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.

Minute-scale detection of SARS-CoV-2 using a low-cost biosensor composed of pencil graphite electrodes.

COVID-19 has led to over 3.47 million deaths worldwide and continues to devastate primarily middle- and low-income countries. High-frequency testing has been proposed as a potential solution to prevent outbreaks. However, current tests are not sufficiently low-cost, rapid, or scalable to enable broad COVID-19 testing. Here, we describe LEAD (Low-cost Electrochemical Advanced Diagnostic), a diagnostic test that detects severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) within 6.5 min and costs $1.50 per unit to produce using easily accessible and commercially available materials. LEAD is highly sensitive toward SARS-CoV-2 spike protein (limit of detection = 229 fg⋅mL-1) and displays an excellent performance profile using clinical saliva (100.0% sensitivity, 100.0% specificity, and 100.0% accuracy) and nasopharyngeal/oropharyngeal (88.7% sensitivity, 86.0% specificity, and 87.4% accuracy) samples. No cross-reactivity was detected with other coronavirus or influenza strains. Importantly, LEAD also successfully diagnosed the highly contagious SARS-CoV-2 B.1.1.7 UK variant. The device presents high reproducibility under all conditions tested and preserves its original sensitivity for 5 d when stored at 4 °C in phosphate-buffered saline. Our low-cost and do-it-yourself technology opens new avenues to facilitate high-frequency testing and access to much-needed diagnostic tests in resource-limited settings and low-income communities.

Repurposing a peptide toxin from wasp venom into antiinfectives with dual antimicrobial and immunomodulatory properties.

Novel antibiotics are urgently needed to combat multidrug-resistant pathogens. Venoms represent previously untapped sources of novel drugs. Here we repurposed mastoparan-L, the toxic active principle derived from the venom of the wasp Vespula lewisii, into synthetic antimicrobials. We engineered within its N terminus a motif conserved among natural peptides with potent immunomodulatory and antimicrobial activities. The resulting peptide, mast-MO, adopted an α-helical structure as determined by NMR, exhibited increased antibacterial properties comparable to standard-of-care antibiotics both in vitro and in vivo, and potentiated the activity of different classes of antibiotics. Mechanism-of-action studies revealed that mast-MO targets bacteria by rapidly permeabilizing their outer membrane. In animal models, the peptide displayed direct antimicrobial activity, led to enhanced ability to attract leukocytes to the infection site, and was able to control inflammation. Permutation studies depleted the remaining toxicity of mast-MO toward human cells, yielding derivatives with antiinfective activity in animals. We demonstrate a rational design strategy for repurposing venoms into promising antimicrobials.

Biologically Active Peptides from Venoms: Applications in Antibiotic Resistance, Cancer, and Beyond.

Peptides are potential therapeutic alternatives against global diseases, such as antimicrobial-resistant infections and cancer. Venoms are a rich source of bioactive peptides that have evolved over time to act on specific targets of the prey. Peptides are one of the main components responsible for the biological activity and toxicity of venoms. South American organisms such as scorpions, snakes, and spiders are important producers of a myriad of peptides with different biological activities. In this review, we report the main venom-derived peptide families produced from South American organisms and their corresponding activities and biological targets.

Molecular Dynamics for Antimicrobial Peptide Discovery.

Although antimicrobial resistance is an increasingly significant public health concern, there have only been two new classes of antibiotics approved for human use since the 1960s. Understanding the mechanisms of action of antibiotics is critical for novel antibiotic discovery, but novel approaches are needed that do not exclusively rely on experiments. Molecular dynamics simulation is a computational tool that uses simple models of the atoms in a system to discover nanoscale insights into the dynamic relationship between mechanism and biological function. Such insights can lay the framework for elucidating the mechanism of action and optimizing antibiotic templates. Antimicrobial peptides represent a promising solution to escalating antimicrobial resistance, given their lesser tendency to induce resistance than that of small-molecule antibiotics. Simulations of these agents have already revealed how they interact with bacterial membranes and the underlying physiochemical features directing their structure and function. In this minireview, we discuss how traditional molecular dynamics simulation works and its role and potential for the development of new antibiotic candidates with an emphasis on antimicrobial peptides.

Antimicrobial and Antibiofilm Activities of Helical Antimicrobial Peptide Sequences Incorporating Metal-Binding Motifs.

