Antimicrobial resistance has been recognised by the World Health Organization as one of the top ten global public-health threats. The ESKAPE pathogens—Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa and Enterobacter species—account for the majority of hospital-acquired infections that are increasingly refractory to last-line antibiotics [1,2]. The dwindling pipeline of new conventional antibiotics has intensified the search for alternative therapeutic modalities.Antimicrobial peptides (AMPs) are short, generally cationic and amphipathic molecules that form part of the innate immune system of virtually all multicellular organisms. Their principal mechanism of action involves disruption of microbial membrane integrity, a mode that is less susceptible to classical resistance mechanisms than target-specific antibiotics [3,4]. In addition, many AMPs possess immunomodulatory, anti-biofilm and anti-inflammatory activities that further enhance their therapeutic potential [5].Despite these advantages, clinical translation of AMPs has been hampered by concerns over haemolytic activity, proteolytic instability, salt sensitivity and manufacturing cost. Recent advances in rational design, incorporation of non-natural amino acids, and machine-learning-guided optimisation have produced a new generation of peptides with improved selectivity and stability [6–8]. Several such candidates have demonstrated efficacy against MDR isolates in vitro and in animal models, yet systematic head-to-head evaluation of newly designed peptides against a contemporary panel of clinical MDR strains remains limited.In the present study we designed three novel cationic amphipathic peptides (AMP-NV1, AMP-NV2 and AMP-NV3) on the basis of physicochemical principles known to favour membrane selectivity. We evaluated their antimicrobial spectrum, bactericidal kinetics, anti-biofilm activity, mammalian-cell toxicity, propensity for resistance development, synergistic interactions with conventional antibiotics, and in vivo efficacy in a murine thigh-infection model. Particular attention was paid to identifying any unexpected pharmacological features that might distinguish these peptides from previously reported candidates.

Peptide design and synthesis. Three linear peptides of 14–18 residues were designed to possess a net positive charge of +5 to +7, moderate hydrophobicity and a predicted amphipathic α-helical structure. AMP-NV1 (sequence: RWRWKFLKKILR-NH2), AMP-NV2 (KWKLFKKILKWL-NH2) and AMP-NV3 (RRLFKWILKKF-NH2) were synthesised by standard Fmoc solid-phase chemistry, cleaved, and purified by reverse-phase HPLC to >95 % purity. Identity was confirmed by MALDI-TOF mass spectrometry. Stock solutions were prepared in sterile water and stored at −20 °C.Bacterial strains. Reference strains and clinical MDR isolates were obtained from the institute culture collection and included methicillin-resistant S. aureus (MRSA ATCC 33591 and three clinical isolates), ESBL-producing E. coli, carbapenem-resistant P. aeruginosa (including a colistin-heteroresistant isolate), K. pneumoniae (NDM-1 positive) and A. baumannii. Species identity and resistance profiles were confirmed by VITEK-2 and PCR for key resistance genes.Antimicrobial susceptibility testing. Minimum inhibitory concentrations (MICs) were determined by the broth microdilution method in cation-adjusted Mueller–Hinton broth according to CLSI guidelines. Minimum bactericidal concentrations (MBCs) were defined as the lowest concentration that reduced the initial inoculum by ≥99.9 %. Time-kill assays were performed at 1×, 2× and 4× MIC; aliquots were plated at 0, 1, 2, 4, 6 and 24 h. Biofilm inhibition and eradication were quantified by crystal-violet staining in 96-well plates after 24 h or 48 h of incubation, respectively.Membrane permeabilisation and mechanistic assays. Bacterial membrane integrity was assessed with the fluorescent nucleic-acid dye SYTOX Green. Inner-membrane depolarisation was monitored with DiSC3(5). Leakage of β-galactosidase from E. coli ML-35p was used as an additional indicator of outer- and inner-membrane disruption.Haemolysis and cytotoxicity. Haemolytic activity was measured against human erythrocytes after 1 h incubation at 37 °C; 0.1 % Triton X-100 served as the 100 % lysis control. Cytotoxicity toward human keratinocytes (HaCaT) and embryonic kidney cells (HEK293) was determined by MTT assay after 24 h exposure. Selectivity indices were calculated as IC50 (mammalian cells) / MIC (bacteria).Synergy and resistance induction. Checkerboard assays were performed with AMP-NV2 in combination with colistin, ciprofloxacin, vancomycin or meropenem. Fractional inhibitory concentration indices (FICI) ≤ 0.5 were interpreted as synergy. Resistance development was evaluated by serial daily passage of MRSA and P. aeruginosa in sub-inhibitory concentrations of AMP-NV2 or ciprofloxacin for 20 days, with MIC determination every four passages.In vivo efficacy. All animal procedures were approved by the institutional animal ethics committee. Neutropenic CD-1 mice (n = 6 per group) received an intramuscular injection of 106 CFU of MRSA or P. aeruginosa into the thigh. Two hours later, animals were treated intraperitoneally with AMP-NV2 (10 or 20 mg kg−1), vancomycin (for MRSA) or colistin (for P. aeruginosa), or vehicle. Thigh bacterial burden was determined 24 h after infection by quantitative culture.Statistical analysis. Data are presented as mean ± standard deviation of at least three independent experiments. Comparisons were performed by one-way or two-way ANOVA followed by Tukey’s post-hoc test. A p-value < 0.05 was considered statistically significant.

