In vitro synergism and anti-biofilm activity of ampicillin, gentamicin, ceftaroline and ceftriaxone against Enterococcus faecalis

In vitro synergism and anti-biofilm activity of ampicillin, gentamicin, ceftaroline and... Abstract Background Enterococci frequently cause severe biofilm-associated infections such as endocarditis. The combination of ampicillin/ceftriaxone has recently been clinically evaluated as non-inferior compared with the standard therapy of ampicillin/gentamicin for treatment of Enterococcus faecalis endocarditis. Ceftaroline is a novel cephalosporin with enhanced activity against Gram-positive bacteria. Objectives To compare the in vitro effectiveness of the ceftaroline/ampicillin combination with those of gentamicin/ampicillin and ceftriaxone/ampicillin in planktonic and biofilm cultures of clinical E. faecalis isolates. Methods Synergistic effects at the planktonic level were analysed by chequerboard assays in 20 E. faecalis isolates. Biofilm-eradicating and biofilm-preventing activities of the antibiotics and their combinations were determined by confocal laser scanning microscopy with quantification by quantitative biofilm analysis (qBA) algorithm and cfu/mL determination. Results Comparable synergistic effects were observed for both β-lactam combinations in most isolates, in contrast to gentamicin/ampicillin. However, none of the antibiotic combinations succeeded in eradicating mature biofilms. Gentamicin showed promising biofilm-preventing activity, but at concentrations above those clinically tolerable. The β-lactams showed a U-shape dose–response relationship in biofilm prevention. Only exposure to cephalosporins caused alterations in cell morphology, which resulted in cell elongation and reclustering in a concentration-dependent manner. Reclustering was associated with high occurrences of small colony variants (SCVs), especially at high ceftriaxone concentrations. Conclusions This study suggests that combinations of cephalosporins or gentamicin with ampicillin may be advantageous only while bacteraemia persists, whereas combinations have no advantage over monotherapy regarding the treatment of mature biofilms. The selection of SCVs at high ceftriaxone concentrations is worth further study. Introduction Enterococci frequently cause biofilm-associated infections such as catheter-related bloodstream infections, urinary tract infections and infective endocarditis (IE).1 Biofilms are matrix-embedded communities that aggregate on artificial or natural surfaces and exhibit increased resistance to attacks from the host’s immune system and antibiotic therapy, often resulting in treatment failure, relapsing infections and increased lethality.2,3 Because the majority of clinical Enterococcus faecalis isolates remain susceptible to β-lactams, the most recommended antibiotic treatment of IE caused by E. faecalis involves ampicillin combined with gentamicin for 4–6 weeks.4 However, this therapy is limited due to the severe side effects of aminoglycosides (nephrotoxicity and ototoxicity) and the emergence of high-level aminoglycoside resistance (HLAR) among E. faecalis isolates.5 A combination of two β-lactam antibiotics, ampicillin and ceftriaxone, was recently proven in a retrospective cohort study as non-inferior and well-tolerated compared with the standard therapy and has therefore been recommended for treatment of infections caused by high-level aminoglycoside-resistant E. faecalis.6 Ceftriaxone is a third-generation cephalosporin and, as most cephalosporins, is ineffective individually in enterococci, but synergistically supports ampicillin by differential and stepwise saturation of PBPs.7 Ceftaroline is a novel, broad-spectrum, fifth-generation cephalosporin that, in contrast to ceftriaxone, shows enhanced in vitro activity against Gram-positive bacteria, including enterococci.8 It is approved for the treatment of acute Staphylococcus aureus-related skin infections and community-acquired pneumonia. In vitro data have demonstrated that ceftaroline has anti-biofilm activity against nascent and mature staphylococcal biofilms.9,10 We thus hypothesized that in line with current antibiotic combination therapies for enterococcal IE, ceftaroline plus ampicillin might be an effective treatment option for E. faecalis biofilm-associated infections. Recently, synergistic effects between ceftaroline and ampicillin against E. faecalis isolates were demonstrated in time–kill experiments11 as well as in pharmacokinetic/pharmacodynamic (PK/PD) models, including simulated endocardial vegetations.12,13 However, the effectiveness of this combination in enterococcal biofilms is still unknown. The aim of this study was to elucidate the in vitro effectiveness of the ceftaroline/ampicillin combination in comparison with the current standard gentamicin/ampicillin treatment and the recommended alternative ceftriaxone/ampicillin treatment in biofilms of clinical E. faecalis isolates. We therefore analysed the biofilm-eradicating and biofilm-preventing activity of these antibiotics and their combinations, the latter in terms of their biofilm prevention concentrations (BPCs), and conducted chequerboard assays to analyse synergistic effects in planktonic cultures and biofilms. Materials and methods Enterococcal strains, liquid cultures and antimicrobials Clinical E. faecalis (n = 20) isolates were acquired from blood cultures or from swabs of mitral valves by the Institute of Medical Microbiology at Jena University Hospital, Germany (Table 1). E. faecalis ATCC 29212 served as a reference strain for the chequerboard assays. Bacterial liquid cultures were prepared in Todd Hewitt (TH) broth (Karl Roth, Karlsruhe, Germany) and incubated at 37°C at constant rotation speed (approximately 13 g) for 2–3 h. Test solutions of ampicillin (Karl Roth, Karlsruhe, Germany), ceftriaxone (TCI Europe, Zwijndrecht, Belgium), gentamicin (TCI Europe, Zwijndrecht, Belgium) and ceftaroline (Forest Laboratories, New York City, USA) were prepared immediately before usage. Table 1. Results of susceptibility testing (MIC), synergy testing (FICI) and clinical data of the patient cohort Isolatea ST Clinical background Sex Age (years) MIC of ampicillin (mg/L) MIC of ceftaroline (mg/L) MIC of ceftriaxone (mg/L) MIC of gentamicin (mg/L) FICI of ampicillin/ ceftarolineb FICI of ampicillin/ ceftriaxone FICI of ampicillin/ gentamicin va67230 579 endocarditis male 39 1 1 16 16 0.50 0.25 1.02 va245 280 endocarditis male 76 0.5 0.125 2 16 0.73 0.50 1.02 bk5597 40 opportunistic infection without clinical signs female 55 1 0.5 8c 16 0.50 0.31c 1.00 bk848 19 opportunistic infection after liver transplant rejection male 60 1 0.25 8 16 0.49 0.31 1.02 bk905 41 endocarditis male 75 1 1 256 8 0.38 0.27 1.03 bk2164 74 opportunistic infection due to acute renal failure female 78 2 0.5 32c 4 0.37 0.25c 1.06 ATCC 29212 30 / / / 1 0.5 4 16 0.50 0.50 1.00 bk8653 6 opportunistic infection after liver transplantation male 56 1 8 >512 >512 0.25 ND ND bk3043 6 opportunistic infection due to respiratory insufficiency, liver cirrhosis female 74 0.5 1 >512 >512 0.37 ND ND bk3062 6 urosepsis due to permanent urinary catheter male 77 1 4 >512 >512 0.31 ND ND bk7183 6 urosepsis (with prostate carcinoma) male 76 1 8 512 >512 0.37 0.38 ND bk6037 6 urosepsis due to permanent urinary catheter male 86 1 2 512c >512 0.37 0.10c ND bk9190 6 wound infection female 59 1 2 >512 >512 0.50 ND ND bk281 498 sepsis due to biliary tract infection female 79 1 1 32c 8 0.50 0.25c 1.03 bk9367 64 recurrent bacteraemia female 85 2 0.5 16c 32 0.37 0.38c 0.63 bk6747 64 biliary tract infection (with Klatskin tumour) male 81 1 0.5 256c 32 0.62 0.10c 1.01 bk1653 179 urosepsis due to urinary tract infection female 87 1 2 16c 64 0.31 0.25c 0.50 bk4497 16 urosepsis due to permanent urinary catheter female 67 1 0.5 64c >512 0.50 0.10c ND bk5187 16 urosepsis post-operative male 79 1 0.25 16 256 0.49 0.19 0.63 bk6886 16 wound infection female 68 2 1 16c 32 0.50 0.19c 1.00 bk8669 23 opportunistic infection due to metastasizing rectal carcinoma male 80 1 4 32c 32 0.50 0.25c 0.56 Isolatea ST Clinical background Sex Age (years) MIC of ampicillin (mg/L) MIC of ceftaroline (mg/L) MIC of ceftriaxone (mg/L) MIC of gentamicin (mg/L) FICI of ampicillin/ ceftarolineb FICI of ampicillin/ ceftriaxone FICI of ampicillin/ gentamicin va67230 579 endocarditis male 39 1 1 16 16 0.50 0.25 1.02 va245 280 endocarditis male 76 0.5 0.125 2 16 0.73 0.50 1.02 bk5597 40 opportunistic infection without clinical signs female 55 1 0.5 8c 16 0.50 0.31c 1.00 bk848 19 opportunistic infection after liver transplant rejection male 60 1 0.25 8 16 0.49 0.31 1.02 bk905 41 endocarditis male 75 1 1 256 8 0.38 0.27 1.03 bk2164 74 opportunistic infection due to acute renal failure female 78 2 0.5 32c 4 0.37 0.25c 1.06 ATCC 29212 30 / / / 1 0.5 4 16 0.50 0.50 1.00 bk8653 6 opportunistic infection after liver transplantation male 56 1 8 >512 >512 0.25 ND ND bk3043 6 opportunistic infection due to respiratory insufficiency, liver cirrhosis female 74 0.5 1 >512 >512 0.37 ND ND bk3062 6 urosepsis due to permanent urinary catheter male 77 1 4 >512 >512 0.31 ND ND bk7183 6 urosepsis (with prostate carcinoma) male 76 1 8 512 >512 0.37 0.38 ND bk6037 6 urosepsis due to permanent urinary catheter male 86 1 2 512c >512 0.37 0.10c ND bk9190 6 wound infection female 59 1 2 >512 >512 0.50 ND ND bk281 498 sepsis due to biliary tract infection female 79 1 1 32c 8 0.50 0.25c 1.03 bk9367 64 recurrent bacteraemia female 85 2 0.5 16c 32 0.37 0.38c 0.63 bk6747 64 biliary tract infection (with Klatskin tumour) male 81 1 0.5 256c 32 0.62 0.10c 1.01 bk1653 179 urosepsis due to urinary tract infection female 87 1 2 16c 64 0.31 0.25c 0.50 bk4497 16 urosepsis due to permanent urinary catheter female 67 1 0.5 64c >512 0.50 0.10c ND bk5187 16 urosepsis post-operative male 79 1 0.25 16 256 0.49 0.19 0.63 bk6886 16 wound infection female 68 2 1 16c 32 0.50 0.19c 1.00 bk8669 23 opportunistic infection due to metastasizing rectal carcinoma male 80 1 4 32c 32 0.50 0.25c 0.56 a Isolates with various backgrounds of infection were obtained from the Institute of Medical Microbiology in Jena, Germany. bk = blood culture. va = mitral valve. / = not applicable to the laboratory standard strain. b FICI values are given as the lowest observed FICI. Synergistic FICI values (lowest FICI ≤0.5) are indicated in bold. ND = not determined due to MICs exceeding 512 mg/L. c Labelled isolates started to regrow as visible aggregates at concentrations above the determined turbid/non-turbid interface of ceftriaxone (alone and in combination with ampicillin). FICIs were calculated using the concentrations in the first non-turbid well found in each row and column, neglecting the regrowth. Table 1. Results of susceptibility testing (MIC), synergy testing (FICI) and clinical data of the patient cohort Isolatea ST Clinical background Sex Age (years) MIC of ampicillin (mg/L) MIC of ceftaroline (mg/L) MIC of ceftriaxone (mg/L) MIC of gentamicin (mg/L) FICI of ampicillin/ ceftarolineb FICI of ampicillin/ ceftriaxone FICI of ampicillin/ gentamicin va67230 579 endocarditis male 39 1 1 16 16 0.50 0.25 1.02 va245 280 endocarditis male 76 0.5 0.125 2 16 0.73 0.50 1.02 bk5597 40 opportunistic infection without clinical signs female 55 1 0.5 8c 16 0.50 0.31c 1.00 bk848 19 opportunistic infection after liver transplant rejection male 60 1 0.25 8 16 0.49 0.31 1.02 bk905 41 endocarditis male 75 1 1 256 8 0.38 0.27 1.03 bk2164 74 opportunistic infection due to acute renal failure female 78 2 0.5 32c 4 0.37 0.25c 1.06 ATCC 29212 30 / / / 1 0.5 4 16 0.50 0.50 1.00 bk8653 6 opportunistic infection after liver transplantation male 56 1 8 >512 >512 0.25 ND ND bk3043 6 opportunistic infection due to respiratory insufficiency, liver cirrhosis female 74 0.5 1 >512 >512 0.37 ND ND bk3062 6 urosepsis due to permanent urinary catheter male 77 1 4 >512 >512 0.31 ND ND bk7183 6 urosepsis (with prostate carcinoma) male 76 1 8 512 >512 0.37 0.38 ND bk6037 6 urosepsis due to permanent urinary catheter male 86 1 2 512c >512 0.37 0.10c ND bk9190 6 wound infection female 59 1 2 >512 >512 0.50 ND ND bk281 498 