The fractional inhibitory concentration index (FICI) was calculated according to Equation (1). If the MIC value was not obtained at the highest concentration measured due to poor antibacterial activity, the FICI was considered to be twice the value of the measurement limit. \* Combinations that showed synergistic effects are shaded in gray. Where there were multiple sets of combinations with low FICI values, they are listed in parentheses.

#### *3.6. Mechanism of Synergistic Activity of PapMA-3 with Antibiotics against CRAB*

3.6.1. Confirmation of Synergistic Effects between PapMA-3 and Antibiotics through Time Killing Assays

Time-killing assays of PapMA-3 alone or in combination with antibiotics against CRAB C1 were performed at those concentrations that showed synergistic effects (FICI < 0.5) in the checkerboard assay. As shown in Figure 6, at the MIC of PapMA-3 (16 <sup>µ</sup>g·mL−<sup>1</sup> ), peptide treatment completely killed CRAB C1 strains. At a PapMA-3 concentration of <sup>4</sup> <sup>µ</sup>g·mL−<sup>1</sup> , for which most combined treatments showed synergistic effects in checkboard assays, the peptide-only treatment did not show any bacterial killing for 4 h. However, when the six antibiotics were incubated at synergistic concentrations in combination with PapMA-3 at 4 <sup>µ</sup>g·mL−<sup>1</sup> (Table 3, Figure S1), meropenem (32 <sup>µ</sup>g·mL−<sup>1</sup> ) exhibited the most synergistic effect—all bacteria were killed within 1 h. Erythromycin (128 <sup>µ</sup>g·mL−<sup>1</sup> ), rifampin (8 <sup>µ</sup>g·mL−<sup>1</sup> ), and vancomycin (16 <sup>µ</sup>g·mL−<sup>1</sup> ) killed all bacteria within 2 h, while imipenem (4 <sup>µ</sup>g·mL−<sup>1</sup> ) and linezolid (64 <sup>µ</sup>g·mL−<sup>1</sup> ) killed all bacteria within 4 h. Therefore, these antibiotics, when combined with PapMA-3 (4 <sup>µ</sup>g·mL−<sup>1</sup> ), could completely and rapidly kill CRAB C1 in a synergistic manner (Figure 6).

#### 3.6.2. Visualization of CRAB C1 Membrane Disruption by PapMA-3 in Combination with Antibiotics Using a Field Emission Scanning Electron Microscope (FE-SEM)

To elucidate the antibacterial mechanism and synergistic effect, the membrane disruption of CRAB by PapMA-3, in combination with antibiotics at concentrations showing synergistic effects, were investigated using an FE-SEM. The changes in the membrane morphology of CRAB C1 were investigated either with PapMA-3 treatment alone or in combination with erythromycin. Figure 7A shows the intact CRAB C1 membrane, revealing that the morphology was maintained at a steady state of membrane integrity, with a smooth surface. As shown in Figure 7B–I, CRAB C1 gradually lost its membrane integrity after 30 min and 1 h as the PapMA-3 concentration increased (4–32 <sup>µ</sup>g·mL−<sup>1</sup> ). PapMA-3

treatment caused the CRAB membrane surface to become severely roughened and wrinkled, in proportion to the concentration of peptide (Figure 7C–I). Peptide treatment at its MIC (16 <sup>µ</sup>g·mL−<sup>1</sup> ) after 1 h caused severe damage, supporting the antibacterial mechanism of PapMA-3 via the membrane disruption of CRAB C1.

**Figure 6.** Time-killing curve of PapMA-3 and antibiotics at synergistic concentration against CRAB C1. *Y*-axis indicates CFU in log scale.

**Figure 7.** FE-SEM images of CRAB C1 treated with PapMA-3. (**A**) Only CRAB C1. (**B**–**E**) after incubation for 30 min with PapMA-3 at 4 (1/4 MIC), 8 (1/2 MIC), 16 (1 MIC), and 32 <sup>µ</sup>g·mL−<sup>1</sup> (2 MIC), respectively. (**F**–**I**) same experiments after incubation for 1 h, respectively.

