Bacterial strains and cell lines
The bacterial species used in this study were Staphylococcus aureus, Staphylococcus epidermidis, Streptococcus pyogenes, Streptococcus pneumoniae, Enterococcus faecalis, Bacillus globisporus, Escherichia coli, Serratia marcescens, Haemophilus influenzae, Klebsiella pneumoniae, Neisseria gonorrhoeae and Pseudomonas aeruginosa (clinical isolates obtained from the Hospital Universitario Central de Asturias) and silver resistant strains Salmonella enterica serovar typhimurium 389/97 (LSP 389/97) [24] and Salmonella enterica serovar typhimurium 207/07 (LSP 207/07) (obtained from the laboratory of Dr. Rosario Rodicio at the University of Oviedo).
All the bacteria were grown in Brain Heart Infusion (BHI) broth (Gibco-Thermo Fischer Scientific, Waltham, MA, USA) except H. Influenzae, which was grown in BHI including Haemophilus Test Medium Supplement (Gibco). All the bacteria were grown at 37ºC in a shaking incubator, except S. pneumoniae, N. gonorrhoeae and H. influenzae, which were grown in a 5% (v/v) CO2 atmosphere without shaking.
Synthesis of silver nanostructures
The synthesis of AgNPs was carried out using a modified polyol-mediated process [22, 23]. This involved the reduction of AgNO3 (Sigma-Aldrich, St. Louis, USA) in ethylene glycol (EG) (Panreac, Barcelona, Spain) in the presence of polyvinylpyrrolidone 360 k (PVP) (Sigma-Aldrich). In the first step, 0.27 mg of CuCl2:2H2O (Panreac) was added to 50 ml of EG followed by the simultaneous addition of a solution of AgNO3 (240 mg of AgNO3 in 15 ml of EG) and a solution of PVP (190 mg of PVP in 15 ml of EG) using a peristaltic pump, at a rate of 1.5 ml/min for 20 min. At the end of this process the solution turned from a transparent to a pearly appearance, denoting the presence of silver nanostructures and, after 5 minutes, the agitation was stopped for another 15 minutes. The reaction was performed at 170ºC (and always below 175ºC) under magnetic stirring. Once the reaction was completed, the flask was quickly submerged in ice water until the contents reached room temperature.
Anhydrous ethanol and pure acetone were used in the separation and purification steps, and all chemical reagents and solvents were of analytical grade and used without further purification. In brief, AgNSs were separated by centrifugation (4000-1000 rpm) from the reaction medium and resuspended for purification. The remaining suspension was left to decant for 3 days, after which a concentrated solution of AgNRs, together with a small amount of AgNWs, were obtained from the supernatant of this suspension, while the remaining suspension provided the majority of the AgNWs. The AgNPs were suspended in water, at a suitable concentration for the subsequent processing. The structural features of the nanostructures obtained through this process were: AgNSs, 40-60 nm in diameter; AgNRs, 80 nm wire diameter, 12-18 μm (average 14 μm) ring diameter; AgNWs, 200 nm in diameter and 50-100 µm in length. FEG-SEM Images of AgNPs, AgNWs and AgNRs are shown in Figure S1 (Additional file 1). To ensure the equivalence between differente batchs, the corresponding analyses (FEG-SEM, UV-spectrophotometry, ATG-ATD) were carried out after each synthesis.
In order to check the initial concentration of the AgNPs, a gravimetric method was used. A cotton swab was heated overnight, at 110ºC to a constant weight (in triplicate); 1 ml of suspension was added to another cold cotton swab and then heated at 110ºC for 4 hours, until the weight remained constant. Calculating the difference in weight between the two swabs enabled the content of nanostructures, measured in grams per millilitre to be obtained. The final concentrations were: 1.7 x 107 AgNSs /µl suspension; 6.0 x 104 AgNRs /µl suspension; 2.4 x 104 AgNWs /µl of suspension.
Effect of AgNPs on cell viability in planktonic bacterial cultures
To analyze the bactericidal effect of the different AgNPs, bacterial cultures grown at an A600 of 0.5 were kept at room temperature and in agitation with the nanomaterials for periods of time between 5 and 30 minutes. Next, the nanostructures were removed by centrifugation at 800 rpm for 2 minutes, and different dilutions of the supernatant were seeded on solid BHI medium plates. After incubation of the plates at 37°C overnight, the colonies obtained were quantified.
