Fifteen K. pneumoniae isolates demonstrated resistance to most antimicrobial agents ( Figure 1 ). Notably, all isolates exhibited a resistant profile against amoxicillin/clavulanic acid (AMC), ceftazidime (CAZ), ceftriaxone (CRO), cefepime (FEP), cefoxitin (FOX), aztreonam (ATM), ertapenem (ETP), chloramphenicol (C) and trimethoprim/sulfamethoxazole (SXT). Additionally, all isolates were resistant to colistin (COL) with MICs ≥4 µg/mL. These results suggested that all isolates were deemed as MDR bacteria and some isolates were possibly XDR bacteria.

Among fifteen MDR/XDR K. pneumoniae clinical isolates, 7 clusters designated A to G and 14 different ERIC-PCR types were identified using a cut-off of 85% and 95% genetic similarity, respectively ( Figure 1 ). Cluster C, the most common cluster, contained an ERIC-PCR type (C1) of two closely related patterns. However, all isolates had different antibiotic resistance profiles, as shown in Figure 1 . Therefore, they were considered clonally unrelated strains, and none of the identical ERIC-PCR patterns was found among these isolates.

MIC and MBC values of K11 were determined against K. pneumoniae reference strains and fifteen MDR/XDR K. pneumoniae clinical isolates. As shown in Table 1 , K11 inhibited the growth of the reference strains and clinical isolates at concentrations of 2-4 μg/mL and 8-512 μg/mL, respectively. Moreover, the MBC values, which represented the bactericidal activity of K11, were identical to the MIC values or not greater than one dilution. Notably, the MIC and MBC values were higher against clinical isolates than reference strains. The MIC₅₀ and MIC₉₀ of K11 were 64 and 256 μg/mL, respectively.

The synergistic effects of the K11 combined with the antibiotics were determined as the fractional inhibitory concentration index (FICI). The results of FICI of K11 in combination with different antibiotics against all MDR/XDR K. pneumoniae clinical isolates are presented in Table 2 . The combination of K11 and chloramphenicol exhibited a synergistic effect against 80% of all the tested isolates. Additionally, the combination of K11 and meropenem showed a synergistic effect against 73% of all isolates. These two combinations were the most promising of those tested against MDR/XDR K. pneumoniae isolates in this study. Of note, combining K11 and rifampicin also resulted in a synergistic effect against 67% of all isolates. Moreover, the combination of K11 and ceftazidime showed a synergistic effect against 53% of all isolates. Additionally, the combination of K11 and ciprofloxacin synergistically affected only 20% of the isolates, whereas the combination of K11 and colistin showed no synergistic effect on any tested bacterial isolates. No antagonism was observed in any combination. Notably, the synergistic effect of colistin and meropenem against MDR/XDR K. pneumoniae clinical isolates was also determined. Intriguingly, the synergistic effect of colistin and meropenem was observed against only 33% of all isolates as shown in Table 2 .

The inhibitory effect of K11 on biofilm formation by strong-biofilm producers of MDR/XDR K. pneumoniae was studied. The results of the antibiofilm efficacy of K11 at different concentrations on initial biofilm formation are shown in Figure 2 . Compared with untreated controls, the different concentrations of K11 exhibited dose-dependent biofilm inhibition against each of four strong-biofilm-forming strains of MDR/XDR K. pneumoniae at 24 h. It was demonstrated that K11 caused approximately 32% to 80% decrease in the biofilm biomass at the concentration of 2 up to 64 μg/mL (0.25×MIC to 1×MIC). Significant biofilm inhibition was observed at especially 0.25×, 0.5× and 1×MIC values of K11.

The effects of K11 in combination with conventional antibiotics on biofilm inhibition were investigated ( Figure 3 ). In this experiment, K11 concentration was fixed at 0.25×MIC, while the concentrations of antibiotics were varied at 0.25×, 0.5×, and 1×MIC. The MIC of meropenem, chloramphenicol, and rifampicin were similar against two representatives of MDR/XDR-strong biofilm producers of K. pneumoniae as 64, 16, and 32 µg/mL, respectively. The results showed that K11, combined with meropenem, chloramphenicol, or rifampicin, enhanced activity against two MDR/XDR-strong biofilm producers of K. pneumoniae. Of note, when K11 was combined with a particular antibiotic, the efficiency in inhibiting biofilm formation was significantly higher than those obtained with K11 or antibiotic alone. These findings indicated that the biofilm-inhibiting activity of K11 against MDR/XDR K. pneumoniae could be augmented by combination with antibiotics.

