Discovery of antimicrobial peptides in the global microbiome with machine learning.

Novel antibiotics are urgently needed to combat the antibiotic-resistance crisis. We present a machine-learning-based approach to predict antimicrobial peptides (AMPs) within the global microbiome and leverage a vast dataset of 63,410 metagenomes and 87,920 prokaryotic genomes from environmental and host-associated habitats to create the AMPSphere, a comprehensive catalog comprising 863,498 non-redundant peptides, few of which match existing databases. AMPSphere provides insights into the evolutionary origins of peptides, including by duplication or gene truncation of longer sequences, and we observed that AMP production varies by habitat. To validate our predictions, we synthesized and tested 100 AMPs against clinically relevant drug-resistant pathogens and human gut commensals both in vitro and in vivo. A total of 79 peptides were active, with 63 targeting pathogens. These active AMPs exhibited antibacterial activity by disrupting bacterial membranes. In conclusion, our approach identified nearly one million prokaryotic AMP sequences, an open-access resource for antibiotic discovery.

Mining human microbiomes reveals an untapped source of peptide antibiotics.

Drug-resistant bacteria are outpacing traditional antibiotic discovery efforts. Here, we computationally screened 444,054 previously reported putative small protein families from 1,773 human metagenomes for antimicrobial properties, identifying 323 candidates encoded in small open reading frames (smORFs). To test our computational predictions, 78 peptides were synthesized and screened for antimicrobial activity in vitro, with 70.5% displaying antimicrobial activity. As these compounds were different compared with previously reported antimicrobial peptides, we termed them smORF-encoded peptides (SEPs). SEPs killed bacteria by targeting their membrane, synergizing with each other, and modulating gut commensals, indicating a potential role in reconfiguring microbiome communities in addition to counteracting pathogens. The lead candidates were anti-infective in both murine skin abscess and deep thigh infection models. Notably, prevotellin-2 from Prevotella copri presented activity comparable to the commonly used antibiotic polymyxin B. Our report supports the existence of hundreds of antimicrobials in the human microbiome amenable to clinical translation.

TP-LMMSG: a peptide prediction graph neural network incorporating flexible amino acid property representation.

Bioactive peptide therapeutics has been a long-standing research topic. Notably, the antimicrobial peptides (AMPs) have been extensively studied for its therapeutic potential. Meanwhile, the demand for annotating other therapeutic peptides, such as antiviral peptides (AVPs) and anticancer peptides (ACPs), also witnessed an increase in recent years. However, we conceive that the structure of peptide chains and the intrinsic information between the amino acids is not fully investigated among the existing protocols. Therefore, we develop a new graph deep learning model, namely TP-LMMSG, which offers lightweight and easy-to-deploy advantages while improving the annotation performance in a generalizable manner. The results indicate that our model can accurately predict the properties of different peptides. The model surpasses the other state-of-the-art models on AMP, AVP and ACP prediction across multiple experimental validated datasets. Moreover, TP-LMMSG also addresses the challenges of time-consuming pre-processing in graph neural network frameworks. With its flexibility in integrating heterogeneous peptide features, our model can provide substantial impacts on the screening and discovery of therapeutic peptides. The source code is available at https://github.com/NanjunChen37/TP_LMMSG.

PGAT-ABPp: harnessing protein language models and graph attention networks for antibacterial peptide identification with remarkable accuracy.

MOTIVATION: The emergence of drug-resistant pathogens represents a formidable challenge to global health. Using computational methods to identify the antibacterial peptides (ABPs), an alternative antimicrobial agent, has demonstrated advantages in further drug design studies. Most of the current approaches, however, rely on handcrafted features and underutilize structural information, which may affect prediction performance. RESULTS: To present an ultra-accurate model for ABP identification, we propose a novel deep learning approach, PGAT-ABPp. PGAT-ABPp leverages structures predicted by AlphaFold2 and a pretrained protein language model, ProtT5-XL-U50 (ProtT5), to construct graphs. Then the graph attention network (GAT) is adopted to learn global discriminative features from the graphs. PGAT-ABPp outperforms the other fourteen state-of-the-art models in terms of accuracy, F1-score and Matthews Correlation Coefficient on the independent test dataset. The results show that ProtT5 has significant advantages in the identification of ABPs and the introduction of spatial information further improves the prediction performance of the model. The interpretability analysis of key residues in known active ABPs further underscores the superiority of PGAT-ABPp. AVAILABILITY AND IMPLEMENTATION: The datasets and source codes for the PGAT-ABPp model are available at https://github.com/moonseter/PGAT-ABPp/.

