D by 100 ethanol for 20 min. Then the cells were coated with

D by 100 ethanol for 20 min. Then the cells were coated with gold and imaged by scanning electron microscopy.Preparation of AFM SamplesFor AFM imaging in air, aliquots of bacterial suspension were added to glass cover slips and rinsed gently in deionized water. The glass cover slips were then dried in a covered Petri dish (50 mm 6 50 mm) for several minutes and immediately moved to AFM for imaging.AFM OperationThe AFM images were taken in air with a commercial AFM (Dimension 3100 with a Nanoscope III controller, Digital Instruments). All images were collected in tapping mode using silicon cantilevers (RTESP, Veeco Probes) with resonance frequencies of approximately 300 kHz and spring constants of approximately 50 Nm21. Cantilevers were not reused to prevent cross-contamination. Height and amplitude images were simultaneously acquired at scan rates of 0.521 Hz with 5126512 resolutions. All visible images were given as amplitude images. To describe the topography of the 25331948 bacterial surface quantitatively, section analyses and root-mean-square roughnesses (Rrms) were taken from the height images. Once AFM images of single cells were acquired, we Title Loaded From File selectively magnified areas (500 nm 6 500 nm) of the cells except their both ends. Then, we randomly selectedDetermination of MIC and Confirmation of the Synergistic EffectMICs were determined by broth microdilution method [11]. Tubes containing various concentrations of EGCG or cefotaxime (or both) in MHB media were prepared. The total volume of media in each tube was 5 ml. The media were then inoculated with 50 ml of E. coli Title Loaded From File suspensions (OD600 = 4) and incubated at 37uC with aeration for 18 h. MIC is defined as the lowestTable 1. MIC and FIC indices of cefotaxime in combination with EGCG against ESBL-EC.MIC (mg/L) A Cefotaxime 128 8 4 4 B C DFIC Indexa B C DEffect0.0.0.SynergybA, Cefotaxime alone; B, plus EGCG (50 mg/L); C, plus EGCG (100 mg/L); D, plus EGCG (250 mg/L). The MIC of EGCG alone was 1500 mg/L. a Fractional inhibitory concentration (FIC) was calculated as MIC of antibiotics alone or EGCG in com-bination divided by MIC of antibiotics or EGCG alone, and the FIC Index was obtained by adding theFICs. b FIC indices were interpreted as below: #0.5, synergy; .0.5 to 1, addition; and .1, indifference. doi:10.1371/journal.pone.0048880.tAFM Study of Effects between EGCG and CefotaximeFigure 1. Time-kill curves of ESBL-EC treated with EGCG and cefotaxime at sub-MICs. doi:10.1371/journal.pone.0048880.gvery small regions (500 nm2) from the magnified areas to reduce the effect of bacterial half cylindrical structure on Rrms calculation. Membrane roughness was an average of these values taken from at least 20 cells.Oxidative Stress Detection using Flow CytometryDihydrorhodamine 123 (DHR 123, Sigma), an oxidative stress probe, was used to measure intracellular oxidation levels in bacteria [23] after chemical treatment. EGCG, cefotaxime, or combinations thereof were added to the culture and incubated for either 4 h or 8 h. Then, cells were washed in phosphate-buffered saline (PBS) and incubated with 5 mg/L of DHR 123 for 1 h.Figure 2. SEM images of ESBL-EC treated with sub-MICs of EGCG, cefotaxime or their combinations. Cells were: treated with 250 mg/L of EGCG for 4 h (A) and 8 h (B); treated with 4 mg/L of cefotaxime for 4 h (C) and 8 h (D); and treated with 250 mg/L of EGCG and 4 mg/L of cefotaxime in combination for 4 h (E) and 8 h (F). Scale bar: 1 mm. doi:10.1371/journal.pone.0048880.gAFM Study o.D by 100 ethanol for 20 min. Then the cells were coated with gold and imaged by scanning electron microscopy.Preparation of AFM SamplesFor AFM imaging in air, aliquots of bacterial suspension were added to glass cover slips and rinsed gently in deionized water. The glass cover slips were then dried in a covered Petri dish (50 mm 6 50 mm) for several minutes and immediately moved to AFM for imaging.AFM OperationThe AFM images were taken in air with a commercial AFM (Dimension 3100 with a Nanoscope III controller, Digital Instruments). All images were collected in tapping mode using silicon cantilevers (RTESP, Veeco Probes) with resonance frequencies of approximately 300 kHz and spring constants of approximately 50 Nm21. Cantilevers were not reused to prevent cross-contamination. Height and amplitude images were simultaneously acquired at scan rates of 0.521 Hz with 5126512 resolutions. All visible images were given as amplitude images. To describe the topography of the 25331948 bacterial surface quantitatively, section analyses and root-mean-square roughnesses (Rrms) were taken from the height images. Once AFM images of single cells were acquired, we selectively magnified areas (500 nm 6 500 nm) of the cells except their both ends. Then, we randomly selectedDetermination of MIC and Confirmation of the Synergistic EffectMICs were determined by broth microdilution method [11]. Tubes containing various concentrations of EGCG or cefotaxime (or both) in MHB media were prepared. The total volume of media in each tube was 5 ml. The media were then inoculated with 50 ml of E. coli suspensions (OD600 = 4) and incubated at 37uC with aeration for 18 h. MIC is defined as the lowestTable 1. MIC and FIC indices of cefotaxime in combination with EGCG against ESBL-EC.MIC (mg/L) A Cefotaxime 128 8 4 4 B C DFIC Indexa B C DEffect0.0.0.SynergybA, Cefotaxime alone; B, plus EGCG (50 mg/L); C, plus EGCG (100 mg/L); D, plus EGCG (250 mg/L). The MIC of EGCG alone was 1500 mg/L. a Fractional inhibitory concentration (FIC) was calculated as MIC of antibiotics alone or EGCG in com-bination divided by MIC of antibiotics or EGCG alone, and the FIC Index was obtained by adding theFICs. b FIC indices were interpreted as below: #0.5, synergy; .0.5 to 1, addition; and .1, indifference. doi:10.1371/journal.pone.0048880.tAFM Study of Effects between EGCG and CefotaximeFigure 1. Time-kill curves of ESBL-EC treated with EGCG and cefotaxime at sub-MICs. doi:10.1371/journal.pone.0048880.gvery small regions (500 nm2) from the magnified areas to reduce the effect of bacterial half cylindrical structure on Rrms calculation. Membrane roughness was an average of these values taken from at least 20 cells.Oxidative Stress Detection using Flow CytometryDihydrorhodamine 123 (DHR 123, Sigma), an oxidative stress probe, was used to measure intracellular oxidation levels in bacteria [23] after chemical treatment. EGCG, cefotaxime, or combinations thereof were added to the culture and incubated for either 4 h or 8 h. Then, cells were washed in phosphate-buffered saline (PBS) and incubated with 5 mg/L of DHR 123 for 1 h.Figure 2. SEM images of ESBL-EC treated with sub-MICs of EGCG, cefotaxime or their combinations. Cells were: treated with 250 mg/L of EGCG for 4 h (A) and 8 h (B); treated with 4 mg/L of cefotaxime for 4 h (C) and 8 h (D); and treated with 250 mg/L of EGCG and 4 mg/L of cefotaxime in combination for 4 h (E) and 8 h (F). Scale bar: 1 mm. doi:10.1371/journal.pone.0048880.gAFM Study o.