sepsis due to biliary tract infection female 79 1 1 32c 8 0.50 0.25c 1.03 bk9367 64 recurrent bacteraemia female 85 2 0.5 16c 32 0.37 0.38c 0.63 bk6747 64 biliary tract infection (with Klatskin tumour) male 81 1 0.5 256c 32 0.62 0.10c 1.01 bk1653 179 urosepsis due to urinary tract infection female 87 1 2 16c 64 0.31 0.25c 0.50 bk4497 16 urosepsis due to permanent urinary catheter female 67 1 0.5 64c >512 0.50 0.10c ND bk5187 16 urosepsis post-operative male 79 1 0.25 16 256 0.49 0.19 0.63 bk6886 16 wound infection female 68 2 1 16c 32 0.50 0.19c 1.00 bk8669 23 opportunistic infection due to metastasizing rectal carcinoma male 80 1 4 32c 32 0.50 0.25c 0.56 Isolatea ST Clinical background Sex Age (years) MIC of ampicillin (mg/L) MIC of ceftaroline (mg/L) MIC of ceftriaxone (mg/L) MIC of gentamicin (mg/L) FICI of ampicillin/ ceftarolineb FICI of ampicillin/ ceftriaxone FICI of ampicillin/ gentamicin va67230 579 endocarditis male 39 1 1 16 16 0.50 0.25 1.02 va245 280 endocarditis male 76 0.5 0.125 2 16 0.73 0.50 1.02 bk5597 40 opportunistic infection without clinical signs female 55 1 0.5 8c 16 0.50 0.31c 1.00 bk848 19 opportunistic infection after liver transplant rejection male 60 1 0.25 8 16 0.49 0.31 1.02 bk905 41 endocarditis male 75 1 1 256 8 0.38 0.27 1.03 bk2164 74 opportunistic infection due to acute renal failure female 78 2 0.5 32c 4 0.37 0.25c 1.06 ATCC 29212 30 / / / 1 0.5 4 16 0.50 0.50 1.00 bk8653 6 opportunistic infection after liver transplantation male 56 1 8 >512 >512 0.25 ND ND bk3043 6 opportunistic infection due to respiratory insufficiency, liver cirrhosis female 74 0.5 1 >512 >512 0.37 ND ND bk3062 6 urosepsis due to permanent urinary catheter male 77 1 4 >512 >512 0.31 ND ND bk7183 6 urosepsis (with prostate carcinoma) male 76 1 8 512 >512 0.37 0.38 ND bk6037 6 urosepsis due to permanent urinary catheter male 86 1 2 512c >512 0.37 0.10c ND bk9190 6 wound infection female 59 1 2 >512 >512 0.50 ND ND bk281 498 sepsis due to biliary tract infection female 79 1 1 32c 8 0.50 0.25c 1.03 bk9367 64 recurrent bacteraemia female 85 2 0.5 16c 32 0.37 0.38c 0.63 bk6747 64 biliary tract infection (with Klatskin tumour) male 81 1 0.5 256c 32 0.62 0.10c 1.01 bk1653 179 urosepsis due to urinary tract infection female 87 1 2 16c 64 0.31 0.25c 0.50 bk4497 16 urosepsis due to permanent urinary catheter female 67 1 0.5 64c >512 0.50 0.10c ND bk5187 16 urosepsis post-operative male 79 1 0.25 16 256 0.49 0.19 0.63 bk6886 16 wound infection female 68 2 1 16c 32 0.50 0.19c 1.00 bk8669 23 opportunistic infection due to metastasizing rectal carcinoma male 80 1 4 32c 32 0.50 0.25c 0.56 a Isolates with various backgrounds of infection were obtained from the Institute of Medical Microbiology in Jena, Germany. bk = blood culture. va = mitral valve. / = not applicable to the laboratory standard strain. b FICI values are given as the lowest observed FICI. Synergistic FICI values (lowest FICI ≤0.5) are indicated in bold. ND = not determined due to MICs exceeding 512 mg/L. c Labelled isolates started to regrow as visible aggregates at concentrations above the determined turbid/non-turbid interface of ceftriaxone (alone and in combination with ampicillin). FICIs were calculated using the concentrations in the first non-turbid well found in each row and column, neglecting the regrowth. MLST MLST was performed according to Ruiz-Garbajosa et al.14 For DNA isolation, 2–3 colonies of each isolate were resuspended in RNase-free H2O and denatured at 95°C for 10 min. The DNA was separated from cell debris by centrifugation at 12 000 g for 5 min. PCR was performed in a 25 μL reaction mixture containing 2 mM MgCl2, 1×KCl buffer, 1.5 U/μL Taq standard polymerase (all Thermo Fisher Scientific, Massachusetts, USA), 0.2 mM dNTPs (Karl Roth, Karlsruhe, Germany), 0.4 μM of each forward and reverse primer and 150 ng of chromosomal DNA. PCR products were purified using a NucleoSpin® Gel and PCR Clean-up Kit (Macherey-Nagel, Düren, Germany) and quantified in the Infinite® 200 plate reader (Tecan, Männedorf, Switzerland) using a NanoQuant plate, both following the manufacturers’ protocols. Samples were sequenced with forward and reverse primers at the EZ-Seq service (Macrogen, Amsterdam, The Netherlands). Allelic profiles and STs were assigned in accordance with the E. faecalis MLST database (http://efaecalis.mlst.net/) and analysed by the Draw Tree module applying the neighbour joining algorithm. Antimicrobial synergism testing by chequerboard assays The chequerboard assays were performed with 11 double-dilution steps of either ceftaroline, ceftriaxone or gentamicin and 7 double-dilution steps of ampicillin, as described previously.15 Antibiotic concentration ranges were selected based on MIC values determined by the broth microdilution method according to EUCAST guidelines (ISO 20776-1:2006), except that TH broth was used instead of Mueller–Hinton broth. Each chequerboard assay was performed in duplicate. The effects of the combined antibiotics were evaluated by calculating the fractional inhibitory concentration indices (FICI) along the turbidity/non-turbidity interface as described previously.15 The lowest FICI value was used for interpretation and we assumed synergism occurred at FICI ≤0.5, antagonism at FICI ≥4 and no interaction at 0.5 < FICI < 4. Anti-biofilm testing Biofilms were grown in optical microtitre plates (0.54 cm2/well) with glass bottoms (Greiner Bio-one, Kremsmünster, Austria) in triplicate by applying 100 μL of a bacterial suspension (0.5 McFarland) to each well. Plates were placed in a humidified chamber and incubated at 37°C, 5% CO2 without shaking. To assess biofilm-eradicating effects, we grew the biofilms for 24 h, carefully removed the supernatant and added 100 μL (2 × 50 μL each for the combinations) of the antibiotic solution prepared in TH medium. Medium was changed in the untreated growth controls. Plates were incubated for a further 24 h before analysis. For assessment of the BPC, the antibiotics in sub-MICs were directly added to the bacterial suspension (0.5 McFarland/well). The plates were grown for 48 h without the medium being changed. For quantification of antibiotic effects, the viable cell number was determined by counting the number of viable bacteria (corresponding to cfu/cm2). The supernatant of the biofilms was therefore carefully removed; the biofilms were washed twice with 100 μL of sterile NaCl, scraped off and resuspended in TH medium. Selected 10-fold dilutions were plated on TH-agar plates and incubated overnight at 37°C, followed by counting of cfu/mL. The BPC was defined as the first concentration at which the viable cell count decreased significantly by 1 log magnitude compared with the untreated control, referring to Macia et al.16 Biofilm imaging and computed analysis Biofilms were stained using the LIVE/DEAD BacLight Bacterial Viability Kit for microscopy (Life Technologies GmbH, Darmstadt, Germany) according to the manufacturer’s protocol. Stained biofilms were analysed under vital conditions using an inverse confocal laser scanning microscope (CLSM), LSM510 (Carl Zeiss AG, Jena, Germany), as described previously.17 The biofilm images were visualized by ZEN 9.0 software (Carl Zeiss AG, Jena, Germany). The biofilm experiments [eradication and prevention (BPC)] were independently performed in duplicate for each strain and in triplicate for each assay. Quantitative analysis of biofilm images was performed by an algorithm termed qBA (quantitative biofilm analysis) that determined the number of bacterial counts/cm2.17 Statistical analysis The correlation of the MIC was analysed by non-parametric Spearman’s rank correlation coefficient (rs) with a two-tailed CI of 95%. The non-parametric Kruskal–Wallis test followed by Dunn’s multiple comparison post-test with a CI of 95% was used for statistical analysis of the cfu/mL or cfu/cm2 quantification. Differences were considered significant at P values <0.05. Results Characterization of isolates Most of the clinical isolates (14/20) were obtained from patients with biofilm-associated infections, including three isolates from patients with endocarditis (Table 1). Among the 20 isolates, 9 isolates exhibited unique STs and variable resistance profiles. Six isolates, all highly resistant to gentamicin and ceftriaxone (≥512 mg/L), belonged to ST6, whereas ST64 (n = 2) and ST16 (n = 3) also showed resistance to those antibiotics, but had different MIC values. Synergy testing on planktonic bacteria To analyse synergistic effects between ceftaroline/ampicillin, ceftriaxone/ampicillin and gentamicin/ampicillin, we performed chequerboard assays for the 20 clinical E. faecalis isolates and the laboratory standard strain ATCC 29212 (Table 1). All isolates showed ampicillin MIC values of 1 mg/L ± 1× MIC. The gentamicin MICs ranked between 4 and >512 mg/L, with eight isolates exhibiting an HLAR profile (MIC of gentamicin >128 mg/L). One isolate exhibiting low-level resistance against gentamicin (MIC of gentamicin <128 mg/L) experienced synergistic effects from gentamicin plus ampicillin (FICI = 0.5), while all other isolates (non-HLAR and HLAR) showed FICI values from >0.5 to 1, indicating no interaction between these two antibiotics [Table 1 and Figure S2 (available as Supplementary data at JAC Online)]. The MIC values of ceftriaxone ranked between 2 and >512 mg/L, while the MIC values of ceftaroline were generally lower, with values between 0.125 mg/L and 8 mg/L (Table 1). Both the MIC of ceftriaxone and the MIC of ceftaroline showed no correlation with the MIC of ampicillin, but the MIC of ceftaroline positively correlated with the MIC of ceftriaxone (rs = 0.705, P < 0.001). Synergistic effects were observed for ceftaroline and ampicillin in 19 of 21 tested isolates, with FICI values of ampicillin/ceftaroline between 0.25 and 0.50 (Table 1). Ceftriaxone and ampicillin synergistically inhibited the growth of 17 isolates, with FICI values of ampicillin/ceftriaxone between 0.1 and 0.5; 15 of these isolates also showed synergistic effects for ceftaroline and ampicillin. At higher concentrations of ceftriaxone (alone and in combination with ampicillin), visible turbid clusters were observed beyond the turbid/non-turbid interface (greater than the MIC of ceftriaxone) for individual isolates (marked in Table 1), suggesting regrowth. This phenomenon was not observed for ceftaroline. Low concentrations of ceftaroline or ceftriaxone were sufficient to strongly reduce the effective concentrations of ampicillin (Figure S2 and Table 1). This effect was most apparent in isolates with high MICs of ceftriaxone, as shown by a weak correlation between the FICI of ampicillin/ceftriaxone and the MIC of ceftriaxone (rs = −0.54, P = 0.025). No correlations between the other MICs and the FICI values or between FICI values and clonality based on MLST (Figure S1) were observed. Biofilm eradication by ceftaroline, ceftriaxone, ampicillin and gentamicin Biofilm-eradicating effects were assessed by CLSM imaging of five biofilm-associated isolates from patients with endocarditis (va245, va67230, bk905) or urosepsis (bk1653, bk6037); these isolates all exhibited synergistic FICI values for the ceftaroline/ampicillin combination (Table 1). Except for isolate bk6037, the isolates produced strong biofilms after 24 h of growth (data not shown). None of the antibiotics in concentrations up to 1000× MIC showed visible biofilm-eradicating effects in any of the isolates, neither alone nor in combination, compared with the untreated control after 24 h of exposure. The biofilms’ thicknesses did not decrease, and the number of dead bacteria did not increase after antibiotic exposure (Figure S3). Biofilm prevention by ceftaroline, ceftriaxone, ampicillin and gentamicin The BPC was tested at sub-MICs of ceftaroline, ceftriaxone, ampicillin and gentamicin for selected isolates (va245, va67230, bk1653 and bk6037). Subinhibitory concentrations of both cephalosporins (ceftaroline and ceftriaxone) induced morphological alterations at the single-cell level in a concentration-dependent manner (Figure 1). As shown in isolate bk1653, the enterococci elongated with increasing ceftaroline