Thank you for visiting Nature.com. The browser version you are using has limited CSS support. For the best experience, we recommend that you use an updated browser (or disable Compatibility Mode in Internet Explorer). In the meantime, to ensure continued support, we will render the site without styles and JavaScript.
A carousel showing three slides at the same time. Use the Previous and Next buttons to move through three slides at a time, or use the slider buttons at the end to move through three slides at a time.
Recently, a chemical-free antimicrobial platform based on nanotechnology using artificial water nanostructures (EWNS) has been developed. EWNS have a high surface charge and are saturated with reactive oxygen species (ROS) that can interact with and inactivate a number of microorganisms, including foodborne pathogens. Here it is shown that their properties during synthesis can be fine-tuned and optimized to further enhance their antibacterial potential. The EWNS laboratory platform was designed to fine-tune the properties of EWNS by changing the synthesis parameters. Characterization of EWNS properties (charge, size and content of ROS) using modern analytical methods. In addition, they were evaluated for their microbial inactivation potential against foodborne microorganisms such as Escherichia coli, Salmonella enterica, Listeria innocuous, Mycobacterium paraaccidentum and Saccharomyces cerevisiae. The results presented here demonstrate that the properties of EWNS can be fine-tuned during synthesis, resulting in an exponential increase in inactivation efficiency. In particular, the surface charge increased by a factor of four and the reactive oxygen species increased. The microbial removal rate was microbially dependent and ranged from 1.0 to 3.8 log after a 45 minute exposure to an aerosol dose of 40,000 #/cc EWNS.
Microbial contamination is the main cause of foodborne illness caused by the ingestion of pathogens or their toxins. In the United States alone, foodborne illness causes about 76 million illnesses, 325,000 hospital admissions, and 5,000 deaths each year1. In addition, the United States Department of Agriculture (USDA) estimates that increased consumption of fresh produce is responsible for 48% of all reported foodborne illnesses in the United States2. The cost of disease and death caused by foodborne pathogens in the United States is very high, estimated by the Centers for Disease Control and Prevention (CDC) at more than US$15.6 billion per year3.
Currently, chemical4, radiation5 and thermal6 antimicrobial interventions to ensure food safety are mostly carried out at limited critical control points (CCPs) along the production chain (usually after harvest and/or during packaging) rather than continuously. thus, they are prone to cross-contamination. 7. Better control of foodborne illness and food spoilage requires antimicrobial interventions that can potentially be applied across the farm-to-table continuum while reducing environmental impact and costs.
Recently, a chemical-free, nanotechnology-based antimicrobial platform has been developed that can inactivate surface and airborne bacteria using artificial water nanostructures (EWNS). EWNS was synthesized using two parallel processes, electrospray and water ionization (Fig. 1a). Previous studies have shown that EWNS have a unique set of physical and biological properties8,9,10. EWNS have an average of 10 electrons per structure and an average nanoscale size of 25 nm (Fig. 1b,c)8,9,10. In addition, electron spin resonance (ESR) showed that EWNS contains a large amount of reactive oxygen species (ROS), mainly hydroxyl (OH•) and superoxide (O2-) radicals (Fig. 1c)8. EVNS is in the air for a long time and can collide with microorganisms suspended in the air and present on the surface, delivering their ROS payload and causing inactivation of microorganisms (Fig. 1d). These early studies also showed that EWNS can interact with and inactivate various gram-negative and gram-positive bacteria, including mycobacteria, on surfaces and in the air. Transmission electron microscopy showed that the inactivation was caused by disruption of the cell membrane. In addition, acute inhalation studies have shown that high doses of EWNS do not cause lung damage or inflammation 8 .
(a) Electrospray occurs when a high voltage is applied between a capillary tube containing liquid and a counter electrode. (b) The application of high pressure results in two different phenomena: (i) electrospraying of water and (ii) formation of reactive oxygen species (ions) trapped in the EWNS. (c) The unique structure of EWNS. (d) Due to their nanoscale nature, EWNS are highly mobile and can interact with airborne pathogens.
The ability of the EWNS antimicrobial platform to inactivate foodborne microorganisms on the surface of fresh food has also recently been demonstrated. It has also been shown that the surface charge of EWNS in combination with an electric field can be used to achieve targeted delivery. Moreover, preliminary results for organic tomatoes after a 90 minute exposure at an EWNS of about 50,000 #/cm3 were encouraging, with various foodborne microorganisms such as E. coli and Listeria 11 observed. In addition, preliminary organoleptic tests showed no sensory effects compared to control tomatoes. Although these initial inactivation results are encouraging for food safety applications even at very low EWNS doses of 50,000#/cc. see, it is clear that a higher inactivation potential would be more beneficial to further reduce the risk of infection and spoilage.
