How do cyclic antibiotics with activity against Gram-negative bacteria permeate membranes? A machine learning informed experimental study

PmxPred: A data-driven approach for the identification of active polymyxin analogues against gram-negative bacteria

The multidrug-resistant Gram-negative bacteria has evolved into a worldwide threat to human health; over recent decades, polymyxins have re-emerged in clinical practice due to their high activity against multidrug-resistant bacteria. Nevertheless, the nephrotoxicity and neurotoxicity of polymyxins seriously hinder their practical use in the clinic. Based on the quantitative structure-activity relationship (QSAR), analogue design is an efficient strategy for discovering biologically active compounds with fewer adverse effects. To accelerate the polymyxin analogues discovery process and find the polymyxin analogues with high antimicrobial activity against Gram-negative bacteria, here we developed PmxPred, a GCN and catBoost-based machine learning framework. The RDKit descriptors were used for the molecule and residues representation, and the ensemble learning model was utilized for the antimicrobial activity prediction. This framework was trained and evaluated on multiple Gram-negative bacteria datasets, including Acinetobacter baumannii, Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa and a general Gram-negative bacteria dataset achieving an AUROC of 0.857, 0.880, 0.756, 0.895 and 0.865 on the independent test, respectively. PmxPred outperformed the transfer learning method that trained on 10 million molecules. We interpreted our model well-trained model by analysing the importance of global and residue features. Overall, PmxPred provides a powerful additional tool for predicting active polymyxin analogues, and holds the potential elucidate the mechanisms underlying the antimicrobial activity of polymyxins. The source code is publicly available on GitHub (https://github.com/yanwu20/PmxPred).

Linking dual mode of action of host defense antimicrobial peptide thanatin: Structures, lipopolysaccharide and LptAm binding of designed analogs

At present, antibiotics options to cure infections caused by drug resistant Gram-negative pathogens are highly inadequate. LPS outer membrane, proteins involved in LPS transport and biosynthesis pathways are vital targets. Thanatin, an insect derived 21-residue long antimicrobial peptide may be exploited for the development of effective antibiotics against Gram-negative bacteria. As a mode of bacterial cell killing, thanatin disrupts LPS outer membrane and inhibits LPS transport by binding to the periplasmic protein LptA_m. Here, we report structure-activity correlation of thanatin and analogs for the purpose of rational design. These analogs of thanatin are investigated, by NMR, ITC and fluorescence, to correlate structure, antibacterial activity and binding with LPS and LptA_m, a truncated monomeric variant. Our results demonstrate that an analog thanatin M21F exhibits superior antibacterial activity. In LPS interaction analyses, thanatin M21F demonstrate high affinity binding to outer membrane LPS. The atomic resolution structure of thanatin M21F in LPS micelle reveals four stranded β-sheet structure in a dimeric topology whereby the sidechain of aromatic residues Y10, F21 sustained mutual packing at the interface. Strikingly, LptA_m binding affinity of thanatin M21F has been significantly increased with an estimated K_d ~ 0.73 nM vs 13 nM for thanatin. Further, atomic resolution structures and interactions of Ala based thanatin analogs define plausible correlations with antibacterial activity and LPS, LptA_m interactions. Taken together, the current work provides a frame-work for the designing of thanatin based potent antimicrobial peptides for the treatment of drug resistance Gram-negative bacteria.

A review of green solvent extraction techniques and their use in antibiotic residue analysis

Antibiotic residues are being continuously recognized in the aquatic environment and in food. Though the concentration of antibiotic residues is typically low, adverse effects on the environment and human health have been observed. Hence, an efficient method to determine numerous antibiotic residues should be simple, inexpensive, selective, with high throughput and with low detection limits. Liquid-based extractions have been exceedingly used for clean-up and preconcentration of antibiotics prior to chromatographic analysis. In order to make methods more green and environmentally sustainable, conventional hazardous organic solvents can be replaced with green solvents. This review presents sampling strategies as well as comprehensive and up-to-date methods for chemical analysis of antibiotic residues in different sample matrices. Particularly, solvent-based sample preparation techniques using green solvents are discussed along with applications in antibiotic residue analysis.