Molecular mechanism underlying the TLR4 antagonistic and antiseptic activities of papiliocin, an insect innate immune response molecule.

SignificanceSimilar to mammalian TLR4/MD-2, the Toll9/MD-2-like protein complex in the silkworm, Bombyx mori, acts as an innate pattern-recognition receptor that recognizes lipopolysaccharide (LPS) and induces LPS-stimulated expression of antimicrobial peptides such as cecropins. Here, we report that papiliocin, a cecropin-like insect antimicrobial peptide from the swallowtail butterfly, competitively inhibits the LPS-TLR4/MD-2 interaction by directly binding to human TLR4/MD-2. Structural elements in papiliocin, which are important in inhibiting TLR4 signaling via direct binding, are highly conserved among insect cecropins, indicating that its TLR4-antagonistic activity may be related to insect Toll9-mediated immune response against microbial infection. This study highlights the potential of papiliocin as a potent TLR4 antagonist and safe peptide antibiotic for treating gram-negative sepsis.

Exploring pathological link between antimicrobial and amyloid peptides.

Amyloid peptides (AMYs) and antimicrobial peptides (AMPs) are considered as the two distinct families of peptides, characterized by their unique sequences, structures, biological functions, and specific pathological targets. However, accumulating evidence has revealed intriguing pathological connections between these peptide families in the context of microbial infection and neurodegenerative diseases. Some AMYs and AMPs share certain structural and functional characteristics, including the ability to self-assemble, the presence of ß-sheet-rich structures, and membrane-disrupting mechanisms. These shared features enable AMYs to possess antimicrobial activity and AMPs to acquire amyloidogenic properties. Despite limited studies on AMYs-AMPs systems, the cross-seeding phenomenon between AMYs and AMPs has emerged as a crucial factor in the bidirectional communication between the pathogenesis of neurodegenerative diseases and host defense against microbial infections. In this review, we examine recent developments in the potential interplay between AMYs and AMPs, as well as their pathological implications for both infectious and neurodegenerative diseases. By discussing the current progress and challenges in this emerging field, this account aims to inspire further research and investments to enhance our understanding of the intricate molecular crosstalk between AMYs and AMPs. This knowledge holds great promise for the development of innovative therapies to combat both microbial infections and neurodegenerative disorders.

Physiological Microenvironment Dependent Self-Cross-Linking of Multifunctional Nanohybrid for Prolonged Antibacterial Therapy via Synergistic Chemodynamic-Photothermal-Biological Processes.

Herein, a multifunctional nanohybrid (PL@HPFTM nanoparticles) was fabricated to perform the integration of chemodynamic therapy, photothermal therapy, and biological therapy over the long term at a designed location for continuous antibacterial applications. The PL@HPFTM nanoparticles consisted of a polydopamine/hemoglobin/Fe2+ nanocomplex with comodification of tetrazole/alkene groups on the surface as well as coloading of antimicrobial peptides and luminol in the core. During therapy, the PL@HPFTM nanoparticles would selectively cross-link to surrounding bacteria via tetrazole/alkene cycloaddition under chemiluminescence produced by the reaction between luminol and overexpressed H2O2 at the infected area. The resulting PL@HPFTM network not only significantly damaged bacteria by Fe2+-catalyzed ROS production, effective photothermal conversion, and sustained release of antimicrobial peptides but dramatically enhanced the retention time of these therapeutic agents for prolonged antibacterial therapy. Both in vitro and in vivo results have shown that our PL@HPFTM nanoparticles have much higher bactericidal efficiency and remarkably longer periods of validity than free antibacterial nanoparticles.

An efficient hybrid deep learning architecture for predicting short antimicrobial peptides.

