Variable Release of Lipoteichoic Acid From Staphylococcus aureus Bloodstream Isolates Relates to Distinct Clinical Phenotypes, Strain Background, and Antibiotic Exposure

BACKGROUND: Staphylococcus aureus is a leading cause of bacterial bloodstream infections. The heterogeneity in patient outcomes in S. aureus bacteremia (SAB) can be attributed in part to strain characteristics, which may influence host response to infection. We specifically examined the relationship between lipoteichoic acid (LTA) release from S. aureus and disease phenotype, strain background, and antibiotic exposure. METHODS: Seven strains of S. aureus causing different clinical phenotypes of bacteremia and two reference strains (LAC USA 300 and Mu3) were analyzed for LTA release at baseline and following exposure to antibiotics from different pharmacologic classes (vancomycin, ceftaroline, and tedizolid). LTA release was quantified by LTA-specific ELISA. Whole genome sequencing was performed on the clinical strains and analyzed using open-source bioinformatics tools. RESULTS: Lipoteichoic acid release varied by 4-fold amongst the clinical strains and appeared to be related to duration of bacteremia, independent of MLST type. Low LTA releasing strains were isolated from patients who had prolonged duration of bacteremia and died. Antibiotic-mediated differences in LTA release appeared to be associated with MLST type, as ST8 strains released maximal LTA in response to tedizolid while other non-ST8 strains demonstrated high LTA release with vancomycin. Genetic variations related to the LTA biosynthesis pathway were detected in all non-ST8 strains, though ST8 strains showed no variations despite demonstrating differential LTA release. CONCLUSION: Our findings provide the basis for future studies to evaluate the relationship between LTA release-mediated host immune response and clinical outcomes as well as the potential for antibiotic modulation of LTA release as a therapeutic strategy and deserve confirmation with larger number of strains with known clinical phenotypes.