and ceftriaxone concentrations in all Z-layers, adapting a long rod shape at ≤1/8× MIC that resulted in filamentous structures. This effect, albeit not as strongly, was also observed for isolates va245 and va67230 (data not shown). The elongated cells were viable, but their density decreased with increasing cephalosporin concentrations as confirmed by cfu/mL count and qBA (Figure 2). As per definition (see the Materials and methods section), the BPCs of both cephalosporins were reached at the elongated stage at 1/8× MIC (0.125 mg/L ceftaroline and 2 mg/L ceftriaxone). At cephalosporin concentrations higher than the BPC, the elongated enterococci reverted to the coccal shape and appeared as clusters in the CLSM images (Figure 1). Ceftaroline-induced cell clusters disappeared at 4× MIC of ceftaroline (4 mg/L), while ceftriaxone-induced clusters were still present at the highest concentration tested (16× MIC of ceftriaxone, 256 mg/L) (data not shown). Figure 1. View largeDownload slide Effects of antibiotics on biofilm formation. E. faecalis isolate bk1653 was grown for 48 h under static conditions in a 96-well glass-bottom plate in the presence of sub-MICs of ampicillin, ceftaroline, ceftriaxone and gentamicin. CLSM images show viable bacteria in green (SYTO 9) and dead bacteria in red (propidium iodide). The concentration corresponding to the fold MIC is noted in each image. Scales in the images of 1/16× MIC apply for all corresponding images to the right. Figure 1. View largeDownload slide Effects of antibiotics on biofilm formation. E. faecalis isolate bk1653 was grown for 48 h under static conditions in a 96-well glass-bottom plate in the presence of sub-MICs of ampicillin, ceftaroline, ceftriaxone and gentamicin. CLSM images show viable bacteria in green (SYTO 9) and dead bacteria in red (propidium iodide). The concentration corresponding to the fold MIC is noted in each image. Scales in the images of 1/16× MIC apply for all corresponding images to the right. Figure 2. View largeDownload slide Biofilm prevention activity of the antibiotics. E. faecalis isolate bk1653 was grown for 48 h under static conditions in a 96-well glass-bottom plate in the presence of sub-MICs of ampicillin, ceftaroline, ceftriaxone and gentamicin. (a) to (d) Quantification of the cfu/cm2. Statistical analysis was performed by one-way ANOVA (Kruskal–Wallis test) followed by Dunn’s multiple comparison test. (e) to (h) Quantification of the bacteria counts/cm2 in the microscope images by qBA. The quantifications were determined independently twice and then in triplicate for each antibiotic. The mean values and the ranges are shown in the diagrams. Figure 2. View largeDownload slide Biofilm prevention activity of the antibiotics. E. faecalis isolate bk1653 was grown for 48 h under static conditions in a 96-well glass-bottom plate in the presence of sub-MICs of ampicillin, ceftaroline, ceftriaxone and gentamicin. (a) to (d) Quantification of the cfu/cm2. Statistical analysis was performed by one-way ANOVA (Kruskal–Wallis test) followed by Dunn’s multiple comparison test. (e) to (h) Quantification of the bacteria counts/cm2 in the microscope images by qBA. The quantifications were determined independently twice and then in triplicate for each antibiotic. The mean values and the ranges are shown in the diagrams. In contrast to the response to the cephalosporins, cell elongation was not observed with ampicillin exposure (Figure 1). The number of viable bacteria remained stable at increasing concentrations until a sudden drop at the BPC at 1/2× MIC of ampicillin (0.5 mg/L ampicillin) (Figure 2). However, ampicillin concentrations above the BPC led to increased bacterial counts, which correlated with cluster formation observed in the CLSM images as under cephalosporin treatment (Figure 1). The clusters disappeared at 2× MIC of ampicillin (2 mg/L) (data not shown). In contrast to the β-lactams, gentamicin killed bacterial cells in the nascent biofilm in a concentration-dependent manner without the clustering effect. Gentamicin did not prevent biofilm formation below the MIC (64 mg/L), but at the MIC of gentamicin (= BPC) no cfu/mL were detected. Synergism in biofilm prevention by ampicillin combined with ceftaroline or gentamicin Only a minor synergism between ampicillin and ceftaroline in biofilm prevention was detected at ≤1/8× MIC of ampicillin because the viable cell numbers per cm2 from the untreated control value were not reduced by 1 log magnitude (Figure 3a). Combining sub-MICs of ceftaroline with increasing ampicillin concentrations ≤1/8× MIC of ampicillin led to the same cell shape change from elongation to coccal clusters as was observed with only sub-MIC ceftaroline exposure (Figure S4). However, the clustering effect was achieved at lower ceftaroline concentrations in combination than with treatment with ceftaroline alone. The BPC for the ceftaroline/ampicillin combination was first reached at 1/4× MIC of ampicillin and 1/32× MIC of ceftaroline, with a reduction in viable cells/cm2 of at least 1.5 log magnitudes (Figure 3a). No elongated cells or clusters were observed under these conditions (Figure S4). The minimum number of viable cells/cm2 was reached at 1/2× MIC of ampicillin plus 1/32× MIC of ceftaroline. Synergy inversion was surprisingly observed with increasing ceftaroline concentrations, i.e. cell numbers were higher than under 1/2× MIC of ampicillin alone (Figure 3a). Figure 3. View largeDownload slide Quantification of the effects on biofilm formation of combining ampicillin with ceftaroline (a) or gentamicin (b). The viable cell counts were determined by the qBA algorithm based on CLSM images (approximately 100 μm × 100 μm) and scaled up to an area of 1 cm2 (104 cells/cm2 represents the limit of detection of this method). The experiments were performed in triplicate and the means and standard deviations are shown. Figure 3. View largeDownload slide Quantification of the effects on biofilm formation of combining ampicillin with ceftaroline (a) or gentamicin (b). The viable cell counts were determined by the qBA algorithm based on CLSM images (approximately 100 μm × 100 μm) and scaled up to an area of 1 cm2 (104 cells/cm2 represents the limit of detection of this method). The experiments were performed in triplicate and the means and standard deviations are shown. The ampicillin/gentamicin combination indifferently affected biofilm prevention (Figures 3b and S5). The combination of gentamicin at concentrations below the MIC of gentamicin and ampicillin at 1/2× MIC of ampicillin increased the bacterial count by 1 log magnitude, indicating an antagonistic effect (Figure 3b). Formation of small colony variants (SCVs) under antimicrobial treatment The formation of clusters above the BPC of ceftaroline, ceftriaxone and ampicillin led us to quantify the proportion of SCVs18 in the viable bacteria (isolates bk1653 and va67230) (cfu/mL) after biofilm growth in the presence of 0.5×, 1× and 2× MIC of each antibiotic. After 48 h, biofilms were resolved and plated on agar to allow for cfu/mL and SCV determination. SCVs were distinguished from normal bacteria as pinpoint colonies18 at higher dilutions of the bacterial suspension. Isolate bk1653 showed a high proportion of SCVs (between 40% and 70%; Figure 4) in the presence of 0.5× and 1× MIC of each β-lactam. At 2× MIC, only the cephalosporins selected SCVs, while under ampicillin exposure, the SCV proportion was strongly reduced to 1.5%. In contrast to isolate bk1653, isolate va67230 produced SCVs only at 0.5× MIC of ampicillin and at all tested ceftaroline concentrations, but not under ceftriaxone exposure (data not shown). Gentamicin led to low or undetectable levels of SCV formation at all concentrations tested in both isolates. In the untreated control and at concentrations lower than 0.5× MIC, the SCV phenotype was not detected for either isolate. Figure 4. View largeDownload slide Proportion of SCVs in viable bacteria (cfu/mL) in dependence on the antibiotics present during biofilm formation. Biofilms of E. faecalis isolate bk1653 were grown in the presence of 0.5×, 1× and 2× MIC of ampicillin (AMP), ceftaroline (CPT), ceftriaxone (CRO) or gentamicin (GEN). After 48 h, biofilms were resolved and plated on agar to allow for cfu/mL and SCV determination. Experiments were performed in triplicate and the means and ranges are shown. The untreated control biofilms never exhibited SCVs. Figure 4. View largeDownload slide Proportion of SCVs in viable bacteria (cfu/mL) in dependence on the antibiotics present during biofilm formation. Biofilms of E. faecalis isolate bk1653 were grown in the presence of 0.5×, 1× and 2× MIC of ampicillin (AMP), ceftaroline (CPT), ceftriaxone (CRO) or gentamicin (GEN). After 48 h, biofilms were resolved and plated on agar to allow for cfu/mL and SCV determination. Experiments were performed in triplicate and the means and ranges are shown. The untreated control biofilms never exhibited SCVs. Discussion In the present study, 20 clinical E. faecalis isolates and one laboratory standard strain were analysed in vitro for synergism between ampicillin and ceftaroline, ceftriaxone or gentamicin in the planktonic and biofilm-embedded status, respectively. Nine isolates were not related to each other and most likely not of nosocomial origin. In contrast, we cannot clearly exclude a nosocomial source for the three clonal clusters, with ST6 being a prevalent clade often found in wild animals19 and livestock,20 but also spreading endemically in hospitals.19,21 Compared with the current standard therapy of gentamicin/ampicillin for E. faecalis IE, the ceftaroline/ampicillin combination showed a superior synergistic effect in vitro against planktonic cultures of clinical E. faecalis isolates, including all HLAR isolates, but FICI values were similar compared with those of the recently recommended ceftriaxone/ampicillin combination.4 These results are in accordance with those of two recently published smaller studies that reported synergism of ceftaroline/ampicillin combinations in different PK/PD models and concluded that only in a limited number of strains ceftaroline might have better activity than ceftriaxone in combination with ampicillin.12,13 However, E. faecalis infections associated with biofilm formation are less susceptible to antibiotics due to biofilm-specific antimicrobial tolerance mechanisms, such as reduced antibiotic penetration and metabolic dormancy.22 The routinely assessed MICs of antibiotics that are sufficient to kill planktonic bacteria are thus usually insufficient to eradicate their biofilm-embedded counterparts.23 Biofilm susceptibility testing is not currently part of diagnostic routines because neither standards for biofilm analysis nor specific breakpoints have been established by official agencies such as EUCAST. We therefore used a microscopy approach to evaluate the efficacy of ampicillin combined with ceftaroline, ceftriaxone or gentamicin in biofilm eradication and the inhibition of biofilm formation and compared the results with the resource-consuming method of cfu/mL determination. The reduction in viable bacterial counts/cm2 in the CLSM images obtained by qBA corresponded to the cfu/cm2 values, but the qBA values were generally lower because the analysed areas of the images were at the centres of the wells, which bear fewer cells than the edges of the wells; in contrast, the edges were included in the cfu/mL determination. Neither the individual antibiotics nor their combinations were sufficient to eradicate established E. faecalis biofilms in vitro. It is still unclear whether β-lactams and aminoglycosides efficiently penetrate the biofilm matrix of E. faecalis because we are not aware of any studies of E. faecalis biofilm penetration. Both antibiotic classes exhibit limited anti-biofilm activity in many species.24 Because β-lactams are inhibitors of cell wall synthesis, the effectiveness of the β-lactam combinations in biofilm eradication is likely hampered by the strongly slowed cell division in metabolically dormant biofilms.24 Gentamicin alone showed promising biofilm-preventing activity