The bacterium:nanostructure ratio used as standard in the tests had been previously estimated from cultures of E. coli maintained for 30 minutes under agitation with AgNWs in the following ratios (bacteria:AgNWs): 108:1, 106:1, 104:1, 102:1, 1:1, 1: 10, 1: 102, 1: 103, 1: 104. The 1: 1 ratio was chosen as the most suitable for the assays
Effect of AgNPs on cell growth in planktonic bacterial cultures
Their influence on bacterial growth was analyzed by incubating the different AgNPs with bacterial cultures at an A600 of 0.02, and using bacteria:nanostructure proportions of 1:1, 1:2 and 1:3. The effect was quantified by determining absorbance after incubation periods of 1, 2, 3 and 4 hours.
Effect of AgNPs on bacterial biofilms
To analyze the effect of AgNPs on biofilms, bacterial cultures at an A600 of 0.01 were added to 96 well plates (100 µl/well) which were incubated for 24 hours. AgNPs of different morphologies were then added in the proportions (bacteria:AgNP) 1:1, 1:10 and 1:100. The plates were incubated again for 24 hours at 37°C. Biofilms were treated with PBS with 0.25% triton X100 for 1 minute, and different dilutions of these extracts were seeded onto solid BHI medium plates and, after incubation at 37°C overnight, the colonies obtained were quantified.
Effect of AgNO3 on E. coli viability
E. coli cultures at an A600 of 0.5 were treated with AgNO3 at concentrations of 102 µM, 1µM, 10-2 µM, 10-4 µM, 10-6 µM, 10-8 µM, 10-10 µM, 10-12 µM, 10-14 µM and for 30 minutes at 37°C with stirring. After this time, aliquots were inoculated in solid BHI medium, the plates were incubated overnight, and the colonies obtained were quantified
Determination of the concentration of silver ions released by AgNPs
Different samples of E. coli cultures in BHI at an A600 of 0.5, sterile BHI, suspensions of E. coli in deionized water at an A600 of 0.5 or sterile deionized water were individually treated with amounts of AgNSs, AgNWs and AgNRs equivalent to a bacteria:AgNP ratio of 1:1 in all cases. After incubating for 30 minutes with stirring, the suspensions were centrifuged at 3000 rpm for 5 minutes to discard the AgNPs. The concentration of Ag in the particle-free supernatants was determined by inductively coupled plasma atomic emission spectroscopy (ICP-AES) using an Agilent 5110 ICP-OES Instrument (Agilent, Santa Clara, CA, USA).
Study of the toxic effect of media conditioned by AgNPs
To analyze the effect that conditioning through the presence of AgNPs might have on bacterial viability, aliquots of deionized water or BHI were kept in contact with nanostructures at different concentrations (1010, 2 x 1010, 4 x 1010, 8 x 1010, 16 x 1010 Units/l) for a variable time (5, 10, 20, 30, 60 or 120 minutes, depending on the experiment) with stirring. Next, the AgNPs were removed by centrifugation at 3000 rpm for 5 minutes, and E. coli was incubated in the conditioned media at an A600 of 0.5 for 30 minutes with stirring. The amount of nanostructures used in the experiments was adjusted to a bacteria:AgNPs ratio of 1:1, equivalent to that used in the studies of cell viability in planktonic bacterial cultures. Finally, different dilutions of these cultures were seeded onto solid BHI medium plates and, after incubation at 37°C overnight, the colonies obtained were quantified.
In order to establish the temporary durability of the conditioning of the culture medium, aliquots of deionized water and BHI were kept in contact with AgNPs at the same concentrations described in the previous paragraph, after which the AgNPs were removed by centrifugation at 3000 rpm for 5 minutes, and the supernatant reserved at room temperature for 0, 2, 4 and 24 hours. Each sample of conditioned medium was incubated with E. coli at an A600 of 0.5 for 30 minutes with stirring, and the quantification was carried out by dilution in solid BHI medium as described above.
Statistical analysis
The results were analyzed using a Kruskal-Wallis test in the Statistics for Windows program (Statsoft Inc; Tulsa, OK). The differences were considered significant when p <0.05.