The membrane integrity of CRAB C1 was not altered by erythromycin itself (128 <sup>µ</sup>g·mL−<sup>1</sup> ), which was lower than the MIC (Figure 8A,C). However, when CRAB C1 was co-treated with 4 <sup>µ</sup>g·mL−<sup>1</sup> PapMA-3 and 128 <sup>µ</sup>g·mL−<sup>1</sup> erythromycin, severe membrane disruption was observed at 2 h (Figure 8B,D). Therefore, PapMA-3 helped in the

entry of antibiotics through the cell membrane by sufficiently changing the morphology of the membrane. In addition, a combination of PapMA-3 and antibiotics resulted in more efficient membrane damage. These results agree with the result obtained from time killing assay (Figure 6).

**Figure 8.** FE-SEM images of CRAB C1 treated with erythromycin and PapMA-3. (**A**) after incubation for 1 h with erythromcyin (128 <sup>µ</sup>g·mL−<sup>1</sup> ; synergistic concentration) and (**B**) with erythromycin (128 <sup>µ</sup>g·mL−<sup>1</sup> ) + PapMA-3 (4 <sup>µ</sup>g·mL−<sup>1</sup> ). (**C**) After incubation for 2 h with erythromcyin (128 <sup>µ</sup>g·mL−<sup>1</sup> ; synergistic concentration) and (**D**) with erythromycin (128 <sup>µ</sup>g·mL−<sup>1</sup> ) + PapMA-3 (4 <sup>µ</sup>g·mL−<sup>1</sup> ).

To confirm the time-dependent outer membrane permeabilization by PapMA-3, we investigated the time required by PapMA-3 for the membrane permeabilization of outer membrane of CRAB C1 by monitoring NPN uptake at 4, 8, 16, and 32 <sup>µ</sup>g·mL-1 of PapMA-3 (Figure S6). Destabilization of the outer membrane by PapMA-3 caused the dye to enter the damaged CRAB C1 membrane, and fluorescence was increased rapidly in a concentrationdependent manner and saturated after 10 min, confirming that PapMA-3 disrupted rapidly outer membrane of CRAB.

#### *3.7. Synergistic Effects of PapMA-3 on Biofilm Inhibition*

Biofilms confer resistance to bacteria against their environment [57,58]. Biofilm formation can occur on an assortment of surfaces, including living tissues such as wounds and infected skin, as well as on prosthetic implants and various abiotic surfaces [59,60]. The rate of formation of biofilms is high in the case of *A. baumannii*, which is found in urinary catheter, bronchial epithelial cells, as well as abiotic surfaces [61]. Bacterial biofilms confer antibiotic resistance and reduce antibiotic penetrance [62].

Biofilm formation in CRAB C1 was inhibited by PapMA-3 combined with antibiotics (Figure 9). PapMA-3 exhibited a significantly superior biofilm inhibition activity against CRAB C1 compared with that of the other tested antibiotics, in a concentration-dependent manner. Biofilm inhibition was quantified by measuring the absorbance at 595 nm of the crystal violet-stained biofilms. Absorbance treated with 32 µg·mL of PapMA-3, imipenem, meropenem, rifampin, erythromycin, vancomycin, and linezolid were 0.15, 0.19, 0.27, 0.28, 0.74, 0.35, and 1.07, respectively (Figure 9A). The absorbance of biofilm formed by CRAB C1 without peptide or antibiotics served as control was 1.11. The percentage of biofilm inhibition caused by these antibiotics at 32 <sup>µ</sup>g·mL−<sup>1</sup> was 98.5, 88.9, 79.2, 77.5, 37.7, 77.2, and 4.4%, respectively, compared to the control (Figure 9B).