Short-length antimicrobial peptides (AMPs) have been demonstrated to have intensified antimicrobial activities against a wide spectrum of microbes. Therefore, exploration of novel and promising short AMPs is highly essential in developing various types of antimicrobial drugs or treatments. In addition to experimental approaches, computational methods have been developed to improve screening efficiency. Although existing computational methods have achieved satisfactory performance, there is still much room for model improvement. In this study, we proposed iAMP-DL, an efficient hybrid deep learning architecture, for predicting short AMPs. The model was constructed using two well-known deep learning architectures: the long short-term memory architecture and convolutional neural networks. To fairly assess the performance of the model, we compared our model with existing state-of-the-art methods using the same independent test set. Our comparative analysis shows that iAMP-DL outperformed other methods. Furthermore, to assess the robustness and stability of our model, the experiments were repeated 10 times to observe the variation in prediction efficiency. The results demonstrate that iAMP-DL is an effective, robust, and stable framework for detecting promising short AMPs. Another comparative study of different negative data sampling methods also confirms the effectiveness of our method and demonstrates that it can also be used to develop a robust model for predicting AMPs in general. The proposed framework was also deployed as an online web server with a user-friendly interface to support the research community in identifying short AMPs.

Insights into Molecular Interactions between a GAPDH-Related Fish Antimicrobial Peptide, Analogs Thereof, and Bacterial Membranes.

Interactions between SJGAP (skipjack tuna GAPDH-related antimicrobial peptide) and four analogs thereof with model bacterial membranes were studied using Fourier-transform infrared spectroscopy (FTIR) and molecular dynamics (MD) simulations. MD trajectory analyses showed that the N-terminal segment of the peptide analogs has many contacts with the polar heads of membrane phospholipids, while the central α helix interacts strongly with the hydrophobic core of the membranes. The peptides also had a marked influence on the wave numbers associated with the phase transition of phospholipids organized as liposomes in both the interface and aliphatic chain regions of the infrared spectra, supporting the interactions observed in the MD trajectories. In addition, interesting links were found between peptide interactions with the aliphatic chains of membrane phospholipids, as determined by FTIR and from the MD trajectories, and the membrane permeabilization capacity of these peptide analogs, as previously demonstrated. To summarize, the combined experimental and computational efforts have provided insights into crucial aspects of the interactions between the investigated peptides and bacterial membranes. This work thus makes an original contribution to our understanding of the molecular interactions underlying the antimicrobial activity of these GAPDH-related antimicrobial peptides from Scombridae.

Identification and Characterization of CC-AMP1-like and CC-AMP2-like Peptides in Capsicum spp.

Antimicrobial peptides (AMPs) are compounds with a variety of bioactive properties. Especially promising are their antibacterial activities, often toward drug-resistant pathogens. Across different AMP sources, AMPs expressed within plants are relatively underexplored with a limited number of plant AMP families identified. Recently, we identified the novel AMPs CC-AMP1 and CC-AMP2 in ghost pepper plants (Capsicum chinense x frutescens), exerting promising antibacterial activity and not classifying into any known plant AMP family. Herein, AMPs related to CC-AMP1 and CC-AMP2 were identified within both Capsicum annuum and Capsicum baccatum. In silico predictions throughout plants were utilized to illustrate that CC-AMP1-like and CC-AMP2-like peptides belong to two broader AMP families, with three-dimensional structural predictions indicating that CC-AMP1-like peptides comprise a novel subfamily of α-hairpinins. The antibacterial activities of several closely related CC-AMP1-like peptides were compared with a truncated version of CC-AMP1 possessing significantly more activity than the full peptide. This truncated peptide was further characterized to possess broad-spectrum antibacterial activity against clinically relevant ESKAPE pathogens. These findings illustrate the value in continued study of plant AMPs toward characterization of novel AMP families, with CC-AMP1-like peptides possessing promising bioactivity.

Antimicrobial Peptides Increase Line Tension in Raft-Forming Lipid Membranes.