Apolipoprotein Mimetic Peptide Inhibits Neutrophil-Driven Inflammatory Damage via Membrane Remodeling and Suppression of Cell Lysis

Neutrophils are crucial for host defense but are notorious for causing sterile inflammatory damage. Activated neutrophils in inflamed tissue can liberate histone H4, which was recently shown to perpetuate inflammation by permeating membranes via the generation of negative Gaussian curvature (NGC), leading to lytic cell death. Here, we show that it is possible to build peptides or proteins that cancel NGC in membranes and thereby suppress pore formation, and demonstrate that they can inhibit H4 membrane remodeling and thereby reduce histone H4-driven lytic cell death and resultant inflammation. As a demonstration of principle, we use apolipoprotein A-I (apoA-I) mimetic peptide apoMP₁. X-ray structural studies and theoretical calculations show that apoMP₁ induces nanoscopic positive Gaussian curvature (PGC), which interacts with the NGC induced by the N-terminus of histone H4 (H4n) to inhibit membrane permeation. Interestingly, we show that induction of PGC can inhibit membrane-permeating activity in general and "turn off" diverse membrane-permeating molecules besides H4n. In vitro experiments show an apoMP₁ dose-dependent rescue of H4 cytotoxicity. Using a mouse model, we show that tissue accumulation of neutrophils, release of neutrophil extracellular traps (NETs), and extracellular H4 all strongly correlate independently with local tissue cell death in multiple organs, but administration of apoMP₁ inhibits histone H4-mediated cytotoxicity and strongly prevents organ tissue damage.

Harnessing machine learning for development of microbiome therapeutics

The last twenty years of seminal microbiome research has uncovered microbiota's intrinsic relationship with human health. Studies elucidating the relationship between an unbalanced microbiome and disease are currently published daily. As such, microbiome big data have become a reality that provide a mine of information for the development of new therapeutics. Machine learning (ML), a branch of artificial intelligence, offers powerful techniques for big data analysis and prediction-making, that are out of reach of human intellect alone. This review will explore how ML can be applied for the development of microbiome-targeted therapeutics. A background on ML will be given, followed by a guide on where to find reliable microbiome big data. Existing applications and opportunities will be discussed, including the use of ML to discover, design, and characterize microbiome therapeutics. The use of ML to optimize advanced processes, such as 3D printing and in silico prediction of drug-microbiome interactions, will also be highlighted. Finally, barriers to adoption of ML in academic and industrial settings will be examined, concluded by a future outlook for the field.

NMR structure and localization of the host defense antimicrobial peptide thanatin in zwitterionic dodecylphosphocholine micelle: Implications in antimicrobial activity

Antimicrobial peptides (AMPs) are potentially vital as the next generation of antibiotics against multidrug resistant bacterial pathogens. Thanatin, an insect derived pathogen inducible 21-residue long antimicrobial peptide, demonstrates antimicrobial activity toward broad range of pathogens. Thanatin is an excellent candidate for antibiotics development due to potent in vivo activity in animal model and low toxicity to human cells. Recent studies indicated mode of action of thanatin could be intriguing and may comprise bacterial membrane permeabilization and interactions with periplasmic proteins. In order to better understand selectivity and membrane disruption, here, we determined 3-D structure of the thanatin in zwitterionic DPC-d₃₈ micelle by NMR spectroscopy. The depth of insertion of thanatin into micelle structure was investigated by spin labelled doxyl lipids, 5-DSA and 16-DSA. DPC-bound structure of thanatin is defined by a β-hairpin structure and an extended and turn conformations, for residues G1-I8, at the N-terminus. The β-hairpin structure is delineated by two antiparallel β-strands, residues I9-C11 and residues K17-R20, which is connected by loop consisted of residues N12-G16. There are cross β-strands sidechain-sidechain packing interactions among hydrophobic and aromatic residues. Spin labelled lipid studies revealed a set of spatially proximal residues V6, I8, Q19, R20 and M21 may be deeply inserted into the hydrophobic core of the DPC micelle. While, residues including those at the turn/loop are merely surface localized. The atomic resolution structure and orientation of thanatin in zwitterionic DPC micelle may be utilized for understating mode of action in lipid membrane and further development of non-toxic analogs.

Recent advances in the development of anti-infective peptoids

This feature article highlights the progress that has been made towards the development of novel anti-infective peptoids and the key areas for future development within this field.