Regulatory Mutations Impacting Antibiotic Susceptibility in an Established Staphylococcus aureus Biofilm

Abstract

We previously determined the extent to which mutations of different Staphylococcus aureus regulatory loci impact biofilm formation as assessed under in vitro conditions. Here we extend these studies to determine the extent to which those regulatory loci that had the greatest effect on biofilm formation also impact antibiotic susceptibility. The experiments were done under in vitro and in vivo conditions using two clinical isolates of S. aureus (LAC and UAMS-1) and two functionally diverse antibiotics (daptomycin and ceftaroline). Mutation of the staphylococcal accessory regulator (sarA) or sigB was found to significantly increase susceptibilities to both antibiotics and in both strains in a manner that could not be explained by changes in the MICs. The impact of a mutation in sarA was comparable to that of a mutation in sigB and greater than the impact observed with any other mutant. These results suggest that therapeutic strategies targeting sarA and/or sigB have the greatest potential to facilitate the ability to overcome the intrinsic antibiotic resistance that defines S. aureus biofilm-associated infections.

INTRODUCTION

Biofilm formation is a defining factor in the clinical approach to many forms of Staphylococcus aureus infections owing to the fact that the presence of a biofilm confers a therapeutically relevant degree of intrinsic antibiotic resistance irrespective of the acquired resistance status of the offending bacterial strain (1). Thus, adjunct therapeutic approaches capable of limiting biofilm formation would offer a tremendous clinical advantage. A key component in the development of such approaches is to define the mechanism(s) by which S. aureus forms a biofilm, thus opening the door to the development of therapeutic approaches that limit biofilm formation and thereby limit the clinical impact of this intrinsic resistance.

We have focused our efforts in this regard on S. aureus regulatory elements, many of which have been shown to impact biofilm formation both negatively and positively, at least under in vitro conditions (2). This work has led us to place a primary emphasis on the staphylococcal accessory regulator (sarA), mutation of which limits biofilm formation to a degree that can be correlated with increased antibiotic susceptibility as assessed under both in vitro and in vivo conditions (3, 4). Moreover, our recent comparison to the impact of mutating sarA relative to that of mutating other S. aureus regulatory loci implicated in biofilm formation led us to conclude that mutation of sarA imposes a greater limitation on biofilm formation than mutation of any other S. aureus regulatory locus (2). However, these studies were limited to in vitro conditions and did not take into account relative antibiotic susceptibility.

To address this, we used in vitro (3) and in vivo (4) models of catheter-associated biofilm formation to assess the relative antibiotic susceptibility of those regulatory mutants previously shown to have the greatest impact, either positively or negatively, on biofilm formation (2). We did this by using the functionally distinct antibiotics daptomycin and ceftaroline and the genetically and phenotypically distinct S. aureus strains LAC (USA300, methicillin resistant) and UAMS-1 (USA200, methicillin sensitive) (4–7).

MATERIALS AND METHODS

Assessment of relative antibiotic susceptibility in vitro.

Antibiotic susceptibility under in vitro conditions was assessed using a catheter-based model as previously described (3). Briefly, 1-cm segments of fluorinated ethylene propylene catheters (14-gauge Introcan Safety catheter; B. Braun, Bethlehem, PA) were first coated with human plasma before being placed into the wells of a 12-well microtiter plate containing 2 ml of tryptic soy broth supplemented with glucose and sodium chloride (biofilm medium [BM]). Each well was then inoculated with LAC, UAMS-1, or the appropriate isogenic mutant at an optical density at 600 nm of 0.05. The plate was then incubated at 37°C for 24 h before the catheters were removed and transferred to fresh BM with and without the appropriate antibiotic. After an additional 24-h incubation, catheters were removed, rinsed in phosphate-buffered saline (PBS) to remove nonadherent bacteria, and then placed in a test tube containing 5 ml of sterile PBS. To remove adherent bacteria, each catheter was then sonicated to remove adherent bacteria as previously described (3). Appropriately diluted samples were then plated on tryptic soy agar without antibiotic selection to determine the number of viable bacteria per catheter remaining.

Assessment of relative antibiotic susceptibility in vivo.

Biofilm formation was assessed in vivo using a murine model of catheter-associated biofilm formation (4). Briefly, uncoated catheters were implanted into each flank of NIH Swiss mice and inoculated with 10⁵ CFU of the test strain in a total volume of 100 μl of PBS by direct injection into the lumen of each catheter. After 24 h, the mice were randomly divided into experimental groups (n = 5). Because each mouse had two catheters implanted and because previous experiments have confirmed the absence of cross-contamination between catheters in opposite flanks of the same mouse (4), each catheter was treated as an independent data point (n = 10). In untreated mice, 100 μl of sterile PBS was injected in the lumen of each catheter at daily intervals for 5 days. Catheters were then harvested and processed as described above to determine the number of CFU per catheter remaining after antibiotic treatment.

