The interaction of Ag2O nanoparticles with Escherichia coli: inhibition–sterilization process

Silver-based antibacterial agents have obtained wide attention due to the fact that bacteria in the environment is ubiquitous, which has become one of the most difficult problems for human health. However, the antibacterial mechanism and process are still inconclusive. Here, Ag2O nanoparticles (NPs) with uniform spherical morphology and small size (around 30 nm) were prepared. The as-prepared Ag2O NPs induced high antibacterial activity (100% inhibition ratio) against E. coli. A two-step antibacterial process was proposed and confirmed, which divided into inhibition and sterilization steps. The optical density measurement, malondialdehyde concentration detection, morphologic imaging with electronic microscopy and Fourier transform infrared spectroscopic analysis unveiled the interaction of Ag2O NPs with E. coli, which verified the inhibition–sterilization process we proposed.

www.nature.com/scientificreports/ The powder X-ray diffraction (XRD, Panalytical) with Cu Kα radiation was used to obtain composition and phase information. Morphologies were characterized by scanning electron microscopy (SEM, FEI Verios 460). Thermostability was tested by thermogravimetric analyzer (TGA, Q500) under nitrogen atmosphere with heating rate of 10 °C/min. High resolution transmission electron microscopy (HRTEM, JEM 2100F) with an acceleration voltage of 200 kV was carried out to obtain the morphology and crystal facet information, and Fourier transform infrared spectrometer (FTIR, Bruker) was used to record the structural change before and after antibacterial treatment. Biological samples were freeze-dried for 20 h before test.
Antibacterial activity test. Paper-disk diffusion method was used to determine the antibacterial activity of Ag 2 O NPs qualitatively, and to compare with two traditional antimicrobial agents. Gram-negative bacteria E. coli and gram-positive bacteria S. aureus were used as model bacteria for antibacterial test. After twice purification, the bacterial cells were cultured in a broth at a constant temperature of 37 °C to achieve to logarithmic growth phase, and the bacterial solution was dropped to the solid medium by a plate coating method. Filter papers (d = 1.5 cm) were immersed in a 0.01 wt% Ag 2 O NPs suspension, 0.01 vol% H 2 O 2 solution, 0.01 vol% absolute ethanol for 2 min, respectively, taken out and then air-dried. Afterwards, filter papers were placed as an equilateral triangle on the same solid medium coated with a bacterial suspension. Subsequently, after incubated at 37 °C for 24 h, the appearance of inhibition zones was observed.
To quantitatively evaluate the antibacterial activity of Ag 2 O NPs, gram-negative bacteria E. coli was used as the model bacteria. The bacterial cells were cultured in broth at 37 °C for 6 h, then diluted with (phosphate buffer solution) PBS of pH 7.4 until optical density value at 600 nm (OD 600 ) = 1.40-1.60, measured by Bio Tek (synergy/ H1), and then Ag 2 O NPs were added. In sequence, culture medium was taken every 2 h to test OD 600 in 96-well plate, and the growth curve was plotted within 12 h. The first value of OD 600 is treated as zero to correct each curve that begins from the same point. The inhibition ratio can be calculated according to the equation as follows: The minimum inhibitory concentration (MIC) and the minimum antibacterial concentration (MBC) are the important indicators for measuring the antibacterial activity of a bactericidal agent. E. coli was cultured to a logarithmic growth phase firstly, and then Ag 2 O NPs was added in a gradient. (In this experiment, Ag 2 O NPs prepared at 60 °C/10 min were added by 5 μg mL −1 , 10 μg mL −1 , 20 μg mL −1 , 30 μg mL −1 , 40 μg mL −1 and 50 μg mL −1 respectively). After cultured at 37 °C for 24 h, each MIC was determined by visually locating the cells with no bacterial growth and the lowest concentration of the sample solution. By drawing the supernatant of the clarified samples in MIC experiment in solid medium, culturing at 37 °C for 24 h again, the minimum concentration at which no colony occurred or their number less than 10 was obtained as the MBC.
Before experiments, all required supplies (e.g. culture dishes, media, PBS solution, pipette tips, graduated cylinders, deionized water, etc.) were sterilized in autoclave. All above operations were carried out in clean bench, and all experiments were carried out with three parallel tests to eliminated possible error.
