibacterial with a new mechanism of action differing from existing antibiotics is urgently needed. Antimicrobial peptides have distinctive mechanisms and are emerging as new therapeutic agents.
AMPs are classified according to their α-helical, β-sheet, and random coil structures; the α-helical structure is the most common [31]. The predicted magainin 2 structure was confirmed by CD spectroscopy to be a-helical in the bacteria membrane, mimicking the surrounding environment, and appeared as a random coil structure in an aqueous environment. In increasing concentrations of SDS and TFE, the percent of a-helix content of magainin 2 also increased. This suggests that the α-helical structure enabled magainin 2 to interact with the bacterial membrane, which is related to its mechanism of action [32].
Because bacteria isolated from hospital patients are resistant to antibiotics, treatment of A. baumannii infection is becoming increasingly difficult [33]. Therefore, the activity of magainin 2 towards bacteria isolated from patients was evaluated. Magainin 2 exhibited lower antibacterial activity against drug-resistant A. baumannii than that observed against the susceptible A. baumannii (KCTC 2508) and exhibited more potent activity than antibiotics (ciprofloxacin and gentamicin). Thus, magainin 2 is a more effective therapeutic agent than antibiotics, which are not resistant to AMPs. Additionally, magainin 2 is not toxic to mammalian cells.
Biofilms are important for the survival of bacteria in the environment, and biofilm formation has been correlated with drug-resistance and several virulence factors [34,35]. The antibiotic resistance of biofilms is significantly higher than that of planktonic bacteria [36,37]. These biofilm-related infections are more difficult to control and can lead to death. Acinetobacter baumannii is a pathogen that commonly forms biofilm [38,39]. We analyzed the biofilm formation of A. baumannii strains treated with a peptide. Peptides and antibiotics showed similar abilities to inhibit biofilm formation by susceptible A. baumannii. However, in the five resistant strains, magainin 2 showed higher antibiofilm activity than antibiotics at low concentrations. These results suggest that magainin 2 exerts strong antibiofilm activity against drug-resistant A. baumannii. Furthermore, we evaluated the efficacy of magainin 2 on A. baumannii within established biofilms. Removing biofilms formed by pathogenic microorganisms is difficult. We found that magainin 2 was more effective than antibiotics towards removing formed biofilms. Thus, magainin 2 may be effective for treating biofilm-related infectious diseases.
The antimicrobial activity of most AMPs is strongly inhibited by presence of high ionic concentrations, which is a disadvantage as therapeutic agents in the serum or other body fluids [40]. For example, the activity of human β-defensins is influenced by NaCl in the airway surface liquid of cystic fibrosis patients [41]. In the presence of NaCl, MgCl 2 , and FeCl 3 at concentrations similar to those in human bodily fluids, the activity of magainin 2 was maintained. These results demonstrate that magainin 2 is stable in high-salt environments, which did not affect the killing mechanism. Combined with previous toxicity data, this peptide may be useful for therapeutic applications in a physiological environment.
AMPs, which have cationic peptides, bind to the negatively charged bacterial membrane through electrostatic interactions [42]. Peptides can disrupt the membrane after binding or enter the membrane to form pores. Pore-forming mechanisms include models of "barrel-stave", "carpet", and "toroidal-pore" [42,43]. In a previous study, magainin 2 was shown to bind to the anionic lipid bilayers of the bacterial membrane via electrostatic attractions and form a pore, which eventually destroyed the membrane [44]. The peptide was suggested to form toroidal pores~80 Å diameter in DMPC/DMPG liposomes [45]. Thus, we investigated the mechanism of magainin 2 against A. baumannii strains in this study.
To determine the mechanism of action against A. baumannii strains, we used NPN and DiSC 3 -5 dyes. First, damage to the bacterial outer membrane was monitored by measuring the fluorescence intensity of NPN. Magainin 2 induced an increase in fluorescence in both stains. Compared to the drug-resistant strain, the response of susceptible A. baumannii was faster. Depolarization of the bacterial cytoplasmic membrane was monitored by measuring the fluorescence intensity of DiSC 3 -5. Consistent with the results of the NPN test, the fluorescence intensity decreased at 55 min, indicating that magainin 2 rapidly affected the susceptible A. baumannii. Collectively, magainin 2 has an α-helical structure, which can associate with the A. baumannii membrane and uses a mechanism in which both the outer and inner membranes are affected.