Here, we will focus our research on the development of an EWNS generation platform to enable fine tuning of synthesis parameters and optimization of the physicochemical properties of EWNS to enhance their antibacterial potential. In particular, optimization has focused on increasing their surface charge (to improve targeted delivery) and ROS content (to improve inactivation efficiency). Characterize optimized physico-chemical properties (size, charge and ROS content) using modern analytical methods and use common food microorganisms such as E. .
EVNS was synthesized by simultaneous electrospraying and ionization of high purity water (18 MΩ cm–1). The electric nebulizer 12 is typically used for the atomization of liquids and the synthesis of polymer and ceramic particles 13 and fibers 14 of controlled size.
As detailed in previous publications 8, 9, 10, 11, in a typical experiment, a high voltage was applied between a metal capillary and a grounded counter electrode. During this process, two different phenomena occur: i) electrospray and ii) water ionization. A strong electric field between the two electrodes causes negative charges to build up on the surface of the condensed water, resulting in the formation of Taylor cones. As a result, highly charged water droplets are formed, which continue to break up into smaller particles, as in the Rayleigh theory16. At the same time, strong electric fields cause some water molecules to split and strip off electrons (ionize), which leads to the formation of a large amount of reactive oxygen species (ROS)17. Simultaneously generated ROS18 was encapsulated in EWNS (Fig. 1c).
On fig. 2a shows the EWNS generation system developed and used in the EWNS synthesis in this study. Purified water stored in a closed bottle was fed through a Teflon tube (2 mm inner diameter) into a 30G stainless steel needle (metal capillary). The flow of water is controlled by the air pressure inside the bottle, as shown in Figure 2b. The needle is mounted on a Teflon console and can be manually adjusted to a certain distance from the counter electrode. The counter electrode is a polished aluminum disk with a hole in the center for sampling. Below the counter electrode is an aluminum sampling funnel, which is connected to the rest of the experimental setup via a sampling port (Fig. 2b). To avoid charge build-up that could disrupt sampler operation, all sampler components are electrically grounded.
(a) Engineered Water Nanostructure Generation System (EWNS). (b) Cross-section of the sampler and electrospray, showing the most important parameters. (c) Experimental setup for bacteria inactivation.
The EWNS generation system described above is capable of changing key operating parameters to facilitate fine tuning of the EWNS properties. Adjust the applied voltage (V), the distance between the needle and the counter electrode (L), and the water flow (φ) through the capillary to fine-tune the EWNS characteristics. Symbol used to represent different combinations: [V (kV), L (cm)]. Adjust the water flow to get a stable Taylor cone of a certain set [V, L]. For the purposes of this study, the aperture diameter of the counter electrode (D) was kept at 0.5 inches (1.29 cm).
Due to the limited geometry and asymmetry, the electric field strength cannot be calculated from first principles. Instead, the QuickField™ software (Svendborg, Denmark)19 was used to calculate the electric field. The electric field is not uniform, so the value of the electric field at the tip of the capillary was used as a reference value for various configurations.
During the study, several combinations of voltage and distance between the needle and the counter electrode were evaluated in terms of Taylor cone formation, Taylor cone stability, EWNS production stability, and reproducibility. Various combinations are shown in Supplementary Table S1.
The output of the EWNS generation system was connected directly to a Scanning Mobility Particle Size Analyzer (SMPS, Model 3936, TSI, Shoreview, MN) for particle number concentration measurement, as well as to an Aerosol Faraday Electrometer (TSI, Model 3068B, Shoreview, MN). ) for aerosol currents was measured as described in our previous publication. Both the SMPS and the aerosol electrometer sampled at a flow rate of 0.5 L/min (total sample flow 1 L/min). The number concentration of particles and the aerosol flow were measured for 120 seconds. The measurement is repeated 30 times. Based on current measurements, the total aerosol charge is calculated and the average EWNS charge is estimated for a given total number of selected EWNS particles. The average cost of EWNS can be calculated using Equation (1):
where IEl is the measured current, NSMPS is the digital concentration measured with the SMPS, and φEl is the flow rate per electrometer.
Because relative humidity (RH) affects surface charge, temperature and (RH) were kept constant during the experiment at 21°C and 45%, respectively.
