Evolutionary and in silico guided development of novel peptide analogues for antibacterial activity against ESKAPE pathogens

Abstract

According to WHO, to combat the resistant strains, new effective anti-microbial agents are needed on an urgent basis and global researchers should focus their efforts and discovery programs on developing them against antibiotic-resistant pathogens or priority pathogens like ESKAPE. In this context, Cationic antimicrobial peptides (AMPs) are being explored extensively as promising next-generation antimicrobials due to their broad range, fast kinetics and multifunctional role. Despite recent advances, it is still a daunting challenge to identify and design a potent AMP with no cytotoxicity, but with broad specific antimicrobial activity, stability and efficacy under in vivo conditions in a cost-effective and robust manner. In this work, as a proof of concept, we designed novel potent AMPs using artificial intelligence based in silico programs. Shortlisted peptide sequences were synthesized using the fmoc chemistry approach, assessed their antimicrobial activity, cell selectivity, mode of action and in vivo efficacy using a series of experiments. The synthesized peptide analogues demonstrated their antimicrobial activity (MIC in the range of 2.5-80 μM) against bacteria. The identified potential lead molecules showed antibacterial activity in physiological conditions with no signs of cytotoxicity. We further tested the antimicrobial activity of peptide analogues for treating wounds infected with Pseudomonas aeruginosa in the mice burn wound model. In drug-development programs, the identification of lead antimicrobial agents is always challenging and involves screening a large number of molecules which is time-consuming and expensive. This work demonstrates the utility of artificial intelligence based in silico analysis programs in discovering novel antimicrobial agents in an economical, robust way.

Keywords: Antimicrobial peptides; Burn wound model; ESKAPE pathogens; Hydrophobicity and charge relationships; In silico designing; Peptide analoguess.

Graphical abstract

Fig. 1

Antibacterial activity of YRA-15 peptide analoguess. (A) Helical wheel diagram of YRA-15 derived peptides showing the clustering and distribution of amino acids as hydrophobic and hydrophilic planes. hydrophobic amino acids (Yellow), positively charged (blue) and polar amino acids (violet) are color coded. The predicted secondary structures of the peptide are shown in the inlet. (B) Antibacterial activity was assessed by radial diffusion assay (RDA) against P. aeruginosa ATCC 25,668, A. baumannii ATCC 1605, K. pneumoniae ATCC 27,736, and S. aureus ATCC 29,213 at 25, 50, and 100 µM concentrations. The clear zone represents antibacterial activity and the zone of inhibition (ZOI) was measured in mm.

Fig. 2

Activity of peptides in physiological conditions. Analysis of viable counts for P. aeruginosa ATCC 25,668, A. baumannii ATCC 1605, K. pneumoniae ATCC 27,736, and S. aureus ATCC 29,213 after being subjected to different concentrations of peptides in the physiological environment (10 mM Tris. HCl, pH 7.4 supplemented with 10 mM glucose, 150 mM sodium chloride and 20% citrate human plasma) as the antimicrobial activity of peptides is highly influenced by the presence of salts and plasma components. After the treatment with peptides, the samples were serially diluted to 10X, 100X, and 1000X in 10 mM Tris. HCl, pH 7.4 containing 10 mM glucose before plating them on Todd Hewitt agar (THA) media and incubated for 16–18 h at 37 °C.

Fig. 3

Influence of complement factors on antibacterial activity. Time-killing kinetics of YRA15–13 and YRA15–23 against P. aeruginosa ATCC 25,668 and S. aureus ATCC 29,213. The assay was performed by determining the counts of the surviving bacteria.

Fig. 4

Cytotoxicity assay of peptides. (A) THP-1 and BV-2 cells were incubated with peptides at 50 and 100 µM concentrations, and LDH release was measured by using TOX-7 KIT reagent mixture. Absorbance was measured at 592 nm. For all the cytotoxicity assays, Two percent Triton X-100 was taken as the positive control and the results were expressed as mean values from triplicate measurements. (B) To evaluate the cytotoxicity of peptides the MTT assay was used to examine how peptides affected the survival rate of THP-1 and BV2 cells. (C) Hemolytic activity of peptides against human red blood cells (RBCs) was measured in the presence of peptides (50 and 100 µM concentrations). The absorbance of released hemoglobin was measured at 540 nm and expressed as% of Triton X-100 induced hemolysis. (D) Protease resistance assay. Peptides were incubated at 37 °C in the presence of V8 protease, Proteinase K or endoproteinase K (0.1 μg, 25,000 units/mg), or Human neutrophil elastase (0.4 μg, 29 units/mg) in a total volume of 10 μl for 3 h. After incubation the samples were analyzed on 16.5% precast Tris-tricine SDS PAGE.

Fig. 5

Scanning electron microscopy analysis of (A) P. aeruginosa ATCC 25,668, (B) A. baumannii ATCC 1605, (C) K. pneumoniae ATCC 27,736, and (D) S. aureus ATCC 29,213 treated with YRA15–13, YRA15–23, LL37 as (+) control and untreated cells as (-) control. Peptides (100 µM) were incubated with respective bacteria at 37 °C for 30 min. Cells were fixed, dehydrated, dried, and analyzed under emission scanning microscopy (FEI Quanta 250, Netherlands) at 5 kV and 3 mm working distance.

Fig. 6

Liposome leakage assay. (A, B) The peptide's ability to destabilize the membrane was assessed by calcein release from the anionic bacterial membrane (PE/PG) and zwitterionic eukaryotic membrane (PC) liposomes. Calcein released from the liposomes was monitored by emitting fluorescence at 512 nm. An absolute leakage was obtained by using 0.1% Triton X-100. Measurements were performed in triplicate at room temperature by using a Tecan infinite M Nano+ spectrometer. (C, D, E, F) CD spectra of peptides. CD spectra measurements of YRA15–13 and YRA15–23 in Milli Q, Tris buffer, pH 7.4, PE/PG liposomes, and PC liposomes were made on J-1500 Circular Dichroism Spectrophotometer. Measurements were taken in triplicate in a quartz cuvette (10 mm) under stirring with a peptide concentration of 100 µM at 37 °C. The effect on peptide secondary structure in the presence of liposomes was monitored in the range of 190–250 nm at a concentration of 0.1 mM. The background value (detected at 250 nm, where no peptide signal is present) was subtracted to nullify the instrumental differences between measurements. Signals from the bulk solution were also corrected before the experiment.

Fig. 7

(A) Ex vivo eye lens experiment. (B, C) In vivo antibacterial activity of YRA15–13 and YRA15–23. The in vivo antibacterial activity of YRA15–13 and YRA15–23 were tested on the mouse burn wound skin infection model. Untreated mice were taken as a negative control. Colistin formulation in Hypromellose was taken as a positive control.