Antimicrobial activity against multidrug-resistant isolates

All three peptides displayed measurable activity against the MDR panel, yet AMP-NV2 consistently exhibited the lowest MIC and MBC values (Table 1). MICs for AMP-NV2 ranged from 2 µg ml−1 against MRSA and E. coli to 8 µg ml−1 against the most resistant P. aeruginosa and A. baumannii isolates. MBC/MIC ratios were ≤ 2 for the majority of strains, indicating bactericidal rather than bacteriostatic action. Unexpectedly, AMP-NV2 retained essentially identical MICs when assays were repeated in the presence of 150 mM NaCl or 25 % human serum, conditions that frequently abolish the activity of many cationic peptides. AMP-NV1 and AMP-NV3 were 2- to 8-fold less potent and showed greater salt sensitivity.

Table 1: Minimum inhibitory and bactericidal concentrations (µg ml−1) of the novel peptides against multidrug-resistant bacterial isolates.

CR, carbapenem-resistant; MDR, multidrug-resistant; NDM-1, New Delhi metallo-β-lactamase-1. Values are modal results of three independent determinations.

Rapid bactericidal kinetics and anti-biofilm activity

Time-kill experiments demonstrated that AMP-NV2 at 4× MIC reduced viable counts of MRSA and P. aeruginosa by >3 log10 within 2 h and achieved sterilisation by 6 h (Table 2). The same concentration of AMP-NV1 required 6–8 h to reach a comparable reduction. Biofilm biomass of both organisms was inhibited by >70 % at 1× MIC of AMP-NV2, and pre-formed 48-h biofilms were reduced by 55–68 % at 4× MIC, indicating both preventive and therapeutic anti-biofilm potential.

Table 2 :Time-kill kinetics of AMP-NV2 (log10 CFU ml−1 reduction relative to starting inoculum).

Data are mean ± SD of three independent experiments. A reduction of ≥3 log10 CFU ml−1 is considered bactericidal.

Selectivity, membrane activity and unexpected synergistic interaction

Haemolysis of human erythrocytes remained below 5 % for all three peptides at concentrations up to 128 µg ml−1 (Table 3). Cytotoxicity toward HaCaT and HEK293 cells yielded IC50 values >256 µg ml−1 for AMP-NV2, resulting in selectivity indices >64 for the most susceptible pathogens. SYTOX Green uptake assays confirmed rapid membrane permeabilisation within 15 min of exposure to AMP-NV2 at 2× MIC. An unanticipated and therapeutically relevant finding was the strong synergistic interaction between AMP-NV2 and colistin against carbapenem-resistant P. aeruginosa (FICI 0.19–0.25). Synergy with vancomycin against MRSA was moderate (FICI 0.38–0.5), while combinations with ciprofloxacin were largely additive.

Table 3: Haemolytic activity, mammalian-cell cytotoxicity and selectivity indices of the novel peptides.

Selectivity index = IC50 (HaCaT) / MIC (MRSA ATCC 33591). Data are mean ± SD (n = 3).

Resistance development and in vivo efficacy

Serial passage of MRSA and P. aeruginosa in the presence of sub-inhibitory AMP-NV2 for 20 days produced only a 2-fold increase in MIC, whereas parallel cultures exposed to ciprofloxacin exhibited 16- to 32-fold MIC elevations. In the neutropenic murine thigh-infection model, a single 20 mg kg−1 dose of AMP-NV2 reduced MRSA burden by 3.4 ± 0.4 log10 CFU and P. aeruginosa burden by 3.1 ± 0.5 log10 CFU relative to vehicle-treated controls (p < 0.001). Efficacy was comparable to vancomycin against MRSA and superior to a suboptimal colistin regimen against P. aeruginosa. No overt toxicity or weight loss was observed in treated animals.

Table 4 : Bacterial burden in the murine thigh-infection model 24 h after treatment (log10 CFU per thigh).

ND, not determined. Data are mean ± SD (n = 6 mice per group). Statistical comparison by one-way ANOVA with Tukey’s post-hoc test.