Antimicrobial peptides (AMPs) represent alternative strategies to combat the global health problem of antibiotic resistance. However, naturally occurring AMPs are generally not sufficiently active for use as antibiotics. Optimized synthetic versions incorporating additional design principles are needed. Here, we engineered amino-terminal Cu(II) and Ni(II) (ATCUN) binding motifs, which can enhance biological function, into the native sequence of two AMPs, CM15 and citropin1.1. The incorporation of metal-binding motifs modulated the antimicrobial activity of synthetic peptides against a panel of carbapenem-resistant enterococci (CRE) bacteria, including carbapenem-resistant Klebsiella pneumoniae (KpC+) and Escherichia coli (KpC+). Activity modulation depended on the type of ATCUN variant utilized. Membrane permeability assays revealed that the in silico selected lead template, CM15, and its ATCUN analogs increased bacterial cell death. Mass spectrometry, circular dichroism, and molecular dynamics simulations indicated that coordinating ATCUN derivatives with Cu(II) ions did not increase the helical tendencies of the AMPs. CM15 ATCUN variants, when combined with Meropenem, streptomycin, or chloramphenicol, showed synergistic effects against E. coli (KpC+ 1812446) biofilms. Motif addition also reduced the hemolytic activity of the wild-type AMP and improved the survival rate of mice in a systemic infection model. The dependence of these bioactivities on the particular amino acids of the ATCUN motif highlights the possible use of size, charge, and hydrophobicity to fine-tune AMP biological function. Our data indicate that incorporating metal-binding motifs into peptide sequences leads to synthetic variants with modified biological properties. These principles may be applied to augment the activities of other peptide sequences.

Novel bioactive peptides from PD-L1/2, a type 1 ribosome inactivating protein from Phytolacca dioica L. Evaluation of their antimicrobial properties and anti-biofilm activities.

Antimicrobial peptides, also called Host Defence Peptides (HDPs), are effectors of innate immune response found in all living organisms. In a previous report, we have identified by chemical fragmentation, and characterized the first cryptic antimicrobial peptide in PD-L4, a type 1 ribosome inactivating protein (RIP) from leaves of Phytolacca dioica L. We applied a recently developed bioinformatic approach to a further member of the differently expressed pool of type 1 RIPs from P. dioica (PD-L1/2), and identified two novel putative cryptic HDPs in its N-terminal domain. These two peptides, here named IKY31 and IKY23, exhibit antibacterial activities against planktonic bacterial cells and, interestingly, significant anti-biofilm properties against two Gram-negative strains. Here, we describe that PD-L1/2 derived peptides are able to induce a strong dose-dependent reduction in biofilm biomass, affect biofilm thickness and, in the case of IKY31, interfere with cell-to-cell adhesion, likely by affecting biofilm structural components. In addition to these findings, we found that both PD-L1/2 derived peptides are able to assume stable helical conformations in the presence of membrane mimicking agents (SDS and TFE) and intriguingly beta structures when incubated with extracellular bacterial wall components (LPS and alginate). Overall, the data collected in this work provide further evidence of the importance of cryptic peptides derived from type 1 RIPs in host/pathogen interactions, especially under pathophysiological conditions induced by biofilm forming bacteria. This suggests a new possible role of RIPs as precursors of antimicrobial and anti-biofilm agents, likely released upon defensive proteolytic processes, which may be involved in plant homeostasis.

Natural and redesigned wasp venom peptides with selective antitumoral activity.

About 1 in 8 U.S. women (≈12%) will develop invasive breast cancer over the course of their lifetime. Surgery, chemotherapy, radiotherapy, and hormone manipulation constitute the major treatment options for breast cancer. Here, we show that both a natural antimicrobial peptide (AMP) derived from wasp venom (decoralin, Dec-NH2), and its synthetic variants generated via peptide design, display potent activity against cancer cells. We tested the derivatives at increasing doses and observed anticancer activity at concentrations as low as 12.5 µmol L-1 for the selective targeting of MCF-7 breast cancer cells. Flow cytometry assays further revealed that treatment with wild-type (WT) peptide Dec-NH2 led to necrosis of MCF-7 cells. Additional atomic force microscopy (AFM) measurements indicated that the roughness of cancer cell membranes increased significantly when treated with lead peptides compared to controls. Biophysical features such as helicity, hydrophobicity, and net positive charge were identified to play an important role in the anticancer activity of the peptides. Indeed, abrupt changes in peptide hydrophobicity and conformational propensity led to peptide inactivation, whereas increasing the net positive charge of peptides enhanced their activity. We present peptide templates with selective activity towards breast cancer cells that leave normal cells unaffected. These templates represent excellent scaffolds for the design of selective anticancer peptide therapeutics.

