Person:

Rajagopal, Mithila

Staphylococcus aureus is a leading cause of life-threatening infections worldwide. The MIC of an antibiotic against S. aureus, as well as other microbes, is determined by the affinity of the antibiotic for its target in addition to a complex interplay of many other cellular factors. Identifying nontarget factors impacting resistance to multiple antibiotics could inform the design of new compounds and lead to more-effective antimicrobial strategies. We examined large collections of transposon insertion mutants in S. aureus using transposon sequencing (Tn-Seq) to detect transposon mutants with reduced fitness in the presence of six clinically important antibiotics-ciprofloxacin, daptomycin, gentamicin, linezolid, oxacillin, and vancomycin. This approach allowed us to assess the relative fitness of many mutants simultaneously within these libraries. We identified pathways/genes previously known to be involved in resistance to individual antibiotics, including graRS and vraFG (graRS/vraFG), mprF, and fmtA, validating the approach, and found several to be important across multiple classes of antibiotics. We also identified two new, previously uncharacterized genes, SAOUHSC_01025 and SAOUHSC_01050, encoding polytopic membrane proteins, as important in limiting the effectiveness of multiple antibiotics. Machine learning identified similarities in the fitness profiles of graXRS/vraFG, SAOUHSC_01025, and SAOUHSC_01050 mutants upon antibiotic treatment, connecting these genes of unknown function to modulation of crucial cell envelope properties. Therapeutic strategies that combine a known antibiotic with a compound that targets these or other intrinsic resistance factors may be of value for enhancing the activity of existing antibiotics for treating otherwise-resistant S. aureus strains.

Staphylococcus aureus is a highly feared Gram-positive pathogen. The rise in antibiotic resistance has made S. aureus infections intractable. To find new ways to treat S. aureus infections, it is important to understand how this organism protects itself from antibiotics. We probed S. aureus transposon libraries with different classes of antibiotics and used Tn-Seq to identify intrinsic resistance factors that are important in withstanding antibiotics. We identified and validated the importance of a number of previously known intrinsic resistance factors such as mprF, fmtA and the graRS/vraFG multi-component sensing system, as well as a number of novel factors whose involvement in antibiotic resistance has not been previously appreciated. In the course of this work, we realized that Tn-Seq data could be mined to predict antibiotic mechanism of action. We used a machine learning approach to predict that a class of anti-MRSA antibiotics that were thought to cause membrane damage actually bind to lipid II and inhibit cell wall synthesis. This predicted mechanism was validated. Finally, we report the identification of a class of disubstituted urea compounds against which the inactivation of mprF is protective. As the inactivation of mprF usually sensitizes to most antibiotics, these compounds might belong to a new class of inhibitors that could be used as tool compounds to further probe S. aureus cell biology. We have developed a strategy that could be useful in identifying the targets of such compounds. The results described here have opened several interesting avenues for a more in-depth understanding of S. aureus biology and antibiotic resistance.