Biofilm disruption and bactericidal activity of aqueous ozone coupled with ultrasonic dental scaling

Background The COVID-19 pandemic has heightened the awareness of a common hazard encountered in the dental clinic: aerosol transmission of pathogens. Treatment of sources of infection before or during dental procedures is one means of decreasing pathogen load and aerosol transmission. Methods An ultrasonic scaler supplied with aqueous ozone was used to examine the effect of its viability on planktonic cultures and biofilms formed by 2 model bacteria: Rothia mucilaginosa and Escherichia coli. Results Both organisms showed susceptibility to aqueous ozone alone (97% and 99.5% lethality, respectively). When combined with manual scaling using an ultrasonic scaler, a greater than 99% reduction in colony-forming units (CFUs)/mL could be reached with an aqueous ozone concentration of approximately 2 mg/L (R. mucilaginosa) or 0.75 mg/L (E. coli) after 5 through 6 seconds of scaling. Conclusions Aqueous ozone coupled with ultrasonic scaling exhibited a higher efficiency of microbial kill than either method used alone. Both gram-positive and gram-negative species were affected by this treatment. Studies on other oral microbiota constituents, including fungi and viruses, will provide information on the efficacy of this method on a greater biological scale. Studies to verify concomitant reduction of microbial load in dispersed aerosols in clinical settings should be completed to support practical applications of this treatment.


Introduction
Dental handpieces and ultrasonic procedures have been identified previously as contributing large quantities of airborne particulates, microorganisms, and viruses into the local environment via aerosols. [1][2][3] These aerosols are capable of spreading nearly 2 m from the operative site during ultrasonic scaling, resulting in high microbial contamination of the surrounding surfaces and the dental health care provider. 2 With the emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the control of the spread of aerosolized pathogens in dentistry has again come to the forefront. The use of polymer additives such as polyacrylic acid or xanthan gum to reduce or eliminate aerosol production is one potential method, although it does not eliminate pathogens. 4 An alternative option is to use a sterilization method in tandem with dental procedures to reduce the aerosolized pathogen load. Several preprocedural mouthrinses, including aqueous ozone, have been investigated previously and have shown a reduction in surface contamination, 5 although the concurrent use of these compounds with ultrasonic scaling has not been examined. The human oral cavity is a highly diverse microbiome and is host to hundreds of bacterial species, each with its own environmental and physiological importance. 6,7 Many of these bacteria reside in the dental plaque biofilm, a functionally and structurally organized commensal community interlinked via a meshlike substrate of secreted, extracellular polymeric substances. 8 These plaques form in a predictable, ordered fashion but comprise a highly variable microbial composition based on the physical location of each organism. 9 One commonly isolated oral microflora, Rothia mucilaginosa (previously called Micrococcus mucilaginosus or Stomatococcus mucilaginosus) has been commonly identified from the oral cavity of both healthy individuals and patients with underlying conditions such as atherosclerosis, although it has been observed with a higher frequency in healthy populations. 10,11 In biofilms such as dental plaques, this grampositive, encapsulated, nonmotile, bacterium forms multispecies microcolonies or, in the case of high-biomass regions, small islands 12 and shows resilience even after multiple endodontic treatments. 13 Although R. mucilaginosa is typically nonpathogenic in healthy individuals, it can act as an opportunistic pathogen and lead to conditions such as pneumonia, 14 bacteremia, 15 and endocarditis, 16 and cause the buildup of biofilms on prosthetics, 17 among others. 18 Ultrasonic activation of aqueous ozone has been shown to be a means of disinfection of wastewater, 19 and a combination of gaseous ozone and ultrasonic activation has been shown to be effective in disrupting biofilms of Enterococcus faecalis in the root canals of extracted teeth. 20 Similarly, combination treatment using aqueous ozone and ultrasonic activation was effective against planktonic Escherichia coli cultures. 21 Gaseous and aqueous ozone have been explored previously as potential agents for sanitizing dental instruments; however, their efficacy in disrupting dental plaques when used in tandem with an ultrasonic scaler has not been explored. 22 These previous studies have shown the efficacy of ozone as a contact-sterilizing agent, but exposure times and efficacy limit the accessibility of the treatment for routine patient procedures. 22 In this study, the opportunistic pathogen R. mucilaginosa and the gram-negative model E. coli were exposed to aqueous ozone in suspension and in an ultrasonic scaler after production of a monoculture biofilm. The data confirm the effectiveness of aqueous ozone against planktonic bacteria and extend these observations to the enhanced effectiveness of combining ultrasonic activation with aqueous ozone on bacterial biofilms.

