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.

Lung antimicrobial proteins and peptides: From host defense to therapeutic strategies.

Representing severe morbidity and mortality globally, respiratory infections associated with chronic respiratory diseases, including complicated pneumonia, asthma, interstitial lung disease, and chronic obstructive pulmonary disease, are a major public health concern. Lung health and the prevention of pulmonary disease rely on the mechanisms of airway surface fluid secretion, mucociliary clearance, and adequate immune response to eradicate inhaled pathogens and particulate matter from the environment. The antimicrobial proteins and peptides contribute to maintaining an antimicrobial milieu in human lungs to eliminate pathogens and prevent them from causing pulmonary diseases. The predominant antimicrobial molecules of the lung environment include human α- and ß-defensins and cathelicidins, among numerous other host defense molecules with antimicrobial and antibiofilm activity such as PLUNC (palate, lung, and nasal epithelium clone) family proteins, elafin, collectins, lactoferrin, lysozymes, mucins, secretory leukocyte proteinase inhibitor, surfactant proteins SP-A, and SP-D, and RNases. It has been demonstrated that changes in antimicrobial molecule expression levels are associated with regulating inflammation, potentiating exacerbations, pathological changes, and modifications in chronic lung disease severity. Antimicrobial molecules also display roles in both anticancer and tumorigenic effects. Lung antimicrobial proteins and peptides are promising alternative therapeutics for treating and preventing multidrug-resistant bacterial infections and anticancer therapies.

Ancestral ribonucleases back in motion for evolutionary-dynamics guided protein design.

The dynamics behavior of a protein is essential for its functionality. Here, Doucet et al. demonstrate how the evolutionary analysis of conformational pathways within a protein family serves to identify common core scaffolds that accommodate branch-specific functional regions controlled by flexibility switches, offering a model for evolutionary-dynamics based protein design.

Cracking the type 1 diabetes code: Genes, microbes, immunity, and the early life environment.

Type 1 diabetes (T1D) results from a complex interplay of genetic predisposition, immunological dysregulation, and environmental triggers, that culminate in the destruction of insulin-secreting pancreatic ß cells. This review provides a comprehensive examination of the multiple factors underpinning T1D pathogenesis, to elucidate key mechanisms and potential therapeutic targets. Beginning with an exploration of genetic risk factors, we dissect the roles of human leukocyte antigen (HLA) haplotypes and non-HLA gene variants associated with T1D susceptibility. Mechanistic insights gleaned from the NOD mouse model provide valuable parallels to the human disease, particularly immunological intricacies underlying ß cell-directed autoimmunity. Immunological drivers of T1D pathogenesis are examined, highlighting the pivotal contributions of both effector and regulatory T cells and the multiple functions of B cells and autoantibodies in ß-cell destruction. Furthermore, the impact of environmental risk factors, notably modulation of host immune development by the intestinal microbiome, is examined. Lastly, the review probes human longitudinal studies, unveiling the dynamic interplay between mucosal immunity, systemic antimicrobial antibody responses, and the trajectories of T1D development. Insights garnered from these interconnected factors pave the way for targeted interventions and the identification of biomarkers to enhance T1D management and prevention strategies.

Grow now, pay later: When should a bacterium go into debt?

Microbes grow in a wide variety of environments and must balance growth and stress resistance. Despite the prevalence of such trade-offs, understanding of their role in nonsteady environments is limited. In this study, we introduce a mathematical model of "growth debt," where microbes grow rapidly initially, paying later with slower growth or heightened mortality. We first compare our model to a classical chemostat experiment, validating our proposed dynamics and quantifying Escherichia coli's stress resistance dynamics. Extending the chemostat theory to include serial-dilution cultures, we derive phase diagrams for the persistence of "debtor" microbes. We find that debtors cannot coexist with nondebtors if "payment" is increased mortality but can coexist if it lowers enzyme affinity. Surprisingly, weak noise considerably extends the persistence of resistance elements, pertinent for antibiotic resistance management. Our microbial debt theory, broadly applicable across many environments, bridges the gap between chemostat and serial dilution systems.

Hundreds of antimicrobial peptides create a selective barrier for insect gut symbionts.