Atomic force microscopy (AFM), Asylum MFP-3D (Asylum Research, Santa Barbara, CA) and AC260T probe (Olympus, Tokyo, Japan) were used to measure the size and lifetime of the EWNS. The AFM scanning frequency was 1 Hz, the scanning area was 5 μm × 5 μm, and 256 scan lines. All images were subjected to 1st order image alignment using Asylum software (mask range 100 nm, threshold 100 pm).
The test funnel was removed and the mica surface was placed at a distance of 2.0 cm from the counter electrode for an averaging time of 120 s to avoid particle agglomeration and formation of irregular droplets on the mica surface. EWNS was sprayed directly onto the surface of freshly cut mica (Ted Pella, Redding, CA). Image of the mica surface immediately after AFM sputtering. The contact angle of the surface of freshly cut unmodified mica is close to 0°, so EVNS is distributed on the mica surface in the form of a dome. The diameter (a) and height (h) of the diffusing droplets were measured directly from the AFM topography and used to calculate the EWNS domed diffusion volume using our previously validated method. Assuming the onboard EWNS have the same volume, the equivalent diameter can be calculated using Equation (2):
Based on our previously developed method, an electron spin resonance (ESR) spin trap was used to detect the presence of short-lived radical intermediates in EWNS. Aerosols were bubbled through a 650 μm Midget sparger (Ace Glass, Vineland, NJ) containing a 235 mM solution of DEPMPO(5-(diethoxyphosphoryl)-5-methyl-1-pyrroline-N-oxide) (Oxis International Inc.). Portland, Oregon). All ESR measurements were performed using a Bruker EMX spectrometer (Bruker Instruments Inc. Billerica, MA, USA) and a flat panel cell. The Acquisit software (Bruker Instruments Inc. Billerica, MA, USA) was used to collect and analyze the data. Determination of the characteristics of the ROS was carried out only for a set of operating conditions [-6.5 kV, 4.0 cm]. EWNS concentrations were measured using the SMPS after accounting for EWNS losses in the impactor.
Ozone levels were monitored using a 205 Dual Beam Ozone Monitor™ (2B Technologies, Boulder, Co)8,9,10.
For all EWNS properties, the mean value is used as the measurement value, and the standard deviation is used as the measurement error. T-tests were performed to compare the values of the optimized EWNS attributes with the corresponding values of the base EWNS.
Figure 2c shows a previously developed and characterized electrostatic precipitation (EPES) “pull” system that can be used for targeted delivery of EWNS at the surface. EPES uses EVNS charges that can be “guided” directly to the surface of the target under the influence of a strong electric field. Details of the EPES system are presented in a recent publication by Pyrgiotakis et al. 11 . Thus, EPES consists of a 3D printed PVC chamber with tapered ends and contains two parallel stainless steel (304 stainless steel, mirror coated) metal plates at the center 15.24 cm apart. The boards were connected to an external high voltage source (Bertran 205B-10R, Spellman, Hauppauge, NY), the bottom plate was always connected to positive voltage, and the top plate was always connected to ground (floating ground). The chamber walls are covered with aluminum foil, which is electrically grounded to prevent particle loss. The chamber has a sealed front loading door that allows test surfaces to be placed on plastic stands that raise them above the bottom metal plate to avoid high voltage interference.
The deposition efficiency of EWNS in EPES was calculated according to a previously developed protocol detailed in Supplementary Figure S111.
As a control chamber, a second cylindrical flow chamber was connected in series to the EPES system, in which an intermediate HEPA filter was used to remove EWNS. As shown in Figure 2c, the EWNS aerosol was pumped through two built-in chambers. The filter between the control room and EPES removes any remaining EWNS resulting in the same temperature (T), relative humidity (RH) and ozone levels.
Important foodborne microorganisms have been found to contaminate fresh foods such as E. coli (ATCC #27325), fecal indicator, Salmonella enterica (ATCC #53647), foodborne pathogen, Listeria harmless (ATCC #33090), surrogate for pathogenic Listeria monocytogenes, derived from ATCC (Manassas, VA) Saccharomyces cerevisiae (ATCC #4098), a substitute for spoilage yeast, and a more resistant inactivated bacterium, Mycobacterium paralucky (ATCC #19686).
Buy random boxes of organic grape tomatoes from your local market and refrigerate at 4°C until use (up to 3 days). The experimental tomatoes were all the same size, about 1/2 inch in diameter.