The formation of phase separated membrane domains is believed to be essential for the function of the cell. The precise composition and physical properties of lipid bilayer domains play crucial roles in regulating protein activity and governing cellular processes. Perturbation of the domain structure in human cells can be related to neurodegenerative diseases and cancer. Lipid rafts are also believed to be essential in bacteria, potentially serving as targets for antibiotics. An important question is how the membrane domain structure is affected by bioactive and therapeutic molecules, such as surface-active peptides, which target cellular membranes. Here we focus on antimicrobial peptides (AMPs), crucial components of the innate immune system, to gain insights into their interaction with model lipid membranes containing domains. Using small-angle neutron/X-ray scattering (SANS/SAXS), we show that the addition of several natural AMPs (indolicidin, LL-37, magainin II, and aurein 2.2) causes substantial growth and restructuring of the domains, which corresponds to increased line tension. Contrast variation SANS and SAXS results demonstrate that the peptide inserts evenly in both phases, and the increased line tension can be related to preferential and concentration dependent thinning of the unsaturated membrane phase. We speculate that the lateral restructuring caused by the AMPs may have important consequences in affecting physiological functions of real cells. This work thus shines important light onto the complex interactions and lateral (re)organization in lipid membranes, which is relevant for a molecular understanding of diseases and the action of antibiotics.

Medium-sized peptides from microbial sources with potential for antibacterial drug development.

Covering: 1993 to the end of 2022As the rapid development of antibiotic resistance shrinks the number of clinically available antibiotics, there is an urgent need for novel options to fill the existing antibiotic pipeline. In recent years, antimicrobial peptides have attracted increased interest due to their impressive broad-spectrum antimicrobial activity and low probability of antibiotic resistance. However, macromolecular antimicrobial peptides of plant and animal origin face obstacles in antibiotic development because of their extremely short elimination half-life and poor chemical stability. Herein, we focus on medium-sized antibacterial peptides (MAPs) of microbial origin with molecular weights below 2000 Da. The low molecular weight is not sufficient to form complex protein conformations and is also associated to a better chemical stability and easier modifications. Microbially-produced peptides are often composed of a variety of non-protein amino acids and terminal modifications, which contribute to improving the elimination half-life of compounds. Therefore, MAPs have great potential for drug discovery and are likely to become key players in the development of next-generation antibiotics. In this review, we provide a detailed exploration of the modes of action demonstrated by 45 MAPs and offer a concise summary of the structure-activity relationships observed in these MAPs.

Antimalarial activity of cecropin antimicrobial peptides derived from Anopheles mosquitoes.

The emergence of clinically drug-resistant malaria parasites requires the urgent development of new drugs. Mosquitoes are vectors of multiple pathogens and have developed resistance mechanisms against them, which often involve antimicrobial peptides (AMPs). An-cecB is an AMP of the malaria-transmitting mosquito genus Anopheles, and we herein report its antimalarial activity against Plasmodium falciparum 3D7, the artemisinin-resistant strain 803, and the chloroquine-resistant strain Dd2 in vitro. We also demonstrate its anti-parasite activity in vivo, using the rodent malaria parasite Plasmodium berghei (ANKA). We show that An-cecB displays potent antimalarial activity and that its mechanism of action may occur through direct killing of the parasite or through interaction with infected red blood cell membranes. Unfortunately, An-cecB was found to be cytotoxic to mammalian cells and had poor antimalarial activity in vivo. However, its truncated peptide An-cecB-1 retained most of its antimalarial activity and avoided its cytotoxicity in vitro. An-cecB-1 also showed better antimalarial activity in vivo. Mosquito-derived AMPs may provide new ideas for the development of antimalarial drugs against drug-resistant parasites, and An-cecB has potential use as a template for antimalarial peptides.

Whole-genome sequencing and identification of antimicrobial peptide coding genes in parsley (Petroselinum crispum), an important culinary and medicinal Apiaceae species.