Sterilization and sanitation
Before each experimental run, the ultrasonic scaler handpiece and tip, tweezers, water, and media were autoclaved to ensure proper sterilization. Materials that could not be sterilized (the aqueous ozone chamber and ultrasonic instrument reservoir) were sprayed and wiped down with 75% ethanol and allowed to dry fully. Before use, the aqueous ozone chamber, ultrasonic instrument reservoir, and all tubing was rinsed and flushed with sterile water to further ensure proper sanitization. Between individual runs, the ultrasonic scaler and tip and the tweezers were submerged in 75% isopropyl alcohol to limit crosscontamination. All 12-and 96-well plates, lids, and plastic coverslips were exposed to UV light for a minimum of 15 minutes before use.

Bacteria
The M9 minimal media broth was prepared from a 5× solution supplemented with glycerol at a final concentration of 2% (vol/vol), without the addition of exogenous thiamine. The lysogeny broth (LB) agar plates were prepared following the standard recipe for the broth (10 g/L sodium chloride, 10 g/L Bacto tryptone (BD Biosciences), and 5 g/L yeast extract (BD Biosciences)) with 15 g/L agar. The brainheart infusion (BHI) medium and agar plates were prepared using a Bacto reagent (BD Biosciences) and supplemented with agar when necessary. E. coli MG1655 and R. mucilaginosa 5762/67 were originally obtained from the American Type Culture Collection (47076 and 25296, respectively). Strains were grown initially on LB agar (E. coli) or BHI agar (R. mucilaginosa) overnight at 37 • C to obtain single colonies. Both organisms have been passaged repeatedly under laboratory conditions and, as a result, may have accumulated mutations altering their physiology from that described previously. One example of this is the loss of a requirement for exogenous thiamine when grown in M9 medium for E. coli MG1655.

Static biofilm preparation
A previously described static biofilm assay for Agrobacterium species 23 was adapted for use in this study. Overnight cultures of E. coli and R. mucilaginosa were grown at 37 • C in M9 and 2% (vol/vol) glycerol and BHI, respectively, and subcultured to an optical density (OD) of 0.1 at 600 nm (OD 600 ). Cultures were grown at 37 • C with aeration until an OD 600 of 0.4 through 0.6 was reached. The samples were then diluted to an OD 600 of 0.05, and 3 mL were dispensed into each well of a sterile 12-well plate containing a vertically placed, sterile coverslip. All cultures were incubated in a humidified chamber for 48 hours at 37 • C. After incubation, the coverslips were washed vigorously 3 times with sterile water to remove loosely adherent cells and biomass. The remaining tightly adherent cells and biomass were then treated with ultrasonic scaling as described below.