but was only effective outside clinically tolerable concentrations. No reduction in BPC of gentamicin was achieved when it was used in combination with ampicillin. All tested β-lactams reduced biofilm formation at sub-MIC concentrations. The combination of ampicillin with ceftaroline, compared with each β-lactam alone, improved the inhibition of biofilm formation in a concentration-dependent manner, suggesting a synergistic interaction between these β-lactams. PBPs 2 and 3 have been identified as primary targets of cephalosporins, while aminopenicillins target PBPs 4 and 5.7 Ampicillin and ceftaroline, similar to the ampicillin/ceftriaxone combination, are thus likely to synergize by inhibiting complementary PBPs, thereby disrupting the cooperation of the PBPs necessary for cell wall biosynthesis.25 Ceftaroline has a higher affinity than ceftriaxone to PBP 5 while maintaining a high affinity for PBPs 2 and 3,8 which explains the lower MIC values of ceftaroline compared with ceftriaxone. The most effective dosing of ceftaroline/ampicillin was achieved at 1/32× MIC of ceftaroline and 1/4× to 1/2× MIC of ampicillin in the BPC experiments, but synergy inversion was observed with increasing ceftaroline concentrations. The effectiveness of ampicillin was thus increased by very low ceftaroline concentrations but was hampered by higher sub-MIC concentrations of ceftaroline. Lower concentrations of ceftaroline likely saturate PBPs 2 and 3, complementing PBP 4 and 5 saturation by ampicillin, while at higher concentrations, ampicillin and ceftaroline may compete for binding of PBP 5, leading to an antagonistic effect. However, this behaviour was not observed at the planktonic level in the chequerboard assays, which displayed synergistic concave isoboles (Figure S2). This result might be caused by the different inocula that were used, i.e. 5 × 105 cfu/mL for MIC and synergism assays versus 1 × 108 cfu/mL in BPC assays. The cephalosporins exhibited a U-shape dose–response relationship in biofilm prevention, which resulted in cell elongation and reclustering in a concentration-dependent manner. Both ceftaroline and ceftriaxone at sub-MICs led to filamentation, suggesting a similar mode of action for both cephalosporins. This hypothesis is further supported by the significant correlation of the MIC of ceftriaxone with the MIC of ceftaroline and the lack of correlation of both with the MIC of ampicillin. The observed filamentation may be explained by an inhibition of cell separation. Specific PBPs of Streptococcus pneumoniae have been involved in septal ring closure in previous studies.26 PBPs 2 and 3 of E. faecalis, which become saturated at very low concentrations of cephalosporins,7 might also be involved in this process. Elongation of bacterial cells exposed to cephalosporins has been described for different Gram-negative bacterial species,27,28 but to our knowledge this report is the first regarding this behaviour in enterococci. Both cephalosporins strongly induced the formation of SCVs at concentrations close to their respective MICs, but with different intensities and not in all isolates. The observed clusters of viable cells escaping the cephalosporins most likely represented the SCVs, and the clustering might be the result of the reduced SCV growth rate.18 Further, the disappearance of the cell clusters at 2× MIC of ampicillin correlated with the decrease in the SCV rate of some isolates (e.g. bk1653) at 2× MIC of ampicillin. Clusters were still observed at 16× MIC of ceftriaxone, indicating that ceftriaxone is unable to kill SCVs. This study suggests that combinations of cephalosporins or gentamicin with ampicillin may be only advantageous for the treatment of persistent bacteraemia (i.e. planktonic cells) but seem to be non-superior compared with monotherapy against mature biofilms. In this regard, the recommendation to treat patients that have enterococcal endocarditis over the entire 4–6 weeks with an antibacterial combination may be questioned, particularly in the face of increased risks of prolonged cephalosporin (e.g. Clostridium difficile colitis) and gentamicin treatment (e.g. nephrotoxicity).4 Furthermore, high cephalosporin concentrations seem to favour selection of SCVs, suggesting that higher doses of combined ceftriaxone (e.g. 2 × 2 g daily) might even be detrimental, at least for some strains. Clinical studies are therefore mandatory to elucidate whether the selection of SCVs composes a risk of using combined therapies against enterococcal biofilm-associated infections. Acknowledgements Parts of this work were presented at the Twenty-seventh European Congress of Clinical Microbiology and Infectious Diseases, Vienna, Austria, 2017 (Poster 2753).  We acknowledge Forest Laboratories, Inc. for providing us with ceftaroline powder. Funding This work was supported by the Federal Ministry of Education and Research, Germany (grant numbers 01KI1501 and 13GW0096D) and the Argus Foundation. Transparency declarations None to declare.  Forest Laboratories, Inc. supplied the ceftaroline powder, but had no involvement in: the design of the study; the collection, analysis and interpretation of the data; or the decision to present these results. Supplementary data Figures S1 to S5 are available as Supplementary data at JAC Online. References 1 Arias CA , Murray BE. The rise of the Enterococcus: beyond vancomycin resistance . Nat Rev Microbiol 2012 ; 10 : 266 – 78 . Google Scholar CrossRef Search ADS PubMed 2 Bjarnsholt T , Alhede M , Alhede M et al. The in vivo biofilm . Trends Microbiol 2013 ; 21 : 466 – 74 . Google Scholar CrossRef Search ADS PubMed 3 Hall-Stoodley L , Stoodley P , Kathju S et al. Towards diagnostic guidelines for biofilm-associated infections . FEMS Immunol Med Microbiol 2012 ; 65 : 127 – 45 . Google Scholar CrossRef Search ADS PubMed 4 Habib G , Lancellotti P , Antunes MJ et al. 2015 ESC guidelines for the management of infective endocarditis: the Task Force for the Management of Infective Endocarditis of the European Society of Cardiology (ESC). Endorsed by: European Association for Cardio-Thoracic Surgery (EACTS), the European Association of Nuclear Medicine (EANM) . Eur Heart J 2015 ; 36 : 3075 – 128 . Google Scholar CrossRef Search ADS PubMed 5 Nigo M , Munita JM , Arias CA et al. What’s new in the treatment of enterococcal endocarditis? Curr Infect Dis Rep 2014 ; 16 : 431. Google Scholar CrossRef Search ADS PubMed 6 Fernandez-Hidalgo N , Almirante B , Gavalda J et al. Ampicillin plus ceftriaxone is as effective as ampicillin plus gentamicin for treating enterococcus faecalis infective endocarditis . Clin Infect Dis 2013 ; 56 : 1261 – 8 . Google Scholar CrossRef Search ADS PubMed 7 Mainardi JL , Gutmann L , Acar JF et al. Synergistic effect of amoxicillin and cefotaxime against Enterococcus faecalis . Antimicrob Agents Chemother 1995 ; 39 : 1984 – 7 . Google Scholar CrossRef Search ADS PubMed 8 Henry X , Verlaine O , Amoroso A et al. Activity of ceftaroline against Enterococcus faecium PBP5 . Antimicrob Agents Chemother 2013 ; 57 : 6358 – 60 . Google Scholar CrossRef Search ADS PubMed 9 Barber KE , Smith JR , Ireland CE et al. Evaluation of ceftaroline alone and in combination against biofilm-producing methicillin-resistant Staphylococcus aureus with reduced susceptibility to daptomycin and vancomycin in an in vitro pharmacokinetic/pharmacodynamic model . Antimicrob Agents Chemother 2015 ; 59 : 4497 – 503 . Google Scholar CrossRef Search ADS PubMed 10 Lazaro-Diez M , Remuzgo-Martinez S , Rodriguez-Mirones C et al. Effects of subinhibitory concentrations of ceftaroline on methicillin-resistant Staphylococcus aureus (MRSA) biofilms . PLoS One 2016 ; 11 : e0147569 . Google Scholar CrossRef Search ADS PubMed 11 Werth BJ , Abbott AN. The combination of ampicillin plus ceftaroline is synergistic against Enterococcus faecalis . J Antimicrob Chemother 2015 ; 70 : 2414 – 7 . Google Scholar CrossRef Search ADS PubMed 12 Luther MK , Rice LB , LaPlante KL. Ampicillin in combination with ceftaroline, cefepime, or ceftriaxone demonstrates equivalent activities in a high-inoculum Enterococcus faecalis infection model . Antimicrob Agents Chemother 2016 ; 60 : 3178 – 82 . Google Scholar CrossRef Search ADS PubMed 13 Werth BJ , Shireman LM. Pharmacodynamics of ceftaroline plus ampicillin against Enterococcus faecalis in an in vitro pharmacokinetic/pharmacodynamic model of simulated endocardial vegetations . Antimicrob Agents Chemother 2017 ; 61 : e02235-16. Google Scholar CrossRef Search ADS PubMed 14 Ruiz-Garbajosa P , Bonten MJ , Robinson DA et al. Multilocus sequence typing scheme for Enterococcus faecalis reveals hospital-adapted genetic complexes in a background of high rates of recombination . J Clin Microbiol 2006 ; 44 : 2220 – 8 . Google Scholar CrossRef Search ADS PubMed 15 Stein C , Makarewicz O , Bohnert JA et al. Three dimensional checkerboard synergy analysis of colistin, meropenem, tigecycline against multidrug-resistant clinical Klebsiella pneumonia isolates . PLoS One 2015 ; 10 : e0126479 . Google Scholar CrossRef Search ADS PubMed 16 Macia MD , Rojo-Molinero E , Oliver A. Antimicrobial susceptibility testing in biofilm-growing bacteria . Clin Microbiol Infect 2014 ; 20 : 981 – 90 . Google Scholar CrossRef Search ADS PubMed 17 Klinger-Strobel M , Suesse H , Fischer D et al. A novel computerized cell count algorithm for biofilm analysis . PLoS One 2016 ; 11 : e0154937. Google Scholar CrossRef Search ADS PubMed 18 Proctor RA , von Eiff C , Kahl BC et al. Small colony variants: a pathogenic form of bacteria that facilitates persistent and recurrent infections . Nat Rev Microbiol 2006 ; 4 : 295 – 305 . Google Scholar CrossRef Search ADS PubMed 19 Weng PL , Ramli R , Shamsudin MN et al. High genetic diversity of Enterococcus faecium and Enterococcus faecalis clinical isolates by pulsed-field gel electrophoresis and multilocus sequence typing from a hospital in Malaysia . Biomed Res Int 2013 ; 2013 : 938937 . Google Scholar CrossRef Search ADS PubMed 20 Getachew Y , Hassan L , Zakaria Z et al. Genetic variability of vancomycin-resistant Enterococcus faecium and Enterococcus faecalis isolates from humans, chickens, and pigs in Malaysia . Appl Environ Microbiol 2013 ; 79 : 4528 – 33 . Google Scholar CrossRef Search ADS PubMed 21 Alonso CA , Rezusta A , Seral C et al. Persistence of a ST6 clone of Enterococcus faecalis genotype vanB2 in two Hospitals in Aragon (Spain) . Enferm Infecc Microbiol Clin 2017 ; 35 : 578 – 81 . Google Scholar CrossRef Search ADS PubMed 22 Lewis K. Riddle of biofilm resistance . Antimicrob Agents Chemother 2001 ; 45 : 999 – 1007 . Google Scholar CrossRef Search ADS PubMed 23 Anwar H , Dasgupta MK , Costerton JW. Testing the susceptibility of bacteria in biofilms to antibacterial agents . Antimicrob Agents Chemother 1990 ; 34 : 2043 – 6 . Google Scholar CrossRef Search ADS PubMed 24 Stewart PS. Mechanisms of antibiotic resistance in bacterial biofilms . Int J Med Microbiol 2002 ; 292 : 107 – 13 . Google Scholar CrossRef Search ADS PubMed 25 Arbeloa A , Segal H , Hugonnet JE et al. Role of class A penicillin-binding proteins in PBP5-mediated β-lactam resistance in Enterococcus faecalis . J Bacteriol 2004 ; 186 : 1221 – 8 . Google Scholar CrossRef Search ADS PubMed 26 Land AD , Tsui HC , Kocaoglu O et al. Requirement of essential Pbp2x and GpsB for septal ring closure in Streptococcus pneumoniae D39 . Mol Microbiol 2013 ; 90 : 939 – 55 . Google Scholar CrossRef Search ADS PubMed 27 Spratt BG. Distinct penicillin binding proteins involved in the division, elongation, and shape of Escherichia coli K12 . Proc Natl Acad Sci USA 1975 ; 72 : 2999 – 3003 . Google Scholar CrossRef Search ADS PubMed 28 Ito A , Sato T , Ota M et al. In vitro antibacterial properties of cefiderocol, a novel siderophore cephalosporin, against Gram-negative bacteria . Antimicrob Agents Chemother 2017 ; doi: 10.1128/AAC.01454-17. © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Antimicrobial Chemotherapy Oxford University Press

In vitro synergism and anti-biofilm activity of ampicillin, gentamicin, ceftaroline and ceftriaxone against Enterococcus faecalis

Loading next page...