Anti-plasmodial activity of bradykinin and analogs.

To find effective new candidate antimalarial drugs, bradykinin and its analogs were synthesized and tested for effectiveness against Plasmodium gallinaceum sporozoites and Plasmodium falciparum on erythrocytes. Among them, bradykinin and its P2 analog presented high activity against Plasmodium gallinaceum, but they degrade in plasma. On the other hand, RI-BbKI did not degrade and reached high activity. No analog was active against Plasmodium falciparum.

Deep-learning-enabled antibiotic discovery through molecular de-extinction.

Molecular de-extinction aims at resurrecting molecules to solve antibiotic resistance and other present-day biological and biomedical problems. Here we show that deep learning can be used to mine the proteomes of all available extinct organisms for the discovery of antibiotic peptides. We trained ensembles of deep-learning models consisting of a peptide-sequence encoder coupled with neural networks for the prediction of antimicrobial activity and used it to mine 10,311,899 peptides. The models predicted 37,176 sequences with broad-spectrum antimicrobial activity, 11,035 of which were not found in extant organisms. We synthesized 69 peptides and experimentally confirmed their activity against bacterial pathogens. Most peptides killed bacteria by depolarizing their cytoplasmic membrane, contrary to known antimicrobial peptides, which tend to target the outer membrane. Notably, lead compounds (including mammuthusin-2 from the woolly mammoth, elephasin-2 from the straight-tusked elephant, hydrodamin-1 from the ancient sea cow, mylodonin-2 from the giant sloth and megalocerin-1 from the extinct giant elk) showed anti-infective activity in mice with skin abscess or thigh infections. Molecular de-extinction aided by deep learning may accelerate the discovery of therapeutic molecules.

Low-Cost Biosensor Technologies for Rapid Detection of COVID-19 and Future Pandemics.

Many systems have been designed for the detection of SARS-CoV-2, which is the virus that causes COVID-19. SARS-CoV-2 is readily transmitted, resulting in the rapid spread of disease in human populations. Frequent testing at the point of care (POC) is a key aspect for controlling outbreaks caused by SARS-CoV-2 and other emerging pathogens, as the early identification of infected individuals can then be followed by appropriate measures of isolation or treatment, maximizing the chances of recovery and preventing infectious spread. Diagnostic tools used for high-frequency testing should be inexpensive, provide a rapid diagnostic response without sophisticated equipment, and be amenable to manufacturing on a large scale. The application of these devices should enable large-scale data collection, help control viral transmission, and prevent disease propagation. Here we review functional nanomaterial-based optical and electrochemical biosensors for accessible POC testing for COVID-19. These biosensors incorporate nanomaterials coupled with paper-based analytical devices and other inexpensive substrates, traditional lateral flow technology (antigen and antibody immunoassays), and innovative biosensing methods. We critically discuss the advantages and disadvantages of nanobiosensor-based approaches compared to widely used technologies such as PCR, ELISA, and LAMP. Moreover, we delineate the main technological, (bio)chemical, translational, and regulatory challenges associated with developing functional and reliable biosensors, which have prevented their translation into the clinic. Finally, we highlight how nanobiosensors, given their unique advantages over existing diagnostic tests, may help in future pandemics.

Synthetic angiotensin II peptide derivatives confer protection against cerebral and severe non-cerebral malaria in murine models.

Malaria can have severe long-term effects. Even after treatment with antimalarial drugs eliminates the parasite, survivors of cerebral malaria may suffer from irreversible brain damage, leading to cognitive deficits. Angiotensin II, a natural human peptide hormone that regulates blood pressure, has been shown to be active against Plasmodium spp., the etiologic agent of malaria. Here, we tested two Ang II derivatives that do not elicit vasoconstriction in mice: VIPF, a linear tetrapeptide, which constitutes part of the hydrophobic portion of Ang II; and Ang II-SS, a disulfide-bridged derivative. The antiplasmodial potential of both peptides was evaluated with two mouse models: an experimental cerebral malaria model and a mouse model of non-cerebral malaria. The latter consisted of BALB/c mice infected with Plasmodium berghei ANKA. The peptides had no effect on mean blood pressure and significantly reduced parasitemia in both mouse models. Both peptides reduced the SHIRPA score, an assay used to assess murine health and behavior. However, only the constrained derivative (Ang II-SS), which was also resistant to proteolytic degradation, significantly increased mouse survival. Here, we show that synthetic peptides derived from Ang II are capable of conferring protection against severe manifestations of malaria in mouse models while overcoming the vasoconstrictive side effects of the parent peptide.