The spatial organization of gut microbiota is crucial for the functioning of the gut ecosystem, although the mechanisms that organize gut bacterial communities in microhabitats are only partially understood. The gut of the insect Riptortus pedestris has a characteristic microbiota biogeography with a multispecies community in the anterior midgut and a monospecific bacterial population in the posterior midgut. We show that the posterior midgut region produces massively hundreds of specific antimicrobial peptides (AMPs), the Crypt-specific Cysteine-Rich peptides (CCRs) that have membrane-damaging antimicrobial activity against diverse bacteria but posterior midgut symbionts have elevated resistance. We determined by transposon-sequencing the genetic repertoire in the symbiont Caballeronia insecticola to manage CCR stress, identifying different independent pathways, including AMP-resistance pathways unrelated to known membrane homeostasis functions as well as cell envelope functions. Mutants in the corresponding genes have reduced capacity to colonize the posterior midgut, demonstrating that CCRs create a selective barrier and resistance is crucial in gut symbionts. Moreover, once established in the gut, the bacteria differentiate into a CCR-sensitive state, suggesting a second function of the CCR peptide arsenal in protecting the gut epithelia or mediating metabolic exchanges between the host and the gut symbionts. Our study highlights the evolution of an extreme diverse AMP family that likely contributes to establish and control the gut microbiota.

Engineering a nanoantibiotic system displaying dual mechanism of action.

In recent decades, peptide amphiphiles (PAs) have established themselves as promising self-assembling bioinspired materials in a wide range of medical fields. Herein, we report a dual-therapeutic system constituted by an antimicrobial PA and a cylindrical protease inhibitor (LJC) to achieve broad antimicrobial spectrum and to enhance therapeutic efficacy. We studied two strategies: PA-LJC nanostructures (Encapsulation) and PA nanostructures + free LJC (Combination). Computational modeling using a molecular theory for amphiphile self-assembly captures and explains the morphology of PA-LJC nanostructures and the location of encapsulated LJC in agreement with transmission electron microscopy and two-dimensional (2D) NMR observations. The morphology and release profile of PA-LJC assemblies are strongly correlated to the PA:LJC ratio: high LJC loading induces an initial burst release. We then evaluated the antimicrobial activity of our nanosystems toward gram-positive and gram-negative bacteria. We found that the Combination broadens the spectrum of LJC, reduces the therapeutic concentrations of both agents, and is not impacted by the inoculum effect. Further, the Encapsulation provides additional benefits including bypassing water solubility limitations of LJC and modulating the release of this molecule. The different properties of PA-LJC nanostructures results in different killing profiles, and reduced cytotoxicity and hemolytic activity. Meanwhile, details in membrane alterations caused by each strategy were revealed by various microscopy and fluorescent techniques. Last, in vivo studies in larvae treated by the Encapsulation strategy showed better antimicrobial efficacy than polymyxin B. Collectively, this study established a multifunctional platform using a versatile PA to act as an antibiotic, membrane-penetrating assistant, and slow-release delivery vehicle.

Viral afterlife: SARS-CoV-2 as a reservoir of immunomimetic peptides that reassemble into proinflammatory supramolecular complexes.

It is unclear how severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection leads to the strong but ineffective inflammatory response that characterizes severe Coronavirus disease 2019 (COVID-19), with amplified immune activation in diverse cell types, including cells without angiotensin-converting enzyme 2 receptors necessary for infection. Proteolytic degradation of SARS-CoV-2 virions is a milestone in host viral clearance, but the impact of remnant viral peptide fragments from high viral loads is not known. Here, we examine the inflammatory capacity of fragmented viral components from the perspective of supramolecular self-organization in the infected host environment. Interestingly, a machine learning analysis to SARS-CoV-2 proteome reveals sequence motifs that mimic host antimicrobial peptides (xenoAMPs), especially highly cationic human cathelicidin LL-37 capable of augmenting inflammation. Such xenoAMPs are strongly enriched in SARS-CoV-2 relative to low-pathogenicity coronaviruses. Moreover, xenoAMPs from SARS-CoV-2 but not low-pathogenicity homologs assemble double-stranded RNA (dsRNA) into nanocrystalline complexes with lattice constants commensurate with the steric size of Toll-like receptor (TLR)-3 and therefore capable of multivalent binding. Such complexes amplify cytokine secretion in diverse uninfected cell types in culture (epithelial cells, endothelial cells, keratinocytes, monocytes, and macrophages), similar to cathelicidin's role in rheumatoid arthritis and lupus. The induced transcriptome matches well with the global gene expression pattern in COVID-19, despite using <0.3% of the viral proteome. Delivery of these complexes to uninfected mice boosts plasma interleukin-6 and CXCL1 levels as observed in COVID-19 patients.