 
/lp/oxford-university-press/in-vitro-synergism-and-anti-biofilm-activity-of-ampicillin-gentamicin-qHzFam82t1
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com.
ISSN
0305-7453
eISSN
1460-2091
D.O.I.
10.1093/jac/dky051
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Enterococci frequently cause severe biofilm-associated infections such as endocarditis. The combination of ampicillin/ceftriaxone has recently been clinically evaluated as non-inferior compared with the standard therapy of ampicillin/gentamicin for treatment of Enterococcus faecalis endocarditis. Ceftaroline is a novel cephalosporin with enhanced activity against Gram-positive bacteria. Objectives To compare the in vitro effectiveness of the ceftaroline/ampicillin combination with those of gentamicin/ampicillin and ceftriaxone/ampicillin in planktonic and biofilm cultures of clinical E. faecalis isolates. Methods Synergistic effects at the planktonic level were analysed by chequerboard assays in 20 E. faecalis isolates. Biofilm-eradicating and biofilm-preventing activities of the antibiotics and their combinations were determined by confocal laser scanning microscopy with quantification by quantitative biofilm analysis (qBA) algorithm and cfu/mL determination. Results Comparable synergistic effects were observed for both β-lactam combinations in most isolates, in contrast to gentamicin/ampicillin. However, none of the antibiotic combinations succeeded in eradicating mature biofilms. Gentamicin showed promising biofilm-preventing activity, but at concentrations above those clinically tolerable. The β-lactams showed a U-shape dose–response relationship in biofilm prevention. Only exposure to cephalosporins caused alterations in cell morphology, which resulted in cell elongation and reclustering in a concentration-dependent manner. Reclustering was associated with high occurrences of small colony variants (SCVs), especially at high ceftriaxone concentrations. Conclusions This study suggests that combinations of cephalosporins or gentamicin with ampicillin may be advantageous only while bacteraemia persists, whereas combinations have no advantage over monotherapy regarding the treatment of mature biofilms. The selection of SCVs at high ceftriaxone concentrations is worth further study. Introduction Enterococci frequently cause biofilm-associated infections such as catheter-related bloodstream infections, urinary tract infections and infective endocarditis (IE).1 Biofilms are matrix-embedded communities that aggregate on artificial or natural surfaces and exhibit increased resistance to attacks from the host’s immune system and antibiotic therapy, often resulting in treatment failure, relapsing infections and increased lethality.2,3 Because the majority of clinical Enterococcus faecalis isolates remain susceptible to β-lactams, the most recommended antibiotic treatment of IE caused by E. faecalis involves ampicillin combined with gentamicin for 4–6 weeks.4 However, this therapy is limited due to the severe side effects of aminoglycosides (nephrotoxicity and ototoxicity) and the emergence of high-level aminoglycoside resistance (HLAR) among E. faecalis isolates.5 A combination of two β-lactam antibiotics, ampicillin and ceftriaxone, was recently proven in a retrospective cohort study as non-inferior and well-tolerated compared with the standard therapy and has therefore been recommended for treatment of infections caused by high-level aminoglycoside-resistant E. faecalis.6 Ceftriaxone is a third-generation cephalosporin and, as most cephalosporins, is ineffective individually in enterococci, but synergistically supports ampicillin by differential and stepwise saturation of PBPs.7 Ceftaroline is a novel, broad-spectrum, fifth-generation cephalosporin that, in contrast to ceftriaxone, shows enhanced in vitro activity against Gram-positive bacteria, including enterococci.8 It is approved for the treatment of acute Staphylococcus aureus-related skin infections and community-acquired pneumonia. In vitro data have demonstrated that ceftaroline has anti-biofilm activity against nascent and mature staphylococcal biofilms.9,10 We thus hypothesized that in line with current antibiotic combination therapies for enterococcal IE, ceftaroline plus ampicillin might be an effective treatment option for E. faecalis biofilm-associated infections. Recently, synergistic effects between ceftaroline and ampicillin against E. faecalis isolates were demonstrated in time–kill experiments11 as well as in pharmacokinetic/pharmacodynamic (PK/PD) models, including simulated endocardial vegetations.12,13 However, the effectiveness of this combination in enterococcal biofilms is still unknown. The aim of this study was to elucidate the in vitro effectiveness of the ceftaroline/ampicillin combination in comparison with the current standard gentamicin/ampicillin treatment and the recommended alternative ceftriaxone/ampicillin treatment in biofilms of clinical E. faecalis isolates. We therefore analysed the biofilm-eradicating and biofilm-preventing activity of these antibiotics and their combinations, the latter in terms of their biofilm prevention concentrations (BPCs), and conducted chequerboard assays to analyse synergistic effects in planktonic cultures and biofilms. Materials and methods Enterococcal strains, liquid cultures and antimicrobials Clinical E. faecalis (n = 20) isolates were acquired from blood cultures or from swabs of mitral valves by the Institute of Medical Microbiology at Jena University Hospital, Germany (Table 1). E. faecalis ATCC 29212 served as a reference strain for the chequerboard assays. Bacterial liquid cultures were prepared in Todd Hewitt (TH) broth (Karl Roth, Karlsruhe, Germany) and incubated at 37°C at constant rotation speed (approximately 13 g) for 2–3 h. Test solutions of ampicillin (Karl Roth, Karlsruhe, Germany), ceftriaxone (TCI Europe, Zwijndrecht, Belgium), gentamicin (TCI Europe, Zwijndrecht, Belgium) and ceftaroline (Forest Laboratories, New York City, USA) were prepared immediately before usage. Table 1. Results of susceptibility testing (MIC), synergy testing (FICI) and clinical data of the patient cohort Isolatea ST Clinical background Sex Age (years) MIC of ampicillin (mg/L) MIC of ceftaroline (mg/L) MIC of ceftriaxone (mg/L) MIC of gentamicin (mg/L) FICI of ampicillin/ ceftarolineb FICI of ampicillin/ ceftriaxone FICI of ampicillin/ gentamicin va67230 579 endocarditis male 39 1 1 16 16 0.50 0.25 1.02 va245 280 endocarditis male 76 0.5 0.125 2 16 0.73 0.50 1.02 bk5597 40 opportunistic infection without clinical signs female 55 1 0.5 8c 16 0.50 0.31c 1.00 bk848 19 opportunistic infection after liver transplant rejection male 60 1 0.25 8 16 0.49 0.31 1.02 bk905 41 endocarditis male 75 1 1 256 8 0.38 0.27 1.03 bk2164 74 opportunistic infection due to acute renal failure female 78 2 0.5 32c 4 0.37 0.25c 1.06 ATCC 29212 30 / / / 1 0.5 4 16 0.50 0.50 1.00 bk8653 6 opportunistic infection after liver transplantation male 56 1 8 >512 >512 0.25 ND ND bk3043 6 opportunistic infection due to respiratory insufficiency, liver cirrhosis female 74 0.5 1 >512 >512 0.37 ND ND bk3062 6 urosepsis due to permanent urinary catheter male 77 1 4 >512 >512 0.31 ND ND bk7183 6 urosepsis (with prostate carcinoma) male 76 1 8 512 >512 0.37 0.38 ND bk6037 6 urosepsis due to permanent urinary catheter male 86 1 2 512c >512 0.37 0.10c ND bk9190 6 wound infection female 59 1 2 >512 >512 0.50 ND ND bk281 498 sepsis due to biliary tract infection female 79 1 1 32c 8 0.50 0.25c 1.03 bk9367 64 recurrent bacteraemia female 85 2 0.5 16c 32 0.37 0.38c 0.63 bk6747 64 biliary tract infection (with Klatskin tumour) male 81 1 0.5 256c 32 0.62 0.10c 1.01 bk1653 179 urosepsis due to urinary tract infection female 87 1 2 16c 64 0.31 0.25c 0.50 bk4497 16 urosepsis due to permanent urinary catheter female 67 1 0.5 64c >512 0.50 0.10c ND bk5187 16 urosepsis post-operative male 79 1 0.25 16 256 0.49 0.19 0.63 bk6886 16 wound infection female 68 2 1 16c 32 0.50 0.19c 1.00 bk8669 23 opportunistic infection due to metastasizing rectal carcinoma male 80 1 4 32c 32 0.50 0.25c 0.56 Isolatea ST Clinical background Sex Age (years) MIC of ampicillin (mg/L) MIC of ceftaroline (mg/L) MIC of ceftriaxone (mg/L) MIC of gentamicin (mg/L) FICI of ampicillin/ ceftarolineb FICI of ampicillin/ ceftriaxone FICI of ampicillin/ gentamicin va67230 579 endocarditis male 39 1 1 16 16 0.50 0.25 1.02 va245 280 endocarditis male 76 0.5 0.125 2 16 0.73 0.50 1.02 bk5597 40 opportunistic infection without clinical signs female 55 1 0.5 8c 16 0.50 0.31c 1.00 bk848 19 opportunistic infection after liver transplant rejection male 60 1 0.25 8 16 0.49 0.31 1.02 bk905 41 endocarditis male 75 1 1 256 8 0.38 0.27 1.03 bk2164 74 opportunistic infection due to acute renal failure female 78 2 0.5 32c 4 0.37 0.25c 1.06 ATCC 29212 30 / / / 1 0.5 4 16 0.50 0.50 1.00 bk8653 6 opportunistic infection after liver transplantation male 56 1 8 >512 >512 0.25 ND ND bk3043 6 opportunistic infection due to respiratory insufficiency, liver cirrhosis female 74 0.5 1 >512 >512 0.37 ND ND bk3062 6 urosepsis due to permanent urinary catheter male 77 1 4 >512 >512 0.31 ND ND bk7183 6 urosepsis (with prostate carcinoma) male 76 1 8 512 >512 0.37 0.38 ND bk6037 6 urosepsis due to permanent urinary catheter male 86 1 2 512c >512 0.37 0.10c ND bk9190 6 wound infection female 59 1 2 >512 >512 0.50 ND ND bk281 498 sepsis due to biliary tract infection female 79 1 1 32c 8 0.50 0.25c 1.03 bk9367 64 recurrent bacteraemia female 85 2 0.5 16c 32 0.37 0.38c 0.63 bk6747 64 biliary tract infection (with Klatskin tumour) male 81 1 0.5 256c 32 0.62 0.10c 1.01 bk1653 179 urosepsis due to urinary tract infection female 87 1 2 16c 64 0.31 0.25c 0.50 bk4497 16 urosepsis due to permanent urinary catheter female 67 1 0.5 64c >512 0.50 0.10c ND bk5187 16 urosepsis post-operative male 79 1 0.25 16 256 0.49 0.19 0.63 bk6886 16 wound infection female 68 2 1 16c 32 0.50 0.19c 1.00 bk8669 23 opportunistic infection due to metastasizing rectal carcinoma male 80 1 4 32c 32 0.50 0.25c 0.56 a Isolates with various backgrounds of infection were obtained from the Institute of Medical Microbiology in Jena, Germany. bk = blood culture. va = mitral valve. / = not applicable to the laboratory standard strain. b FICI values are given as the lowest observed FICI. Synergistic FICI values (lowest FICI ≤0.5) are indicated in bold. ND = not determined due to MICs exceeding 512 mg/L. c Labelled isolates started to regrow as visible aggregates at concentrations above the determined turbid/non-turbid interface of ceftriaxone (alone and in combination with ampicillin). FICIs were calculated using the concentrations in the first non-turbid well found in each row and column, neglecting the regrowth. Table 1. Results of susceptibility testing (MIC), synergy testing (FICI) and clinical data of the patient cohort Isolatea ST Clinical background Sex Age (years) MIC of ampicillin (mg/L) MIC of ceftaroline (mg/L) MIC of ceftriaxone (mg/L) MIC of gentamicin (mg/L) FICI of ampicillin/ ceftarolineb FICI of ampicillin/ ceftriaxone FICI of ampicillin/ gentamicin va67230 579 endocarditis male 39 1 1 16 16 0.50 0.25 1.02 va245 280 endocarditis male 76 0.5 0.125 2 16 0.73 0.50 1.02 bk5597 40 opportunistic infection without clinical signs female 55 1 0.5 8c 16 0.50 0.31c 1.00 bk848 19 opportunistic infection after liver transplant rejection male 60 1 0.25 8 16 0.49 0.31 1.02 bk905 41 endocarditis male 75 1 1 256 8 0.38 0.27 1.03 bk2164 74 opportunistic infection due to acute renal failure female 78 2 0.5 32c 4 0.37 0.25c 1.06 ATCC 29212 30 / / / 1 0.5 4 16 0.50 0.50 1.00 bk8653 6 opportunistic infection after liver transplantation male 56 1 8 >512 >512 0.25 ND ND bk3043 6 opportunistic infection due to respiratory insufficiency, liver cirrhosis female 74 0.5 1 >512 >512 0.37 ND ND bk3062 6 urosepsis due to permanent urinary catheter male 77 1 4 >512 >512 0.31 ND ND bk7183 6 urosepsis (with prostate carcinoma) male 76 1 8 512 >512 0.37 0.38 ND bk6037 6 urosepsis due to permanent urinary catheter male 86 1 2 512c >512 0.37 0.10c ND bk9190 6 wound infection