Polyproline peptide targets Klebsiella pneumoniae polysaccharides to collapse biofilms.

Hypervirulent Klebsiella pneumoniae is known for its increased extracellular polysaccharide production. Biofilm matrices of hypervirulent K. pneumoniae have increased polysaccharide abundance and are uniquely susceptible to disruption by peptide bactenecin 7 (bac7 (1-35)). Here, using confocal microscopy, we show that polysaccharides within the biofilm matrix collapse following bac7 (1-35) treatment. This collapse led to the release of cells from the biofilm, which were then killed by the peptide. Characterization of truncated peptide analogs revealed that their interactions with polysaccharide were responsible for the biofilm matrix changes that accompany bac7 (1-35) treatment. Ultraviolet photodissociation mass spectrometry with the parental peptide or a truncated analog bac7 (10-35) reveal the important regions for bac7 (1-35) complexing with polysaccharides. Finally, we tested bac7 (1-35) using a murine skin abscess model and observed a significant decrease in the bacterial burden. These findings unveil the potential of bac7 (1-35) polysaccharide interactions to collapse K. pneumoniae biofilms.

Molecular de-extinction of ancient antimicrobial peptides enabled by machine learning.

Molecular de-extinction could offer avenues for drug discovery by reintroducing bioactive molecules that are no longer encoded by extant organisms. To prospect for antimicrobial peptides encrypted within extinct and extant human proteins, we introduce the panCleave random forest model for proteome-wide cleavage site prediction. Our model outperformed multiple protease-specific cleavage site classifiers for three modern human caspases, despite its pan-protease design. Antimicrobial activity was observed in vitro for modern and archaic protein fragments identified with panCleave. Lead peptides showed resistance to proteolysis and exhibited variable membrane permeabilization. Additionally, representative modern and archaic protein fragments showed anti-infective efficacy against A. baumannii in both a skin abscess infection model and a preclinical murine thigh infection model. These results suggest that machine-learning-based encrypted peptide prospection can identify stable, nontoxic peptide antibiotics. Moreover, we establish molecular de-extinction through paleoproteome mining as a framework for antibacterial drug discovery.

Human gut metagenomic mining reveals an untapped source of peptide antibiotics.

Drug-resistant bacteria are outpacing traditional antibiotic discovery efforts. Here, we computationally mined 444,054 families of putative small proteins from 1,773 human gut metagenomes, identifying 323 peptide antibiotics encoded in small open reading frames (smORFs). To test our computational predictions, 78 peptides were synthesized and screened for antimicrobial activity in vitro, with 59% displaying activity against either pathogens or commensals. Since these peptides were unique compared to previously reported antimicrobial peptides, we termed them smORF-encoded peptides (SEPs). SEPs killed bacteria by targeting their membrane, synergized with each other, and modulated gut commensals, indicating that they may play a 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. We report the discovery of hundreds of peptide sequences in the human gut.

Synergistic Combination of Antimicrobial Peptides and Cationic Polyitaconates in Multifunctional PLA Fibers.

Combining different antimicrobial agents has emerged as a promising strategy to enhance efficacy and address resistance evolution. In this study, we investigated the synergistic antimicrobial effect of a cationic biobased polymer and the antimicrobial peptide (AMP) temporin L, with the goal of developing multifunctional electrospun fibers for potential biomedical applications, particularly in wound dressing. A clickable polymer with pendent alkyne groups was synthesized by using a biobased itaconic acid building block. Subsequently, the polymer was functionalized through click chemistry with thiazolium groups derived from vitamin B1 (PTTIQ), as well as a combination of thiazolium and AMP temporin L, resulting in a conjugate polymer-peptide (PTTIQ-AMP). The individual and combined effects of the cationic PTTIQ, Temporin L, and PTTIQ-AMP were evaluated against Gram-positive and Gram-negative bacteria as well as Candida species. The results demonstrated that most combinations exhibited an indifferent effect, whereas the covalently conjugated PTTIQ-AMP displayed an antagonistic effect, potentially attributed to the aggregation process. Both antimicrobial compounds, PTTIQ and temporin L, were incorporated into poly(lactic acid) electrospun fibers using the supercritical solvent impregnation method. This approach yielded fibers with improved antibacterial performance, as a result of the potent activity exerted by the AMP and the nonleaching nature of the cationic polymer, thereby enhancing long-term effectiveness.