Antibacterial process study. To in-depth investigate the bactericidal process of Ag 2 O NPs, four tests were taken: (1) environmental scanning electron microscope (E-SEM, FEI Q45) was applied to observe the morphological changes of bacteria; (2) malondialdehyde (MDA) will be generated inside the bacteria cells once membrane is destroyed. MDA kit (Nanjing Jiancheng Bioengineering Institute) was used to measure the content change; (3) XRD planes information and HRTEM lattice fringes information were used to represent how Ag 2 O NPs natural properties influenced antibacterial activity; and (4) FTIR was used to reflect the structural changes of E. coli cell and Ag 2 O NPs. After antimicrobial process, Ag 2 O NPs were collected from broth, washed three times by deionized water, and then dried in vacuum oven at 60 °C for 6 h. The cells as collected after sterilization were freeze-dried at least 20 h before FTIR test.

Results and discussion
Structure and morphology of Ag 2 O NPs. Synthesis temperature was adjusted carefully to control the crystal structure. The crystalline characters of Ag 2 O NPs were influenced by the reaction temperatures, which can be inspected from XRD patterns as shown in Fig. 1. Overall, five distinct diffraction peaks of Ag 2 O NPs appear at 2θ = 32.98°, 38.24°, 55.24°, 65.76° and 69.09° in four XRD patterns. The upper two XRD patterns show that Ag 2 O NPs have good crystallinity at 25 °C and 60 °C due to no obviously miscellaneous peaks appearing. Although the position of characteristic diffraction peaks are not obviously changing in 80 °C and 100 °C samples, slight miscellaneous peaks near the main diffraction peaks appear, especially around the (111), (200) and (220), which indicates undesirable microscopic defects therein. The grain sizes can be derived from XRD patterns, which are 29.7 nm (25 °C), 36.4 nm (60 °C), 37.2 nm (80 °C), 38.8 nm (100 °C), respectively. With the gradual increase of the temperature, the grain size increases slightly. However, the aggregation is more likely to occur due to too small size as shown in Fig. 2, which may have negative impact on the activity. Moreover, when the Ag 2 O NPs preparation temperature was increased to 60 °C, higher intense (200) diffraction of Ag 2 O NPs was observed than that of the sample synthesized at 25 °C. The (200) diffraction peak should be ascribed to high orientation of {100} crystal facet of Ag 2 O NPs 28 . From the previous work 28 , it has been proved the {100} crystal facet of Ag 2 O NPs is more active, which may makes better performance.
Preparation temperature also influences the morphology of Ag 2 O NPs as reflected by the SEM images shown in Fig. 2. With the synthesis times and temperatures changing, the morphology and dispersibility of the Ag 2 O NPs were changed gradually. It can be seen from the Fig. 2a www.nature.com/scientificreports/ and clear spherical shape. In contrast, although Ag 2 O NPs (60 °C/10 min) also have uniform spherical shape, most of them flock together (Fig. 2b), which may arise from their much smaller particle sizes and higher surface activity. When the temperatures continue rising, Ag 2 O NPs (80 °C/10 min) show much rough crystal surfaces with non-uniform sizes (Fig. 2c), while much severely for Ag 2 O NPs (100 °C/10 min) (Fig. 2d). From the result of the size distribution of Ag 2 O NPs (Fig. S1), it can be known the average size of each sample, which are 29.80 nm (25 °C/10 min), 37.60 nm (60 °C/10 min), 38.03 nm (80 °C/10 min), 39.28 nm (100 °C/10 min), respectively. The results above are nearly consistent with the results from XRD patterns in Fig. 1. Furthermore, the uniform morphology of Ag 2 O NPs (60 °C/10 min) and spherical crystal shape with quite small size were observed by HRTEM images in Fig. 3. The smaller particle size is usually more beneficial to the antibacterial reaction. Because the particle size is smaller, much more particles will be easily adsorbed on the surface of the bacterial cell membrane, and then successfully attack the cell to destroy the physiological function in the cell. It may be a key reason to explain that Ag 2 O NPs (prepared at 60 °C/10 min) have the best antibacterial performance among all as prepared Ag 2 O NPs.