The stability of AMPs is crucial for storage and applications. Different temperatures and pH environments might influence the sensitivity of bacteria to some peptides. Hence, the thermal- and pH-resistant stabilities of K11 were evaluated by exposure to different temperatures and pH buffers for 1 h. Interestingly, K11 maintained its antimicrobial activity against K. pneumoniae ATCC 70063 and K. pneumoniae ATCC BAA-1706 when treated at 50, 70, or 90°C ( Table 3 ). Furthermore, K11 also exhibited similar antimicrobial activity against the reference strains over a relatively wide pH range from 2.0 to 10.0 ( Table 4 ). These results revealed that K11 had excellent thermal- and pH-resistant stabilities.

The sensitivity of AMPs to physiological salts and serum has been regarded as an inevitable problem in clinical applications. Therefore, the salt- and serum-resistant stabilities of K11 were further assessed by exposure to different physiological salts and fetal bovine serum (FBS) concentrations. The results showed that treatment with various concentrations of NaCl, MgCl₂, and FeCl₃ had no apparent influence on the antimicrobial activity of K11 against tested strains ( Table 5 ). The ability of K11 to inhibit bacterial growth was not significantly decreased even at final concentrations of salts. Additionally, K11 did not exhibit any change in MIC after treatment in 25% and 50% FBS for 1 h ( Table 6 ). These results indicated that K11 showed good resistance to physiological salts, such as NaCl, MgCl₂, and FeCl₃, as well as serum.

The tendencies of antimicrobial resistance induced by ciprofloxacin, colistin, and K11 against K. pneumoniae ATCC BAA-1706 reference strain was monitored by determining the MIC values of drugs and peptide after 30 days of consecutive treatments of bacteria at the concentration of 0.5×MIC. The MIC of ciprofloxacin significantly increased to 16-fold after 7 serial passages and reached 64-fold after 12 passages, then maintained this fold change until 30 passages ( Figure 4 ). Noteworthy, the MIC of colistin remarkably increased to 64-fold after a few passages and continue to rise dramatically over 30 passages (4,096-fold). Fortunately, slight variations in the MIC of K11 were observed over 30 passages, and the MIC did not rise above 2-fold. This result suggested that K11 showed a negligible drug resistance induction tendency.

In this study, fifteen non-duplicate clinical isolates of MDR/XDR K. pneumoniae were obtained from the bacterial repository of the Division of Infectious Diseases and Tropical Medicine, Department of Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University. The species were confirmed using standard biochemical tests. Antimicrobial susceptibility testing was also carried out by the disk diffusion method according to CLSI guidelines (Clinical and Laboratory Standards Institute, 2021). Furthermore, all isolates were tested for colistin resistance by broth microdilution based on CLSI recommendations. Following the CLSI breakpoints, colistin resistance was defined as a MIC of ≥4 µg/mL (Clinical and Laboratory Standards Institute, 2021). The MDR bacteria was defined as bacteria that are resistant to at least one agent in three or more antimicrobial categories, while the XDR bacteria refer to bacteria that are resistant to at least one agent in all but two or fewer antimicrobial categories based on laboratory results. E. coli ATCC 25922 was used as a reference control strain for susceptibility testing. Other reference strains, including K. pneumoniae ATCC 70063 and K. pneumoniae ATCC BAA-1706 (bla _KPC negative), were used as control strains throughout the study.

The clonal relatedness of all K. pneumoniae clinical isolates was determined by enterobacterial repetitive intergenic consensus PCR (ERIC-PCR) using primers, ERIC-1 (5’-ATGTAAGCTCCTGGGGATTCAC-3’) and ERIC-2 (5’-AAGTAAGTGACTGGGGTGAGCG-3’) as those previously described (Versalovic et al., 1991). The genomic DNA of all isolates was extracted using a TIANamp bacteria DNA kit (Tiangen Biotech, Beijing, China) according to the manufacturer’s protocol. The PCR amplification was performed with slight modification as follows: initial denaturation at 94°C for 5min, followed by 35 cycles of denaturation at 94°C for 1 min, annealing at 38°C for 1 min, and extension at 72°C for 3 min, with a final extension at 72°C for 10 min. ERIC-PCR patterns were compared by InfoQuest™FP Software, version 4.5 (Bio-Rad, Hercules, USA) using the Dice coefficient. The dendrogram was generated by the unweighted pair group method with arithmetic means (UPGMA) using 1.0% optimization and 1.0% band position tolerance. Percentage similarities were used to assess the relatedness of the clones (Widerström et al., 2007).