Experimental setup
Aqueous ozone was prepared using a method similar to the one outlined by César and colleagues. 22 Pure oxygen was passed through an ozone generator (high power ozone generator machine Ozonator, [Dr O Solutions]) and bubbled into 750 mL of sterile Milli-Q water. Aqueous ozone concentrations were measured using the Vacu-vials ozone test kit (CHEMetrics K-7423). For the initial analysis of the effect of aqueous ozone on R. mucilaginosa, a minimum ozone concentration of 1.0 mg/L was used. Overnight cultures of both E. coli and R. mucilaginosa were subcultured to an OD 600 of 0.01 and grown for 2 hours at 37 • C to ensure active growth. Five microcentrifuge tubes containing 1.0 mL of culture for each strain were spun at maximum speed (16,100g) for 2 minutes, aspirated, and washed with 1.0 mL of sterile water. This process was performed 3 times, and then the pelleted cells were stored at room temperature. The pellets were stored for a maximum of 4 hours before use. The pelleted cells were resuspended by vortexing after 1.0 mL of aqueous ozone was added to 4 tubes for each strain and 1.0 mL sterile water was added to the fifth. Two hundred μL of sample was aliquoted into 3 wells of a 96-well plate from the first tube (T 0 being time at which culture is first exposed to aqueous ozone ). Five 10-fold dilutions were prepared from the initial wells. One hundred μL of each dilution plated onto the LB media for E. coli and BHI for R. mucilaginosa and incubated overnight at 37 • C. The resulting colonies were counted to determine total colonyforming units (CFU) per mL.
An ultrasonic dental scaler (Newtron P5 XS B.LED; ACTEON) using either sterile water or aqueous ozone was used to disrupt 48-hour old-biofilms of E. coli MG1655 and R. mucilaginosa 5762/67. Aqueous ozone was prepared as detailed earlier until an ozone concentration greater than 4.0 mg/L was reached; 300 mL of aqueous ozone was transferred to the ultrasonic instrument reservoir. Using the ultrasonic scaler, water was dispensed for 2 through 3 seconds during the abrasion of coverslips to disrupt the biofilm ( Figure 1A, B). This process was repeated on the opposite side to ensure maximum biofilm disruption ( Figure 1C). Four coverslips were processed for each ozone concentration per run. New ozone concentrations were used every 15 through 20 minutes to account for the decomposition of the ozone and to mark the start of the next run. Target ozone concentrations were 4.00, 2.00, 1.00, and 0.50 mg/L. A separate run was performed using sterile water that was not exposed to any ozone and accounted for the less than 0.01 mg/L ozone concentration values. Six 10-fold dilutions were prepared for each sample. One hundred μL of each dilution was plated on LB or BHI and incubated overnight at 37 • C after which individual colonies were counted and CFUs/mL calculated.

Statistical analysis
Bacterial concentrations were normalized to the sterile water-treated conditions of each strain for both sets of data before statistical analysis. For all biological variables, the mean values were compared first with an analysis of variance followed by the t test. Any differences and correlations were considered significant when the P values were scored below .05.

Results
Direct exposure of E. coli and R. mucilaginosa to aqueous ozone results in a greater than 97% reduction in CFUs/mL after 30 minutes Gaseous and aqueous ozone have been shown previously to be effective at killing several bacterial genera pathogenic to humans, including Staphylococcus, Streptococcus, and Escherichia. Because of this interaction across both grampositive and gram-negative species, it was predicted that a gram-positive Rothia species would be equally affected by this disinfecting agent.
To test this, pure cultures of the previously studied E. coli and the organism of interest, R. mucilaginosa, were exposed to aqueous ozone at a concentration greater than 1.0 mg/L. The survivability of the culture was examined every 10 minutes and compared with that of a sample that had been resuspended in nonozonated water ( Figure 2). These data show that E. coli is highly susceptible to the effects of aqueous ozone, resulting in a statistically significant 99.70% (standard deviation [SD] 0.07%) decrease in CFU formation immediately after the exposure (T 0 ) and a 99.95% (SD 0.01%) decrease after 30 minutes of exposure ( Figure 2A). The data for R. mucilaginosa, however, show a higher resistance to the lethal effects of ozone, with only an 89.74% (SD 1.82%) decrease in CFUs/mL after the initial T 0 exposure, although this difference remains significant compared with the untreated values. Likewise, the data after 30 minutes show an increased resistance, resulting in a 96.99% (SD 0.74%) decrease in CFUs/mL ( Figure 2B) suggesting that R. mucilaginosa is more resistant to ozone exposure or better adapted for handling oxidative stress.