The thermogravimetric analysis of Ag 2 O prepared at 60 °C/10 min is shown in Fig. S2. According to the outermost electronic arrangement of Ag, it is the most stable when the outermost electronic arrangement is 4d 10 , which is the reason why the order of stability of silver oxides is Ag 2 O > Ag 2 O 2 . At 438.85 °C, as shown in   Paper-disk diffusion test. As shown in Fig. S3, under the same concentration, among three different antibacterial agents, Ag 2 O NPs, ethanol and H 2 O 2, Ag 2 O NPs resulted in the most obvious and well-proportioned inhibition zone on the solid medium of E. coli (Fig. S3a). The diameter of inhibition zone were 1.91 cm (E. coli) and 1.96 cm (S. aureus), respectively. In contrast, no clear inhibition zone appeared when using ethanol and H 2 O 2 soaked filter paper. Such better antibacterial action of Ag 2 O NPs than that of other two traditional disinfectants was also observed in S. aureus group (Fig. S3b) (Table S1, Figs. S4, S5), strongly evidencing the advantage of the quite low using dose.
Quantitative antibacterial activity test. The inhibition growth curves were used to study the dynamics of bacterial growth, and evaluated the antibacterial properties of Ag 2 O NPs with different preparation conditions. Overall, as-prepared Ag 2 O NPs showed noteworthy antibacterial activities in Fig. 4a, which shows the growth curves of the culture medium of E. coli co-cultured with different Ag 2 O NPs (60 °C , 80 °C, 100 °C, 120 °C, 140 °C/10 min) at 37 °C for varied time intervals (up to 12 h). Particularly, considering the balance of efficiency and economic, the best antibacterial activity was observed at 60 °C/10 min group with almost 100% antibacterial ratio was obtained after only 2 h (Fig. 4b,c). The higher antibacterial efficiency is likely owing to the smaller particles and uniform morphology, which is consistent with the results of XRD analysis (Fig. 1), SEM images (Fig. 2) and HRTEM measurements (Fig. 3). Some quantitative results of antibacterial activity are provided in supplementary information (Figs. S6-S9).
To understand why Ag 2 O NPs (60 °C/10 min) have higher activity, their crystalline features and interaction with cells were studied. It can be known that the {100} facet is much more exposed in Ag 2 O NPs (60 °C/10 min) than that in other Ag 2 O NPs as inferred XRD patterns (Fig. 1). As shown in Fig. 6a, Ag 2 O NPs (60 °C/10 min) have {100} facet exposure via analyzing crystal lattice, and it can be clearly seen that a single Ag 2 O NP presents a uniform spherical shape. In addition, the results of selected area electron diffraction (SAED) pattern are shown in Fig. 5b, presenting six obvious diffraction spots connected in a circle and are symmetrical, which are corresponding to the presence of {100}, {111}, {110} plane axis, as well as the XRD results. In the present work, due to the different atomic arrangement, Ag 2 O NPs have several crystal planes, which may cause differences in the release rate and concentration of Ag + , further influencing their antibacterial performances of antibacterial. According to the previous work, the {100} facets are more reactive 28 . Moreover, the Ag + ions are more likely to adsorb on the {100} surface. Because the charged ions can be preferentially adsorbed on the cell membrane, and the high reactivity of the {100} facet makes Ag 2 O NPs more easily adsorbed on the surface of E. coli membrane www.nature.com/scientificreports/ to participate in the afterwards antibacterial reaction. From Fig. 5a, it can be concluded that higher reactive {100} facet indeed highly exposed in the 60 °C/10 min Ag 2 O NPs sample.