Parsley is a commonly cultivated Apiaceae species of culinary and medicinal importance. Parsley has several recognized health benefits and the species has been utilized in traditional medicine since ancient times. Although parsley is among the most commonly cultivated members of Apiaceae, no systematic genomic research has been conducted on parsley. In the present work, parsley genome was sequenced using the long-read HiFi (high fidelity) sequencing technology and a draft contig assembly of 1.57 Gb that represents 80.9% of the estimated genome size was produced. The assembly was highly repeat-rich with a repetitive DNA content of 81%. The assembly was phased into a primary and alternate assembly in order to minimize redundant contigs. Scaffolds were constructed with the primary assembly contigs, which were used for the identification of AMP (antimicrobial peptide) genes. Characteristic AMP domains and 3D structures were used to detect and verify antimicrobial peptides. As a result, 23 genes (PcAMP1-23) representing defensin, snakin, thionin, lipid transfer protein and vicilin-like AMP classes were identified. Bioinformatic analyses for the characterization of peptide physicochemical properties indicated that parsley AMPs are extracellular peptides, therefore, plausibly exert their antimicrobial effects through the most commonly described AMP action mechanism of membrane attack. AMPs are attracting increasing attention since they display their fast antimicrobial effects in small doses on both plant and animal pathogens with a significantly reduced risk of resistance development. Therefore, identification and characterization of AMPs is important for their incorporation into plant disease management protocols as well as medicinal research for the treatment of multi-drug resistant infections.

First vertebrate BRICHOS antimicrobial peptides: ß-hairpin host defense peptides in limbless amphibia lung resemble those of marine worms.

Innate immunity of invertebrates offers potent antimicrobial peptides (AMPs) against drug-resistant infections. To identify new worm ß-hairpin AMPs, we explored the sequence diversity of proteins with a BRICHOS domain, which comprises worm AMP precursors. Strikingly, we discovered new BRICHOS AMPs not in worms, but in caecilians, the least studied clade of vertebrates. Two precursor proteins from Microcaecilia unicolor and Rhinatrema bivittatum resemble SP-C lung surfactants and bear worm AMP-like peptides at C-termini. The analysis of M. unicolor tissue transcriptomes shows that the AMP precursor is highly expressed in the lung along with regular SP-C, suggesting a different, protective function. The peptides form right-twisted ß-hairpins, change conformation upon lipid binding, and rapidly disrupt bacterial membranes. Both peptides exhibit broad-spectrum activity against multidrug-resistant ESKAPE pathogens with 1-4 µM MICs and remarkably low toxicity, giving 40-70-fold selectivity towards bacteria. These BRICHOS AMPs, previously unseen in vertebrates, reveal a novel lung innate immunity mechanism and offer a promising antibiotics template.

Casein phosphopeptide/antimicrobial peptide co-modified SrTiO3 nanotubes for prevention of bacterial infections and repair of bone defects.

Endowing titanium surfaces with multifunctional properties can reduce implant-related infections and enhance osseointegration. In this study, titanium dioxide nanotubes with strontium doping (STN) were first created on the titanium surface using anodic oxidation and hydrothermal synthesis techniques. Next, casein phosphopeptide (CCP) and an antimicrobial peptide (HHC36) were loaded into the STN with the aid of vacuum physical adsorption (STN-CP-H), giving the titanium surface a dual function of "antimicrobial-osteogenic". The surface of STN-CP-H has a suitable roughness and good hydrophilicity, which is conducive to osteoblasts. STN-CP-H had a 99 % antibacterial rate against S. aureus and E. coli and effectively prevented the growth of bacterial biofilm. Meanwhile, the antibacterial mechanism of STN-CP-H was initially explored with the help of transcriptome sequencing technology. STN-CP-H could greatly increase osteoblast adhesion, proliferation, and expression of osteogenic markers (alkaline phosphatase, runt-related transcription) when CCP and Sr worked together synergistically. In vivo, the STN-CP-H coating could effectively promote new osteogenesis around titanium implant bone and had no toxic effects on heart, liver, spleen, lung and kidney tissues. A potential anti-infection bone healing material, STN-CP-H bifunctional coating developed in this work efficiently inhibited bacterial infection of titanium implants and encouraged early osseointegration.

Boosting Membrane Interactions and Antimicrobial Effects of Photocatalytic Titanium Dioxide Nanoparticles by Peptide Coating.