Antibiotics and S. aureus strains tested.

For both in vitro and in vivo assays, daptomycin was tested at 5 times the Clinical and Laboratory Standards Institute (CLSI)-defined breakpoint MIC, while ceftaroline was tested at 10 times its CLSI-defined breakpoint concentration. The use of different concentrations of each antibiotic was based on preliminary studies indicating that ceftaroline exhibits somewhat reduced efficacy in the context of a biofilm in comparison to daptomycin (data not shown) and a desire to employ an antibiotic concentration that would allow us to detect differences in susceptibility that would not be apparent with antibiotic concentrations that were either too low or too high. The LAC mutants included were sarA, atl, codY, fur, mgrA, rot, rsbU, and sigB. UAMS-1 mutants examined were more limited but included sarA, codY, mgrA, rot, and sigB. As previously described (2), all of these mutants were generated by phage-mediated transduction from primary mutants available as part of the Nebraska Transposon Mutant Library (NTML).

Impact of regulatory mutations on MICs of relevant mutants.

The relative daptomycin and ceftaroline susceptibilities of each strain were assessed by Etest (bioMérieux SA, Marcy l'Etoile, France) using tryptic soy agar as the growth medium.

Statistical analysis.

Statistical comparisons were made between each parent strain and its isogenic mutant with and without antibiotic exposure. For each experimental setting, a set of contrasts that defined the comparisons of interest were created. Permutation tests, as described in Pallmann et al. (8), were performed to obtain the adjusted P values for each contrast. Briefly, using the observed data, t test statistics were calculated for each individual contrast, and the absolute values of the statistics were recorded. The data were then randomly permuted. The statistics from the permuted set were calculated, and the one resulting in the minimum P value across all contrasts was recorded. The data were permuted 50,000 times, presumably resulting in a distribution of test statistics from the null distribution. The number of times the permuted test statistic was larger than the observed test statistic was calculated for each contrast. The adjusted P value for a contrast is the aforementioned number divided by 50,001. A logarithmic transformation was applied to the CFU data prior to analysis. Adjusted P values of <5% were considered statistically significant. This analysis was performed using SAS 9.4 (SAS Institute Inc., Cary, NC).

RESULTS AND DISCUSSION

Using a microtiter plate-based assay, we previously examined the relative impact of mutating individual regulatory loci on S. aureus biofilm formation in vitro (2). These studies identified a number of mutants that exhibited either a decreased or increased capacity to form a biofilm, but they did not address the issue of whether these changes were sufficient to have an impact on antibiotic susceptibility in the context of an established biofilm. In this report, we examined this issue with a focus on those regulatory loci shown to have the greatest impact on biofilm formation in our previous study. This included LAC mutants sarA, atl, codY, fur, mgrA, rot, rsbU, and sigB as well as sarA, codY, mgrA, rot, and sigB mutants generated in the osteomyelitis isolate UAMS-1. To facilitate the ability to focus on the relative antibiotic susceptibility in a quantitative fashion, these studies were done using in vitro and in vivo models of catheter-associated biofilm formation (3, 4).

The only mutation that imposed a significant limitation on biofilm formation under in vitro conditions in the absence of antibiotic exposure was the sarA mutation, and this was true in both LAC and UAMS-1 (Fig. 1). However, exposure to 5X daptomycin was associated with significantly increased susceptibility in both UAMS-1 and LAC sarA and sigB mutants relative to the isogenic parent strain. Mutation of rsbU also resulted in a significant increase in susceptibility in LAC, but we did not have a UAMS-1 rsbU mutant. These reductions were reflected in colony counts per catheter and the percentage of catheters cleared of viable bacteria at least as defined by the limit of detection of our experimental method (50 CFU per catheter) (Fig. 1). Similar results were observed under in vivo conditions, although in this case none of the mutations, including the sarA mutation, were found to have a statistically significant impact in either strain in the absence of antibiotic exposure (Fig. 2). Additionally, under in vivo conditions, mutation of sigB had a significant impact in LAC but not in UAMS-1 (Fig. 2). Interestingly, mutation of codY resulted in a significant increase in daptomycin susceptibility in vivo in LAC but not in UAMS-1. This is consistent with the observation that mutation of codY resulted in a significant increase in biofilm formation in UAMS-1 in vivo but not in LAC (Fig. 2). To the extent that the goal is to identify potential S. aureus targets that can be exploited to therapeutic advantage, this emphasizes the importance of considering diverse clinical isolates in studies focusing on biofilm formation and relative antibiotic susceptibility.