Microorganisms and Mouse Red Blood Cells
Acinetobacter baumannii KCTC 2508 was obtained from the KCTC. Other Acinetobacter baumannii 244752, 409081, 719705, 892199, and 907233 were antibiotic-resistant bacteria isolated from patients at Eulji University Hospital (Seoul, Korea). The study protocol was reviewed and approved by the institutional review board of Eulji Hospital (No. EMCS 2016). All patients gave their informed consent to participate in this study or the informed consent process was waived in accordance with the decision of the ethics committee of each hospital. The mouse used in this study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, and approved by the Committee on the Ethics of Animal Experiments (CIACUC2017-S0042; Chosun University, Gwangju, South Korea).

Peptide Synthesis and Sequence Analysis
The peptides were synthesized using the solid-phase-9-fluorenylmethoxycarbonyl method as reported previously [46] on a Rink amide 4-methylbenzhydrylamine resin using a Liberty microwave peptide synthesizer (CEM, Matthews, NY, USA). The purity and molecular weight of the peptide was confirmed by reversed-phase high-performance liquid chromatography and matrix-assisted laser desorption ionization-time of flight mass spectrometry. Projections of the predicted three-dimensional structures were constructed online using the Mobyle@RPBS bioinformatics portal (http://mobyle.rpbs. univ-paris-diderot.fr/cgi-bin/portal.py#welcome), whereas the HeliQuest site (http://heliquest.ipmc. cnrs.fr) was used to create helical wheel diagrams and determine the relative hydrophobic moments of the peptides.

Antimicrobial Activity Assay
The MIC of peptides was determined by the standard micro-dilution method [48] in 96-well microtiter plates. Briefly, A. baumannii strains were cultured in MHB media and prepared at 2 × 10 5 colony-forming units per milliliter (CFU/mL). The peptides were serially diluted to concentrations between 1 and 32 µM in 10 mM sodium phosphate buffer in a 96-well plate. The bacteria were mixed with serially diluted peptide in the 96-well plate and incubated at 37 • C for 18-24 h. The absorbance of the sample at 600 nm was measured using a microplate reader and repeated three times.

Hemolysis Assay
The hemolytic activity of peptides was evaluated using mouse red blood cells (RBCs). RBCs were washed three times with PBS (Phosphate Buffered Saline) until the supernatant was clear. The peptides were diluted to 64 µM and added to a 96-well plate. RBCs were added at a final concentration of 8% (v/v). After incubation for 1 h, the plate was centrifuged for 10 min and the absorbance of the supernatant was measured at 414 nm. The percentage of hemolysis was calculated using the following formula: RBCs suspended in PBS and 1% Triton X-100 represented zero hemolysis and 100% hemolysis, respectively.

Cytotoxicity Assay
The MTT assay was conducted to measure the cytotoxicity of the peptides towards HaCaT cells.
HaCaT cells were cultured in DMEM supplemented with 1% penicillin and 10% FBS at 37 • C with CO 2 . Briefly, a total of 2 × 10 4 cells/well were seeded into a 96-well plate, which was incubated overnight. Peptides serially diluted with DMEM at 0-64 µM were added to each well and reacted for 23 h. Next, 10 µL of 5 mg/mL MTT was added to each well, followed by incubation for 1 h. The supernatants were removed and dissolved by adding DMSO. Absorbance was measured at 570 nm [49]. The control was DMEM media without peptide. Cytotoxicity was calculated using the following formula: Cytotoxicity (%) = 100 − [(A 570 of peptide treated cells/A 570 of control) × 100]