**Figure 9.** Anti-biofilm activity of PapMA-3 in combination with antibiotics. Biofilms were quantified by staining with crystal violet. (**A**) Absorbance of crystal violet-stained biofilms with treatment of PapMA-3 and antibiotics at a concentration range of 32 to 512 <sup>µ</sup>g·mL−<sup>1</sup> , assessed at 595 nm. (**B**) Confirmation of anti-biofilm activity against CRAB C1 at log scale concentrations (from 1 to 512 <sup>µ</sup>g·mL−<sup>1</sup> ) of PapMA-3 and antibiotics, comparative calculation result with CTL of 0% without peptide or antibiotics. (**C**) Synergistic anti-biofilm activities of PapMA-3 and antibiotics against CRAB C1, assessed based on absorbance at 595 nm. CRAB C1 without peptides or antibiotics served as the control (red). Statistical analysis was performed using two-way ANOVA with Dunnett's comparison test. The values are expressed as the mean ± SEM of three independent experiments and are statistically significant at \*\* *p* < 0.01 and \*\*\* *p* < 0.001. ns, not significant.

However, co-treatments comprising 4 <sup>µ</sup>g·mL−<sup>1</sup> PapMA-3 with antibiotics (4 <sup>µ</sup>g·mL−<sup>1</sup> imipenem, 32 <sup>µ</sup>g·mL−<sup>1</sup> meropenem, 8 <sup>µ</sup>g·mL−<sup>1</sup> rifampin, 128 <sup>µ</sup>g·mL−<sup>1</sup> erythromycin, <sup>16</sup> <sup>µ</sup>g·mL−<sup>1</sup> vancomycin, or 64 <sup>µ</sup>g·mL−<sup>1</sup> linezolid) showed synergistic effects (Table 3); the absorbance at 595 nm for these co-treatments were less than 0.20. Thus, it can be concluded that combining PapMA-3 with antibiotics can deliver superior therapeutic effects compared to using antibiotics alone, regarding the inhibition of biofilm formation. This occurred due to the effect of PapMA-3 on inducing the permeabilization of the bacterial membrane (Figure 9C).

## *3.8. Stability and Effects of PapMA-3 on Mammalian Cells Compared to That of Melittin*

3.8.1. Stability of PapMA-3 Compared to That of Melittin in the Presence of Human Serum

High stability is necessary for the in vivo efficacy of peptides. Peptides are degraded by proteases and other components in the serum; therefore, we measured the stability of PapMA-3 alone or in combination with imipenem in human serum to confirm its potential as an AMP candidate [25]. The antibacterial activity of PapMA-3 was reduced four-fold in the presence of 50% human serum in MH media (Table 4), while melittin lost antibacterial activity considerably. Checkerboard assays revealed that 4 <sup>µ</sup>g·mL−<sup>1</sup> PapMA-3 displayed an outstanding synergistic effect with 4 <sup>µ</sup>g·mL−<sup>1</sup> imipenem, exhibiting FICI value of 0.31 against CRAB C1 (Table 3). PapMA-3 in combination with 16 <sup>µ</sup>g·mL−<sup>1</sup> imipenem retained its antibacterial activity at 16 <sup>µ</sup>g·mL−<sup>1</sup> , even in the presence of 50% serum (Table 4). Even though PapMA-3 contains all L-amino acids in the sequence, these results ascertain the potential of PapMA-3 for therapeutic applications and combinational therapy can compensate the problems caused by the instability of peptide antibiotics in the serum.

**Table 4.** Measurement of serum stability of PapMA-3 and melittin against *E.coli*, *A.baumannii*, and CRAB C1.


3.8.2. Effects of PapMA-3 Compared to That of Melittin on Mammalian Cells

We investigated the effect of PapMA-3 on the mammalian cells, HEK-293, and HaCaT for 48 h to evaluate its cytotoxicity (Figure 10). Cell activities were monitored at 24 h and 48 h following the peptide treatment. At 32 <sup>µ</sup>g·mL−<sup>1</sup> , the cell proliferation and viability remained unaltered at 24 h and 48 h compared to that of the blank control. Even at 64 <sup>µ</sup>g·mL−<sup>1</sup> , viability was reduced to less than 20% at 24 h and at 48 h compared to the control. In contrast, treatment with melittin caused severe toxicity and significantly reduced viability at 24 h and 48 h, even at its MIC. Therefore, PapMA-3 could be a potent antibiotic peptide.

