Reduced vancomycin susceptibility in an in vitro catheter-related biofilm model correlates with poor therapeutic outcomes in experimental endocarditis due to methicillin-resistant Staphylococcus aureus

Abstract

Staphylococcus aureus is the most common cause of endovascular infections, including catheter sepsis and infective endocarditis (IE). Vancomycin (VAN) is the primary choice for treatment of methicillin-resistant S. aureus (MRSA) infections. However, high rates of VAN treatment failure in MRSA infections caused by VAN-susceptible strains have been increasingly reported. Biofilm-associated MRSA infections are especially prone to clinical antibiotic failure. The present studies examined potential relationships between MRSA susceptibility to VAN in biofilms in vitro and nonsusceptibility to VAN in endovascular infection in vivo. Using 10 "VAN-susceptible" MRSA bloodstream isolates previously investigated for VAN responsiveness in experimental IE, we studied the mechanism(s) of such in vivo VAN resistance, including: (i) VAN binding to MRSA organisms; (ii) the impact of VAN on biofilm formation and biofilm composition; (iii) VAN efficacy in an in vitro catheter-related biofilm model; (iv) effects on cell wall thickness. As a group, the five strains previously categorized as VAN nonresponders (non-Rsp) in the experimental IE model differed from the five responders (Rsp) in terms of lower VAN binding, increased biofilm formation, higher survival in the presence of VAN within biofilms in the presence or absence of catheters, and greater biofilm reduction upon proteinase K treatment. Interestingly, sub-MICs of VAN significantly promoted biofilm formation only in the non-Rsp isolates. Cell wall thickness was similar among all MRSA strains. These results suggest that sublethal VAN levels that induce biofilm formation and reduce efficacy of VAN in the in vitro catheter-associated biofilms may contribute to suboptimal treatment outcomes for endovascular infections caused by "VAN-susceptible" MRSA strains.

Fig 1

(A) Population analyses of two control S. aureus strains, ATCC 25923 (♢) and MU50 (■) upon exposure to a range of VAN concentrations (0 to 16 μg/ml). (B and C) VAN population analyses of 10 MRSA study strains, including nonresponders (B) and responders (C). Nonresponder strains: ⧫, 300-087; ■, 300-169; ▲, 324-136; ×, 300-103; ○, 300-246. Responder strains: ⧫, 301-188; ■, 010-016; ▲, 077-107; ×, 088-180; ○, 088-237. These data represent the means (± SD) for two separate assays.

Fig 2

In vitro time-kill curve of VAN (15 μg/ml) versus 10 MRSA study strains at a high initial inoculum (∼10⁸ CFU/ml) of exponential-phase cells. Data are expressed as the Δlog₁₀ CFU/ml (± SD) at 2, 4, 6, and 24 h of incubation versus the initial inoculum. Non-Rsp control, white; non-Rsp strains with VAN exposure, black; Rsp control, light gray; Rsp with VAN exposure, dark gray. *, P < 0.05 between non-Rsp versus Rsp comparisons with VAN exposure.

Fig 3

In vitro VAN binding levels of the 10 MRSA study strains, using Bodipy-labeled VAN. The mean level of VAN binding for each group is indicated by a horizontal dashed line.

Fig 4

Biofilm formation of the 10 MRSA strains under static condition with (■) and without (□) 0.5× the MIC of VAN, based on absorbance values (A₄₉₀ ± SD). *, P < 0.05 versus the respective control without VAN exposure.

Fig 5

Biofilm formation of two selected non-Rsp and Rsp strain pairs under flow conditions. Confocal microscopy was used to obtain images of MRSA isolates grown in a flow cell biofilm model.

Fig 6

Biofilm stability of preformed biofilms by the 10 MRSA strains following sodium metaperiodate, proteinase K, or DNase I treatment (as quantified by the percentage of A₄₉₀ reduction versus controls). (A) Without VAN exposure; (B) 0.5× the MIC of VAN during overnight biofilm formation. White bars, non-Rsp group; black bars, Rsp group. *, P < 0.05 versus non-Rsp group.

Fig 7

Effect of VAN (15 μg/ml) in the in vitro catheter-associated biofilm formation model. Catheters infected with 10 MRSA strains without VAN exposure (white bars) and with VAN exposure (black bars) were harvested at daily intervals. The MRSA counts (log₁₀ CFU/catheter) were compared to the initial density for each strain. D1, D2, and D3 represent 1, 2, and 3 days with or without VAN exposure on an established biofilm, respectively. *, P < 0.05 versus the respective control without VAN exposure.