female 59 1 2 >512 >512 0.50 ND ND bk281 498 sepsis due to biliary tract infection female 79 1 1 32c 8 0.50 0.25c 1.03 bk9367 64 recurrent bacteraemia female 85 2 0.5 16c 32 0.37 0.38c 0.63 bk6747 64 biliary tract infection (with Klatskin tumour) male 81 1 0.5 256c 32 0.62 0.10c 1.01 bk1653 179 urosepsis due to urinary tract infection female 87 1 2 16c 64 0.31 0.25c 0.50 bk4497 16 urosepsis due to permanent urinary catheter female 67 1 0.5 64c >512 0.50 0.10c ND bk5187 16 urosepsis post-operative male 79 1 0.25 16 256 0.49 0.19 0.63 bk6886 16 wound infection female 68 2 1 16c 32 0.50 0.19c 1.00 bk8669 23 opportunistic infection due to metastasizing rectal carcinoma male 80 1 4 32c 32 0.50 0.25c 0.56 Isolatea ST Clinical background Sex Age (years) MIC of ampicillin (mg/L) MIC of ceftaroline (mg/L) MIC of ceftriaxone (mg/L) MIC of gentamicin (mg/L) FICI of ampicillin/ ceftarolineb FICI of ampicillin/ ceftriaxone FICI of ampicillin/ gentamicin va67230 579 endocarditis male 39 1 1 16 16 0.50 0.25 1.02 va245 280 endocarditis male 76 0.5 0.125 2 16 0.73 0.50 1.02 bk5597 40 opportunistic infection without clinical signs female 55 1 0.5 8c 16 0.50 0.31c 1.00 bk848 19 opportunistic infection after liver transplant rejection male 60 1 0.25 8 16 0.49 0.31 1.02 bk905 41 endocarditis male 75 1 1 256 8 0.38 0.27 1.03 bk2164 74 opportunistic infection due to acute renal failure female 78 2 0.5 32c 4 0.37 0.25c 1.06 ATCC 29212 30 / / / 1 0.5 4 16 0.50 0.50 1.00 bk8653 6 opportunistic infection after liver transplantation male 56 1 8 >512 >512 0.25 ND ND bk3043 6 opportunistic infection due to respiratory insufficiency, liver cirrhosis female 74 0.5 1 >512 >512 0.37 ND ND bk3062 6 urosepsis due to permanent urinary catheter male 77 1 4 >512 >512 0.31 ND ND bk7183 6 urosepsis (with prostate carcinoma) male 76 1 8 512 >512 0.37 0.38 ND bk6037 6 urosepsis due to permanent urinary catheter male 86 1 2 512c >512 0.37 0.10c ND bk9190 6 wound infection female 59 1 2 >512 >512 0.50 ND ND bk281 498 sepsis due to biliary tract infection female 79 1 1 32c 8 0.50 0.25c 1.03 bk9367 64 recurrent bacteraemia female 85 2 0.5 16c 32 0.37 0.38c 0.63 bk6747 64 biliary tract infection (with Klatskin tumour) male 81 1 0.5 256c 32 0.62 0.10c 1.01 bk1653 179 urosepsis due to urinary tract infection female 87 1 2 16c 64 0.31 0.25c 0.50 bk4497 16 urosepsis due to permanent urinary catheter female 67 1 0.5 64c >512 0.50 0.10c ND bk5187 16 urosepsis post-operative male 79 1 0.25 16 256 0.49 0.19 0.63 bk6886 16 wound infection female 68 2 1 16c 32 0.50 0.19c 1.00 bk8669 23 opportunistic infection due to metastasizing rectal carcinoma male 80 1 4 32c 32 0.50 0.25c 0.56 a Isolates with various backgrounds of infection were obtained from the Institute of Medical Microbiology in Jena, Germany. bk = blood culture. va = mitral valve. / = not applicable to the laboratory standard strain. b FICI values are given as the lowest observed FICI. Synergistic FICI values (lowest FICI ≤0.5) are indicated in bold. ND = not determined due to MICs exceeding 512 mg/L. c Labelled isolates started to regrow as visible aggregates at concentrations above the determined turbid/non-turbid interface of ceftriaxone (alone and in combination with ampicillin). FICIs were calculated using the concentrations in the first non-turbid well found in each row and column, neglecting the regrowth. MLST MLST was performed according to Ruiz-Garbajosa et al.14 For DNA isolation, 2–3 colonies of each isolate were resuspended in RNase-free H2O and denatured at 95°C for 10 min. The DNA was separated from cell debris by centrifugation at 12 000 g for 5 min. PCR was performed in a 25 μL reaction mixture containing 2 mM MgCl2, 1×KCl buffer, 1.5 U/μL Taq standard polymerase (all Thermo Fisher Scientific, Massachusetts, USA), 0.2 mM dNTPs (Karl Roth, Karlsruhe, Germany), 0.4 μM of each forward and reverse primer and 150 ng of chromosomal DNA. PCR products were purified using a NucleoSpin® Gel and PCR Clean-up Kit (Macherey-Nagel, Düren, Germany) and quantified in the Infinite® 200 plate reader (Tecan, Männedorf, Switzerland) using a NanoQuant plate, both following the manufacturers’ protocols. Samples were sequenced with forward and reverse primers at the EZ-Seq service (Macrogen, Amsterdam, The Netherlands). Allelic profiles and STs were assigned in accordance with the E. faecalis MLST database (http://efaecalis.mlst.net/) and analysed by the Draw Tree module applying the neighbour joining algorithm. Antimicrobial synergism testing by chequerboard assays The chequerboard assays were performed with 11 double-dilution steps of either ceftaroline, ceftriaxone or gentamicin and 7 double-dilution steps of ampicillin, as described previously.15 Antibiotic concentration ranges were selected based on MIC values determined by the broth microdilution method according to EUCAST guidelines (ISO 20776-1:2006), except that TH broth was used instead of Mueller–Hinton broth. Each chequerboard assay was performed in duplicate. The effects of the combined antibiotics were evaluated by calculating the fractional inhibitory concentration indices (FICI) along the turbidity/non-turbidity interface as described previously.15 The lowest FICI value was used for interpretation and we assumed synergism occurred at FICI ≤0.5, antagonism at FICI ≥4 and no interaction at 0.5 < FICI < 4. Anti-biofilm testing Biofilms were grown in optical microtitre plates (0.54 cm2/well) with glass bottoms (Greiner Bio-one, Kremsmünster, Austria) in triplicate by applying 100 μL of a bacterial suspension (0.5 McFarland) to each well. Plates were placed in a humidified chamber and incubated at 37°C, 5% CO2 without shaking. To assess biofilm-eradicating effects, we grew the biofilms for 24 h, carefully removed the supernatant and added 100 μL (2 × 50 μL each for the combinations) of the antibiotic solution prepared in TH medium. Medium was changed in the untreated growth controls. Plates were incubated for a further 24 h before analysis. For assessment of the BPC, the antibiotics in sub-MICs were directly added to the bacterial suspension (0.5 McFarland/well). The plates were grown for 48 h without the medium being changed. For quantification of antibiotic effects, the viable cell number was determined by counting the number of viable bacteria (corresponding to cfu/cm2). The supernatant of the biofilms was therefore carefully removed; the biofilms were washed twice with 100 μL of sterile NaCl, scraped off and resuspended in TH medium. Selected 10-fold dilutions were plated on TH-agar plates and incubated overnight at 37°C, followed by counting of cfu/mL. The BPC was defined as the first concentration at which the viable cell count decreased significantly by 1 log magnitude compared with the untreated control, referring to Macia et al.16 Biofilm imaging and computed analysis Biofilms were stained using the LIVE/DEAD BacLight Bacterial Viability Kit for microscopy (Life Technologies GmbH, Darmstadt, Germany) according to the manufacturer’s protocol. Stained biofilms were analysed under vital conditions using an inverse confocal laser scanning microscope (CLSM), LSM510 (Carl Zeiss AG, Jena, Germany), as described previously.17 The biofilm images were visualized by ZEN 9.0 software (Carl Zeiss AG, Jena, Germany). The biofilm experiments [eradication and prevention (BPC)] were independently performed in duplicate for each strain and in triplicate for each assay. Quantitative analysis of biofilm images was performed by an algorithm termed qBA (quantitative biofilm analysis) that determined the number of bacterial counts/cm2.17 Statistical analysis The correlation of the MIC was analysed by non-parametric Spearman’s rank correlation coefficient (rs) with a two-tailed CI of 95%. The non-parametric Kruskal–Wallis test followed by Dunn’s multiple comparison post-test with a CI of 95% was used for statistical analysis of the cfu/mL or cfu/cm2 quantification. Differences were considered significant at P values <0.05. Results Characterization of isolates Most of the clinical isolates (14/20) were obtained from patients with biofilm-associated infections, including three isolates from patients with endocarditis (Table 1). Among the 20 isolates, 9 isolates exhibited unique STs and variable resistance profiles. Six isolates, all highly resistant to gentamicin and ceftriaxone (≥512 mg/L), belonged to ST6, whereas ST64 (n = 2) and ST16 (n = 3) also showed resistance to those antibiotics, but had different MIC values. Synergy testing on planktonic bacteria To analyse synergistic effects between ceftaroline/ampicillin, ceftriaxone/ampicillin and gentamicin/ampicillin, we performed chequerboard assays for the 20 clinical E. faecalis isolates and the laboratory standard strain ATCC 29212 (Table 1). All isolates showed ampicillin MIC values of 1 mg/L ± 1× MIC. The gentamicin MICs ranked between 4 and >512 mg/L, with eight isolates exhibiting an HLAR profile (MIC of gentamicin >128 mg/L). One isolate exhibiting low-level resistance against gentamicin (MIC of gentamicin <128 mg/L) experienced synergistic effects from gentamicin plus ampicillin (FICI = 0.5), while all other isolates (non-HLAR and HLAR) showed FICI values from >0.5 to 1, indicating no interaction between these two antibiotics [Table 1 and Figure S2 (available as Supplementary data at JAC Online)]. The MIC values of ceftriaxone ranked between 2 and >512 mg/L, while the MIC values of ceftaroline were generally lower, with values between 0.125 mg/L and 8 mg/L (Table 1). Both the MIC of ceftriaxone and the MIC of ceftaroline showed no correlation with the MIC of ampicillin, but the MIC of ceftaroline positively correlated with the MIC of ceftriaxone (rs = 0.705, P < 0.001). Synergistic effects were observed for ceftaroline and ampicillin in 19 of 21 tested isolates, with FICI values of ampicillin/ceftaroline between 0.25 and 0.50 (Table 1). Ceftriaxone and ampicillin synergistically inhibited the growth of 17 isolates, with FICI values of ampicillin/ceftriaxone between 0.1 and 0.5; 15 of these isolates also showed synergistic effects for ceftaroline and ampicillin. At higher concentrations of ceftriaxone (alone and in combination with ampicillin), visible turbid clusters were observed beyond the turbid/non-turbid interface (greater than the MIC of ceftriaxone) for individual isolates (marked in Table 1), suggesting regrowth. This phenomenon was not observed for ceftaroline. Low concentrations of ceftaroline or ceftriaxone were sufficient to strongly reduce the effective concentrations of ampicillin (Figure S2 and Table 1). This effect was most apparent in isolates with high MICs of ceftriaxone, as shown by a weak correlation between the FICI of ampicillin/ceftriaxone and the MIC of ceftriaxone (rs = −0.54, P = 0.025). No correlations between the other MICs and the FICI values or between FICI values and clonality based on MLST (Figure S1) were observed. Biofilm eradication by ceftaroline, ceftriaxone, ampicillin and gentamicin Biofilm-eradicating effects were assessed by CLSM imaging of five biofilm-associated isolates from patients with endocarditis (va245, va67230, bk905) or urosepsis (bk1653, bk6037); these isolates all exhibited synergistic FICI values for the ceftaroline/ampicillin combination (Table 1). Except for isolate bk6037, the isolates produced strong biofilms after 24 h of growth (data not shown). None of the antibiotics in concentrations up to 1000× MIC showed visible biofilm-eradicating effects in any of the isolates, neither alone nor in combination, compared with the untreated control after 24 h of exposure. The biofilms’ thicknesses did not decrease, and the number of dead bacteria did not increase after antibiotic exposure (Figure S3). Biofilm prevention by ceftaroline, ceftriaxone, ampicillin and gentamicin The BPC was tested at sub-MICs of ceftaroline, ceftriaxone, ampicillin and gentamicin for selected isolates (va245, va67230, bk1653 and bk6037). Subinhibitory concentrations of both cephalosporins (ceftaroline and ceftriaxone) induced morphological alterations at the single-cell level in a concentration-dependent manner (Figure 1). As shown in isolate