**Figure 10.** Cytotoxicity of PapMA-3 (**A**) Cytotoxicity of PapMA-3 against HEK-293 cells at 24 h and 48 h. (**B**) Cytotoxicity of PapMA-3 against HaCaT cell at 24 h and 48 h. Statistical analysis was performed

using two-way ANOVA with Dunnett's comparison test. The values are expressed as the mean ± SEM of three independent experiments and are statistically significant at \* *p* < 0.05; \*\*\* *p* < 0.001; ns, not significant.

#### *3.9. Binding Interactions of PapMA-3 with LPS as Studied by STD-NMR Spectroscopy and ITC*

STD NMR experiments were conducted to clarify the antibacterial mechanism of PapMA-3. To determine which residues in PapMA-3 were the most favorable to LPS binding, they were compared to <sup>1</sup>D <sup>1</sup>H NMR spectra of PapMA-2 (with Ala18) and PapMA-3 (with Trp18); a previously obtained spectrum of PapMA was also used [63]. The STD effect was determined using the spectral differences; it primarily constituted resonances belonging to peptide protons bound to LPS. Significant STD effects were identified in the aromatic ring region for Trp<sup>2</sup> , Phe<sup>5</sup> , and Trp<sup>18</sup> (in the region of 7.8–7.4 ppm). This confirmed that all aromatic residues at both the N- and C-termini had direct molecular interactions with LPS (Figure 11A,B). Furthermore, protons in aliphatic regions also showed an STD effect with LPS, confirming that PapMA-3 enacted antibacterial activity via strong LPS interactions, resulting in disruption of CRAB bacterial membrane.

**Figure 11.** Binding interaction of PapMA-3 with LPS. Saturation transfer difference (STD) NMR analysis of interaction between PapMA-3 and LPS in D2O at 298 K. (**A**) <sup>1</sup>D <sup>1</sup>H NMR spectra of 0.5 mM PapMA-3 plus 15 µM LPS (sample A), (**B**) STD NMR spectrum obtained on sample A at 298 K. (**C**) Isothermal titration calorimetry (ITC) measurement showing the binding affinity of 0.2 mM PapMA-3 to 0.025 mM LPS from E. coli O55:B5.

The binding affinity of PapMA-3 to LPS was further investigated using ITC, revealing that an exothermic process with strong electrostatic interactions occurred between PapMA-3 and LPS, with a binding affinity of 1.47 × 10−6 M at 37 ◦C (Figure 11C). The

STD-NMR spectroscopy and ITC results together confirmed that PapMA-3 exhibited antibacterial activity via its strong interaction with LPS; thereby, it can enhance the membrane permeability of conventional antibiotics.

#### **4. Discussion**

The discovery and advancement of antibiotics initially seemed to have effectively combated diseases caused by bacterial infections; however, the overuse of antibiotics has led to the emergence of MDR bacterial strains. As a countermeasure against resistant strains, multiple antibiotics can be used in combination. In clinical settings, this strategy is advantageous, as it can broaden the target spectra against pathogens and prevent the development of drug resistance by reducing the amounts of antibiotics used. Furthermore, combination therapy can decrease the toxicity by allowing lower doses of the combined harmful drugs to be used. Combination therapies for antibiotics that have recently been approved by the US Food and Drug Administration (FDA) include ceftolozane/tazobactam, ceftazidime/avibactam, and meropenem/vaborbactam; furthermore, imipenem/relebactam and aztreonam/avibactam remain under clinical research [64].

