Distinguishing Natural Infections of the Bovine Mammary Gland by Staphylococcus from Streptococcus spp. Using Quantitative Milk Proteomics.

Bovine mastitis is the most frequent disease on dairy farms, which leads to a decrease in the health welfare of the animals and great economic losses. This study was aimed at determining the quantitative variations in the milk proteome caused by natural infection by Staphylococcus and Streptococcus species in order to gain further understanding of any discrepancies in pathophysiology and host immune responses, independent of the mastitis level. After identification of Staphylococcus (N = 51) and Streptococcus (N = 67) spp., tandem mass tag (TMT)-labeled quantitative proteomic and liquid chromatography-mass spectrometry (LC-MS/MS) techniques on a modular Ultimate 3000 RSLCnano system coupled to a Q Exactive Plus was applied on aseptically sampled milk from Holstein cows. Proteome Discoverer was used for protein identification and quantitation through the SEQUEST algorithm. Statistical analysis employing R was used to identify differentially abundant proteins between the groups. Protein classes, functions and functional-association networks were determined using the PANTHER and STRING tools and pathway over-representation using the REACTOME. In total, 156 master bovine proteins were identified (two unique peptides, p < 0.05 and FDR < 0.001), and 20 proteins showed significantly discrepant abundance between the genera ( p < 0.05 and FDR < 0.5). The most discriminatory proteins per group were odorant-binding protein (higher in staphylococci) and fibrinogen beta chain protein (higher in streptococci). The receiver operating characteristic (ROC) curve showed that protein kinase C-binding protein NELL2, thrombospondin-1, and complement factor I have diagnostic potential for differentiating staphylococci and streptococci intramammary infection and inflammation. Improved understanding of the host response mechanisms and recognition of potential biomarkers of specific-pathogen mastitis, which may aid prompt diagnosis for control implementation, are potential benefits of this study.