Integrated Genome Mining, Bacterial Co-Culture Activation, and Peptidomic Analyses Identify Antimicrobial Peptide Candidates from South American Bacteria

Background/Objectives: Antimicrobial resistance (AMR) is a major global health threat that requires the discovery of new antimicrobial agents. Environmental microbiomes from understudied regions represent a valuable source of antimicrobial peptide (AMP) candidates. This study aimed to identify and prioritize AMP candidates from South American genomic and metagenomic datasets and to investigate the antimicrobial potential of bioactive secretomes obtained through bacterial co-culture. Methods: A total of 853 genomes and 360 metagenomes were analyzed using a reproducible genome- and metagenome-mining pipeline combined with machine learning-based AMP prediction. Predicted AMP candidates were further characterized using complementary bioinformatic tools to assess physicochemical, structural, hemolytic, toxicological, anti-inflammatory, and anticancer properties. Selected environmental isolates were subjected to bacterial co-culture, followed by SPE-C18 and HPLC fractionation. Antimicrobial activity, antioxidant activity, hemolysis, minimum inhibitory concentration (MIC), and LC-MS/MS peptidomic analyses were performed on bioactive secretome fractions. Results: Genome and metagenome mining identified diverse AMP candidate sequences associated with bacterial genera including Streptomyces, Bacillus, Burkholderia, and Shewanella. Structural predictions revealed a predominance of α-helical conformations among prioritized candidates. Several secretome fractions obtained from co-cultures displayed antimicrobial activity against Gram-positive and Gram-negative bacteria, including methicillin-resistant Staphylococcus aureus (MRSA). Active fractions showed no detectable hemolytic activity and exhibited antioxidant activity in DPPH assays. MIC analyses indicated broad-spectrum activity against Escherichia coli ATCC 11229, Pseudomonas aeruginosa ATCC 27853, Klebsiella pneumoniae, carbapenem-resistant Acinetobacter baumannii, and MRSA, with an apparent MIC of 10,000 mg/L. LC-MS/MS analysis of bioactive fractions identified peptide sequences by de novo sequencing, including KTESHHK, KRVGPRR, GLFPRLGVSPR, and HHAEHLVHFR. Conclusions: Integrated genome mining, bacterial co-culture activation, and peptidomic analyses provide a useful framework for prioritizing antimicrobial peptide candidates from environmental microbiomes. The identification of peptide-containing bioactive fractions with antimicrobial and antioxidant activities highlights the potential of South American bacterial resources for the discovery of novel antimicrobial compounds. Further purification, peptide synthesis, and biological validation will be required to determine the contribution of individual peptides to the observed activities.