Many studies have explored the combination of AMPs and antibiotics. The emergence of resistant strains to carbapenem, which is an important antibiotic against Gram-negative bacteria, has intensified the need for new alternatives for the treatment of CRAB pathogens classified as critical MDR bacteria by WHO [2]. However, few studies have synergistically investigated the combined effects of AMPs and antibiotics against Gram-negative bacteria, due to complications posed by the bacterial membranes. For example, Ω76 has been studied regarding its synergistic effects on CRAB; an FICI value of 0.56 was obtained with colistin [54], demonstrating a partial synergistic effect via a synergistic mechanism that enhanced the membrane permeability of antibiotics. The combination of melittin and doripenem has also shown a very good synergistic combination, achieving a FICI value of <0.1 against CRAB, whereas melittin was found not to exhibit a synergistic effect with doxycycline and colistin [25]. However, the severe toxicity of melittin can limit the clinical application. SET-M33 has showed synergistic effects with aztreonam, meropenem, rifampin, and tobramycin against CRAB strain [24].

In clinical trials, combinations of colistin and conventional antibiotics are mainly used to treat MDR Gram-negative bacteria [65,66]. Although colistin itself has excellent antibacterial activities, its high nephrotoxicity is a factor that limits its use alone; the appearance of colistin-resistant bacteria also limits its usage. For example, a randomized clinical trial of colistin in combination with meropenem is currently ongoing in Europe and the United States (ClinicalTrials.gov IDs NCT01732250 and NCT01597973) [67]. Additionally, clinical trials of colistin and rifampin in Korea have confirmed the presence of a partial synergistic effect (NCT03622918) [68]. However, in these studies, the combination treatments have not been shown to be superior to colistin monotherapy, as no similar or significant differences have been obtained [65,66].

Mechanisms of antibiotic resistance in bacteria include thickening the membrane to lower the permeability of antibiotics, creating an efflux pump to re-release antibiotics, modifying the target of antibiotics, and inactivating antibiotics by decomposing them [69]. Carbapenem antibiotics are members of β-lactam antibiotics, which inhibit synthesis of bacterial cell wall by binding to penicillin-binding proteins. Furthermore, carbapenem resistance mechanisms have been described in *A. baumannii*, including the alteration or loss of outer membrane proteins and efflux modifications [70]. Among many carbapenemhydrolyzing oxacillinase-encoding genes, OXA-23 is widespread in Korea, and the number of antibiotics available to treat CRAB are decreasing [71]. The present study aimed to find an efficient treatment method for CRAB infections using combinational therapy of the newly designed PapMA-3 and six conventional antibiotics, which included antibiotics that are potent against Gram-negative or Gram-positive bacteria. PapMA-3-antibiotic combinations were assessed against five clinical isolates, OXA-23-producting CRAB (C1–C5), and the underlying mechanism was explored.

To facilitate the uptake of antibiotics through the LPS outer membrane, PapMA-3 showed strong interactions with LPS and depolarized the CRAB outer membrane, while demonstrating low cytotoxicity. Its binding interactions with LPS were investigated using BC displacement assays, ITC, and STD-NMR experiments, confirming that membrane permeabilization via strong binding to LPS was the major antibacterial mechanism. PapMA-3 showed a superior BC displacement to a well-known LPS-neutralizing peptide, polymyxin B, by binding the core part of LPS, lipid A [72,73]. The therapeutic potential of PapMA-3 against CRAB was examined in combination with imipenem and meropenem, which are effective against Gram-negative bacteria. Furthermore, PapMA-3 was also combined with four antibiotics that have demonstrated antibacterial activity against Gram-positive bacteria. Outstanding synergistic effects (FICI < 0.5) between PapMA-3 and all six antibiotics were confirmed against both CRAB C1 and C4 clinical isolates. In particular, combining PapMA-3 with rifampin, vancomycin, and erythromycin achieved efficient synergistic effects against CRAB C4, with FICI values of <0.25, implying that PapMA-3 disrupted the membrane integrity of CRAB, allowing the antibiotics that are effective against Gram-positive bacteria to enter and reach their intracellular targets in the CRAB cells. Additionally, PapMA-3 might help imipenem and meropenem to overcome the CRAB membrane; however, underlying mechanism is not yet clearly understood.