The culture, inoculation, exposure, and colony count protocols are detailed in our previous publication and detailed in the Supplementary Data. The effectiveness of EWNS was evaluated by exposing inoculated tomatoes to 40,000 #/cm3 for 45 minutes. Briefly, three tomatoes were used to evaluate the surviving microorganisms at time t = 0 min. Three tomatoes were placed in EPES and exposed to EWNS at 40,000 #/cc (EWNS exposed tomatoes) and the remaining three were placed in the control chamber (control tomatoes). Additional processing of tomatoes in both groups was not carried out. EWNS-exposed tomatoes and control tomatoes were removed after 45 minutes to evaluate the effect of EWNS.
Each experiment was carried out in triplicate. Data analysis was performed according to the protocol described in Supplementary Data.
Inactivation mechanisms were assessed by sedimentation of exposed EWNS samples (45 min at 40,000 #/cm3 EWNS aerosol concentration) and non-irradiated samples of harmless bacteria E. coli, Salmonella enterica and Lactobacillus. The particles were fixed in 2.5% glutaraldehyde, 1.25% paraformaldehyde and 0.03% picric acid in 0.1 M sodium cacodylate buffer (pH 7.4) for 2 hours at room temperature. After washing, post-fix with 1% osmium tetroxide (OsO4)/1.5% potassium ferrocyanide (KFeCN6) for 2 hours, wash 3 times in water and incubate in 1% uranyl acetate for 1 hour, then wash twice in water, then dehydrate in for 10 minutes in 50%, 70%, 90%, 100% alcohol. The samples were then placed in propylene oxide for 1 hour and impregnated with a 1:1 mixture of propylene oxide and TAAP Epon (Marivac Canada Inc. St. Laurent, CA). The samples were embedded in TAAB Epon and polymerized at 60°C for 48 hours. The cured granular resin was cut and visualized by TEM using a conventional transmission electron microscope JEOL 1200EX (JEOL, Tokyo, Japan) equipped with an AMT 2k CCD camera (Advanced Microscopy Techniques, Corp., Woburn, Massachusetts, USA).
All experiments were carried out in triplicate. For each time point, bacterial washes were seeded in triplicate, resulting in a total of nine data points per point, the average of which was used as the bacterial concentration for that particular microorganism. The standard deviation was used as the measurement error. All points count.
The logarithm of the decrease in the concentration of bacteria compared to t = 0 min was calculated using the following formula:
where C0 is the concentration of bacteria in the control sample at time 0 (i.e. after the surface has dried but before being placed in the chamber) and Cn is the concentration of bacteria on the surface after n minutes of exposure.
To account for the natural degradation of bacteria during the 45-minute exposure, the log reduction compared to the control after 45 minutes was also calculated as follows:
where Cn is the concentration of bacteria in the control sample at time n and Cn-Control is the concentration of control bacteria at time n. Data are presented as a log reduction compared to control (no EWNS exposure).
During the study, several combinations of voltage and distance between the needle and the counter electrode were evaluated in terms of Taylor cone formation, Taylor cone stability, EWNS production stability, and reproducibility. Various combinations are shown in Supplementary Table S1. Two cases showing stable and reproducible properties (Taylor cone, EWNS generation, and stability over time) were selected for comprehensive study. On fig. Figure 3 shows the results for the charge, size, and content of ROS in both cases. The results are also shown in Table 1. For reference, both Figure 3 and Table 1 include the properties of the previously synthesized non-optimized EWNS8, 9, 10, 11 (baseline-EWNS). Statistical significance calculations using a two-tailed t-test are republished in Supplementary Table S2. In addition, additional data include studies of the effect of counter electrode sampling hole diameter (D) and distance between ground electrode and tip (L) (Supplementary Figures S2 and S3).
(ac) Size distribution measured by AFM. (df) Surface charge characteristic. (g) ROS characterization of the EPR.
It is also important to note that for all of the above conditions, the measured ionization current was between 2 and 6 μA and voltage between -3.8 and -6.5 kV, resulting in a power consumption of less than 50 mW for this single EWNS generation contact module. Although EWNS was synthesized under high pressure, ozone levels were very low, never exceeding 60 ppb.
Supplementary Figure S4 shows the simulated electric fields for the [-6.5 kV, 4.0 cm] and [-3.8 kV, 0.5 cm] scenarios, respectively. For the [-6.5 kV, 4.0 cm] and [-3.8 kV, 0.5 cm] scenarios, the field calculations are 2 × 105 V/m and 4.7 × 105 V/m, respectively. This is expected, since in the second case the voltage-distance ratio is much higher.