bk1653, the enterococci elongated with increasing ceftaroline and ceftriaxone concentrations in all Z-layers, adapting a long rod shape at ≤1/8× MIC that resulted in filamentous structures. This effect, albeit not as strongly, was also observed for isolates va245 and va67230 (data not shown). The elongated cells were viable, but their density decreased with increasing cephalosporin concentrations as confirmed by cfu/mL count and qBA (Figure 2). As per definition (see the Materials and methods section), the BPCs of both cephalosporins were reached at the elongated stage at 1/8× MIC (0.125 mg/L ceftaroline and 2 mg/L ceftriaxone). At cephalosporin concentrations higher than the BPC, the elongated enterococci reverted to the coccal shape and appeared as clusters in the CLSM images (Figure 1). Ceftaroline-induced cell clusters disappeared at 4× MIC of ceftaroline (4 mg/L), while ceftriaxone-induced clusters were still present at the highest concentration tested (16× MIC of ceftriaxone, 256 mg/L) (data not shown). Figure 1. View largeDownload slide Effects of antibiotics on biofilm formation. E. faecalis isolate bk1653 was grown for 48 h under static conditions in a 96-well glass-bottom plate in the presence of sub-MICs of ampicillin, ceftaroline, ceftriaxone and gentamicin. CLSM images show viable bacteria in green (SYTO 9) and dead bacteria in red (propidium iodide). The concentration corresponding to the fold MIC is noted in each image. Scales in the images of 1/16× MIC apply for all corresponding images to the right. Figure 1. View largeDownload slide Effects of antibiotics on biofilm formation. E. faecalis isolate bk1653 was grown for 48 h under static conditions in a 96-well glass-bottom plate in the presence of sub-MICs of ampicillin, ceftaroline, ceftriaxone and gentamicin. CLSM images show viable bacteria in green (SYTO 9) and dead bacteria in red (propidium iodide). The concentration corresponding to the fold MIC is noted in each image. Scales in the images of 1/16× MIC apply for all corresponding images to the right. Figure 2. View largeDownload slide Biofilm prevention activity of the antibiotics. E. faecalis isolate bk1653 was grown for 48 h under static conditions in a 96-well glass-bottom plate in the presence of sub-MICs of ampicillin, ceftaroline, ceftriaxone and gentamicin. (a) to (d) Quantification of the cfu/cm2. Statistical analysis was performed by one-way ANOVA (Kruskal–Wallis test) followed by Dunn’s multiple comparison test. (e) to (h) Quantification of the bacteria counts/cm2 in the microscope images by qBA. The quantifications were determined independently twice and then in triplicate for each antibiotic. The mean values and the ranges are shown in the diagrams. Figure 2. View largeDownload slide Biofilm prevention activity of the antibiotics. E. faecalis isolate bk1653 was grown for 48 h under static conditions in a 96-well glass-bottom plate in the presence of sub-MICs of ampicillin, ceftaroline, ceftriaxone and gentamicin. (a) to (d) Quantification of the cfu/cm2. Statistical analysis was performed by one-way ANOVA (Kruskal–Wallis test) followed by Dunn’s multiple comparison test. (e) to (h) Quantification of the bacteria counts/cm2 in the microscope images by qBA. The quantifications were determined independently twice and then in triplicate for each antibiotic. The mean values and the ranges are shown in the diagrams. In contrast to the response to the cephalosporins, cell elongation was not observed with ampicillin exposure (Figure 1). The number of viable bacteria remained stable at increasing concentrations until a sudden drop at the BPC at 1/2× MIC of ampicillin (0.5 mg/L ampicillin) (Figure 2). However, ampicillin concentrations above the BPC led to increased bacterial counts, which correlated with cluster formation observed in the CLSM images as under cephalosporin treatment (Figure 1). The clusters disappeared at 2× MIC of ampicillin (2 mg/L) (data not shown). In contrast to the β-lactams, gentamicin killed bacterial cells in the nascent biofilm in a concentration-dependent manner without the clustering effect. Gentamicin did not prevent biofilm formation below the MIC (64 mg/L), but at the MIC of gentamicin (= BPC) no cfu/mL were detected. Synergism in biofilm prevention by ampicillin combined with ceftaroline or gentamicin Only a minor synergism between ampicillin and ceftaroline in biofilm prevention was detected at ≤1/8× MIC of ampicillin because the viable cell numbers per cm2 from the untreated control value were not reduced by 1 log magnitude (Figure 3a). Combining sub-MICs of ceftaroline with increasing ampicillin concentrations ≤1/8× MIC of ampicillin led to the same cell shape change from elongation to coccal clusters as was observed with only sub-MIC ceftaroline exposure (Figure S4). However, the clustering effect was achieved at lower ceftaroline concentrations in combination than with treatment with ceftaroline alone. The BPC for the ceftaroline/ampicillin combination was first reached at 1/4× MIC of ampicillin and 1/32× MIC of ceftaroline, with a reduction in viable cells/cm2 of at least 1.5 log magnitudes (Figure 3a). No elongated cells or clusters were observed under these conditions (Figure S4). The minimum number of viable cells/cm2 was reached at 1/2× MIC of ampicillin plus 1/32× MIC of ceftaroline. Synergy inversion was surprisingly observed with increasing ceftaroline concentrations, i.e. cell numbers were higher than under 1/2× MIC of ampicillin alone (Figure 3a). Figure 3. View largeDownload slide Quantification of the effects on biofilm formation of combining ampicillin with ceftaroline (a) or gentamicin (b). The viable cell counts were determined by the qBA algorithm based on CLSM images (approximately 100 μm × 100 μm) and scaled up to an area of 1 cm2 (104 cells/cm2 represents the limit of detection of this method). The experiments were performed in triplicate and the means and standard deviations are shown. Figure 3. View largeDownload slide Quantification of the effects on biofilm formation of combining ampicillin with ceftaroline (a) or gentamicin (b). The viable cell counts were determined by the qBA algorithm based on CLSM images (approximately 100 μm × 100 μm) and scaled up to an area of 1 cm2 (104 cells/cm2 represents the limit of detection of this method). The experiments were performed in triplicate and the means and standard deviations are shown. The ampicillin/gentamicin combination indifferently affected biofilm prevention (Figures 3b and S5). The combination of gentamicin at concentrations below the MIC of gentamicin and ampicillin at 1/2× MIC of ampicillin increased the bacterial count by 1 log magnitude, indicating an antagonistic effect (Figure 3b). Formation of small colony variants (SCVs) under antimicrobial treatment The formation of clusters above the BPC of ceftaroline, ceftriaxone and ampicillin led us to quantify the proportion of SCVs18 in the viable bacteria (isolates bk1653 and va67230) (cfu/mL) after biofilm growth in the presence of 0.5×, 1× and 2× MIC of each antibiotic. After 48 h, biofilms were resolved and plated on agar to allow for cfu/mL and SCV determination. SCVs were distinguished from normal bacteria as pinpoint colonies18 at higher dilutions of the bacterial suspension. Isolate bk1653 showed a high proportion of SCVs (between 40% and 70%; Figure 4) in the presence of 0.5× and 1× MIC of each β-lactam. At 2× MIC, only the cephalosporins selected SCVs, while under ampicillin exposure, the SCV proportion was strongly reduced to 1.5%. In contrast to isolate bk1653, isolate va67230 produced SCVs only at 0.5× MIC of ampicillin and at all tested ceftaroline concentrations, but not under ceftriaxone exposure (data not shown). Gentamicin led to low or undetectable levels of SCV formation at all concentrations tested in both isolates. In the untreated control and at concentrations lower than 0.5× MIC, the SCV phenotype was not detected for either isolate. Figure 4. View largeDownload slide Proportion of SCVs in viable bacteria (cfu/mL) in dependence on the antibiotics present during biofilm formation. Biofilms of E. faecalis isolate bk1653 were grown in the presence of 0.5×, 1× and 2× MIC of ampicillin (AMP), ceftaroline (CPT), ceftriaxone (CRO) or gentamicin (GEN). After 48 h, biofilms were resolved and plated on agar to allow for cfu/mL and SCV determination. Experiments were performed in triplicate and the means and ranges are shown. The untreated control biofilms never exhibited SCVs. Figure 4. View largeDownload slide Proportion of SCVs in viable bacteria (cfu/mL) in dependence on the antibiotics present during biofilm formation. Biofilms of E. faecalis isolate bk1653 were grown in the presence of 0.5×, 1× and 2× MIC of ampicillin (AMP), ceftaroline (CPT), ceftriaxone (CRO) or gentamicin (GEN). After 48 h, biofilms were resolved and plated on agar to allow for cfu/mL and SCV determination. Experiments were performed in triplicate and the means and ranges are shown. The untreated control biofilms never exhibited SCVs. Discussion In the present study, 20 clinical E. faecalis isolates and one laboratory standard strain were analysed in vitro for synergism between ampicillin and ceftaroline, ceftriaxone or gentamicin in the planktonic and biofilm-embedded status, respectively. Nine isolates were not related to each other and most likely not of nosocomial origin. In contrast, we cannot clearly exclude a nosocomial source for the three clonal clusters, with ST6 being a prevalent clade often found in wild animals19 and livestock,20 but also spreading endemically in hospitals.19,21 Compared with the current standard therapy of gentamicin/ampicillin for E. faecalis IE, the ceftaroline/ampicillin combination showed a superior synergistic effect in vitro against planktonic cultures of clinical E. faecalis isolates, including all HLAR isolates, but FICI values were similar compared with those of the recently recommended ceftriaxone/ampicillin combination.4 These results are in accordance with those of two recently published smaller studies that reported synergism of ceftaroline/ampicillin combinations in different PK/PD models and concluded that only in a limited number of strains ceftaroline might have better activity than ceftriaxone in combination with ampicillin.12,13 However, E. faecalis infections associated with biofilm formation are less susceptible to antibiotics due to biofilm-specific antimicrobial tolerance mechanisms, such as reduced antibiotic penetration and metabolic dormancy.22 The routinely assessed MICs of antibiotics that are sufficient to kill planktonic bacteria are thus usually insufficient to eradicate their biofilm-embedded counterparts.23 Biofilm susceptibility testing is not currently part of diagnostic routines because neither standards for biofilm analysis nor specific breakpoints have been established by official agencies such as EUCAST. We therefore used a microscopy approach to evaluate the efficacy of ampicillin combined with ceftaroline, ceftriaxone or gentamicin in biofilm eradication and the inhibition of biofilm formation and compared the results with the resource-consuming method of cfu/mL determination. The reduction in viable bacterial counts/cm2 in the CLSM images obtained by qBA corresponded to the cfu/cm2 values, but the qBA values were generally lower because the analysed areas of the images were at the centres of the wells, which bear fewer cells than the edges of the wells; in contrast, the edges were included in the cfu/mL determination. Neither the individual antibiotics nor their combinations were sufficient to eradicate established E. faecalis biofilms in vitro. It is still unclear whether β-lactams and aminoglycosides efficiently penetrate the biofilm matrix of E. faecalis because we are not aware of any studies of E. faecalis biofilm penetration. Both antibiotic classes exhibit limited anti-biofilm activity in many species.24 Because β-lactams are inhibitors of cell wall synthesis, the effectiveness of the β-lactam combinations in biofilm eradication is likely hampered by the strongly slowed cell division in metabolically dormant biofilms.24 