Antimicrobial resistance (AMR) is predicted to be the leading cause of death around the world by 2050. Among the main causes contributing to AMR are the over-prescription of antibiotics in clinics and the over-use of antibiotics in farming. Therefore, it is imperative to explore new strategies to slow down and even reverse the evolution of AMR and preserve our ability to fight infectious diseases. A critical area for intervention includes the rapid diagnosis of microbes in biological samples. The gold-standard method [microbial culturing followed by matrix assisted laser desorption ionization (MALDI) - TOF analysis] takes approximately 24-48 hours, lacks specificity for closely related species and is not quantitative.
For several years, our team has developed new approaches for microbial diagnosis using high-sensitivity/ high-specificity LC-MS/MS combined to Artificial Intelligence in order to provide a microbial identification in less than four hours. We applied this strategy to two problems related to microbial infections: urinary tract infection (UTI) in humans, accounting for 14% of all antibiotic prescriptions in Canada, and bovine mastitis (BM), which causes significant losses to the dairy industry. We demonstrated that combining LC-MS/MS data to machine learning algorithms allows the extraction of a peptide signature able to distinguish between 15 bacterial species; and we improved this method using crude signal from ultrafast LC-MS/MS runs (< 5 min) without peptide identification.
We also demonstrated the transfer of the method to cost-effective triple quadrupole mass spectrometer instruments capable of operating in routine applications. Overall, our work paves the way toward the new generation of diagnosis methods for infectious diseases that could, in the end, have an important contribution to reducing the use of antibiotics.
Florence Roux-Dalvai, Research Assistant and Project Manager, Centre de Recherche du CHU de Québec – Université Laval
Florence Roux-Dalvai is an expert in proteomics and mass spectrometry and the co-founder of Canadian Proteomics & AI consortium. She has more than ten years of experience with Thermo Scientific Orbitrap instruments and used them for deep proteome analyses of complex biological samples.
* Required field