On fig. 3a,b shows the EWNS diameter measured with the AFM8. The calculated average EWNS diameters were 27 nm and 19 nm for the [-6.5 kV, 4.0 cm] and [-3.8 kV, 0.5 cm] schemes, respectively. For the [-6.5 kV, 4.0 cm] and [-3.8 kV, 0.5 cm] scenarios, the geometric standard deviations of the distributions are 1.41 and 1.45, respectively, indicating a narrow size distribution. Both the mean size and the geometric standard deviation are very close to the baseline EWNS, at 25 nm and 1.41, respectively. On fig. 3c shows the size distribution of the base EWNS measured using the same method under the same conditions.
On fig. 3d,e shows the results of charge characterization. Data are average measurements of 30 simultaneous measurements of concentration (#/cm3) and current (I). The analysis shows that the average charge on the EWNS is 22 ± 6 e- and 44 ± 6 e- for [-6.5 kV, 4.0 cm] and [-3.8 kV, 0.5 cm], respectively. They have significantly higher surface charges compared to baseline EWNS (10 ± 2 e-), two times greater than the [-6.5 kV, 4.0 cm] scenario and four times greater than the [-3 .8 kV, 0.5 cm]. Figure 3f shows the charge. data for Baseline-EWNS.
From the concentration maps of the EWNS number (Supplementary Figures S5 and S6), it can be seen that the [-6.5 kV, 4.0 cm] scenario has significantly more particles than the [-3.8 kV, 0.5 cm] scenario. It is also worth noting that the EWNS number concentration was monitored up to 4 hours (Supplementary Figures S5 and S6), where the EWNS generation stability showed the same levels of particle number concentration in both cases.
On fig. 3g shows the EPR spectrum after subtraction of the optimized EWNS control (background) at [-6.5 kV, 4.0 cm]. The ROS spectra were also compared with the Baseline-EWNS scenario in a previously published work. The number of EWNS reacting with spin traps was calculated to be 7.5 × 104 EWNS/s, which is similar to the previously published Baseline-EWNS8. The EPR spectra clearly showed the presence of two types of ROS, with O2- being the predominant species and OH• being less abundant. In addition, a direct comparison of the peak intensities showed that the optimized EWNS had a significantly higher ROS content compared to the baseline EWNS.
On fig. 4 shows the deposition efficiency of EWNS in EPES. The data are also summarized in Table I and compared with the original EWNS data. For both cases of EUNS, the deposition is close to 100% even at a low voltage of 3.0 kV. Typically, 3.0 kV is sufficient for 100% deposition, regardless of surface charge change. Under the same conditions, the deposition efficiency of Baseline-EWNS was only 56% due to their lower charge (average 10 electrons per EWNS).
On fig. 5 and in table. 2 summarizes the inactivation value of microorganisms inoculated on the surface of tomatoes after exposure to about 40,000 #/cm3 EWNS for 45 minutes at the optimum mode [-6.5 kV, 4.0 cm]. Inoculated E. coli and Lactobacillus innocuous showed a significant reduction of 3.8 logs during the 45 minute exposure. Under the same conditions, S. enterica had a 2.2-log decrease, while S. cerevisiae and M. parafortutum had a 1.0-log decrease.
The electron micrographs (Figure 6) depict the physical changes induced by EWNS on harmless Escherichia coli, Streptococcus, and Lactobacillus cells leading to their inactivation. The control bacteria had intact cell membranes, while the exposed bacteria had damaged outer membranes.
Electron microscopic imaging of control and exposed bacteria revealed membrane damage.
The data on the physicochemical properties of the optimized EWNS collectively show that the properties (surface charge and ROS content) of the EWNS were significantly improved compared to the previously published EWNS baseline data8,9,10,11. On the other hand, their size remained in the nanometer range, very similar to the results previously reported, allowing them to remain in the air for long periods of time. The observed polydispersity can be explained by surface charge changes that determine the size of EWNS, the randomness of the Rayleigh effect, and potential coalescence. However, as detailed by Nielsen et al. 22, high surface charge reduces evaporation by effectively increasing the surface energy/tension of the water drop. In our previous publication8 this theory was experimentally confirmed for microdroplets 22 and EWNS. Loss of charge during overtime can also affect the size and contribute to the observed size distribution.