CFTR is required for zinc-mediated antibacterial defense in human macrophages.

Cystic fibrosis transmembrane conductance regulator (CFTR) is an anion transporter required for epithelial homeostasis in the lung and other organs, with CFTR mutations leading to the autosomal recessive genetic disease CF. Apart from excessive mucus accumulation and dysregulated inflammation in the airways, people with CF (pwCF) exhibit defective innate immune responses and are susceptible to bacterial respiratory pathogens such as Pseudomonas aeruginosa. Here, we investigated the role of CFTR in macrophage antimicrobial responses, including the zinc toxicity response that is used by these innate immune cells against intracellular bacteria. Using both pharmacological approaches, as well as cells derived from pwCF, we show that CFTR is required for uptake and clearance of pathogenic Escherichia coli by CSF-1-derived primary human macrophages. CFTR was also required for E. coli-induced zinc accumulation and zinc vesicle formation in these cells, and E. coli residing in macrophages exhibited reduced zinc stress in the absence of CFTR function. Accordingly, CFTR was essential for reducing the intramacrophage survival of a zinc-sensitive E. coli mutant compared to wild-type E. coli. Ectopic expression of the zinc transporter SLC30A1 or treatment with exogenous zinc was sufficient to restore antimicrobial responses against E. coli in human macrophages. Zinc supplementation also restored bacterial killing in GM-CSF-derived primary human macrophages responding to P. aeruginosa, used as an in vitro macrophage model relevant to CF. Thus, restoration of the zinc toxicity response could be pursued as a therapeutic strategy to restore innate immune function and effective host defense in pwCF.

Development of 2nd generation aminomethyl spectinomycins that overcome native efflux in Mycobacterium abscessus.

Mycobacterium abscessus (Mab), a nontuberculous mycobacterial (NTM) species, is an emerging pathogen with high intrinsic drug resistance. Current standard-of-care therapy results in poor outcomes, demonstrating the urgent need to develop effective antimycobacterial regimens. Through synthetic modification of spectinomycin (SPC), we have identified a distinct structural subclass of N-ethylene linked aminomethyl SPCs (eAmSPCs) that are up to 64-fold more potent against Mab over the parent SPC. Mechanism of action and crystallography studies demonstrate that the eAmSPCs display a mode of ribosomal inhibition consistent with SPC. However, they exert their increased antimicrobial activity through enhanced accumulation, largely by circumventing efflux mechanisms. The N-ethylene linkage within this series plays a critical role in avoiding TetV-mediated efflux, as lead eAmSPC 2593 displays a mere fourfold susceptibility improvement against Mab ΔtetV, in contrast to the 64-fold increase for SPC. Even a minor shortening of the linkage by a single carbon, akin to 1st generation AmSPC 1950, results in a substantial increase in MICs and a 16-fold rise in susceptibility against Mab ΔtetV. These shifts suggest that longer linkages might modify the kinetics of drug expulsion by TetV, ultimately shifting the equilibrium towards heightened intracellular concentrations and enhanced antimicrobial efficacy. Furthermore, lead eAmSPCs were also shown to synergize with various classes of anti-Mab antibiotics and retain activity against clinical isolates and other mycobacterial strains. Encouraging pharmacokinetic profiles coupled with robust efficacy in Mab murine infection models suggest that eAmSPCs hold the potential to be developed into treatments for Mab and other NTM infections.

A comparative analysis of peptide-delivered antisense antibiotics using diverse nucleotide mimics.

Antisense oligomer (ASO)-based antibiotics that target mRNAs of essential bacterial genes have great potential for counteracting antimicrobial resistance and for precision microbiome editing. To date, the development of such antisense antibiotics has primarily focused on using phosphorodiamidate morpholino (PMO) and peptide nucleic acid (PNA) backbones, largely ignoring the growing number of chemical modalities that have spurred the success of ASO-based human therapy. Here, we directly compare the activities of seven chemically distinct 10mer ASOs, all designed to target the essential gene acpP upon delivery with a KFF-peptide carrier into Salmonella. Our systematic analysis of PNA, PMO, phosphorothioate (PTO)-modified DNA, 2'-methylated RNA (RNA-OMe), 2'-methoxyethylated RNA (RNA-MOE), 2'-fluorinated RNA (RNA-F), and 2'-4'-locked RNA (LNA) is based on a variety of in vitro and in vivo methods to evaluate ASO uptake, target pairing and inhibition of bacterial growth. Our data show that only PNA and PMO are efficiently delivered by the KFF peptide into Salmonella to inhibit bacterial growth. Nevertheless, the strong target binding affinity and in vitro translational repression activity of LNA and RNA-MOE make them promising modalities for antisense antibiotics that will require the identification of an effective carrier.

