Development of New Antimicrobial Peptides by Directional Selection

Abstract

Background/Objectives: The global rise in antibiotic resistance necessitates the development of novel antimicrobial agents. Antimicrobial peptides (AMPs), key components of innate immunity, are promising candidates. This study aimed to develop novel therapeutic peptides with enhanced properties through the mutagenesis of natural AMPs and high-throughput screening. Methods: We constructed mutant libraries of three broad-spectrum AMPs-melittin, cecropin, and Hm-AMP2-using mutagenesis with partially degenerate oligonucleotides. Libraries were expressed in Escherichia coli, and antimicrobial activity was assessed through bacterial growth kinetics and droplet serial dilution assays. Candidate molecules were identified by DNA sequencing, and the most promising variants were chemically synthesized. Antimicrobial activity was determined by minimal inhibitory concentration (MIC) against E. coli and Bacillus subtilis, while cytotoxicity was evaluated in human Expi293F cells (IC₉₀) viability. The therapeutic index was calculated as the ratio of an AMP's cytotoxic concentration to its effective antimicrobial concentration. Results: Mutant forms of melittin (MR1P7, MR1P8) showed significantly reduced cytotoxicity while retaining antimicrobial activity. Cecropin mutants exhibited reduced efficacy against E. coli, but variants CR2P2, CR2P7, and CR2P8 gained activity against Gram-positive bacteria. Mutagenesis of Hm-AMP2 generally decreased activity against E. coli, though two variants (A2R1P5 and A2R3P6) showed retained or enhanced efficacy against B. subtilis while maintaining low cytotoxicity. Conclusions: The proposed strategy successfully generated peptides with improved therapeutic profiles, including reduced toxicity or a broader spectrum of antimicrobial activity, despite not improving all parameters. This approach enables the discovery of novel bioactive peptides to combat antibiotic-resistant pathogens.

Keywords: Hm-AMP2; antibacterial activity; antimicrobial peptide; cecropin; expression system; library; melittin; mutagenesis.

Figure 1

Schematic overview of the experimental approach. A library of expression plasmids encoding random mutants of well-characterized antimicrobial peptides (Hm-AMP2, melittin and cecropin) was generated by random mutagenesis with partially degenerate oligonucleotides. The plasmid library was transformed into E. coli, followed by induction of recombinant AMP expression. Clones with the strongest growth inhibition were selected for subsequent rounds of mutagenesis and screening. The most promising variants were synthesized and tested for their properties.

Figure 2

Calculated and observed numbers of mutations for mutant plasmid libraries for (A) Hm-AMP2, (B) melittin and (C) cecropin. X-axis: number of nucleotide substitutions; Y-axis: fraction of clones with the corresponding number of mutations (%). Estimated value—predicted number of mutations for the “88%” library. This library was synthesized using a phosphoramidite mixture containing 88 vol% of the correct (primary) phosphoramidite and 4 vol% of each of the other three nucleotides at every position.

Figure 3

Analysis of the antimicrobial activity of libraries encoding Hm-AMP2 variants. (A)—first round of mutagenesis. (B)—second round of mutagenesis. (C)—third round of mutagenesis. For each round, the five most active peptides were selected. Amino acid substitutions are indicated in red. Activity was assessed using the R_OD value and verified by droplet serial dilution. DPSD—Droplet Serial Dilution—the first dilution at which no visible colonies were observed on agar plates. Boxplots show R_OD values from experiments evaluating the antibacterial activity of the five most active peptides. The edges of the boxes represent the 25th and 75th percentiles, the line inside the box indicates the median (50th percentile), and the whiskers correspond to the limits of the statistically significant sample. Whisker length is defined as 1.5× the interquartile range below the first quartile and 1.5× above the third quartile. The vertical axis represents R_OD values measured 5 h after the start of the kinetic assay.

Figure 4

Analysis of the antimicrobial activity of library encoding melittin variants. The five most active peptides were selected. Amino acid substitutions are indicated in red. Activity was evaluated using the R_OD value and verified by droplet serial dilution. DPSD—Droplet Serial Dilution—the first dilution at which no visible colonies were observed on agar plates. Boxplots show R_OD values from experiments assessing the antibacterial activity of the five most active peptides. The edges of the boxes represent the 25th and 75th percentiles, the line inside the box indicates the median (50th percentile). The vertical axis shows R_OD values measured 5 h after the start of the kinetic assay.

Figure 5

Analysis of the antimicrobial activity of libraries encoding cecropin variants. (A)—first round of mutagenesis. (B)—second round of mutagenesis. For each round, the five most active peptides were selected. Amino acid substitutions are indicated in red. Activity was evaluated using the R_OD value and verified by droplet serial dilution. DPSD—Droplet Serial Dilution—the first dilution at which no visible colonies were observed on agar plates. Boxplots show R_OD values from experiments assessing the antibacterial activity of the five most active peptides. The edges of the boxes represent the 25th and 75th percentiles, the line inside the box indicates the median (50th percentile). The vertical axis shows R_OD values measured 5 h after the start of the kinetic assay.

Figure 6

Analysis of position-dependent amino acid substitution frequencies. For each histogram, the frequency of mutations leading to a change in amino acid class is shown for each residue in the sequence of the investigated antimicrobial peptide. Blue indicates substitutions at positions associated with increased activity, while pink indicates substitutions associated with decreased activity. Hatched bars, as well as coloured residues in the peptide sequence, indicate substitutions that occur exclusively in either the more active or the less active peptides. (A)—First round of mutagenesis of Hm-AMP2. The original Hm-AMP2 sequence is shown below. (B)—Second round of mutagenesis of Hm-AMP2. The sequence of peptide A2R1P5, which showed the highest activity in experiments, is shown below. (C)—Third round of mutagenesis of Hm-AMP2. The sequence of peptide A2R2P3, which showed the highest activity in the second round of mutagenesis, is shown below. (D)—Results of mutation frequency analysis for melittin mutant libraries. The sequence of melittin is shown below. (E)—First round of mutagenesis of cecropin. The original cecropin sequence is shown below. (F)—Second round of mutagenesis of cecropin. The sequence of peptide CR1P5, which showed the highest activity in experiments, is shown below.

Figure 7

Activity of synthetic peptides melittin (A), cecropin (B), and Hm-AMP2 (C) in relation to introduced mutations. In the alignment on the left, shades of blue indicate amino acid sequence similarities. On the right, antimicrobial activity against E. coli and B. subtilis, cytotoxicity toward Expi293F cells, and the calculated therapeutic index based on peptide activities are shown. MEL—melittin, CECR—cecropin, AMP2—Hm-AMP2 peptide. AMP’s cytotoxicity was measured as the IC₉₀, the concentration that inhibits 90% of cell growth. NA (not applicable) indicates that peptide showed no activity within tested concentration range. Minimal inhibitory concentration (MIC) and IC₉₀ are given in µM. The therapeutic index (TI) was defined as the lowest IC₉₀ divided by the highest MIC.