Biofilm Inhibition Assay
To investigate the inhibitory effect of peptide on biofilm formation, A. baumannii strains were cultured in MHB [50]. Bacteria were diluted to 5 × 10 5 CFU/mL in MHB supplemented with 0.2% glucose, and then 90 µL of the bacterial suspension was mixed with 10 µL of the peptide in a 96-well plate for 24 h. The supernatant was then carefully removed, and the formed biofilm was fixed with 100% methanol for 10 min. After removing the methanol, the biofilms were stained with 0.1% crystal violet for 30 min. The plates were then washed with distilled water three times. Finally, the biofilms were completely dissolved in 95% ethanol and absorbance was measured at 595 nm using a Versa-Max microplate ELISA reader (Molecular Devices, Sunnyvale, CA, USA).

Biofilm Eradication Assay
To measure the removal effect of the formed biofilms, 100 µL aliquots of A. baumannii 907233 (5 × 10 5 CFU/mL) were incubated in MHB with 0.2% glucose and incubated for 24 h. The culture medium was removed, and the wells were carefully washed with PBS to remove planktonic bacteria. Peptide or antibiotics were added at up to 256 µM in MHB supplemented with 0.2% glucose for 24 h. The biofilms were stained with crystal violet for 30 min, washed three times with PBS, and dissolved in 95% ethanol.

Visualization of Biofilms
To visualize the inhibition and elimination effect of the peptide on the biofilm, SYTO9 dye was used [51]. The formed A. baumannii 907233 biofilm was fixed with 100% methanol and stained with SYTO9 dye for 30 min in the dark. Images were obtained using an EVOS FL Auto 2 fluorescence microscope (Invitrogen, Carlsbad, CA, USA).

Salts Sensitivity Assay
Acinetobacter baumannii KCTC 2508 and A. baumannii 907233 strains were diluted to 2 × 10 5 CFU/mL in MHB media. The peptide was serially diluted from 0 to 32 µM in the presence of physiological salts. The final concentrations of physiological salts were as follows: 50, 100, 150 mM NaCl, 0.5, 1, 2 mM MgCl 2 , and 2, 4, 8 µM FeCl 3 . After these treatments, the procedures were same as used for the MIC assay described above.

Mechanism of Action Analysis
Outer membrane permeabilization assay. Permeation of the bacterial outer membrane by the peptide was measured by conducting a 1-N-phenylnaphthylamine (NPN) uptake assay [52]. Acinetobacter baumannii KCTC 2508 and A. baumannii 907233 strains were cultured in MHB, washed, and suspended to 0.25 at OD 600 nm in 5 mM HEPES buffer. Each strain was mixed with NPN to a final concentration of 10 µM. Next, 50 µL of peptide (1×, 2×, or 4× MIC) was added to the mixture in a 96-well plate. The relative fluorescence intensity was measured over time using a Spectramax M3 spectrophotometer (Molecular Devices) at 420 nm.
Cytoplasmic membrane depolarization assay. The membrane potential-sensitive dye DiSC 3 -5 was used to measure cytoplasmic membrane depolarization [53]. Acinetobacter baumannii KCTC 2508 and A. baumannii 907233 strains were cultured in MHB and washed three times with 5 mM HEPES (pH 7.3) containing 20 mM glucose. The bacteria were resuspended to 0.05 at OD 600 in buffer (5 mM HEPES, 20 mM glucose, and 100 mM KCl) and incubated with 1 µM DiSC 3 -5. The fluorescence was stabilized for 1 h and the peptides were added to the mixture. The fluorescence was measured using a Spectramax M3 spectrophotometer with excitation at 622 nm and emission at 670 nm.

Conclusions
In this study, we confirmed that magainin 2 has an α-helix structure and showed strong antibacterial activity against A. baumannii including multidrug-resistant strains. Magainin 2 is not toxic towards mammalian cells and maintained its stability and excellent antibacterial activity in a biologically relevant salt environment. Magainin 2 is also effective for treating infections by inhibiting or eliminating biofilm formation. The mechanism of action of standard and drug-resistant strains is that magainin 2 acts on both the outer and inner membranes. Taken together, magainin 2 may be useful as a new antibacterial and antibiofilm agent.