FIG 1.

Relative daptomycin susceptibility in vitro. Daptomycin susceptibility in LAC, UAMS-1, and the indicated mutants was assessed using a catheter-based model of biofilm formation. The results indicate individual data points. The horizontal bar and error bars indicate the means ± standard errors of the mean (SEM) based on CFU per catheter remaining after antibiotic exposure. An asterisk above an experimental group indicates the statistical significance of each mutant relative to the isogenic parent strain in the absence of antibiotic exposure. An asterisk below a group indicates the significance of each mutant relative to the parent strain after antibiotic exposure. The number above a strain indicates the percentage of catheters cleared of bacteria as defined by the level of detection of our assay. NSR, no significant reduction.

FIG 2.

Relative daptomycin susceptibility in vivo. Daptomycin susceptibility was assessed using a murine model of a catheter-based model of biofilm formation. The results indicate individual data points. The horizontal bar and error bars indicate the means ± standard errors of the mean (SEM) based on CFU per catheter remaining after antibiotic exposure. An asterisk below a group indicates the significance after antibiotic exposure. The number above a mutant indicates the percentage of catheters cleared of bacteria below the level of detection. NSR, no significant reduction.

To determine whether the results observed with daptomycin might be generalized to those with other antibiotics and thus likely to be a function of the impact of individual mutations on biofilm formation itself rather than a daptomycin-specific effect, we repeated the in vivo studies using ceftaroline. We chose ceftaroline rather than the more commonly used anti-methicillin-resistant Staphylococcus aureus (MRSA) antibiotic vancomycin because ceftaroline is a functionally distinct antibiotic in comparison to daptomycin and because our previous results have confirmed that vancomycin has relatively little efficacy in the context of an established biofilm (9). The results were essentially identical to those observed with daptomycin except that in this case mutation of sarA and sigB resulted in a significant increase in antibiotic susceptibility in both UAMS-1 and LAC (Fig. 3). As with daptomycin, this increased susceptibility was evident both in the average colony counts per catheter and the percentage of catheters cleared of viable bacteria as defined by the limit of detection of our experimental method. Importantly, mutation of sarA or sigB did not significantly alter the MIC of LAC or UAMS-1 to daptomycin or ceftaroline (Fig. 4), thus providing support for the hypothesis that the increased susceptibility we observed is a function of the impact of each mutation on the relative capacity to form a biofilm.

FIG 3.

Relative ceftaroline susceptibility in vivo. Ceftaroline susceptibility was assessed using a murine model of a catheter-based model of biofilm formation. The results indicate individual data points. The horizontal bar and error bars indicate the means ± standard errors of the mean (SEM) based on CFU per catheter remaining after antibiotic exposure. An asterisk above an experimental group indicates the statistical significance relative to the isogenic parent strain in the absence of antibiotic exposure. An asterisk below a group indicates the significance after antibiotic exposure. The number above a mutant indicates the percentage of catheters cleared of bacteria below the level of detection. NSR, no significant reduction.

FIG 4.

Impact of mutating sarA and sigB on MICs. The MICs of LAC and UAMS-1 sarA and sigB mutants was determined by Etest. DPC, daptomycin; CPT, ceftaroline.

In summary, the primary clinical problem with S. aureus biofilm-associated infections is their intrinsic resistance to conventional antibiotic therapy, thus making the experimentally critical parameter the degree to which mutation of genes that impact biofilm formation also impact this intrinsic resistance. The results we report are significant in that they provide further support for the hypothesis that the staphylococcal accessory regulator (sarA) plays a critical role in this regard and that it does so in diverse strains of S. aureus irrespective of their methicillin-resistance status. At the same time, the results demonstrate that elements within the sigB regulon also play a critical role. There is a report suggesting that sigB increases expression of sarA (9), and we confirmed that mutation of sigB in LAC results in a significant decrease in the accumulation of SarA (2), thus suggesting that the impact of sigB may be mediated at least in part through its impact on sarA. However, a recent report demonstrated that sigB is required for the establishment of chronic S. aureus infections and suggested that this was due to the fact that a functional SigB regulon promotes both the development of small colony variants (SCVs) and increased intracellular persistence (10). In contrast, mutations of sarA had relatively little impact on SCV formation. An independent report also concluded that sarA, but not sigB, is essential for biofilm development in S. aureus (11). Thus, the specific mechanistic basis that defines the comparable impact of sarA and sigB on antibiotic susceptibility in vivo remains unclear. Having said this, we previously reported one commonality: that protease production is increased in both sarA and sigB mutants to a degree that limits biofilm formation (2). However, the more important point in the context of this report is that these results confirm that both of these regulatory loci are potentially important targets for therapeutic intervention.