K11 with C-terminal amidation was synthesized by GenScript (Piscataway, NJ, USA) using solid-phase Fmoc chemistry and purified to >90% purity using HPLC. The mass was confirmed by MALDI-TOF mass spectroscopy. The antibiotics were acquired from TCI (Tokyo, Japan), except colistin sulfate, which was obtained from Chem-Impex Int’l Inc. (Wood Dale, IL, USA). Antibiotic disks were purchased from Oxoid (Oxoid Ltd., UK).

The ability of K11 to inhibit bacterial growth was determined by the broth microdilution method (Clinical and Laboratory Standards Institute, 2018). Briefly, a series of 2-fold dilutions of K11 were prepared from 1,024 to 2 µg/mL in cation-adjusted Mueller-Hinton broth (CAMHB), Then 50 µL of each peptide dilution was added to each well of polystyrene 96-well U-bottom microplates. The log-phase cultures were adjusted to 0.5 McFarland standard and diluted 1:100 in CAMHB to provide the density of 10⁶ CFU/mL. Consequently, 50 µL of diluted bacterial suspension was added to wells containing 50 µL of peptide solution, followed by incubation for 20-24 h at 37°C. The lowest concentration of peptide that completely inhibited bacterial growth was regarded to be the MIC. To evaluate MBC, a volume of 10 μL was taken from each well with no visible growth, inoculated on Mueller-Hinton agar (MHA) plates, and incubated at 37°C. The MBC was defined as the lowest concentration of antimicrobials that killed at least 99.9% of the initial inoculums (Clinical and Laboratory Standards Institute, 1999).

The synergistic effects of K11 in combination with conventional antibiotics, including meropenem, chloramphenicol, rifampicin, colistin, ciprofloxacin, and ceftazidime, were assessed using the checkerboard method (Zhang et al., 2019). Briefly, K11 and antibiotics were prepared at 4×of desired concentrations by a two-fold dilution method. Next, 25 µL of K11 and antibiotic dilutions with different concentrations were added to the designed wells on the plate to obtain different proportions with K11 or antibiotic concentrations. Bacterial cultures were grown to log phase and diluted to 10⁶ CFU/mL in CAMHB. Then, 50 µL of diluted bacterial suspension was added to each dilution of K11 and antibiotic combinations to give the final desired concentrations and subsequently incubated for 20-24 h at 37°C. The FICI values were calculated using the concentration combinations with the highest combination effects: FICI = MIC of drug A in combination/MIC of drug A alone + MIC of drug B in combination/MIC of drug B alone. The antimicrobial combination was defined as synergy when the FICI was ≤ 0.5, no interaction when 0.5 <FICI <4, and antagonism when the FICI was ≥ 4.

The ability of K11 to inhibit biofilm formation against four representatives of strong biofilm-producing MDR/XDR K. pneumoniae was examined by the crystal violet staining as previously described with few modifications (Gopal et al., 2014; Liu et al., 2022). Overnight cultures of bacteria were conducted at 37°C in CAMHB. Subsequently, cultures were diluted to 10⁸ CFU/mL in CAMHB supplemented with 0.5% glucose. A 50 µL aliquot of bacterial suspension was transferred to the wells of a sterile polystyrene 96-well flat-bottom plate. The stock solutions (0.5×, 1×, and 2×MIC) of K11 were prepared, adding 50 µL of each stock solution to the wells to a final concentration of 0.25×, 0.5×, and 1×MIC of the peptide. The mixtures were incubated for 24 h at 37°C to allow biofilm formation. After incubation, the medium was discarded from the wells, and any free-floating planktonic cells were removed by washing the wells with sterile deionized water. The biofilms were fixed with 100% methanol for 10 min, and then, the plates were allowed to dry. After that, wells were stained with 1% (w/v) crystal violet for 30 min. The excess stain was then thoroughly rinsed away with deionized water, and the plates were left to dry. Once dry, 95% ethanol was added into wells to solubilize the stain. The optical density (OD) was measured at 595 nm. The percentage of biofilm formation was calculated based on the growth control.