Rapid and accurate detection of herpes simplex virus type 2 using a low-cost electrochemical biosensor.

Herpes simplex virus type 2 (HSV-2) infection, which is almost exclusively sexually transmitted, causes genital herpes. Although this lifelong and incurable infection is extremely widespread, currently there is no readily available diagnostic device that accurately detects HSV-2 antigens to a satisfactory degree. Here, we report an ultrasensitive electrochemical device that detects HSV-2 antigens within 9 min and costs just $1 (USD) to manufacture. The electrochemical biosensor is biofunctionalized with the human cellular receptor nectin-1 and detects the glycoprotein gD2, which is present within the HSV-2 viral envelope. The performance of the device is tested in a guinea pig model that mimics human biofluids, yielding 88.9% sensitivity, 100.0% specificity, and 95.0% accuracy under these conditions, with a limit of detection of 0.019 fg mL-1 for gD2 protein and 0.057 PFU mL-1 for titered viral samples. Importantly, no cross-reactions with other viruses were detected, indicating the adequate robustness and selectivity of the sensor. Our low-cost technology could facilitate more frequent testing for HSV-2.

Autonomous Treatment of Bacterial Infections in Vivo Using Antimicrobial Micro- and Nanomotors.

The increasing resistance of bacteria to existing antibiotics constitutes a major public health threat globally. Most current antibiotic treatments are hindered by poor delivery to the infection site, leading to undesired off-target effects and drug resistance development and spread. Here, we describe micro- and nanomotors that effectively and autonomously deliver antibiotic payloads to the target area. The active motion and antimicrobial activity of the silica-based robots are driven by catalysis of the enzyme urease and antimicrobial peptides, respectively. These antimicrobial motors show micromolar bactericidal activity in vitro against different Gram-positive and Gram-negative pathogenic bacterial strains and act by rapidly depolarizing their membrane. Finally, they demonstrated autonomous anti-infective efficacy in vivo in a clinically relevant abscess infection mouse model. In summary, our motors combine navigation, catalytic conversion, and bactericidal capacity to deliver antimicrobial payloads to specific infection sites. This technology represents a much-needed tool to direct therapeutics to their target to help combat drug-resistant infections.

Mining for encrypted peptide antibiotics in the human proteome.

The emergence of drug-resistant bacteria calls for the discovery of new antibiotics. Yet, for decades, traditional discovery strategies have not yielded new classes of antimicrobial. Here, by mining the human proteome via an algorithm that relies on the sequence length, net charge, average hydrophobicity and other physicochemical properties of antimicrobial peptides, we report the identification of 2,603 encrypted peptide antibiotics that are encoded in proteins with biological function unrelated to the immune system. We show that the encrypted peptides kill pathogenic bacteria by targeting their membrane, modulate gut and skin commensals, do not readily select for bacterial resistance, and possess anti-infective activity in skin abscess and thigh infection mouse models. We also show, in vitro and in the two mouse models of infection, that encrypted antibiotic peptides from the same biogeographical area display synergistic antimicrobial activity. Our algorithmic strategy allows for the rapid mining of proteomic data and opens up new routes for the discovery of candidate antibiotics.

Synthetic Antibiotic Derived from Sequences Encrypted in a Protein from Human Plasma.

Encrypted peptides have been recently found in the human proteome and represent a potential class of antibiotics. Here we report three peptides derived from the human apolipoprotein B (residues 887-922) that exhibited potent antimicrobial activity against drug-resistant Klebsiella pneumoniae, Acinetobacter baumannii, and Staphylococci both in vitro and in an animal model. The peptides had excellent cytotoxicity profiles, targeted bacteria by depolarizing and permeabilizing their cytoplasmic membrane, inhibited biofilms, and displayed anti-inflammatory properties. Importantly, the peptides, when used in combination, potentiated the activity of conventional antibiotics against bacteria and did not select for bacterial resistance. To ensure translatability of these molecules, a protease resistant retro-inverso variant of the lead encrypted peptide was synthesized and demonstrated anti-infective activity in a preclinical mouse model. Our results provide a link between human plasma and innate immunity and point to the blood as a source of much-needed antimicrobials.