Assessing computational predictions of antimicrobial resistance phenotypes from microbial genomes.

The advent of rapid whole-genome sequencing has created new opportunities for computational prediction of antimicrobial resistance (AMR) phenotypes from genomic data. Both rule-based and machine learning (ML) approaches have been explored for this task, but systematic benchmarking is still needed. Here, we evaluated four state-of-the-art ML methods (Kover, PhenotypeSeeker, Seq2Geno2Pheno and Aytan-Aktug), an ML baseline and the rule-based ResFinder by training and testing each of them across 78 species-antibiotic datasets, using a rigorous benchmarking workflow that integrates three evaluation approaches, each paired with three distinct sample splitting methods. Our analysis revealed considerable variation in the performance across techniques and datasets. Whereas ML methods generally excelled for closely related strains, ResFinder excelled for handling divergent genomes. Overall, Kover most frequently ranked top among the ML approaches, followed by PhenotypeSeeker and Seq2Geno2Pheno. AMR phenotypes for antibiotic classes such as macrolides and sulfonamides were predicted with the highest accuracies. The quality of predictions varied substantially across species-antibiotic combinations, particularly for beta-lactams; across species, resistance phenotyping of the beta-lactams compound, aztreonam, amoxicillin/clavulanic acid, cefoxitin, ceftazidime and piperacillin/tazobactam, alongside tetracyclines demonstrated more variable performance than the other benchmarked antibiotics. By organism, Campylobacter jejuni and Enterococcus faecium phenotypes were more robustly predicted than those of Escherichia coli, Staphylococcus aureus, Salmonella enterica, Neisseria gonorrhoeae, Klebsiella pneumoniae, Pseudomonas aeruginosa, Acinetobacter baumannii, Streptococcus pneumoniae and Mycobacterium tuberculosis. In addition, our study provides software recommendations for each species-antibiotic combination. It furthermore highlights the need for optimization for robust clinical applications, particularly for strains that diverge substantially from those used for training.

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.

Diff-AMP: tailored designed antimicrobial peptide framework with all-in-one generation, identification, prediction and optimization.

Antimicrobial peptides (AMPs), short peptides with diverse functions, effectively target and combat various organisms. The widespread misuse of chemical antibiotics has led to increasing microbial resistance. Due to their low drug resistance and toxicity, AMPs are considered promising substitutes for traditional antibiotics. While existing deep learning technology enhances AMP generation, it also presents certain challenges. Firstly, AMP generation overlooks the complex interdependencies among amino acids. Secondly, current models fail to integrate crucial tasks like screening, attribute prediction and iterative optimization. Consequently, we develop a integrated deep learning framework, Diff-AMP, that automates AMP generation, identification, attribute prediction and iterative optimization. We innovatively integrate kinetic diffusion and attention mechanisms into the reinforcement learning framework for efficient AMP generation. Additionally, our prediction module incorporates pre-training and transfer learning strategies for precise AMP identification and screening. We employ a convolutional neural network for multi-attribute prediction and a reinforcement learning-based iterative optimization strategy to produce diverse AMPs. This framework automates molecule generation, screening, attribute prediction and optimization, thereby advancing AMP research. We have also deployed Diff-AMP on a web server, with code, data and server details available in the Data Availability section.

How plants manage pathogen infection.

To combat microbial pathogens, plants have evolved specific immune responses that can be divided into three essential steps: microbial recognition by immune receptors, signal transduction within plant cells, and immune execution directly suppressing pathogens. During the past three decades, many plant immune receptors and signaling components and their mode of action have been revealed, markedly advancing our understanding of the first two steps. Activation of immune signaling results in physical and chemical actions that actually stop pathogen infection. Nevertheless, this third step of plant immunity is under explored. In addition to immune execution by plants, recent evidence suggests that the plant microbiota, which is considered an additional layer of the plant immune system, also plays a critical role in direct pathogen suppression. In this review, we summarize the current understanding of how plant immunity as well as microbiota control pathogen growth and behavior and highlight outstanding questions that need to be answered.