The antibiofilm efficiency of combination K11 and meropenem, chloramphenicol, or rifampicin was conducted by crystal violet assay as described above. Two strong biofilm-producer strains of MDR/XDR K. pneumoniae were selected for this study. The concentrations of 0.25×, 0.5×, and 1×MIC of antibiotics were tested in the presence of a fixed concentration of peptide (0.25×MIC).

The temperature and pH stabilities of K11 were evaluated by the MIC assay, as mentioned above. To assess the thermal stability, Aliquots of 2 mg/mL of K11 in sterile deionized water were incubated for 1 h at 25°C, 37°C, 50°C, 70°C, and 90°C. Then, a series of 2-fold dilutions of K11 were prepared in CAMHB, 50 µL of each peptide solution was transferred to the well of a 96-well microplate. An equal volume of K. pneumoniae ATCC 70063 or K. pneumoniae ATCC BAA-1706 (10⁶ CFU/mL) suspension was added, and the mixture was incubated at 37°C for 20-24 h (Li et al., 2021). To examine the pH stability, the pH was adjusted to range from 2.0 to 10.0 with the following 100 mM buffers: glycine-HCl buffer (pH 2.0), sodium acetate buffer (pH 4.0), sodium phosphate buffer (pH 6.0), Tris-HCl buffer (pH 8.0), and glycine-NaOH buffer (pH 10.0). Then, 2 mg/mL of K11 was prepared in buffers with different pH values (pH 2.0-10.0) and incubated for 1 h at 37°C. Afterward, the pH of the reaction solution was neutralized by HCl or NaOH to terminate the reaction. The serially diluted peptide solutions were conducted, then treated against the tested strains, and the mixtures were incubated at 37°C for 20-24 h (Cao et al., 2015; Zong et al., 2016). The MICs were observed.

The effects of various cations and serum on the antimicrobial activity of K11 were investigated against K. pneumoniae ATCC 70063 and K. pneumoniae ATCC BAA-1706 by the MIC assay, as described above. For salt stability, 2 mg/mL of K11 was dissolved in solutions containing different salts at their physiological concentrations; NaCl (100, 150, and 200 mM), MgCl₂ (0.5, 1 and 2 mM), and FeCl₃ (1, 4, and 8 mM), then incubated for 1 h at 37°C. Subsequently, serially 2-fold diluted K11 solutions were prepared in CAMHB, and aliquots of each peptide solution were added to a 96-well plate. An equal volume of bacterial suspension was added, and the mixtures were incubated at 37°C for 20-24 h (Ko et al., 2019). For serum stability, 2 mg/mL of K11 was prepared in solution with 25% and 50% FBS and incubated for 1 h at 37°C, then heat activated for 30 min at 60°C. A series of 2-fold dilutions of K11 were performed in CAMHB and transferred to a 96-well plate. The diluted bacterial suspension was added to the well, and mixtures were incubated for 20-24 h at 37°C (Oliva et al., 2018). The antimicrobial activity was determined by measuring the MICs.

The resistance development of K11, ciprofloxacin, and colistin was assessed according to the previously described protocol (Pollard et al., 2012). Briefly, MIC values for a K. pneumoniae ATCC BAA-1706 were first examined and recorded. After incubation at 37°C for 20-24 h, serial passaging was initiated by harvesting bacterial cells growing at 0.5×MIC values and inoculating them into fresh CAMHB. This inoculum was subjected to another MIC assay. After a 20-24 h incubation period, cells growing at 0.5×MIC from the previous passage were once again harvested and assayed for the MIC. The process was repeated for 30 passages. The MIC values for K11 and antibiotics were recorded and plotted as the MIC fold change compared to day 1 of the experiment.

All experiments were performed with at least two reproductions with no less than two replicates each. The quantitative data are presented as mean ± standard deviation (SD). The results were analyzed for normal distribution using the Shapiro-Wilk test (P value <0.05). When no significant difference from normality was identified, the significance of differences among groups was analyzed by one-way analysis of variance (ANOVA). All statistical calculations were performed using GraphPad Prism version 8.4.3. Values of P <0.05 were considered statistically significant.