Inhibition-sterilization process. An interesting phenomenon is the OD 600 value achieved around 0 at 2 h (Fig. 4a,b). In order to further studying the antibacterial process in the first 2 h, the growth curve of Ag 2 O NPs  www.nature.com/scientificreports/ (60 °C/10 min) within 2 h was conducted. As shown in Fig. 4d, the concentration of bacteria maintains the same level for the first 80 min. In the next 40 min, the antibacterial agent begins to play a significant role, and the number of the bacteria cell is damaged to the initial concentration level in the end of sterilization step. The result above reveals that the Ag 2 O NPs takes effect after at least 1 h. More importantly, it can be concluded that the whole antimicrobial process may consist of two steps, the first one is inhibition procedure, which means, the reproduced capacity of bacteria was destroyed by Ag 2 O NPs, thus the number of cells no longer increases. Following this step, Ag 2 O NPs sterilized the rest level of bacteria, which is the sterilization step. The morphological changes of E. coli cells can be represented via E-SEM images in Fig. 6a-c. At the original state, the cells had the full membrane with smooth surface, and showed good dispersity (Fig. 6a). However, after co-culturing with Ag 2 O NPs (60 °C/10 min) for 12 h, it can be evidently seen that the membranes of all cells became indistinct and messy, and severe adhesion occurred among them (Fig. 6b). The destruction of cell membrane belongs to the "inhibition" part, which is the initial antibacterial action of Ag 2 O NPs against E. coli cells. And then the intercellular dissolved matters flowed out gradually such as proteins, sugars, and so on. Compared to the average size of 2 µm of E. coli cells, the size of Ag 2 O NPs is quite smaller, which makes more react sites on the membrane of E. coli 13 . From Fig. 6d, it can be clearly seen that some of the Ag 2 O NPs attached to the surface of E. coli cell after the reaction, and the others entered the inside of the bacterial cell. After the reaction, the Ag 2 O NPs showed agglomeration, and the bacterial cell membrane was obviously incomplete. In a word, the antibacterial agent destroyed the membrane, which was seen as the "protective shield" of intact cells, making the bacteria lose the basic living viability, then leading to the ultimate death of the cells. This result is consistent with the two-step inhibition-sterilization antibacterial process as we proposed above.
In order to help analyze the changes of the groups of E. coli before and after the bactericidal reaction, the FTIR measurement of E. coli before and after the reaction were conducted and the results were shown in Fig. S10. The infrared spectrum showed that phospholipid bilayer, which is the important component of cell membrane, protein and DNA were destroyed to some extent, suggesting that Ag 2 O NPs attacked these parts of E. coli, leading to cell apoptosis in the end. However, the changes of the stretching vibration summit of the FTIR appear superposition. For example, the stretching vibration peak of MDA is around 1725 cm −1 , which coincides with the area of the phosphate diester group, thus the changes cannot be accurately observed. Therefore, infrared spectra can only be used for reference.
Overall, the inhibition-sterilization antimicrobial process was explained from the results above, which is showed in Scheme 1b,c. It can be inferred that when Ag 2 O NPs began to contact with bacterial cells, the destruction of DNA and proteins may not effectively weaken the reproductive capacity of the bacteria. The number of damaged cells is equal to the number of reproduced cells, which is defined as the antibacterial inhibition step. However, after a period of continuous contact, the reproductive capacity of E. coli drops dramatically, even the newly reproduced bacterial cells begin to be destroyed, thereby the growth curve shows a sharp decline trend. This stage is sterilization step.
When the cell membrane subjected to oxidative stress, MDA will produce spontaneously from the cell. Therefore, MDA is regarded as a symbol of the destruction of cell membrane. In order to further proof that the Ag 2 O NPs damaged the cell membrane and then entered into the cell to interact with the intracellular substance, the MDA concentrations in the cell were evaluated 32,33 . As shown in Fig. 7, in the sterilization group, after exposing to Ag 2 O NPs (60 °C/10 min), the concentration of MDA in E. coli cells (0.00926 nmol/mg prot) is around 3 times higher than the initial concentration (around 0.003 nmol/mg prot), which represents the MDA concentration before adding the antibacterial agent. Therefore, it is a strong evidence to prove the destruction of cell membrane from the MDA concentration change, which is consistent with the results of E-SEM and FTIR analyses as well as the inhibition-sterilization process we proposed above.

Conclusion
To sum up, a series of Ag 2 O NPs with small size, uniform morphology and excellent thermostability were synthesized. The synthesis reaction temperature and reaction time critical influence on the crystalline and morphological properties of Ag 2 O NPs and thus impact their antibacterial performances. A high antibacterial ratio of 100% can be reached within 40 min after an 80 min inhibition stage. More importantly, the inhibition-sterilization  www.nature.com/scientificreports/