Biofilm formation by MDR bacteria aids antibiotic resistance; it needs to be overcome due to its effects in causing pneumonia, meningitis, bacteremia, wounds, and soft-tissue infections [74]. PapMA-3 itself was able to suppress biofilm formation at its MIC, but it was also able to suppress sufficiently biofilm formation at lower concentrations when combined with antibiotics. This implies that combinational therapies constituting PapMA-3 and conventional antibiotics could be applied clinically. FE-SEM images suggested that PapMA-3 destabilized the morphology of the bacterial membrane even at concentrations below the MIC. Importantly, the CRAB membrane was destroyed when PapMA-3 was applied in combination with erythromycin, which alone are only effective against Gram-positive bacteria. Time killing assays suggested that the combinations of PapMA-3 with meropenem or erythromycin completely and rapidly killed CRAB C1 (within 1 h). Therefore, this combinational therapy could be applied to the enable the usage of potent antibiotics against both Gram-negative and Gram-positive bacteria by facilitating membrane permeability.

However, several problems persist that need to be addressed in future studies. First, the resistance to protease needs to be improved in our peptides by introducing D-amino acids, non-natural amino acids, or cyclization [75–77]. Additionally, these synergistic effects need to be confirmed by in vivo animal experiments before this combinational therapy can be applied clinically. Additionally, underlying mechanisms for synergistic effect on combination therapy should be investigated further in our future studies.

#### **5. Conclusions**

In this study, PapMA-3, a novel peptide, was designed and demonstrated potent anti-microbial activity against CRAB without notable cytotoxicity against mammalian cells. PapMA-3 was shown to target the outer bacterial membrane of CRAB via a strong interaction with LPS. At synergistic concentrations, PapMA-3 was found to cause the partial depolarization of the CRAB membrane, which changed the membrane morphology sufficiently to allow the antibiotics to penetrate intracellularly. This synergistic usage of PapMA-3 with well-known antibiotics resulted in the killing of CRAB and the inhibition of their biofilm formation. This was even achieved when the antibiotics used had previously only demonstrated potency against Gram-positive bacteria. This study may provide insights regarding the development of alternative therapies that utilize novel peptide antibiotics in combination with classical antibiotics to treat CRAB infections.

#### **6. Patents**

Patent applications for these peptides have been registered in Korea (101875057). These peptides have given rise to patent number PCT/KR2017/006650, and patent applications have been completed in United State (SOP114552US) and China (201780039278.1).

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/pharmaceutics13111800/s1, Figure S1: PapMA-3 and antibiotic checkboard assay results, showing fractional inhibitory concentration index (FICI) calculated against CRAB C1 according to Equation (1), Figure S2: PapMA-3 and antibiotic checkboard assay results, showing FICI calculated against CRAB C2., Figure S3: PapMA-3 and antibiotic checkboard assay results, showing FICI calculated against CRAB C3, Figure S4: PapMA-3 and antibiotic checkboard assay results, showing FICI calculated against CRAB C4, Figure S5: PapMA-3 and antibiotic checkboard assay results, showing FICI calculated against CRAB C5.

**Author Contributions:** Conceptualization, Y.K.; methodology, J.C. and Y.K.; data analysis, J.C. and A.J.; resources, Y.K.Y.; writing—original draft preparation, J.C. and Y.K.; writing—review and editing, J.C. and Y.K.; visualization, J.C. and A.J.; supervision, Y.K.; project administration, Y.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2020R1A2C2005338).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Pharmaceutics* Editorial Office E-mail: pharmaceutics@mdpi.com www.mdpi.com/journal/pharmaceutics

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34

www.mdpi.com ISBN 978-3-0365-7200-0