Combination of ultrasonic scaling and aqueous ozone amplifies the efficacy of biofilm disruption
The formation of bacterial biofilms poses an increased risk owing to the inherent difficulty in inhibiting or killing the bacteria present within the biofilm. 8,9 While dental plaques are not a significant health concern, some oral microbes, including Rothia can form biofilms elsewhere in the body. In the light of this, a second assay was performed to determine the efficacy of disrupting R. mucilaginosa and E. coli biofilms using aqueous ozone and an ultrasonic scaler.
After 5 through 6 seconds of scaling using the ultrasonic scaler and nonozonated water, 1,950.00 (SD 593.55 CFUs/ mL) (Run 1) and 700.00 (SD 320.78 CFUs/mL) (Run 2) were observed. The addition of ozone significantly decreased the observed CFUs/mL in each sample with an ozone concentration of 0.52 mg/L resulting in a 97.06% (SD 1.52%) decrease in CFUs/mL and a concentration of 4.53 mg/L resulting in a 99.9995% (SD 0.0008%) decrease ( Figure 3A, Table). For R. mucilaginosa, initial scaling with nonozonated water resulted in a recovery of 4,866.67 (SD 2,450.17 CFUs/mL) (Run 1) and 4,293.33 ± 244.40 CFUs/mL (Run 2). An ozone concentration of 0.36 mg/L showed no significant decrease in activity; however, increased concentrations did show a noticeable drop in the CFUs/mL recovered. An ozone concentration of 0.58 mg/L decreased the total recovered CFUs/mL by 34.16% ± 27.31%, and an ozone concentration of 4.67 mg/L decreased by 99.32% ± 0.13% ( Figure 3B, Table).