Gentamicin alone showed promising biofilm-preventing activity but was only effective outside clinically tolerable concentrations. No reduction in BPC of gentamicin was achieved when it was used in combination with ampicillin. All tested β-lactams reduced biofilm formation at sub-MIC concentrations. The combination of ampicillin with ceftaroline, compared with each β-lactam alone, improved the inhibition of biofilm formation in a concentration-dependent manner, suggesting a synergistic interaction between these β-lactams. PBPs 2 and 3 have been identified as primary targets of cephalosporins, while aminopenicillins target PBPs 4 and 5.7 Ampicillin and ceftaroline, similar to the ampicillin/ceftriaxone combination, are thus likely to synergize by inhibiting complementary PBPs, thereby disrupting the cooperation of the PBPs necessary for cell wall biosynthesis.25 Ceftaroline has a higher affinity than ceftriaxone to PBP 5 while maintaining a high affinity for PBPs 2 and 3,8 which explains the lower MIC values of ceftaroline compared with ceftriaxone. The most effective dosing of ceftaroline/ampicillin was achieved at 1/32× MIC of ceftaroline and 1/4× to 1/2× MIC of ampicillin in the BPC experiments, but synergy inversion was observed with increasing ceftaroline concentrations. The effectiveness of ampicillin was thus increased by very low ceftaroline concentrations but was hampered by higher sub-MIC concentrations of ceftaroline. Lower concentrations of ceftaroline likely saturate PBPs 2 and 3, complementing PBP 4 and 5 saturation by ampicillin, while at higher concentrations, ampicillin and ceftaroline may compete for binding of PBP 5, leading to an antagonistic effect. However, this behaviour was not observed at the planktonic level in the chequerboard assays, which displayed synergistic concave isoboles (Figure S2). This result might be caused by the different inocula that were used, i.e. 5 × 105 cfu/mL for MIC and synergism assays versus 1 × 108 cfu/mL in BPC assays. The cephalosporins exhibited a U-shape dose–response relationship in biofilm prevention, which resulted in cell elongation and reclustering in a concentration-dependent manner. Both ceftaroline and ceftriaxone at sub-MICs led to filamentation, suggesting a similar mode of action for both cephalosporins. This hypothesis is further supported by the significant correlation of the MIC of ceftriaxone with the MIC of ceftaroline and the lack of correlation of both with the MIC of ampicillin. The observed filamentation may be explained by an inhibition of cell separation. Specific PBPs of Streptococcus pneumoniae have been involved in septal ring closure in previous studies.26 PBPs 2 and 3 of E. faecalis, which become saturated at very low concentrations of cephalosporins,7 might also be involved in this process. Elongation of bacterial cells exposed to cephalosporins has been described for different Gram-negative bacterial species,27,28 but to our knowledge this report is the first regarding this behaviour in enterococci. Both cephalosporins strongly induced the formation of SCVs at concentrations close to their respective MICs, but with different intensities and not in all isolates. The observed clusters of viable cells escaping the cephalosporins most likely represented the SCVs, and the clustering might be the result of the reduced SCV growth rate.18 Further, the disappearance of the cell clusters at 2× MIC of ampicillin correlated with the decrease in the SCV rate of some isolates (e.g. bk1653) at 2× MIC of ampicillin. Clusters were still observed at 16× MIC of ceftriaxone, indicating that ceftriaxone is unable to kill SCVs. This study suggests that combinations of cephalosporins or gentamicin with ampicillin may be only advantageous for the treatment of persistent bacteraemia (i.e. planktonic cells) but seem to be non-superior compared with monotherapy against mature biofilms. In this regard, the recommendation to treat patients that have enterococcal endocarditis over the entire 4–6 weeks with an antibacterial combination may be questioned, particularly in the face of increased risks of prolonged cephalosporin (e.g. Clostridium difficile colitis) and gentamicin treatment (e.g. nephrotoxicity).4 Furthermore, high cephalosporin concentrations seem to favour selection of SCVs, suggesting that higher doses of combined ceftriaxone (e.g. 2 × 2 g daily) might even be detrimental, at least for some strains. Clinical studies are therefore mandatory to elucidate whether the selection of SCVs composes a risk of using combined therapies against enterococcal biofilm-associated infections. Acknowledgements Parts of this work were presented at the Twenty-seventh European Congress of Clinical Microbiology and Infectious Diseases, Vienna, Austria, 2017 (Poster 2753).  We acknowledge Forest Laboratories, Inc. for providing us with ceftaroline powder. Funding This work was supported by the Federal Ministry of Education and Research, Germany (grant numbers 01KI1501 and 13GW0096D) and the Argus Foundation. Transparency declarations None to declare.  Forest Laboratories, Inc. supplied the ceftaroline powder, but had no involvement in: the design of the study; the collection, analysis and interpretation of the data; or the decision to present these results. Supplementary data Figures S1 to S5 are available as Supplementary data at JAC Online. References 1 Arias CA , Murray BE. The rise of the Enterococcus: beyond vancomycin resistance . Nat Rev Microbiol 2012 ; 10 : 266 – 78 . Google Scholar CrossRef Search ADS PubMed 2 Bjarnsholt T , Alhede M , Alhede M et al. The in vivo biofilm . Trends Microbiol 2013 ; 21 : 466 – 74 . Google Scholar CrossRef Search ADS PubMed 3 Hall-Stoodley L , Stoodley P , Kathju S et al. Towards diagnostic guidelines for biofilm-associated infections . FEMS Immunol Med Microbiol 2012 ; 65 : 127 – 45 . Google Scholar CrossRef Search ADS PubMed 4 Habib G , Lancellotti P , Antunes MJ et al. 2015 ESC guidelines for the management of infective endocarditis: the Task Force for the Management of Infective Endocarditis of the European Society of Cardiology (ESC). Endorsed by: European Association for Cardio-Thoracic Surgery (EACTS), the European Association of Nuclear Medicine (EANM) . Eur Heart J 2015 ; 36 : 3075 – 128 . Google Scholar CrossRef Search ADS PubMed 5 Nigo M , Munita JM , Arias CA et al. What’s new in the treatment of enterococcal endocarditis? Curr Infect Dis Rep 2014 ; 16 : 431. Google Scholar CrossRef Search ADS PubMed 6 Fernandez-Hidalgo N , Almirante B , Gavalda J et al. Ampicillin plus ceftriaxone is as effective as ampicillin plus gentamicin for treating enterococcus faecalis infective endocarditis . Clin Infect Dis 2013 ; 56 : 1261 – 8 . Google Scholar CrossRef Search ADS PubMed 7 Mainardi JL , Gutmann L , Acar JF et al. Synergistic effect of amoxicillin and cefotaxime against Enterococcus faecalis . Antimicrob Agents Chemother 1995 ; 39 : 1984 – 7 . Google Scholar CrossRef Search ADS PubMed 8 Henry X , Verlaine O , Amoroso A et al. Activity of ceftaroline against Enterococcus faecium PBP5 . Antimicrob Agents Chemother 2013 ; 57 : 6358 – 60 . Google Scholar CrossRef Search ADS PubMed 9 Barber KE , Smith JR , Ireland CE et al. Evaluation of ceftaroline alone and in combination against biofilm-producing methicillin-resistant Staphylococcus aureus with reduced susceptibility to daptomycin and vancomycin in an in vitro pharmacokinetic/pharmacodynamic model . Antimicrob Agents Chemother 2015 ; 59 : 4497 – 503 . Google Scholar CrossRef Search ADS PubMed 10 Lazaro-Diez M , Remuzgo-Martinez S , Rodriguez-Mirones C et al. Effects of subinhibitory concentrations of ceftaroline on methicillin-resistant Staphylococcus aureus (MRSA) biofilms . PLoS One 2016 ; 11 : e0147569 . Google Scholar CrossRef Search ADS PubMed 11 Werth BJ , Abbott AN. The combination of ampicillin plus ceftaroline is synergistic against Enterococcus faecalis . J Antimicrob Chemother 2015 ; 70 : 2414 – 7 . Google Scholar CrossRef Search ADS PubMed 12 Luther MK , Rice LB , LaPlante KL. Ampicillin in combination with ceftaroline, cefepime, or ceftriaxone demonstrates equivalent activities in a high-inoculum Enterococcus faecalis infection model . Antimicrob Agents Chemother 2016 ; 60 : 3178 – 82 . Google Scholar CrossRef Search ADS PubMed 13 Werth BJ , Shireman LM. Pharmacodynamics of ceftaroline plus ampicillin against Enterococcus faecalis in an in vitro pharmacokinetic/pharmacodynamic model of simulated endocardial vegetations . Antimicrob Agents Chemother 2017 ; 61 : e02235-16. Google Scholar CrossRef Search ADS PubMed 14 Ruiz-Garbajosa P , Bonten MJ , Robinson DA et al. Multilocus sequence typing scheme for Enterococcus faecalis reveals hospital-adapted genetic complexes in a background of high rates of recombination . J Clin Microbiol 2006 ; 44 : 2220 – 8 . Google Scholar CrossRef Search ADS PubMed 15 Stein C , Makarewicz O , Bohnert JA et al. Three dimensional checkerboard synergy analysis of colistin, meropenem, tigecycline against multidrug-resistant clinical Klebsiella pneumonia isolates . PLoS One 2015 ; 10 : e0126479 . Google Scholar CrossRef Search ADS PubMed 16 Macia MD , Rojo-Molinero E , Oliver A. Antimicrobial susceptibility testing in biofilm-growing bacteria . Clin Microbiol Infect 2014 ; 20 : 981 – 90 . Google Scholar CrossRef Search ADS PubMed 17 Klinger-Strobel M , Suesse H , Fischer D et al. A novel computerized cell count algorithm for biofilm analysis . PLoS One 2016 ; 11 : e0154937. Google Scholar CrossRef Search ADS PubMed 18 Proctor RA , von Eiff C , Kahl BC et al. Small colony variants: a pathogenic form of bacteria that facilitates persistent and recurrent infections . Nat Rev Microbiol 2006 ; 4 : 295 – 305 . Google Scholar CrossRef Search ADS PubMed 19 Weng PL , Ramli R , Shamsudin MN et al. High genetic diversity of Enterococcus faecium and Enterococcus faecalis clinical isolates by pulsed-field gel electrophoresis and multilocus sequence typing from a hospital in Malaysia . Biomed Res Int 2013 ; 2013 : 938937 . Google Scholar CrossRef Search ADS PubMed 20 Getachew Y , Hassan L , Zakaria Z et al. Genetic variability of vancomycin-resistant Enterococcus faecium and Enterococcus faecalis isolates from humans, chickens, and pigs in Malaysia . Appl Environ Microbiol 2013 ; 79 : 4528 – 33 . Google Scholar CrossRef Search ADS PubMed 21 Alonso CA , Rezusta A , Seral C et al. Persistence of a ST6 clone of Enterococcus faecalis genotype vanB2 in two Hospitals in Aragon (Spain) . Enferm Infecc Microbiol Clin 2017 ; 35 : 578 – 81 . Google Scholar CrossRef Search ADS PubMed 22 Lewis K. Riddle of biofilm resistance . Antimicrob Agents Chemother 2001 ; 45 : 999 – 1007 . Google Scholar CrossRef Search ADS PubMed 23 Anwar H , Dasgupta MK , Costerton JW. Testing the susceptibility of bacteria in biofilms to antibacterial agents . Antimicrob Agents Chemother 1990 ; 34 : 2043 – 6 . Google Scholar CrossRef Search ADS PubMed 24 Stewart PS. Mechanisms of antibiotic resistance in bacterial biofilms . Int J Med Microbiol 2002 ; 292 : 107 – 13 . Google Scholar CrossRef Search ADS PubMed 25 Arbeloa A , Segal H , Hugonnet JE et al. Role of class A penicillin-binding proteins in PBP5-mediated β-lactam resistance in Enterococcus faecalis . J Bacteriol 2004 ; 186 : 1221 – 8 . Google Scholar CrossRef Search ADS PubMed 26 Land AD , Tsui HC , Kocaoglu O et al. Requirement of essential Pbp2x and GpsB for septal ring closure in Streptococcus pneumoniae D39 . Mol Microbiol 2013 ; 90 : 939 – 55 . Google Scholar CrossRef Search ADS PubMed 27 Spratt BG. Distinct penicillin binding proteins involved in the division, elongation, and shape of Escherichia coli K12 . Proc Natl Acad Sci USA 1975 ; 72 : 2999 – 3003 . Google Scholar CrossRef Search ADS PubMed 28 Ito A , Sato T , Ota M et al. In vitro antibacterial properties of cefiderocol, a novel siderophore cephalosporin, against Gram-negative bacteria . Antimicrob Agents Chemother 2017 ; doi: 10.1128/AAC.01454-17. © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Antimicrobial ChemotherapyOxford University Press

Published: Feb 28, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off