Abstract
Bovine mastitis, a multi-etiological disease, is driven by complex microbial consortia; however, the transcriptional activity of pathogens and their underlying molecular mechanisms remains insufficiently explored. To the best of our knowledge, no metatranscriptome study on bovine mastitis is available in the public domain that identifies transcriptionally active pathogens and their associated molecular signatures. In this study, an in silico metatranscriptomics approach is employed on publicly available bovine mastitis RNA sequencing (RNA-Seq) datasets to identify transcriptionally active pathogens and their gene expression signatures. The analysis of unmapped reads (those not mapped to the bovine genome) identified 25 transcriptionally active pathogenic genera, accounting for 8,995 sequences, approximately from 500 bacterial strains of different species. Major findings of the study includes: (I) list of emerging pathogens "Pseudomonas, Stenotrophomonas, Comamonas, and Sphingomonas" actively contributing to disease development alongside well-known pathogens; (II) expression profiling of 4,121 virulence proteins, 484 peptidases, 432 secretory proteins, and 74 antimicrobial resistance genes; (III) identification of numerous hypothetical proteins in Staphylococcus (112), Mycoplasma (69), and Escherichia (32), representing potential source for diagnostics and multi-epitope vaccine candidates; and (IV) negative correlations between beneficial bacteria (Blautia, Bacillus, Lactobacillus) and pathogenic species in microbial co-occurrence interaction networks, suggesting opportunities for microbiome-based therapeutic strategies to treat subclinical mastitis. This study demonstrated the advantages of the metatranscriptomics approach and publicly available dual RNA-Seq datasets in unraveling the complexity of polymicrobial infectious diseases.