Efficacy of ozone on established biofilms
Ozone has long been used as a disinfecting agent, owing to both its efficacy against a broad range of organisms and viruses, and its rapid degradation. [24][25][26][27] It has been used widely in water sanitation and remediation, 28 food safety 29 and as a contact disinfectant on surfaces. 30 By 2009, aqueous ozone and gaseous ozone have been shown to properly sterilize surgical and dental tools, expanding its use into the medical fields. In the light of the SARS-CoV-2 pandemic, ozone was once again evaluated for its ability to sterilize N95 respirators. It was found that, at concentrations effective for killing the influenza virus, the N95 respirators were properly sterilized of bacterial contaminants and their filtration rate or integrity was not affected. 31 As such, the inclusion of aqueous ozone in an ultrasonic scaler reservoir has the potential for managing aerosolized microbes and possibly, airborne viral particles and may offer alternative sterilization procedures for dental and medical facilities to deal with future pandemics similar to the SARS-CoV-2 outbreak. A 2020 molecular modeling study by Tizaoui 32 suggests that ozone would be an effective oxidant for SARS-CoV-2.
While exposure to gaseous ozone can induce oxidative stress in human respiratory tissue, 33 aqueous ozone does not appear to cause damage to human oral cells or tissue. [34][35][36] Short, controlled exposures have shown potential efficacy in treating infection and periodontal disease 25,34,37 as well as aiding in pain management, wound healing, and cosmetic treatments. 25,36 As such, use of aqueous ozone alongside ultrasonic scaling to remove dental plaque is not expected to cause any harm to the patient. Furthermore, ultrasonic scaling has long been shown to increase the bacterial load in the local atmosphere 1-3 and has the potential for spreading pathogens through airborne droplets, direct inhalation, ocular membranes, or contact lens; this contamination poses increased risk to immunocompromised patients and dental personnel. 1,38 Despite these risks, however, routine plaque control and debridement lead to an overall reduction in oral bacterial load and, as a result, a clinical improvement in periodontal disease. 39 Previous data suggest that ozone treatment as an adjunct to regular scaling for patients with chronic periodontitis can aid in reducing overall bacterial load 37 ; however, this interaction is reliant highly on individual methods of ozone preparation and application and cannot be relied on readily to act as antimicrobial treatment. 40,41 Likewise, ozone shows no benefit over the traditionally used sodium hypochlorite in decreasing bacterial load when used during root canal disinfection, 42 but it does appear to aid in the overall Table Raw and normalized descriptive statistics of the colony-forming units (CFUs)/mL for Escherichia coli and Rothia mucilaginosa when exposed to different aqueous ozone concentrations applied via an ultrasonic scaler for 5 through 6 s. wound healing after the procedure. 36 Furthermore, application of gaseous ozone shows no significant antimicrobial activity when used during nonsurgical periodontal treatments. 43 Our results suggest that the addition of ozone to the water supply of an ultrasonic scaler at concentrations capable of reducing the bacterial load (> 1.0 mg/L) of plaque biofilms is predicted to aid in the control of aerosolized microorganisms and viruses, though additional experimentation and validation will be performed in future studies. Our experimental setup used ultrapure Milli-Q water. The stability of aqueous ozone is known to depend on water source and purity. 44 At a minimum, the use of deionized water should be considered in future studies and, if translated, into clinical use.
Variation between E. coli and R. mucilaginosa Much of the variation in the efficacy of ozone on the 2 model organisms we present likely can be attributed to the morphologic and physiological differences between the 2 organisms. It has been suggested previously that the thick peptidoglycan wall provides increased, but incomplete, resistance to ozone in gram-positive bacteria. 45 Some grampositive organisms also show increased resistance to ozone due to adaptation to the evolutionary pressure of consistent oxidant exposure, such as oxidizing disinfectants. 46 Based on this, the Rothia species may exhibit resistance to ozone treatment as a by-product of resistance to oxidative stress in general. The R. mucilaginosa genome codes for several σ factors that are upregulated in response to oxidative stress. 13 In addition, R. mucilaginosa can generate acetaldehyde from ethanol; increased levels of acetaldehyde are capable of inducing oxidative stress, 47 suggesting that R. mucilaginosa has increased resistance to either acetaldehyde or the resultant oxidative stress it induces. This factor, combined with the thick peptidoglycan cell wall, may explain the variation in efficacy between the 2 organisms we studied. Previous data examining the efficacy of ozone on the disruption of gram-positive biofilms suggest that ozone used in conjunction with another disruptive treatment results in significantly higher reduction in viable cells 48 ; however, the effect of aqueous ozone alone on R. mucilaginosa biofilms was not examined during the course of our work. Furthermore, both species selected for our study are capable of aerobic respiration, and therefore, are more capable of oxidative stress responses than some anaerobic microorganisms commonly found in the oral cavity. These anaerobic species, as a result, would be expected to have a higher rate of mortality when exposed to aqueous ozone, as shown by Nagayoshi and colleagues 34 using Streptococcus species. Although both E. coli and R. mucilaginosa have been shown to be susceptible to ozone and the biofilms produced by both organisms can be eliminated readily by the use of aqueous ozone dispensed through an ultrasonic scaler, further investigation into mixed culture biofilms and in vivo plaque disruption will still need to be explored to validate the procedure used in our study. Although our study thus establishes proof of concept for the combined use of aqueous ozone and ultrasonic scaling in the dental setting, there remain important limitations. For one, the tested strains are laboratory adapted domesticated strains and may have accumulated mutations making them more susceptible to the treatment modalities we tested. In addition, both strains serve only as representative gram-negative and grampositive organisms and do not necessarily reflect the full range of physiological adaptations that may be found in a wider range of clinically relevant periodontal pathogens, either when grown in monoculture or in a mixed culture biofilm.

Conclusion
Our study demonstrates that both gram-positive and gramnegative bacterial species are sensitive to the effects of aqueous ozone. Planktonic and biofilm-associated bacteria of both gram-negative E. coli and gram-positive R. mucilaginosa exhibited a decrease in viable colony-forming units following exposure to aqueous ozone. When combined with ultrasonic scaling biofilms formed by these organisms exhibited a further decrease in viable colonyforming units. Importantly, the working concentrations of aqueous ozone required for significant reduction in microbial load is readily achievable with commercially available equipment.
Disclaimer. The sponsor had no role in the design, execution, interpretation, or publication of the study.
Disclosure. None of the authors reported any disclosures.
Funding. This work was performed using the start-up funding provided by the University of the Sciences in Philadelphia, Pennsylvania to Jason E. Heindl.