Photocatalytic nanoparticles offer antimicrobial effects under illumination due to the formation of reactive oxygen species (ROS), capable of degrading bacterial membranes. ROS may, however, also degrade human cell membranes and trigger toxicity. Since antimicrobial peptides (AMPs) may display excellent selectivity between human cells and bacteria, these may offer opportunities to effectively "target" nanoparticles to bacterial membranes for increased selectivity. Investigating this, photocatalytic TiO2 nanoparticles (NPs) are coated with the AMP LL-37, and ROS generation is found by C11-BODIPY to be essentially unaffected after AMP coating. Furthermore, peptide-coated TiO2 NPs retain their positive ζ-potential also after 1-2 h of UV illumination, showing peptide degradation to be sufficiently limited to allow peptide-mediated targeting. In line with this, quartz crystal microbalance measurements show peptide coating to promote membrane binding of TiO2 NPs, particularly so for bacteria-like anionic and cholesterol-void membranes. As a result, membrane degradation during illumination is strongly promoted for such membranes, but not so for mammalian-like membranes. The mechanisms of these effects are elucidated by neutron reflectometry. Analogously, LL-37 coating promoted membrane rupture by TiO2 NPs for Gram-negative and Gram-positive bacteria, but not for human monocytes. These findings demonstrate that AMP coating may selectively boost the antimicrobial effects of photocatalytic NPs.

In Silico Study of the Early Stages of Aggregation of ß-Sheet Forming Antimicrobial Peptide GL13K.

Antimicrobial peptides (AMPs) are of growing interest as potential candidates that may offer more resilience against antimicrobial resistance than traditional antibiotic agents. In this article, we perform the first in silico study of the synthetic ß sheet-forming AMP GL13K. Through atomistic simulations of single and multi-peptide systems under different conditions, we are able to shine a light on the short timescales of early aggregation. We find that isolated peptide conformations are primarily dictated by sequence rather than charge, whereas changing charge has a significant impact on the conformational free energy landscape of multi-peptide systems. We demonstrate that the loss of charge-charge repulsion is a sufficient minimal model for experimentally observed aggregation. Overall, our work explores the molecular biophysical underpinnings of the first stages of aggregation of a unique AMP, laying necessary groundwork for its further development as an antibiotic candidate.

Endogenous Antimicrobial-Immunomodulatory Molecules: Networking Biomolecules of Innate Immunity.

Endogenous antimicrobial-immunomodulatory molecules (EAIMs) are essential to immune-mediated human health and evolution. Conventionally, antimicrobial peptides (AMPs) have been regarded as the dominant endogenous antimicrobial molecule; however, AMPs are not sufficient to account for the full spectrum of antimicrobial-immunomodulatory duality occurring within the human body. The threat posed by pathogenic microbes is pervasive with the capacity for widespread impact across many organ systems and multiple biochemical pathways; accordingly, the host needs the capacity to react with an equally diverse response. This can be attained by having EAIMs that traverse the full range of molecular size (small to large molecules) and structural diversity (including molecules other than peptides). This review identifies multiple molecules (peptide/protein, lipid, carbohydrate, nucleic acid, small organic molecule, and metallic cation) as EAIMs and discusses the possibility of cooperative, additive effects amongst the various EAIM classes during the host response to a microbial assault. This comprehensive consideration of the full molecular diversity of EAIMs enables the conclusion that EAIMs constitute a previously uncatalogued structurally diverse and collectively underappreciated immuno-active group of integrated molecular responders within the innate immune system's first line of defence.

A novel antimicrobial peptide S24 combats serious wound infections caused by Pseudomonas aeruginosa and Acinetobacter baumannii.

OBJECTIVES: Pseudomonas aeruginosa and Acinetobacter baumannii are ranked as top-priority organisms by WHO. Antimicrobial peptides (AMPs) are promising antimicrobial agents that are highly effective against serious bacterial infections. METHODS: In our previous study, a series of α-helical AMPs were screened using a novel multiple-descriptor strategy. The current research suggested that S24 exhibited strong antimicrobial activity against major pathogenic bacteria, and displayed minimal haemolysis, good serum stability and maintained salt resistance. RESULTS: We found that S24 exerted an antimicrobial effect by destroying outer membrane permeability and producing a strong binding effect on bacterial genomic DNA that inhibits genomic DNA migration. Furthermore, S24 exerted a strong ability to promote healing in wound infected by P. aeruginosa, A. baumannii and mixed strains in a mouse model. CONCLUSIONS: Overall, S24 showed good stability under physiological conditions and excellent antimicrobial activity, suggesting it may be a potential candidate for the development of serious bacterial infection treatment.