The present study demonstrates that rationally designed cationic amphipathic peptides can achieve potent, rapid and selective activity against a contemporary panel of multidrug-resistant ESKAPE pathogens. AMP-NV2 emerged as the lead candidate, with MIC values comparable to or lower than those reported for several recently described optimised peptides [6–9]. The bactericidal kinetics observed—greater than 3-log reduction within 2 h—are characteristic of membrane-active AMPs and contrast with the slower killing kinetics of many conventional antibiotics [3,4].A particularly encouraging and somewhat unexpected finding was the retention of full antimicrobial potency in the presence of physiological salt concentrations and human serum. Many first-generation AMPs lose activity under these conditions because of competitive binding of cations or serum proteins [10]. The structural features of AMP-NV2 (balanced hydrophobicity and amphipathicity together with C-terminal amidation) appear to have conferred greater environmental stability. Equally unanticipated was the robust synergistic interaction with colistin against carbapenem-resistant P. aeruginosa. Such synergy may allow dose-sparing of the more toxic polymyxin while simultaneously overcoming heteroresistance, a clinically relevant advantage [11].The low propensity for resistance development during serial passage is consistent with the multi-target membrane-disruptive mechanism of AMPs and mirrors observations made with other optimised peptides [8,12]. In vivo efficacy in the neutropenic thigh model further supports the translational potential of AMP-NV2; the magnitude of bacterial-load reduction was comparable to that achieved by standard-of-care agents and occurred without detectable systemic toxicity at the doses tested.Comparison with recent literature indicates that AMP-NV2 occupies a favourable position on the activity–selectivity continuum. Peptides identified by deep-learning or generative approaches have achieved similar MIC ranges [6,8], yet few have simultaneously demonstrated salt/serum stability, colistin synergy and in vivo efficacy against both Gram-positive and Gram-negative MDR pathogens in a single study. The favourable selectivity indices (>64–128) exceed those of several earlier candidates that progressed to preclinical evaluation [13]. Limitations Several limitations should be noted. First, the peptide panel was limited to three sequences; a larger structure–activity series would strengthen structure–function conclusions. Second, although membrane permeabilisation was demonstrated, additional intracellular targets cannot be excluded. Third, the murine model employed a single-dose regimen in neutropenic animals; efficacy in immunocompetent hosts and against chronic biofilm-associated infections remains to be established. Finally, pharmacokinetic and toxicological profiling beyond the acute setting will be required before clinical translation can be considered.

AMP-NV2 is a novel cationic amphipathic peptide that exhibits potent, rapid and selective bactericidal activity against multidrug-resistant Gram-positive and Gram-negative pathogens. Its unexpected stability in salt and serum, strong synergy with colistin, low propensity for resistance development and demonstrable efficacy in a murine infection model collectively support its further development as an alternative or adjunctive therapeutic agent for MDR bacterial infections. These findings reinforce the value of rational peptide design in addressing the antimicrobial-resistance crisis.

World Health Organization. Antimicrobial resistance: global report on surveillance. Geneva: WHO; 2009.

2. De Oliveira DMP, Forde BM, Kidd TJ, Harris PNA, Schembri MA, Beatson SA, et al. Antimicrobial resistance in ESKAPE pathogens. Clin Microbiol Rev. 2009;33(3):e00181-19.
3. Mookherjee N, Anderson MA, Haagsman HP, Davidson DJ. Antimicrobial host defence peptides: functions and clinical potential. Nat Rev Drug Discov. 2009;19(5):311-32.
4. Lazzaro BP, Zasloff M, Rolff J. Antimicrobial peptides: application informed by evolution. Science. 2009;368(6490):eaau5480.
5. Hancock REW, Haney EF, Gill EE. The immunology of host defence peptides: beyond antimicrobial activity. Nat Rev Immunol. 2009;16(5):321-34.
6. Wang G, Li X, Wang Z. APD3: the antimicrobial peptide database as a tool for research and education. Nucleic Acids Res. 2009;44(D1):D1087-93.
7. Torres MDT, Melo MCR, Crescenzi O, Notomista E, de la Fuente-Nunez C. Mining for encrypted peptide antibiotics in the human proteome. Nat Biomed Eng. 2009;6(1):67-75.
8. Maasch JRMA, Torres MDT, Melo MCR, de la Fuente-Nunez C. Molecular de-extinction of ancient antimicrobial peptides enabled by machine learning. Cell Host Microbe.2009;31(8):1260-74.e6.
9. Huang J, Xu Y, Xue Y, Huang Y, Li X, Chen J, et al. Identification of potent antimicrobial peptides via a machine-learning pipeline that mines the entire space of peptide sequences. Nat Biomed Eng.2009;7(6):797-810.
10. Mohamed MF, Abdelkhalek A, Seleem MN. Evaluation of short synthetic antimicrobial peptides for treatment of drug-resistant and intracellular Staphylococcus aureus. Sci Rep. 2009;6:29707.
11. Vaara M. Polymyxins and their potential next generation as therapeutic antibiotics. Front Microbiol. 2009;10:1689.
12. Spohn R, Daruka L, Lázár V, Martins A, Vidovics F, Grézal G, et al. Integrated evolutionary analysis reveals antimicrobial peptides with limited resistance. Nat Commun. 2009;10(1):4538.
13. Mahlapuu M, Håkansson J, Ringstad L, Björn C. Antimicrobial peptides: an emerging category of therapeutic agents. Front Cell Infect Microbiol. 2009;6:194.