Drosophila eIF3f1 mediates host immune defense by targeting dTak1.

Eukaryotic translation initiation factors have long been recognized for their critical roles in governing the translation of coding RNAs into peptides/proteins. However, whether they harbor functional activities at the post-translational level remains poorly understood. Here, we demonstrate that eIF3f1 (eukaryotic translation initiation factor 3 subunit f1), which encodes an archetypal deubiquitinase, is essential for the antimicrobial innate immune defense of Drosophila melanogaster. Our in vitro and in vivo evidence indicate that the immunological function of eIF3f1 is dependent on the N-terminal JAMM (JAB1/MPN/Mov34 metalloenzymes) domain. Mechanistically, eIF3f1 physically associates with dTak1 (Drosophila TGF-beta activating kinase 1), a key regulator of the IMD (immune deficiency) signaling pathway, and mediates the turnover of dTak1 by specifically restricting its K48-linked ubiquitination. Collectively, these results provide compelling insight into a noncanonical molecular function of a translation initiation factor that controls the post-translational modification of a target protein.

Non-faecium non-faecalis enterococci: a review of clinical manifestations, virulence factors, and antimicrobial resistance.

SUMMARYEnterococci are a diverse group of Gram-positive bacteria that are typically found as commensals in humans, animals, and the environment. Occasionally, they may cause clinically relevant diseases such as endocarditis, septicemia, urinary tract infections, and wound infections. The majority of clinical infections in humans are caused by two species: Enterococcus faecium and Enterococcus faecalis. However, there is an increasing number of clinical infections caused by non-faecium non-faecalis (NFF) enterococci. Although NFF enterococcal species are often overlooked, studies have shown that they may harbor antimicrobial resistance (AMR) genes and virulence factors that are found in E. faecium and E. faecalis. In this review, we present an overview of the NFF enterococci with a particular focus on human clinical manifestations, epidemiology, virulence genes, and AMR genes.

Challenges for global antibiotic regimen planning and establishing antimicrobial resistance targets: implications for the WHO Essential Medicines List and AWaRe antibiotic book dosing.

SUMMARYThe World Health Organisation's 2022 AWaRe Book provides guidance for the use of 39 antibiotics to treat 35 infections in primary healthcare and hospital facilities. We review the evidence underpinning suggested dosing regimens. Few (n = 18) population pharmacokinetic studies exist for key oral AWaRe antibiotics, largely conducted in homogenous and unrepresentative populations hindering robust estimates of drug exposures. Databases of minimum inhibitory concentration distributions are limited, especially for community pathogen-antibiotic combinations. Minimum inhibitory concentration data sources are not routinely reported and lack regional diversity and community representation. Of studies defining a pharmacodynamic target for ß-lactams (n = 80), 42 (52.5%) differed from traditionally accepted 30%-50% time above minimum inhibitory concentration targets. Heterogeneity in model systems and pharmacodynamic endpoints is common, and models generally use intravenous ß-lactams. One-size-fits-all pharmacodynamic targets are used for regimen planning despite complexity in drug-pathogen-disease combinations. We present solutions to enable the development of global evidence-based antibiotic dosing guidance that provides adequate treatment in the context of the increasing prevalence of antimicrobial resistance and, moreover, minimizes the emergence of resistance.

Intestinal biofilms: pathophysiological relevance, host defense, and therapeutic opportunities.

SUMMARYThe human intestinal tract harbors a profound variety of microorganisms that live in symbiosis with the host and each other. It is a complex and highly dynamic environment whose homeostasis directly relates to human health. Dysbiosis of the gut microbiota and polymicrobial biofilms have been associated with gastrointestinal diseases, including irritable bowel syndrome, inflammatory bowel diseases, and colorectal cancers. This review covers the molecular composition and organization of intestinal biofilms, mechanistic aspects of biofilm signaling networks for bacterial communication and behavior, and synergistic effects in polymicrobial biofilms. It further describes the clinical relevance and diseases associated with gut biofilms, the role of biofilms in antimicrobial resistance, and the intestinal host defense system and therapeutic strategies counteracting biofilms. Taken together, this review summarizes the latest knowledge and research on intestinal biofilms and their role in gut disorders and provides directions toward the development of biofilm-specific treatments.