The escalating global health crisis of multi-drug resistant bacteria necessitates the urgent development of novel antibiotics. Macrocyclic peptides, encompassing both nonribosomal peptides (NRPs) and ribosomally synthesized and post-translationally modified peptides (RiPPs), represent a promising alternative source for antibiotic discovery. Their unique macrocyclic structures confer resistance to proteolysis and maintain active conformations for effective target binding. Several examples of successful RiPP and NRP antibiotics highlight this potential. Traditional antibiotic discovery methods are time-consuming and inefficient. Genome mining, leveraging the vast amount of microbial genome data available, offers a more efficient approach. RiPPs, with their structural diversity and bioactive potential, are a particularly attractive target for genome mining. While previous studies have focused on genomics-guided RiPP discovery, effectively prioritizing BGCs and linking them to metabolites remains a challenge. Aminovinyl-(methyl-)cysteine-containing peptides (ACyPs), a class of compounds with unique cyclic peptide rings and antibacterial properties, are of particular interest. The Avi(Me)Cys unit provides structural rigidity, protecting the peptides from degradation and conferring drug-like properties. Existing ACyPs, such as microbisporicin and lexapeptide, demonstrate potent activity against drug-resistant bacteria. However, the chemical space of ACyPs remains largely unexplored. This study aims to systematically investigate ACyPs through large-scale bioinformatic analysis of publicly available genomes, employing a rule-based genome mining pipeline to discover novel antibiotics.

The paper extensively reviews the existing literature on antimicrobial resistance, macrocyclic peptides as antibiotic sources (specifically RiPPs and NRPs), and genome mining techniques for discovering novel antibiotics. It highlights the limitations of traditional screening methods and emphasizes the potential of genome mining, particularly focusing on RiPPs, for accelerating antibiotic discovery. The review also focuses on ACyPs, detailing their unique structural features, biosynthesis, and known bioactivities. Several known ACyPs, like microbisporicin and lexapeptide, are mentioned as examples of their potential. The conserved biosynthetic machinery of ACyPs, despite their structural diversity, is noted as a key feature facilitating genome mining strategies. The review provides a background on various ACyPs families and their modifications, emphasizing the conserved biosynthetic pathway towards the Avi(Me)Cys unit.

The study employed a three-pronged approach: rule-based genome mining, biosynthetic rule-guided metabolic analysis, and heterologous expression. First, the researchers used the SPECO pipeline to analyze 21,911 actinobacteria and 30,666 firmicute genomes for ACyP BGCs, identifying 1172 unique putative precursor-flavoprotein pairs. Sequence similarity network (SSN) and sequence logo analyses were used to categorize these BGCs into 67 ACyP families and prioritize candidates with unknown precursors or tailoring enzymes. Two families, *mat* and *sis*, were selected for further investigation. Second, a mass mapping pipeline was implemented to link identified BGCs with their corresponding metabolites through a rule-based metabolomic analysis. This involved predicting the theoretical masses of peptide fragments with various modifications and comparing them to experimental HRMS data. For example, in the *mat* cluster analysis, theoretical m/z values of precursor peptide fragments with predicted modifications were compared with experimental data from the *Streptomyces leeuwenhoekii* DSM42122 culture extract. Third, heterologous expression of the identified BGCs in a *Streptomyces* host was performed to confirm the BGC-metabolite association and assess the antibiotic potential of the produced compounds. The complete *mat* BGC was cloned and expressed in *Streptomyces albus*, confirming the production of massatide A. Similar heterologous expression was performed for the *sis* and *keb* BGCs to obtain sistertides and kebanetides. Antibacterial activity was assessed using Kirby-Bauer assays and MIC determinations against various Gram-positive and Gram-negative bacterial strains. Furthermore, in vitro stability assays (trypsin, chymotrypsin, heat) and in vivo acute toxicity studies in mice were conducted for massatide A. In vivo efficacy of massatide A was evaluated in a mouse septicemia model using MRSA ATCC43300.

The study successfully identified and characterized several novel ACyPs. The *mat* BGC from *S. leeuwenhoekii* was shown to produce massatide A, a class V lanthipeptide with a unique structure featuring D-amino acids and an AviMeCys ring. Massatide A exhibited potent antibacterial activity against various Gram-positive bacteria, including drug-resistant clinical isolates (MIC of 0.25 µg/mL against linezolid-resistant *S. aureus* and methicillin-resistant *S. aureus*). In vitro experiments showed that massatide A is bactericidal and does not cause bacterial lysis. Importantly, massatide A demonstrated low resistance development and a favorable safety profile in mice. The *sis* BGC from *S. kasugaensis* produced sistertide A1 and A2, which also showed antibacterial activity. The *keb* BGC from *S. kebangsaanensis* produced kebanetide A1 and A2, again demonstrating antibiotic properties. The study further demonstrated that the *sis* family maturase displays broad substrate tolerance, producing diverse crosslinking patterns and allowing the generation of new-to-nature analogs through leader-core mixing and matching experiments. The study showcases the power of a combination of bioinformatics and experimental techniques in identifying novel antibiotics.

The findings address the research question by successfully identifying novel ACyPs with potent antibacterial activity. The rule-based omics approach proved effective in prioritizing BGCs and linking them to bioactive metabolites, improving the efficiency of antibiotic discovery. The discovery of massatide A, with its potent activity against drug-resistant Gram-positive pathogens and favorable safety profile, represents a significant advancement in the search for new antibiotics. The study’s approach offers a model for targeted genome mining and circumvents the issues of rediscovery. The discussion also touches upon the limitations of the rule-based approach in identifying compounds with unanticipated modifications and the need for better enzyme function annotation. The potential mode of action of massatide A, possibly targeting lipid II or membrane proteins, is discussed. The observed low resistance development of *S. aureus* to massatide A warrants further investigation. The promiscuity of the *sis* family maturases opens avenues for generating new antibiotic analogs through bioengineering.

This study demonstrates a successful strategy for discovering novel antimicrobial peptides using rule-based omics mining. The identification and characterization of massatide A, a potent antibiotic with a low resistance profile and favorable safety profile, highlights the potential of ACyPs as valuable antibiotic candidates. The work emphasizes the efficiency of combining bioinformatics, metabolomics, and heterologous expression for targeted antibiotic discovery. Future research should focus on elucidating the precise mode of action of massatide A, further optimizing its drug-like properties, and exploring the potential of the *sis* family maturases for generating new antibiotic analogs.

The study primarily focused on Gram-positive bacteria, with limited activity observed against Gram-negative strains. The rule-based metabolomic analysis might have overlooked novel compounds with unpredictable modifications. While the in vivo toxicity studies in mice showed a favorable safety profile for massatide A, further research is needed to completely assess its toxicity in human cells and potential long-term effects. The relatively small sample size in some of the in vivo experiments should also be considered.